Proceedings of the 6th International Conference on Civil, Offshore and Environmental Engineering (ICCOEE2020)

Citation preview

CQ

Cea

MGMT Me

SEL

ar S. Mohammed - Nasir Shafiq sul Rahman M. Kuttyham Mohamad TU Re cced msdeUt OL C0Y RS

[GG0] a PUPA Proceedings of the 6th International Cy CaCaeeO eT OAKTON PINOMINTIOIILC lM mae iLI\0) (ICCOEE2020)

a Santas

Lecture Notes in Civil Engineering Volume

132

Series Editors Marco di Prisco, Politecnico di Milano, Milano, Italy Sheng-Hong Chen, School of Water Resources and Hydropower Engineering, Wuhan University, Wuhan, China Toannis Vayas, Institute of Steel Structures, National Technical University of Athens, Athens, Greece Sanjay Kumar Shukla, School of Engineering, Edith Cowan University, Joondalup, WA, Australia Anuj Sharma, Iowa State University, Ames, IA, USA Nagesh Kumar, Department of Civil Engineering, Indian Institute of Science Bangalore, Bengaluru, Karnataka, India Chien Ming Wang, School of Civil Engineering, The University of Queensland, Brisbane, QLD, Australia

Lecture Notes in Civil Engineering

(LNCE)

publishes the latest developments

in

Civil Engineering - quickly, informally and in top quality. Though original research reported in proceedings and post-proceedings represents the core of LNCE, edited volumes of exceptionally high quality and interest may also be considered for publication. Volumes published in LNCE embrace all aspects and subfields of, as well as new challenges in, Civil Engineering. Topics in the series include: Construction and Structural Mechanics Building Materials Concrete, Steel and Timber Structures Geotechnical Engineering Earthquake Engineering Coastal Engineering Ocean and Offshore Engineering; Ships and Floating Structures Hydraulics, Hydrology and Water Resources Engineering Environmental Engineering and Sustainability Structural Health and Monitoring Surveying and Geographical Information Systems Indoor Environments Transportation and Traffic Risk Analysis Safety and Security To submit a proposal or request further information, please contact the appropriate Springer Editor: — —

Mr. Pierpaolo Riva at [email protected] (Europe and Americas); Ms. Swati Meherishi at [email protected] (Asia - except China, and Australia, New

—

Zealand);

Dr. Mengchu Huang at mengchu.huang @springer.com (China).

All books in the series now indexed by Scopus and EI Compendex database!

More information about this series at http://www.springer.com/series/15087

Bashar S. Mohammed - Nasir Shafiq Shamsul Rahman M. Kutty Hisham Mohamad - Abdul-Lateef Balogun Editors

ICCOEE2020 Proceedings of the 6th International Conference on Civil, Offshore

and Environmental Engineering

(ICCOEE2020)

DQ Springer

Editors Universiti Teknologi PETRONAS. Seri Iskandar, Perak, Malaysia

Nasir Shafiq Civil and Environmental Engineering Universiti Teknologi PETRONAS Seri Iskandar, Malaysia

Shamsul Rahman M. Kutty Civil and Environmental Engineering Universiti Teknologi PETRONAS Seri Iskandar, Perak, Malaysia

Hisham Mohamad Civil and Environmental Engineering Universiti Teknologi PETRONAS Seri Iskandar, Perak, Malaysia

Bashar S. Mohammed

Civil and Environmental Engineering

Abdul-Lateef Balogun Civil and Environmental Engineering Universiti Teknologi PETRONAS Seri Iskandar, Perak, Malaysia

ISSN 2366-2557 ISSN 2366-2565 (electronic) Lecture Notes in Civil Engineering ISBN 978-981-33-6310-6 ISBN 978-981-33-6311-3 (eBook) https://doi.org/10.1007/978-981-33-63 11-3 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concemed, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, ete. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be tue and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,

Singapore

Preface

This book contains papers presented in the 6th International Conference on Civil, Offshore

and

Environmental

Engineering

(ICCOEE2020)

under

the

banner

of

World Engineering, Science and Technology Congress (ESTCON2020) held on 13-15 July 2021 at Borneo Convention Centre, Kuching, Malaysia. The ICCOEE series of conferences started in Kuala Lumpur, Malaysia, in 2012. The main objective of the ICCOEE is to provide a platform for academia and industry to showcase their latest advancements and findings in the broad disciplines of civil, offshore and environmental engineering with an emphasis on the looming Industrial Revolution 4.0. The conference also provides great opportunities for participants to exchange new ideas and experience as well as to forge research and business relations with global partners for future collaborations. The articles in this book were accepted after a rigorous review process. All accepted papers are categorized based on the following themes and areas of research: ¢ e ¢ e ¢

Green Environment and Smart Water Resource Management Systems Advanced Coastal and Offshore Engineering Resilient Structures and Smart Materials Advanced Construction and Building Information Modelling Smart and Sustainable Infrastructure

We would like to express our gratitude to the Technical Programme Committee and Advisory Committee who undertook the biggest responsibility in the paper reviewing process. We are also grateful to the additional reviewers who helped the authors deliver better papers by providing them with constructive comments. We hope that this process contributed to a consistently good level of the papers that are included in the book. Bashar Sami Mohammed Nasir Shafiq Shamsul Rahman M. Kutty Hisham Mohamad Abdul-Lateef Balogun

v

Organization

Organizing Committee Conference Chair Bashar S. Mohammed

Universiti Teknologi PETRONAS,

Malaysia

Conference Co-chair Universiti Teknologi PETRONAS, Malaysia

Zahiraniza Mustaffa

Secretary Ng Cheng

Yee

Universiti Teknologi PETRONAS, Universiti Teknologi PETRONAS,

(L)

Yani Rahmawati

Treasurer Ho Yeek

Universiti Teknologi PETRONAS, Universiti Teknologi PETRONAS, Malaysia

Chia (L)

Muslich Sutanto

Technical Committee Montasir Osman

Ahmed

Ali

Universiti Teknologi PETRONAS, Malaysia

(L) Nasir Shafiq Shamsul Rahman M. Kutty Siti Habibah Bt Shafiai Abdul-Lateef Babatunde Balogun

Universiti Teknologi PETRONAS,

vii

Organization

viii

Publication Committee Khamaruzaman

Universiti Teknologi PETRONAS, Malaysia

Wan

Yusof (L)

Ehsan Nikbakht Jarghouyeh Wesam Salah Alaloul Dimas Bayu Endrayana Idris Othman

siti Teknologi i Teknologi ‘sit Teknologi Universiti Teknologi

PETRONAS, PETRONAS, PETRONAS, PETRONAS, Malaysia

Logistics Committee Universiti Teknologi PETRONAS, Malaysia

Niraku Rosmawati Ahmad

(L)

Universiti Teknologi PETRONAS, Malaysia

Husna Takaijudin

IT And Media Mohamed

Universiti Teknologi PETRONAS, Malaysia

Latheef (L)

Publicity and Protocol Committee Muhammad

Universiti Teknologi PETRONAS, Malaysia

Raza Ul

Mustafa (L)

Ahmad Al-Yacouby

F&B

Mahamad

Universiti Teknologi PETRONAS, Malaysia

Committee

Aslinda Jamaluddin

Universiti Teknologi PETRONAS, Malaysia

(L)

Event Management Committee Teh Hee

Min

Universiti Teknologi PETRONAS, Universiti Teknologi PETRONAS, Malaysia

(L)

Lavania Baloo

Co-organizer Division Dimas

Bayu

Endrayana

(L)

Universiti Teknologi PETRONAS, Malaysia

Sponsorship Hisham

B Mohamad

Ho Yeek Chia Wesam Salah Alaloul

(L)

Universiti Teknologi PETRONAS, Univers iti Teknologi PETRONAS, Universiti Teknologi PETRONAS, Malaysia

Contents

Green Environment and Smart Water Resource Management Systems Study on Monthly Rainfall Trend Impact on Reservoir Simulation in Greater Bandung . . wee S. Sanjaya, D. Yudianto, and Willy Aulia Study of Saturation Flow at Signalized Intersection on Sunny Weather and Rainy Weather .................0..0..0..0 0.000. Risdiyanto and Syaripin

12

Deep Learning Neural Network for Time Series Water Level Forecasting. ........0. 000.000.0000 02 ceeeeeeeeeeeee Nuratiah Zaini, Marlinda Abdul Malek, Shuhairy Norhisham, and Nurul Hani Mardi

22

Optimization Study of n-ZVI Oxidation for Organic Pollutants Removal from Wastewater... ...... 2-2-6006. 0202052020 e eee Muhammad Raza UI Mustafa, Tahir Haneef, Brenda Tan Pei Jian, Khamaruzaman Wan Yusof, and Hifsa Khurshid

30

The Effectiveness of Cascaded Bioretention System in Treating Urban Stormwater Runoff Husna Takaijudin, Manal Osman, Khamaruzaman Wan Yusof, Aminuddin Ab Ghani, and Goh Hui Weng An Evaluation of Hydrological Simulation of Extensive Green Roof . . Siti Fatin Mohd Razali, Hasrul Hazman Hasan, Siti Aminah Osman, Melisa Ismail, Mohd Reza Azmi, Muhamad Nazri Borhan, Azman Mohd Jais, Rohaya Abdullah, and Suhayya Rofik

39

.

47

x

Contents

Assessment of SRTM, ASTER and IFSAR Digital Elevation Model

(DEM)

in Oil Palm

Plantation River Derivation

and Basin Delineation..................-.....0.0002 022020555 Siti Hajar Md Nor Azam, Wardah Tahir, and Jazuri Abdullah Features and Geomorphic Response of Mountainous River by Reach Scale... 2.2.00. Nor Azidawati Haron, Badronnisa Yusuf, Siti Nurhidayu, Mohd Sofiyan Sulaiman, and Mohd Shahrizal Ab Razak

56

eee

66

........

74

Application of Inhibition Model to Prevent Nitrification Upset in Petrochemical Wastewater Treatment Plant...................Idzham Fauzi M. Ariff

81

Water Infiltration in Salt Land in Nagekeo Flores, Indonesia Trihono Kadri

Design Expert Application for Optimization of Ag/AgBr/TiO2 Visible Light Photocatalyst Preparation......................... Augustine Chioma Affam, Wong Chee Chung, Poh Lin Lau, Olufemi Adebayo Johnson, Khor Cheng Seong, Lavania Baloo, Bryan Wong Lee Peng, and Fung Xinru

93

Hydrodynamics of Flow over Axonopus Compressus (Cow Grass) as a Flexible Vegetation Muhammad Mujahid Muhammad, Khamaruzaman Wan Yusof, Muhammad Raza UI Mustafa, Aminuddin Ab. Ghani, Abdurrasheed Said Abdurrasheed, Abdulkadir Taofeeq Sholagberu, Abdullahi Sule Argungu, and Umar Alfa Abubakar Greenhouse Gas Emission from Domestic Wastewater Treatment and Discharge in East Java Province - Indonesia ................. Yatnanta Padma Devia and Dian Tristi Agustini Removal of Cadmium from Aqueous Solution by Optimized Magnetic Biochar Using Response Surface Methodology............ Anwar Ameen Hezam Saeed, Noorfidza Yub Harun, Mohamed Mahmoud Nasef, Haruna Kolawole Afolabi, and Aiban Abdulhakim Saeed Ghaleb Usage of Seaweed as a Biocomposite Material in Green Construction ..... 2.2.2.2... 660062 e eee eee eee eee Lavania Baloo, Mubarak Usman Kankia, and Oh Jia Wei Performance of Permeable Pavement with Subsurface Micro Detention Storage as Rainwater Harvesting Device ................ Norazlina Bateni, Lai Sai Hin, Md Abdul Mannan, Jethro Henry Adam, Kuryati Kipli, and Rosmina Ahmad Bustami

lll

119

127

139

Contents

xi

A Proposed Framework of Life Cycle Cost Analysis for Petrochemical Wastewater Treatment Plants .................Muhammad Ilyas, Freselam Mulubrhan Kassa, and Mohd Ridzuan Darun A Concise Review of Major Desalination Techniques: Features and Limitations . . sees sees .. Tijani Oladoyin Abimbola, Khamaruzaman Wan Yusof, Husna Takaijudin, Abdurrasheed Said Abdurrasheed, Ebrahim Hamid Hussein Al-Qadami, Samiat Abike Ishola, Tunji Adetayo Owoseni, and Suleiman Akilu Impact of Treating Ammonia-Nitrogen Contamination from Chemical Fertilizer Plant Using Extended Aeration Activated Sludge System.... Mohammad Fakhuma Ubaidillah Bin Md Hafiz, Shamsul Rahman Bin Mohamed Kutty, and Shekhah Norafizah Binti Shekh Imaduddin Hakmi Study Effectiveness Sabo Dam on Reducing Flood in Way Leman River ...... 2.2.2.6 2 0202s Reja Putra Jaya

147

154

163

174

The Effectiveness of the Use of Tripikon-S in Tofu industry Wastewater Treatment Sardi, Edy Sriyono, Tania Edna Bhakty, and Ganang Azas Hayininda Emerging Coagulant in Water Treatment: A Review and a Preliminary Study ...................00..02.00.0 0.0000. Jia-Shen Lau, Wei-Jing Lee, Hoe-Guan Beh, Wawan Sujarwo, Krishnan Hariharan, Balamurugan Panneerselvam, and Yeek-Chia Ho Eco-composite Porous Concrete Drainage Systems: An Alternative Mitigation for Urban Flood Management Feroz Hanif Mohamed Ahmad, Mohamad Hidayat Jamal, Abdul Rahman Mohd. Sam, and Nuryazmeen Farhan Haron

187

...........

195

Tapping the Potential of Shallow Water Model for Wave Simulations... ........ 2.22... 6.600002 e eee ee eee Bobby Minola Ginting, Doddi Yudianto, and Albert Wicaksono

205

Impact of Coastal Development on Hydrodynamic Change of the Mangrove Coastline in Tanjung Piai, Malaysia.............. Iwan Tan Sofian Tan, Nik Mohd. Kamel Nik Hassan, and Teh Hee Min

213

ZEEPod Reshaping the Future of Oil & Gas Marginal Field Development in Malaysia ....................0 0.02.0 020 00000. Herman Perera and Mohd Izzuan Zaharudin

221

Advanced

Coastal and Offshore Engineering

xii

Contents

Effect of Underwater Sill Height Against Flow Patterns in Order to Reduce Sedimentation in Navigation Channel and Basins ......... Tania Edna Bhakty, Nur Yuwono, Bambang Triatmodjo, and Ahmad Faramarz Ghalizhan

232

Numerical Investigation of an Efficient Blade Design for a Flow Driven Horizontal Axis Marine Current Turbine.................Nauman Maldar, Cheng Yee Ng, Ahmad Fitriadhy and Hooi Siang Kang

241

Numerical

249

Studies on the Stability of Offshore Wind

Turbine

(OCS)...

Nur Shahira Fazira Binti Shamsul Ariffin and Montasir Osman Ahmed Ali Numerical Simulation to Assess Floating Instability of Small Passenger Vehicle Under Sub-critical Flow . Ebrahim Hamid Hussein Al-Qadami, Zahiraniza Mustaffa, Eduardo Martinez-Gomariz, Khamaruzaman Wan Yusof, Abdurrasheed S. Abdurrasheed, and Syed Muzzamil Hussain Shah Interaction of Wave-Induced Motion and Bioelectricity Generation for Floating Microalgal Biophotovoltaic System................... Jia-Chun Chin, William Chong Woei Fong, Kee-Quen Lee, Cheng-Yee Ng, Kiat Moon Lee, Wah Yen Tey, and Hooi-Siang Kang Managing a HTHP Pipeline: Detailed Integrity Assessment of Deepwater Oil Line Displacement Near Continental Shelf... ..... . Ir. Hayati Hussien, Azam Syah Jaafar, Md Anuar Desa, and Mohd Faisal Aziz Investigation on Motion Responses of a Floating Wave Barrier in a Wave Flume Subjected to Regular Wave Action............... Hee Min Teh, Thinagran Silavaraj, Syed Shuja Ul Ha and Eric Joseph Pereira Numerical Assessment of Flow Around Circular Cylinder........... Malakonda Reddy Lekkala, Mohamed Latheef, Do Kyun Kim, and Mubarak Bin A. Wahab Corrosion Resistant Alloy Pipeline Installation for High Pressure High Temperature Requirement........................0..005. Khairan Syuhada Kassim A Short Review on Numerical Simulation of Floating Debris Migration ..........0..0.00 0000000200 2c eee eee ee Lavine Wong, Mohamad Hidayat bin Jamal, and Erwan Hafizi bin Kasiman Response of Corroded Offshore Structural Plate at Topsides Due to Blast Loading.............00.00 0000.00.00 0002. e eee eee Mohamed Mubarak Abdul Wahab, Nurfadhilah Ali Anwar, Bashar Mohammed, Ahmad Rizal Abdul Rahman, and Zahari Razak

266

274

281

289

302

310

318

Contents

xiii

Hybrid Floating Structures Case Study: Marisco’s Floating Dry Dock... 1.0.22 Sugeng Wijanto, Vivian Sabas, and Takim Andriono

eee

326

Platform Conductor Integrity Management in Life Extension of Ageing Offshore Wells ....... Loganathan Radzakrishnan, Mohd Khairi Abu Husain, Roslina Mohammad, Astuty Amrin, and Mohd Akmal Resilient Structures and Smart Materials Soil Improvement Using Xanthan Gum Biopolymer for Loose Sand: Experimental Study ..... 2.00.00... 20.0000 ..00 000200020 00000. Aswin Lim, Yohanes Albrecht Montol, and Siska Rustiani The Effect of Sky Bridge Modeling on Structural Behavior.......... Wivia Octarena Nugroho, Lidya Fransisca Tjong, and Aditya

349 356

Effect of Slag on Chloride Resistance of Concrete Saeed Ahmad, Nasir Shafiq, Hafiz Waheed Iqbal, Raja Zaheer Ahmad, Zulqurnain Abbas, Anees-ur Rehman, and Muhammad Ali Anomaly Phenomena on the New Indonesian Seismic Code SNI 1726:2019 Design Response Spectra .................0..00.. Suradjin Sutjipto and Indrawati Sumeru

375

Experimental Study on Blast Furnace Nickel Slag Powder and Fly Ash as a Binder for Geopolymer Concrete ................ Lisa Oksri Nelfia, Mutia Rahmawati, and Sotya Astutiningsih

385

Application of Response Surface Methodology for the Optimization of Mix Design Concrete Using Coal Bottom Ash as Cement Replacement Material ................0.020.00 0.02.0 220 00000. Nur Liyana Mohd Kamal, Nasir Shafiq, Wesam Sallah Alaloul, Salmia Beddu, and Teh Sabariah Binti Abd Manan

396

Performance of Oil Palm Shell Lightweight Concrete Incorporated with Bamboo Fiber..............-.-...0 0000020 e eee cee eens Siew Choo Chin, Qi Hao Roger Wong, Kar Sing Lim, and Shu Ing Doh

405

Experimental Study on Structural Behaviour of Corbels with Hybrid

Fibre Reinforced

Concrete

(HyFRC)

.................

Muhammad Aswin, Ehsan Nikbakht, Nazihah Shahirah Bt Nazir, Noor Syamimi Bt Tajuddin, and Nor Amirah Bt Ahmad Impact of Elevated Temperature on Rubberized Concrete: A Review Wesam Salah Alaloul, Muhammad Ali Musarat, and Chan Jia Hui

413

xiv

Contents

A Numerical and Experimental Study on Crack Propagation of the Abutment Back Wall - Unified Wing Wall Under Incremental Load. ............2.-2-.6 0000002 e eee ee eee eee Desy Setyowulan, Ming Narto Wijaya, and Lilya Susanti Shear Capacity of Lightly Reinforced Concrete Columns Ari Wibowo

...........

428 435

Effect of Twisted Soft Drink Can Waste Fiber on the FRC Flexural Behavior................ Christin Remayanti Nainggolan, Indradi W: Performance of Soil Composite Cement Base Layers with Additive Matos Soil Stabilizer on Suka Bumi-Kedang Ipil Road Section Kutai Kertanegara East Kalimantan Indonesia ................... Teguh Widodo An Experimental Assessment on the Performance of Fly Ash in Concrete... 22.2.2. ee Nasir Shafiq and Muhammad Afiq Ammar

450

eee

458

......

468

Rigidity Boundaries of Floor Reinforced Concrete Diaphragm ....... Hadi R. Tanuwidjaja, Grace K. Santoso, and Euricky Tanuwidjaja

476

Effect of Sand Proportion on Fineness Modulus of Combined Aggregate, Workability, and Compressive Strength of Concrete Arusmalem Ginting

Effect of Graphene Oxide on Mechanical Properties of Rubberized Concrete: A Review... 2... 0.020.222 6 06 ee cece eee eee Muhammad Wihardi Tjaronge, Muhammad Ali Musarat, Kevin Law, Wesam Salah Alaloul, and Saba Ayub Compressive and Flexural Strengths of Mortar with Silica Aerogel Powder .. 1.2... 20. Lee Thin Tay, Yee Yong Lee, Yeong Huei Lee, and Ahmad Beng Hong Kueh Study of C-S-H Formation of Cemented Sediment Brick. ........... L. W. Ean, M. A. Malek, B. S. Mohammed, Chao-Wei Tang, and C. Y. Ng Flexural Behaviour of Glass Fiber Reinforced

Polymer

484

493

501

(GFRP)

Tubes Subjected to Static Load ...............0 0.02.0 ..0 0.000. Agusril Syamsir, Abdulrahman Alhayek, Audrey Yeow Yee Keng, Daud Mohamad, Mohamad Zakir Abd Rashid, and Shuhairy Norhisham Investigation on Behavior of Concrete Slab Due to Low Velocity Impact Using Numerical Modeling......................0..02.. Agusril Syamsir, Syahidatul Islamiah, Shuhairy Norhisham, Nur Liyana Mohd Kamal, Norazman Mohamad Nor, and Vivi Anggraini

510

517

Contents

xv

Investigating an Optimum Mixing Method to Produce Foam Concrete Fulfilling the Workability, Density, Shrinkage, Strength and Total Volume M. I. Safawi, S. N. L. Taib, L. P. Hua, and A. Rashidi Experimental Study of Two Stages on the Use of Local Rubber as Base Isolator for Dwelling Houses .......................02.. Usman Wijaya and Elly Kusumawati Advanced

534

Construction and Building Information Modelling

Prospects of a Sustainable EOL - Carbon Footprint Assessment of a Tropical Housing Habitat ..................0..0.....0..2. Syed Shujaa Safdar Gardezi and Nasir Shafiq

545

Analytical Investigation of Failure Behavior of Beam-Column Knee Joint with External Steel Plates Anchorage Using 3D RBSM......... Liyanto Eddy, Kohei Nagai, and Punyawut Jiradilok

555

Dominant Success Factors of Managing Subcontractors by Main Contractors in Sustainable Development Project ........... Bambang Endro Yuwono, Yuhana, and Raflis

564

Effect of Fire Flame Exposure on Basalt and Carbon Fiber-Reinforced Concrete ................600-2 020525000 Siti Nooriza Abd Razak, Laurent Guillaumat, and Nasir Shafiq

20000-

573

Human Factor Engineering in Oil and Gas Construction Works — A Case Study to Mitigate Safety Risk ................2.....0.... Mat Saaud Nor Arinee, Othman Idris, and Ir Baharuddin A. Rahim

590

Causes of Construction Accidents and the Provisions of Safety Regulations in Construction Industry in Malaysia ................. Ku Adenan Ku Ismail and Idris Othman

602

Delay and Cost Overrun of Palm Oil Refinery Construction Projects: Artificial Neural Network

Muhammad Sani Abdullah, Wesam and Muhammad Ali Musarat

(ANN)

Model

.

Salah Alaloul, M. S. Liew,

Development of Framework for BIM-Based Tools to Minimize the Causes of Accidents in Construction .......................Idris Othman, Aminu Darda’uRafindadi, Madzlan Napiah, MiljanMikic¢, Hayroman Ahmad, Nura Shehu AliyuYaro, Balarabe Wada Isah, Ahmed Farouk Kineber, and Muhanad Kamil Buniya Factors of Safety Misconduct Affecting Safety Performance of Tall Building Construction Site ...............0...2.....0..0. Varunesh Thinakaran and Idris Othman

608

620

xvi

Contents

Characteristics of Interlocking Concrete Bricks Incorporated Crumb Rubber and Fly Ash .... 2.00.20. 00..00 0002.00.50 ..000. Bashar S. Mohammed, Amin Al-Fakih, and M. S. Liew Image - Based Change Detection in Concrete Beam ............... Krisada Chaiyasam, Apichat Buatik, Kuntapit Jirakasemsuk, Pakkapong Khuangsimma, and Suraparb Keawsawasvong Mechanical Properties of Steel Fibre Reinforced Lightweight Concrete by Incorporating Recycled Car Waste Tyres Aggregate ..... Yeow Kah Niam, Ming Kun Yew, Siong Kang Lim, and Chuan Fang Ong

631 640

648

The Effectiveness Implementation of Project Risk Management Plan in Property Development in Malaysia . . Idris Othman, Nor Haslinayati Abdul Ghafé Review of Unmanned Aerial Vehicle Photogrammetry for Aerial Mapping Applications N. M. Zahari, Mohammad Arif Abdul Karim, F. Nurhikmah, Nurhanani A. Aziz, M. H. Zawawi, and Daud Mohamad Effects of Quartz Powder on the Compressive Strength of High Performance Engineered Cementitious Composites ................ M. S. Liew, Bashar S. Mohammed, Kamaluddeen Usman Danyaro, A. M. Al-Yacouby, and Sani Haruna Methodology Review on Multi Stakeholders Decision of Urban Market Land Use... .... 2.6.0.2 - 0226.0 e eee eee eee eee Christiono Utomo, Yani Rahmawati, and O. L. Sari

eee

Sustainability Criteria for Green Building Material Selection in the Malaysian Construction Industry.....................0... Ezzaddin Al-Atesh, Yani Rahmawati, and Noor Amila Wan Abdullah Zawawi Role of Inflation in Construction: A Systematic Review............. Indra Jaya, Wesam Salah Alaloul, and Muhammad Ali Musarat Sources of Risk and Related Effects in the Malaysian Construction Industry ...........00..0000 2.000.020.0020 Sim Nee Ting and Beatrice Jarit

05000.

677

685

693

701

709

Smart and Sustainable Infrastructure Accuracy of Bus Timetable Using Information Communication and Technology GPS to Inform Trans Metro Bandung Bus Passenger ... 2.2.20. 2 00 eee Anastasia Caroline Sutandi and Aldian Dermawan

725

Contents

xvii

Decision Making of Retrofitting Alternatives of Cikeusik Bridge Pillars.... 2.2.00. 2 000 eee Altho Sagara, Andreas F. V. Roy, and Gisella Liviana

735

Study on Bus Rapid Transit (BRT) in Medan Based on Coverage Area and Accessibility Index Ridwan Anas, Ami Kholis Hasibuan, Medis S. Surbakti, and Ika Puji Hastuty The Impact of Growth in Vehicle Ownership on Commuter Travel Time ..... 2.2.6.6 62022 Nindyo Cahyo Kresnanto and Bayu Kunto Wicaksono

749

Measurement of User Interest in Public Transport Performance Variables Using AHP ...... 2.00.00... 0000002 eee Fatsyahrina Fitriastuti and Nindyo Cahyo Kresnanto

756

Evaluation of Road Maintenance Program Based on International Roughness

Index

(IRI) and Surface Distress Index

(SDI)............

764

Medis Surbakti, Saleh Samsuri, Ridwan Anas, and Ahmad Perwira Tarigan Interpretation Method of Distributed Fibre Optic Strain Sensor in Instrumented Static Pile Load Test Nur Hidayah Mahadi and Hisham Mohamad The Effect of the Placement of the Roughometer III Sensor on the Result of the IRI Values on National Roads ................ Medis Surbakti, Doly Manurung, Ridwan Anas, and Irwan Sembiring

7719

Role of Project Governance in Managing Projects Sustainability: A Theoretical Perspective Mehfooz Ullah, Muhammad Waris Ali Khan, and Lee Chia Kuang Investigation of Traffic Noise Pollution at Puchong Residential Areas ..... 2.2... 022.2226 e eee eee eee Nik Nur Ifadah Mohamed Nazi Nur’atiah Zaini, Shuhairy Norhisham, and Nurul Natasha Nabila Naim Online Shopping and Travel Behaviour Based on Information and Communication Technology Activity....................0... Jeanly Syahputri, Tri Basuki Joewono, Muhamad Rizki, and Dimas B.E. Dharmowijoyo

799

807

Remediation of Soft Soil by Hydrated Lime Yulvi Zaika, As’ad Munawir, and Alwafi Pujiraharjo Truck Accident Risk Model for East Java, Indonesia .............. Achmad Wicaksono, Muhammad Zainul Arifin, Meriana Wahyu Nugroho, and Yayan Rachmadi Utomo

828

xviii

Contents

Effect of Phase Change Material on Rheological Properties of Asphalt Mastic... 2.2.0.2... eee I K Mizwar, Madzlan Napiah, and Muslich H Sutanto The Future of Wind Power in Malaysia: A Review................ Shamsan Alsubal, M. S. Liew, E. S. Lim, Indra S. H. Harahap, and Ahmed M. M. Nasser Investigating the Ride-Hailing Users and Their Perception of the Usefulness of Its Services: A Case from Bandung, Indonesia.... Tri Basuki Joewono, Muhamad Rizki, Dimas Endrayana Dharmowijoyo, and Dwi Prasetyanto Exploring the Ride-Hailing Drivers’ Characteristics and Their Order Rejection Behavior in Bandung City .......................0... Muhamad Rizki, Tri Basuki Joewono, Prawira F. Belgiawan, and Dwi Prasetyanto Spatial Analysis for Sustainable Campus Transportation: A Case Study of UTP.... 00.2002 Umira Binti Ayub and Abdul-Lateef Babatunde Balogun

836 844

852

861

eee

870

Indirect Bridge Health Monitoring Employing Contact-Point Response of Instrumented Stationary Vehicle .................... Ibrahim Hashlamon, Ehsan Nikbakht, and Ameen Topa

883

Ground Response Analysis for Stiff and Soft Soil Under Different Earthquake Events: A Comparison M. Mazlina, M. S. Liew, A. Adnan, I. S. H. Harahap, and N. H. Hamid Bearing Capacity of Residual Soil Treated with Fine Demolished Concrete

Waste

(DCW)

Under

Nur Masyitah Osman and Ahmad

Soaked

and

Unsoaked

Condition. .... .

899

Syauqi Md Hasan

The Influence of Socio-Demographic and Activity-Travel Participation Variables on Mode Choice for the New Railway Development in South

Sulawesi, Indonesia (Case:

Makassar-Parepare

Line)

.......

907

Assessment of Earth Dam Critical Failure Using Numerical Method... Aniza Ibrahim, Nurul Amirah Osman, and Zulkifli Abu Hassan

920

Syahreza Alvan, Muhammad Isran Ramli, Hajriyanti Yatmar, Muralia Hustim, and Ridwan Anas

S-Curve Rubble Mound Breakwater . Muhammad Arsyad Thaha, Andi Ildha Dwipuspita, and Dimas Bayu Endrayana Dharmowijoyo Road Traffic Noise Analysis at the U-Turn in Makassar City Muralia Hustim, Rasdiana Zakaria, Muhammad Isran Ramli, and Nurul Azizah Syafruddin

........

936

Contents

xix

The Impact of Socio-Demographic and Activity-Travel Participation Variables on Mode Choice Preference of Sub-urban Commuters: A Case Study on the New Railway Operation in Maros-Makassar Line .........-. 2-0-0600 202050 e eee eee Hajriyanti Yatmar, Muhammad Isran Ramli, Mubassirang Pasra, Gusfiadi, and Dimas B. E. Dharmowijoyo The Relationship Between Vertical Stress Due to Attraction Load and Time for Asphalt Mixture Containing Pet Plastic Albert Meraudje, Muhammad Isran Ramli, Mubassirang Pasra, A. Arwin Amiruddin, and Hajriyanti Yatmar

.........

Application of Surface Response Methodology (RSM) for Improving the Marshall Quotient of AC-WC Mixtures Containing PET Plastic Waste ..............-........Franky E. P. Lapian, Muhammad Isran Ramli, Mubassii Ardy Arsyad, and Hajriyanti Yatmar

945

956

966

The Effect of Polypropylene (Pp) Plastic Waste on Horizontal Deformation of Concrete Asphalt... . Sukrislistarto, Muhammad Isran Ramli, Mubassirang Pasta, A. Arwin Amiruddin, and Hajriyanti Yatmar Assessing the Seasonal Variation of Particulate Matters Concentration in Perak, Malaysia ......................0..00.. Mohd Fadhil Musbah, Muralia Hustim, and Wesam Al Madhoun

983

The Perception and Behavior of Coal Truck Drivers on the Travel Timeliness ...............-.-.-000000-022000555 Ludfi Djakfar, Dewi Yuniar, Achmad Wicaksono, and Achmad Efendi

992

Foundation Testing on Large Bored Pile in Indonesia.............. Aksan Kawanda and Dennis Rio Perdana Measurement of Static Compaction Energy in Laboratory for Simulating Actual Field Condition on Cohesive Soils............ Anastasia Sri Lestari, Moh. Reinaldo, Lawrence Adrian, and Paulus Pramono Rahardjo

1001

1010

A Sensitivity Parameters on Inelastic Response of Interaction Soil-Pile-Structure System Under Lateral Loading................. Khadidja Sekhri, Djarir Yahiaoui, and Indra Sati Hamonangan Harahap

1020

Greenfield Settlement of KVMRT SSP Line 2 in Kuala Lumpur Limestone Formation .................6..6 0020205200 e eens Noor Azlina Azhari and Hisham Mohamad

1027

XX

Contents

A State-of-the-Art Shafie Rahim, Siti Mohd Reza Azmi, Rohaya Abdullah,

Review on Green Roof Implementation .......... Aminah Osman, Siti Fatin Mohd Razali, Muhamad Nazri Borhan, Azman Mohd Jais, and Suhayya Rofik

Effects of Bitumen Modification on Pavement Performance Subjected to High Vehicular Speed and Extreme Temperature Conditions: A Review. ....................0..004. Abdul Muhaimin Memon, Muslich Hartadi Sutanto, Madzlan Napiah, Fadhli Wong, and Mastura Bujang

1035

1044

Review on the Effect of Curing on Cold Recycled Asphalt Mixture... . 1052 Saeed Modibbo Saeed, Muslich Hartadi Sutanto, Madzlan Napiah, Fadhli Wong, and Mastura Bujang Back-Analysis of Ground Movement Based on Displacement Matching Approach: A Case Study of Landslide at Bridge Abutment Using 3D Finite Element Method Aflizal Arafianto and Paulus Pramono Rahardjo Influence of Raw Precursor on Mechanical Strength and Durability of Geopolymer Pastes in Sulfates . . . Mohamad Yusry Harman, Idawati Ismail, and Yusuf Sahari Critical Parameters for the Development of Web-Based Knowledge System for Low Volume Rural Roads in Sarawak................. Ron Aldrino Chan @ Ron Buking, Muhamad Nazri Borhan, Riza Atiq Abdullah O. K. Rahmat, and Wan Hashim Wan Ibrahim

1082

Toward Applying Complexity Perspective on Travel Behaviour Research ... . . Muhammad Ridwan Anas, Dimas B. E. Dharmowijoyo, and Nindyo C. Kresnanto The Use of Geotechnical Instrumentation and CPTu for Investigation of Geotechnical Failures During Construction in Civil Engineering Projects Paulus P. Rahardjo and Bondan W. Anggoro Factors that Affect Sustainability of Bridges in Jayapura ........... Rezky Aprilyanto Wibowo, Christiono Utomo, and Moh. Arif Rohman

1116

Assessing Landslide Exposure from Soil Characteristics Using Unmanned

Aerial Vehicle (UAV)

...................-..000-00055

1123

Munirah Radin Mohd Mokhtar, Suriani Ngah Abdul Wahab, and Khin Cho Myint Author Index...

2.2.0... 6 020s

1131

Green Environment and Smart Water

Resource Management Systems

®

Check updatesfor

Study on Monthly Rainfall Trend Impact on Reservoir Simulation in Greater Bandung S. Sanjaya'®®, D. Yudianto', and Willy Aulia? ' Civil Engineering Department, Parahyangan Catholic University, Bandung, Indonesia {sanjaya. stephen, doddi_yd}@unpar.ac. id > Graduate Student of Civil Engineering, Parahyangan Catholic University, Bandung, Indonesia willyau. 95@gmail. com Abstract. Greater Bandung lies on the heart of upstream part of biggest water catchment in West Java with abundant of water resources. However, it is facing

an imbalance in supplying its water demand, due to population growth concentrated mostly in urban center. Consequently, it forces to provide the water

supply from its outskirts using reservoir which additionally should be able to cope with climate variability. This

study is aimed to understand the trend of

water availability in greater Bandung and applying such trend towards a hypothetical reservoir. The performance of the hypothetical reservoir will be evaluated using reservoir simulation. The result shows that there is a downward

trend of monthly rainfall trend in greater Bandung. The annual down rates for dry and wet season are 16% and 10%, respectively. Consequently, the utilization of reservoir in the upper area of the study demonstrates only 82% of success rate

to supply the water in urban region, which is lower than recommended rate. Keywords: allocation

1

Rainfall trend - Greater bandung

- Reservoir simulation - Water

Introduction

‘Water is an essential aspect of human life. In many parts of the world, its supply does not meet the demand, and foreseen to spread even more. The imbalance between its supply and demand happens because of the inevitable population growth, particularly that concentrated

in urban

areas

[1].

Consequently,

this circumstance

creates

more

demands upon water in urban settlement transported from its outskirts. Similar situation is expected to occur in Bandung, the capital city of West Java Province in Indonesia. It is expected that in 2040 the population in Bandung and its surrounding districts will grow to 2.924.331 with 0,69% annual rate [2]. In order to supply the bulk water in this city, reservoirs are proposed amongst other alternatives with the highest potential yield of fresh water [2]. To only construct such vast structure will not sufficiently answer the

challenge of abovementioned problem without ensuring its re: nifies the capabilities to cope with the considered risks and results in certain level of performances using limited information, also to endure and compensate from the

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 3-11, 2021. https://doi.org/10.1007/978-981-33-6311-3_1

4

S. Sanjaya et al.

impacts [3]. One of the risks that is threatening infrastructures is climate variability. Hence, it is important to build a climate-resilient infrastructure. Climate-resilient infrastructure will not solely provide its function, but also the potential to enhance the reliability of its function, service life and protect asset returns

[4].

To initiate such concept, one should understand the behavior and future trend of climate condition. Several studies have been carried out focusing solely on extreme climate events [5] but fail to elaborate more about the behavior of seasonal trend which

is useful for water allocation. Therefore, this study is aimed to understand the trend of water availability from a climate condition point of view in Greater Bandung. The research will set up a statistical method to analyze the regional rainfall in Greater Bandung and carry forward supply investigation using hypothetical reservoir given the trend condition. It will be round off by assessing the performance of reservoir simulation using the historical data. 1.1

Study Area and Available Data

Greater Bandung lies on the heart of the upstream area of Citarum River which one of the biggest water catchments in West Java. Bandung people rely its fresh water from surface, spring water and groundwater. A study proposed that, Bandung will not be able to accommodate the needs if it only relies on the same existing water supply in year 2034 [6]. There are 14 stations spread across Greater Bandung from 2001 until 2015, which depicts in Fig. 1. The annual precipitation in Bandung Area is around 2.000 mm and contributes most of its water source to the most important dams in West Java.

Fig. 1.

Rainfall station distribution in Bandung and its outskirts

The hypothetical reservoir is proposed to be built in the West Bandung district, northern part of Bandung city. The dam location is designed in the —6,825 Long and

Study on Monthly Rainfall Trend Impact on Reservoir Simulation

107,661 Lat, which shows in Fig. 2. It lies in Cikawari tributaries, after Cikawari 2A. Occupying around 15,74 km? of water catchment, projected to result around 0,34 m/s bulk water production. The dam approximately 40 m in height, and able to supply water around reservoir is also located one kilometer away from Lembang’s fault.

5

hence, is named Cikawari 2A is is designed for 0,59 m*/s. The

Fig. 2. Location of hypothetical reservoir in the West Bandung district 2

Methods

The analysis will be divided into two parts, the trend of monthly rainfall to check the trend of effective monthly rainfall and reservoir simulation to check the bulk water supply using the hypothetical reservoir. 2.1

Trend of Monthly Rainfall

Using a simple approach, the evaluation will be done by fitting the monthly rainfall data under different scenarios. Monthly rainfall data is the accumulated daily rainfall of each month. These data are fitted to follow a normal distribution. Standard normal distribution is selected due to its simplicity, and since the analysis is not towards extreme events that predominantly shows in other types of distribution. These fitted data will be drawn in a duration curve, which indicates the percentage of exceedance of an arbitrary magnitude. The effective rainfall with 80 percentiles will be taken from the curve to express the monthly rainfall value. These procedures will be performed for each month of the data, to overcome the seasonality behavior. Furthermore, to understand the trend, the scenario will be made by adding up each year of the historical data and repeating the same procedures for all stations. Suggesting the aim of this study, the monthly rainfall trend in Greater Bandung will be analyzed by combining all data using regional rainfall. A regional rainfall is an

6

S. Sanjaya et al.

analysis of representative rainfall data based on relative weights. There are several methods to obtain this value, one of which is the thiessen method. It assumes that rainfall within the watershed shares the same value with the recorded rainfall gauge to a halfway of surrounding gauges [7].

2.2

NRECA Model

NRECA

(Non-Recorded

Catchment

Area)

is

a

conceptual

rainfall-runoff

model

developed by Crawford. The advantage of this model is that it uses monthly rainfall interval, since this data is easier to obtain. In general, NRECA uses two parameters that have a major impact to the simulated discharge, and other three parameters are considered to have less impact. Those two important parameters are PSUB and GWE. In this case, the PSUB and GWF are 0,90 and 0,14, respectively. The first parameter exemplifies excess moisture that seeps into the groundwater storage, meaning that soil with low permeability will have low infiltration, hence low in PSUB value. The latter parameter indicates the contribution of groundwater flow to the discharge in rivers. The input of this model is monthly rainfall and potential evaporation (PET). In this case, the model will be calibrated using the gauged discharge on the downstream area of Cikawari, considering there is no discharge measurement in Cikawari tributary. 2.3

Reservoir Simulation

In principle, reservoir routing is estimated using mass conservation. The input of reservoir routing is the inflow discharge, outflow discharge and storage function. Given all the mention information, one would be able to estimate the water supply using the storage equation. The mass conservation equation for reservoir routing is as follow: I1-O=

AS

ar

where: : Inflow oO : Outflow AS _ : Volume storage change I

AT:

Time

In performing reservoir simulation, for analyzing water supply, the use timestep is monthly based. The reservoir inflow consists of river discharge and rainfall that fall in the reservoir. The reservoir outflow consists of discharge overflow on the spillway, evapotranspiration and the water supply.

Study on Monthly Rainfall Trend Impact on Reservoir Simulation 3

7

Results and Discussion

3.1.

Rainfall Trend Analysis

Figure 3 depicts the result of monthly rainfall in greater Bandung From

the ouput,

it is shown

that there

is a downward

trend

of the

from 2001-2015. Rgo of monthly

rainfall in greater Bandung. It is contrast in comparison to the extreme rainfall that were detected occurred in short period with higher intensity [5]. A good water management and planning should also consider the fact that certain areas are subjected to a high natural variability such as seasonality [8]. Therefore, the result in Fig. 3 will be dissected into two seasons, dry and wet. Subsequently, those months in each category will be averaged and the average values are shown in Table 1. In general, there is a shifting tendency towards low average value during wet season (Dec-Apr) from 2001-2015. It is shown in Fig. 3 that the Rgo in March, which is the peak-wet month, experienced almost 50 mm reduction in overall Rgo from 2001-2015.

Besides, the percent decrease in wet season rainfall exacerbates every year, and in average appear to have 10% annual reduction (Table 1). Unlike wet season, there is no significant change in the amount of Rgo in dry season (May-Oct). However,

the annual

percent decrease in dry period is 16%, which is higher than wet season. The percent decrease for both seasons indicates a strong evidence of downward trend in monthly average rainfall in greater Bandung.

Rp Analysis on Regional Rainfall

300

2001 === 2001 - 2 = & = 2001 2001 — = 2001

250 200

100

0 Jan

Feb

Mar

Apr

May

Jun

Jul

°

Aug

Sep

Oct

Nov

Dec

Fig. 3. Regional monthly rainfall in greater Bandung from 2001-2015 If the downward trend persists to happen in the future, it will affect the management of water supply in the downstream of Citarum River, especially Saguling dam. Saguling dam needs to accommodate certain level of elevation to produce 750 MW electricity, while at the same time also has to share its water to the downstream dams [9]. While this evidence shows a depletion of monthly rainfall solely based on climate

8

S. Sanjaya et al. Table 1.

Year

Percent decrease of average rainfall for dry and wet seasons

‘Average rainfall dry season [mm]

| Percent decrease dry

Average rainfall

Percent

wet season [mm]

decrease wet

season -

season

2001-2011

47.41

2001-2012

| 40.62

14%

195.89

8%

2001-2013

34.54

15%

178.25

9%

2001-2014

| 28.69

17%

159.63

10%

2001-2015

| 23.49

18%

139.19

13%

Annual

213.42

16%

-

10%

point of view, many studies have suggested that combining the climatic and also land conditions will lead to a better water supply management system [10]. 3.2

Hypothetical

(Cikawari

2A)

Reservoir Simulation

The downward trend in precipitation rate is also captured locally in Lembang station, which is depicted in Fig. 4 (red bar chart). Combined between the precipitation rate and calculated PET as the input of NRECA model, Fig. 4 presents the simulated discharge of Cikawari 2001-2015. The simulated discharge may seems in a lower order of magnitude, due to its location in the very upstream tributaries. Hence, the river’s cross is also small. The remarkable results show a low estimation of discharge in the year 2006-2009 and 2015. These low discharges point out one of many the drought symptoms, due to El-Nino that happened several years in Indonesia (2004, 2006, 2009 and 2015)

[11]. Hence,

the performance

of potential reservoir, Cikawari

2A,

will be

tested using these data. In the reservoir simulation, the abovementioned parameters are the main input. Furthermore, to compute the storage function, the area-volumeelevation curve was also used. Along those parameters, some technical parameters were also assumed, such as spillway elevation, dead storage etc. Figure 5 presents the result of Cikawari 2A reservoir simulation for 2001-2015. The dotted line indicates water overflows on the spillway, the solid line signifies the water yields from the storage capacity, whereas the dashed line signifies the water elevation. The simulation shows that the reservoir was not able to supply its maximum capacity in several episodes. First episode was happened after El-Nino in 2006 which caused the water elevation to drop dramatically (see circle 1 in Fig. 5). This dramatic downfall affects the water level to recover to its initial state even in the following years. This was also severely affected by the El-Nino in 2009 and consider causing several succeeding episodes of drought (see circle 2 in Fig. 5). One last remarkable point was shown

in the last circle (circle 3) that reproduce a gradual decrease in water level from

2011-2015. However, the water supply fails to meet the demand only in 2015. Additionally, further studies are suggested to be carried out to find more correlation between the reservoir performance and several El-nino episodes.

Study on Monthly Rainfall Trend Impact on Reservoir Simulation

mmm Precipitation

9

——Simulated Discharge 600

500 400 300

0.8 0.6

04 02

200

|

| | Lh | | | HeAWS| IMS 20 408 05 ga 400% 00° 2087 400% y00° ga go g01 qa1? gov go18 o Wl

Wh

0

Year

Fig. 4. Monthly discharge and precipitation in Lembang station 2001-2015 The monthly water available from the reservoir were evaluated again whether it is able to yield the designated supply. Since the reservoir was designed to supply 0,59 m*/s, therefore, any water available below that threshold is considered fail. Table 2 shows the performance of water supply from the Cikawari 2A reservoir for 2001-2015, expressed in the success rate. Thirty-three out of 180 months are unsuccessful to provide the designated water supply. Thus, the reservoir has only 82% success rate which is still lower than the recommended rate, 95% for bulkwater. To increase the reliability of its water supply, especially during critical periods, one could also consider to occupy intermittent water supply system or other water management systems

[1].

Water Supply - ~~ ~ Elevation 1

1,135 1,130

1

=

1

@ S2

t

= & 5

1,125 =S

1 Ly '

3

s 3 1,120 3

a

8

1115 1,110 om er DY ShYG SPSS

OFM Sf Hf HG F

Se DM SPSSDMPL %, GPG PPL ear

DY LK

LK NS

Fig. 5. Overflow, water supply and water elevation of Cikawari 2A reservoir from 2001-2015

10S. Sanjaya et al. Table 2.

Success rate of Cikawari 2A simulation

Condition | Q (m/s) 0,59 20 mm/h [12] (Table 1).

Study of Saturation Flow at Signalized Intersection

3

15

Methodology

The survey was conducted at the Monjali signal intersection in Yogyakarta. The Monjali Intersection is a crossing of the North Ringroad Road (west and east sides), with the Palagan Tentara Pelajar Road (north side) and with Monjali Road (south side).

Signal control at the intersection is designed with four phases. lane is the lane designated for traffic “left turn on red” so that not used in the calculation of saturation flow. Saturation movements in the lane that experience red lights and green observed when the vehicle is moving at the beginning of the when

On the west side, the left the left-turn movement is flow is carried out on lights. Saturation flow is green light until it stops

the red light is on (Fig. 2).

North 3.5m 3.5m 3.5m 3.5m

Fig. 2.

Geometric conditions of the intersection studied

The traffic volume data used for this study was obtained from the movement of vehicles released from the west side of North Ring Road Road. Vehicle movement from the west side consists of one left tum lane, two straight lanes, and one right turn lane. The intersection geometric survey is conducted at night when the vehicles do not pass much. Each lane has a width of 3.5 m. Signal settings at Monjali intersection use fixed time so that the phase time does not change. The green time of the movement of the vehicle from the west side for 45 s with a yellow time of 2 s. 2 surveyors are needed for recording moving vehicles and 1 surveyor responsible for measuring rainfall. The camcorders are used to record the traffic of vehicles passing through the intersection. The camcorders operate on the east side rather north and on the south side

16

Risdiyanto and Syaripin

with a height that allows all movement of vehicles from the west moving across the stop line to be seen. This tool must also be protected when it rains so an umbrella is needed as a protector. Rainfall intensity data is obtained by using a simple tool. Rainfall measurement is measuring the height of rain water falling in one particular area. BMKG Climatology Station Mlati released a simple tool made from used paint cans with a can diameter of about 10 cm, which was given a buffer board for a higher position. For practical purposes, in this study used a 4D PVC pipe (4 in. diameter or 10.16 cm) as high as 30 cm, a PVC pipe mounting tool in the form of a flat wooden board, a ruler with a length of 30 cm, and a stopwatch. PVC pipe is placed at the edge of the Monjali intersection, not blocked by trees and does not interfere with the movement of motorized vehicles/pedestrians. Location requirements for rainfall gauge placement are: ¢ ¢ ¢

Open area/without shade, Make sure the rainfall reservoir is installed at a minimum height of 1.2 m from the ground. This is to avoid splashing water from the soil into the reservoir. Rainfall catcher is placed on a flat/flat surface and does not fall/tilt when the wind blows strong.

The stopwatch tool is used to calculate the duration of rain when it falls. After the rain stops, the water level in the pipe is measured with a ruler to get the thickness of the rain when the survey is conducted. The study was conducted for 2 days (one day in sunny conditions and one day in rainy conditions) on a weekday peak hour in the morning/evening for 2 h. The survey was conducted in a span of time with heavy traffic at the Monjali signal intersection in Yogyakarta. Based on observations during the preliminary survey, the survey range was used at 16.00-18.00 WIB. If on the first day of the survey, sunny weather conditions, we get saturation flow sunny weather conditions. However, on the first day of rainy weather, a rain saturation flow survey was conducted as well as a rainfall intensity survey. The survey plan on H + | is canceled if the weather conditions are the same as the previous day. However, if the H + 1 weather conditions differ from the previous day, then the survey can be conducted. Thus the survey was carried out, until the saturation flow data obtained for sunny and rainy conditions including the intensity of the rain. Field saturation currents in sunny and rainy conditions are obtained by counting the number of vehicles that cross the stop line at the west side of Monjali intersection per 6s period in the green signal period. The passing vehicles are grouped into 3 types namely Motorcycle (MC), Light Vehicle (LV), and Heavy Vehicle (HV), with passenger car equivalence value of 1, MC of 0.2, and for HV of 1.3 [6]. The calculation

process is repeated every time the green signal lights up, to obtain many vehicle volume data of the stop line. Not all vehicle volume data that are separated from the intersection stop line in the green period are used in the calculation of saturation flow. Saturation flow is defined as the steady state discharge rate of queued vehicles from an approach at a signalized intersection with continuous green and an infinite queue [8]. By looking at this definition, the data chosen is data with the state of the vehicle at the intersection of the western approach which has the following three characteristics:

Study of Saturation Flow at Signalized Intersection

¢ ¢

¢

17

When the red light is on, the vehicle queue is quite long and the distance between the vehicles is quite tight There are still vehicles left when the red light comes on from the movement of the vehicle when it was green before. This condition shows the flow of vehicles in a saturated condition because the flow does not stop The number of vehicles crossing the stop line - as seen from the camcorders - does not differ greatly from the highest saturation flow value at the same 6-s time interval.

Some selected data then total the number of vehicles in each period of six seconds and then divided by the number of intervals of 6 s. For example in the 9th second (midpoint of the second 6-s range), the number of vehicles passing by 6 selected data is 14; 15; 15; 14; 15.5; and 15 junior high, then the average is 14.75 junior high/6 s. This

value is then entered into the table. This is also done for the range of the first 6 s, third, fourth, etc.

4

Results and Discussion

By using the equivalence values of passenger cars for MC, LV, and HV each of 0.2; 1; and 1.3, then the number of each type of vehicle surveyed is multiplied by its equivalence

so that traffic flow is obtained (pcu/time).

After the reduction

of current

data under sunny conditions, 6 valid data of saturated currents were obtained in each period of six seconds. By averaging six saturation flow values the following results are obtained.

Table 1. No_

Rain intensity category

Rain intensity (mm/h) | Rain category

1

20

Light Medium Heavy Very heavy

By looking at Table 2, 14.0 + 14.47 + 14.65 + 14.9 + saturation flow of 13.86 pcu/6 From the graph, it appears green time to the end of green pew/6 s which occurs at a time end of green time.

we get saturated current in sunny conditions of 13.42 + 11.72 = 83.16. Average of 6 data obtained s or 8,316 pewh (Fig. 3). that the current departing from 0 at the beginning of time reaches a peak saturation current value of 13.86 after 9-39 s. This value will decrease slightly until the

18

Risdiyanto and Syaripin Table 2.

Traffic flow in sunny weather conditions

Seconds _| Traffic flow (pcu/time) MC

LV

HV

Q)

Q

@)

0-3

15,20

13,00

3-9

33,00

51,00

9-15

1,30

Amount (pcu/time)

Saturation flow

(1) + (2) + (3)

(peu/time)

29,50

4,92

-

84,00

14,00

35,80

51,00

-

86,80

14,47

15-21

24,00

60,00

3,90

87,90

14,65

21-27 27-33

14,80 11,60

72,00 65,00

2,60 3,90

89,40 80,50

14,90 13,42

33-39

7,20

54,00

9,10

70,30

11,72

39-45,

3,20

36,00

1,30

40,50

6,75

16 zou 8 &

12

xg 2

10

2 2

8

26 5 8g

4

3 °

2

6

0 0

3

9

15

21

27

33

39

«45°

«48

green time Fig. 3, Saturated current graph in sunny weather conditions Meanwhile, saturated currents during rainy conditions are obtained at H + | at 16:30-17:30 WIB. The height of the rainfall intensity was also measured in this H + 1 survey. As a calculation in sunny conditions, an average of intersection saturation flow results in 8 relevant data. The average of the 8 samples produced saturated currents in rain conditions of 14.74 pcu/6 s or 8,844 pcu/h. The height of rainfall intensity is 3 mm and included in the category of light rain intensity (Fig. 4 and Table 3).

The two saturation flow data in sunny and rainy conditions are compared to the following table. According to Table 4, in each 6 s, saturation flow in rain conditions is mostly above the saturation flow value in sunny conditions except for the second six seconds where saturation flow when it rains is 13.78 mm/6s, while in sunny time is

Study of Saturation Flow at Signalized Intersection Table 3.

Seconds

Traffic flow in rainy weather conditions

| Traffic flow (pcu/time) MC Q@

LV Q)

HV @)

Amount (pcu/time)

Saturation flow

() +2) +6)

(pcu/time)

0-3

24,40

23,00

3,90

51,30

6,41

3-9

37,40

65,00

7,80

110,20

13,78

9-15

43,20

77,00

9,10

129,30

16,16

15-21

27,00

96,00

5,20

128,20

16,03

21-27 27-33

18,20 14,80

98,00 85,00

5,20 9,10

121,40 108,90

15,18 13,61

33-39

13,20

91,00

5,20

109,40

13,68

39-45,

8,40

72,00

3,90

84,30

10,54

=z

18 16

g

14

$

12 +

2

10

2

= < & 6

5&

19

8) 6

4]

2 0. o

3

9

15

21

27

33

39

45

48

green time Fig. 4, Saturated current graph in light rainy weather conditions Table 4.

Comparison of sunny and rain weather saturation flow values

Weather

Saturation flow (pcu/6 seconda)

Sunny

4,92 | 14,00)

Rain (3 mnvh)

14,47

14,65 | 14,90 | 13,42

11,72,

6,75

6,41 | 13,78 | 16,16

16,03 | 15,18 | 13,61

13,68

10,54

14.00 mm/6 s. To test the differences in the saturation flow data group, an Analysis of Variance (ANOVA)

was performed. With the ANOVA

difference test on two paired data) using SPSS following matters:

Test Paired T

software,

test (parametric

it can be concluded

the

20 ¢

Risdiyanto and Syaripin The significance level of the relationship is 0,000, which means it is significant at the 0.01 level. Df (degree of freedom) for T Paired analysis is always N — 1. Where N is the number of samples. So Df is worth 8 — 1 = 7 T value of 1,000. Compared with t table in DF 7 of 0.711, we get t count > t table, which means significant. The probability value or p-value of the Paired T-test is 0.022. This means that there is a difference in value between sunny and rainy weather conditions because the p-

e ¢ ¢

value is < 0.05

¢

(95% confidence).

Mean negative value of —1.32. This means that there is a tendency for an increase in saturation flow values when the weather is light rain. The average increase is 1.32

The analysis shows that there is a difference between saturation flow in sunny weather conditions and saturation flow in rain conditions. Saturation flow in rain conditions is higher than saturation flow in sunny conditions. This result is quite surprising because previous studies show that when it rains, the speed and volume of traffic will fall [13], so that the

saturation

flow

when

it rains

is also lower than

the

saturation flow when it is sunny. However, by seeing that the rain category is light rain, it can be understood that the vehicle driver has not experienced obstacles in driving the vehicle. Precisely when it starts to rain lightly, the driver will be in a hurry to get to the destination quickly. Especially by seeing that most drivers ride motorbikes. This rush affects the increase in saturation flow at the signal intersection.

5

Conclusions and Recommendations

Based on the results of research on saturation flow on the west side of the Monjali Yogyakarta Indonesia intersection during sunny and rainy conditions, the following conclusions can be drawn: ¢

The saturation flow value that occurs in sunny conditions is 13.86 pcu/6 s or 8,316 peu/h in green and the saturation flow value in the rain condition is 14.74 pcu/6 s or 8,844 pcu/h in green. Rainfall that occurs during a saturation flow survey is included in the category of light rain The saturation flow value in light rain conditions is higher than the saturation flow value in sunny weather

¢ ¢

By looking at the conclusions, the advice that can be given from the results of this study is the need to do a saturation flow study in various conditions of rainfall intensity.

References 1. Messer, C.J.: A Guide for Designing and Operating Signalized Intersections in Texas. Texas v

(1975) . Shao, C.Q.,

Liu,

X.M.:

Estimation

of saturation

Discret. Dyn. Nat. Soc. 2012, 1-0 (2012)

flow

rates at signalized

intersections.

Study of Saturation Flow at Signalized Intersection

21

. Agent, K.R., Crabtree, D.: Analysis of Saturation Flow at Signalized Intersections, no. 2, p. 10 (1982) . Sun, H., Yang, J., Wang,

L., Li, L., Wu, B.: Saturation flow rate and start-up lost time of

dual-left lanes at signalized intersection in rainy weather condition. Procedia - Soc. Behav. Sci. 96, 270-279 (2013)

. Shin, C.-H., Choi, K.: Saturation flow rate estimation under rainy weather conditions for on-

line traffic control purpose. . MKJI: Manual Kapasitas Indonesia, vol. 1, no. I, p. . McMahon, J.W., Krane, Transp. Res. Board (1997) . Arkatkar,

S., Mitra,

KSCE J. Civ. Eng. 2(3), 211-222 (1998) Jalan Indonesia (MKJI). In: Mkji Manual Kapasitas Jalan 564 (1997) J.P. Federico, A.P.: Saturation flow rates by facility type.

S., Mathew,

T.:

Global

Practices on

Road Traffic

Signal Control,

pp. 217-242 (2019) . Salter, R.: Highway Traffic Analysis and Design, Revised Ed. London: The Macmillan Press Ltd (1976) . Critchley, W., Siegert, K., Chapman, C.: A Manual for the Design and Construction of Water Harvesting Schemes for Plant Production. Food and Agriculture Organization of the United Nations (1991). https://www.fao.org/3/u3160e/u3160e00.htm#Contents. . BMKG: Daftar Istilah Klimatologi (2020). https://balai3 .denpasar.bmkg.go.id/daftar-istilahmusim. Accessed 08 Jan 2020

. BMKG: Peraturan Kepala Badan Meteorologi, Klimatologi dan Geofisika, no. 497, p. 4246703 (2010) . Xu, F., He, Z., Sha, Z., Zhuang, L., Sun, W.: Assessing the impact of rainfall on traffic operation of urban road network. Procedia - Soc. Behav. Sci. 96, 82-89 (2013)

® ‘upaates

Deep Learning Neural Network for Time Series Water Level Forecasting Nuratiah Zaini®, Marlinda Abdul Malek'?, Shuhairy Norhisham!, and Nurul Hani Mardi! ' Department of Civil Engineering, Universiti Tenaga Nasional,

Kajang, Malaysia {Nur_Atiah, Marlinda, Shuhairy, NHani}@uniten. edu. my > Institute of Sustainable Energy, Universiti Tenaga Nasional, Kajang, Malaysia Abstract.

Reliable forecasting of water level is essential for flood prevention,

future planning and warning. This study proposed to forecast daily time series water level for Malaysia’s rivers based on deep learning technique namely long short-term memory (LSTM). The deep learning neural network is based on artificial neural network (ANN)

and part of broader machine learning. In this

study, forecasting models are developed for I-h ahead of time at multiple lag time which are 1-h, 2-h and 3-h lag time denoted as LSTM,1, LSTM,2 and LSTM,.3, respectively. Forecasted water level is significant for determination of

effected area, future planning and warning. Root mean square error (RMSE) and coefficient of determination

(R?) are utilized to evaluate the performance of

proposed forecasting models. An analysis of error in term of RMSE and R? show

that the proposed LSTM,3

model outperformed other models for water

level forecasting during training and testing phase.

Keywords: Water level - Forecasting - Deep learning - Neural network Machine learning 1

Introduction

Several areas in Malaysia are vulnerable to flood due to heavy rain after the water level tise abruptly. Thus, reliable water level forecasting is essential for flood control, future planning and management. However, to forecast water level, which known tend to vary abruptly according to rainfall, precise decisions must be declared within a few hours. Therefore, alternative methodology should be chosen which is believes to be able to forecast water level efficiently

[1].

Recently, various methods artificial

intelligence

(AI)

as an

have

been proposed

alternative

to forecast water level utilizing

to traditional

statistical

method

such

as

autoregressive integrated moving average (ARIMA) and autoregressive moving average (ARMA)

[2]. Machine

learning

techniques

namely

artificial

neural

network

(ANN),

support vector machine (SVM), and hybrid techniques are widely applied in time series forecasting [3-5]. It is found that, the application of machine learning for time series forecasting are able to overcome the drawback in traditional statistical method in various

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 22-29, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_3

Deep Learning Neural Network

23

field such as air quality forecasting [6, 7], river flow forecasting [8, 9] and price forecasting [10,

11].

A new technology in machine learning namely deep learning has gain its popularity to solve time series problems in the past few years. Deep learning is a branch of broad machine learning techniques that is based on ANN that consists of multiple hidden layers [12]. Deep learning architectures such as recurrent neural network (RNN), convolutional neural network (CNN) and gated recurrent unit (GRU) have been successfully applied in

numerous fields. For instance, deep learning NN is proved to be successful in field of economy such as price forecasting and stock forecasting [10, 13], environmental engineering such as air quality forecasting and weather forecasting [14-16], water resources engineering [1] and medicine [17, 18]. Besides that, the application of deep learning NN can be divided into another two domain which are classification and regression. Classification is a method to identify the categories of certain observations while regression is a method to estimate the relationships among variables [19-21].

This study aims to explore deep learning neural network namely

long short-term

memory (LSTM) for time series water level forecasting. LSTM is a variation of recurrent neural network (RNN) that developed to overcome the shortcoming in RNN

which has difficulties in learning long term dependencies [22]. On top of that, it is found that LSTM has been broadly applied in non-linear time series forecasting problems and yield outstanding forecasting performances [14, 15]. Therefore, this study introduced the application of LSTM to forecast time series water level. Proposed LSTM forecasting model is developed using combination of rainfall and historical water level at several time lags. Hourly water level and rainfall data for Sungai Kuantan are used as input variables to the forecasting models. The models are designed to forecast water level for 1-h step ahead of forecasting at 1-h, 2-h and 3-h time lag denoted as LSTM,.,, LSTM,.. and LSTM_-;, respectively. Performances of proposed model are evaluated using statistical approach in term of coefficient of determination

(R?)

and

root

mean

square

(RMSE).

In depth

analysis

based

on

the

statistical approaches are preformed to show the improvement of forecasting models for time series water level forecasting.

2 2.1

Materials and Methods Study Area and Dataset

The study area is Sungai Kuantan, located in the state of Pahang, Malaysia. Hourly data of historical water level and rainfall datasets for three different rainfall stations from 1 January 2016 to 31 December 2016 are used as input data to the proposed forecasting models. Total 8784 number of data are divided into two dataset which are 70% for training and remaining 30% for testing dataset. Mean value of rainfall depth for three rainfall stations considered in this study namely Station 1 (Rainfall)), Station 2 (Rainfallz) and Station 3 (Rainfall3) are 0.3 mm, 0.2 mm and 0.3 mm respectively.

Meanwhile, maximum rainfall depth for the three stations are 47.1 mm, 67.7 mm and 67.3 mm respectively. Water level station used in this study has mean value of 16.063 m and maximum level of 18.69 m. Table 1 summarizes the mean, standard deviation, minimum and maximum values of input parameters.

24

N. Zaini et al. Table 1.

Statistics of input parameters

Water level (m) | Rainfall, (mm) | Rainfall, (mm)

Rainfall; (mm)

Mean 16.063 Standard deviation | 0.3639

0.3 1.9232

0.2 2.1265

0.3 2.3115

Minimum

15.70

0

0

0

Maximum

18.69

47.1

8784

8784

Total number

2.2.

67.7

67.3

8784

8784

Long Short-Term Memory

Long short term memory (LSTM) is a variation of RNN that capable to learn long term dependencies by including architecture special units called memory blocks [14]. An LSTM unit performs self-loop memory block and consists of four gate units which are input gate to write information, update gate to update information, forget gate to forget information and output gate to read information from memory cell [22]. Therefore, self-loop memory helps the gradient flow along the long sequences and solve the gradient problem. While gate units help to control the flow information and allow the network to learn over many time steps. The gates in hidden units own unique characteristics that help in holding relevant data and forgetting irrelevant data to provide constant error [14, 22]. Figure

1 shows general architecture of LSTM.

©

output Layer

Fig. 1. LSTM architecture [14] LSTM

have

been

widely

used

forecasting problems in various medicine [21] and energy-related

and gain

its popularity

for solving

time

series

field such as environmental [15], financial [13], problems [22]. For instance, Navares and Aznarte

[14] used four different architecture of LSTM namely fully connected LSTM, group of pollutant LSTM, individual groups of pollutant LSTM and single pollutant LSTM to predict air quality. Besides that, Salman et al. [23] utilized single layer and multi-layer LSTM for forecasting univariate weather variable. In financial field, Wu, Wu and Zhu

Deep Learning Neural Network [13], Safari and Davallou

[10] and Chen,

He and Tso

[11] applied LSTM

25

to forecast

crude oil price and oil price. This study proposed to forecast water level at 1-h step ahead using three LSTM models which are LSTM,.,, LSTM,5, and LSTM,.; with the combinations of water level and three rainfall data from different gauge at several time lags. The data were prepared for LSTM forecasting model involves dataset framing as supervised learning problem and normalizing the input variables. Next, the data were divided into two datasets called training and testing before the model was defined and fitted. Mean absolute error (MAE) is used as loss function for the proposed models shown in Eq. 1. Both training and test loss is tracked during training by setting validation argument to the algorithm. The models are fitted for 100 training epochs with a batch size of 1 [22, 24, 25]. The parameter setting of forecasting models is summarized in Table 2.

Table 2. No

Parameter setting for forecasting model Parameter

Value

1 | Learning rate 0.001 2 | Exponential decay 0.9, 0.999 3. Number of hidden units 50 4 Batch size 1 5 Epoch 100 Forecasting

of water level is performed

after the model

fitted and forecasting

performances are evaluated in term of root mean square error (RMSE) of determination (R?).

and coefficient

(1) where ¥; is forecasted water level ith data point, y; is actual water n denotes as total number of data points in the test set. As an altemative, adaptive moment estimation (ADAM)

of parameters

for proposed forecasting technique

is used as an optimization

with a learning rate « = 0.001, an

exponential decay of B; = 0.9 and Bz = 0.999 [14, 26]. ADAM

the unknowns

in the proposed

adaptive estimates

of lower-order

deep

learning

moments

level and

forecasting

algorithm used to train

model

[14, 27, 28]. Besides

that is based that, ADAM

on

algo-

rithm has benefit in solving optimization problem as it is easy to implement, efficient computation, required small computer memory, suitable for non-stationary objectives and little tuning on hyper-parameters [14, 29, 30].

26

2.3,

N. Zaini et al.

Models Evaluation

To evaluate the performance of proposed models for water level forecasting, different Statistical criteria are

used called root mean

square

error (RMSE)

and coefficient

of

determination (R’) as in Eq. 2 and Eq. 3, respectively.

RMSE =

2 R=

(2)

[DLO = Yang)Or = Fave)]” 3

Thi = Yang)” x Dia

Yang)

3 ”

where J; is forecasted water level i data point, y; is actual water level and n denotes as total number of data points. Vavg and Yay are average value of actual and forecasted water level. R’ indicates square of correlation coefficient and determine the degree of regression function [31]. In this study, proposed LSTM water level forecasting models were compared to multilayer perceptron (MLP) forecasting model in order to evaluate forecasting model’s performances.

3

Performance Results

This section discusses the performances comparison of proposed long short term memory (LSTM) forecasting models with multilayer perceptron (MLP) for combination of water level and rainfall data at different time lags for time series water level forecasting. The efficiency of LSTM technique is studied by carrying out a comparative analysis of LSTM forecasting model over an alternative deep learning technique namely MLP for water level forecasting at several time lags. The main purpose of this study is to develop an efficient forecasting model with minimal complexity and high

accuracy. Table 3 summarized forecasting error and accuracy of developed forecasting models in term of RMSE and R° at different time lags of forecasting. LSTM models are evaluated for three different time lags which are 1-h, 2-h and 3-h denoted as LSTM,.1,

LSTM,. and LSTM,.3, respectively. Multilayer perceptron (MLP) is introduced in order to compare and evaluate its performances over proposed LSTM models. An in-depth analysis of error analysis in term of RMSE shows that proposed LSTM model with 3-h time lag outperformed other models for water level forecasting during training and testing. Similar result is obtained for performance accuracy in term of R. The results show that LSTM with 3-h lag time yield better R’ as compared to other forecasting models for both during training and testing.

Deep Learning Neural Network Table 3. RMSE and R? for Model | RMSE (m) Training | Testing LSTM,; |0.049 |0.119 LSTM,2 |0.038 0.074 LSTM,3 | 0.035 0.073 MLP 0.050 0.113

forecasting R Training | 0.941 | 0.967 0.972 0.941

27

models Testing 0.899 0.961 0.963 0,909

To further study the performances of proposed LSTM, percentage of improvement for RMSE and R? are utilized to exhibit performances improvement of LSTM over MLP forecasting model. From Table 3, it is found that, RMSE increased by 30% for training and 35% for testing when compare the performances of LSTM with 3-h lag time and the lowest performance of MLP forecasting model. Moreover, performance accuracy of LSTM increased by 3% and 6% for training and testing, respectively. From analysis study, it is found that, there is significant improvements of water level forecasting for the three LSTM models as compared to MLP forecasting model for both RMSE and R’. However, the small improvement of LSTM over MLP illustrates that LSTM not really showing good advantage in skill over MLP model. Therefore, exploring MLP with a large window is recommended in order to solve for autoregression problems.

4

Conclusion

This study presents the performance of LSTM forecasting model for time series water level. LSTM models are designed for combination of historical water level and rainfall at three different station as model input to forecast water level. The proposed model is evaluated at several time lags which are 1-h, 2-h and 3-h lag time namely LSTM,.1, LSTM,.2

and LSTM_>3, respectively. Performances

forecasting model are evaluated

in

term of forecasting error and its accuracy. LSTM forecasting model with 3-h lag time yielded better performance as compared to other developed model for water level forecasting at training and testing dataset. LSTM,.; shows improvement of forecasting performance in both RMSE and R?. However, it is recommended that further study can be conducted to explore other method in deep learning such as convolutional neural network, gated recurrent unit and hybrid techniques for time series forecasting.

Acknowledgement. The author would like to acknowledge Universiti Tenaga Nasional, Malaysia for financially support this research under UNITEN BOLD 2020 Grant: RJO10517844/077. References

1. Sung, J.Y., Lee, J., Chung, LM., Heo, J.H.: Hourly water level forecasting at tributary affecteby main river condition. Water 9(9), 1-7 (2017)

v

N. Zaini et al. . Arbain, S.H., Wibowo, A.: Time series methods for water level forecasting of Dungun River

in Terengganu Malaysia. Int. J. Eng. Sci. Technol. 4(4), 1803-1811 (2012) . Shang, Y., Xu, Y., Shang, L., Fan, Q., Wang, Y., Liu, Z.: A method of direct, real-time forecasting of downstream water levels via hydropower station reregulation: a case study from Gezhouba Hydropower Plant, China. J. Hydrol. 573, 895-907 (2019) . Zhu, S., Hrnjica, B., Ptak, M., Choitiski, A., Sivakumar, B.: Forecasting of water level in

multiple temperate lakes using machine learning models. J. Hydrol. 585, 124819 (2020)

. Zhang, X., Liu, P., Zhao, Y., Deng, C., Li, Z., Xiong, M.

ror correction-based forecasting

of reservoir water levels: improving accuracy over multiple lead times. Environ. Model. Softw. 104, 27-39 (2018)

. Wang, J., Niu, T., Wang, R.: Research and application of an air quality early warning system based on a modified least squares support vector machine and a cloud model. Int. J. Environ. Res. Public Health 14(3), 249 (2017) . Wang, J., Bai, L., Wang, S., Wang, C.: Research and application of the hybrid forecasting model based on secondary denoising and multi-objective optimization for air pollution early warning system. J. Clean. Prod. 234, 54-70 (2019) . Sahoo, A., Samantaray, S., Ghose, D.K.: Stream flow forecasting in mahanadi river basin us g artificial neural networks. Procedia Comput. Sci. 157, 168-174 (2019) . Zaini, N., Malek, M.A., Yusoff, M., Mardi, N.H., Norhisham, S.: Daily river flow forecasting with hybrid support vector machine - particle swarm optimization. In: IOP Conference Series: Earth and Environmental Science, vol. 140, no. 1 (2018)

. Safari, A., Davallou, M.: Oil price forecasting using a hybrid model. Energy 148, 49-58 (2018) . Chen, Y., He, K., Tso, G.K.F.: Forecasting crude oil prices: a deep learning based model. Procedia Comput. Sci. 122, 300-307 (2017) . Chen, Z., Chen, S.: Applicability of deep neural networks on production in Bakken shale reservoirs. J. Pet. Sci. Eng. 179, 112-125 (2019) . Wa, Y.-X., Wu, Q.-B., Zhu, J.-Q.: Improved EEMD-based crude oil price forecasting using LSTM networks. Phys. A Stat. Mech. its Appl. 516, 114-124 (2019) . Navares, R., Aznarte, J.L: Predicting air quality with deep learning LSTM: towards comprehensive models. Ecol. Inform. 55, 101019 (2020)

. Wu, Q., Lin, H.: Daily urban air quality index forecasting based on variational mode decomposition, sample entropy and LSTM neural network. Sustain. Cities Soc. 50, 101657 (2019) . Zhou, Y., Chang, F.-J., Chang, L.-C., Kao, L-F. . Wang, Y~S.: Explore a deep learning multi-output neural network for regional mul (ep-ahead air quality forecasts. J. Clean. Prod. 209, 134-145 (2019)

. Saba, L., et al.: The present and future of deep learning in radiology. Eur. J. Radiol. 114, 14— 24 (2019) . Wang, J., Li, R., Li, R., Fu, B.: A knowledge-based deep learning method for ECG signal delineation. Futur. Gener. Comput.

Syst. 109, 56-66 (2020)

. Ma, J., Li, Z., Cheng, J.CP., Ding, Y., Lin, C., Xu, Z.: Air quality prediction at new stations using spatially transferred bi-directional long short-term memory network. Sci. Total Environ. 705, 135771 (2019) . Shen, Z., Zhang, Y., Lu, J., Xu, J., Xiao, G.: A novel time series forecasting model with deep learning. Neurocomputing 396, 302-313 (2019) . Majd, M., Safabakhsh, R.: Correlational Convolutional LSTM for human action recognition.

Neurocomputing 396,

224-229 (2019)

Deep Learning Neural Network 22.

Somu, N., Raman, M.R.G., Ramamritham, K.: A hybrid model for building energy consumption

23.

forecasting using long short term memory

114131 (2020) Salman, A.G., Heryadi, Y., Abdurahman, short-term memory (LSTM) model with Procedia Comput. Sci. 135, 89-98 (2018)

24.

networks.

26. 27. 28. 29.

Appl.

Energy

261,

E., Suparta, W.: Single layer & multi-layer long intermediate

variables

for weather

forecasting.

Liang, Z., Zou, R., Chen, X., Ren, T., Su, H., Liu, Y.: Simulate the forecast capacity of a complicated water quality model using the long short-term memory

25.

29

approach. J. Hydrol.

581, 124432 (2020) Kumar, J., Goomer, R., Singh, AJ Long short term memory recurrent neural network (LSTM-RNN) based workload forecas ing model for cloud datacenters. Procedia Comput. Sci. 125, 676-682 (2018) Ruder, S.: An overview of gradient descent optimization algorithms, p. 14 (2016) YY Hong CLPP Rioflorido: A hybrid deep learning-based neural network for 24-h ahead wind power forecasting. Appl. Energy 250, 530-539 (2019) Athira,

V., Geetha,

P., Vinayakumar,

R., Soman,

K.P.: DeepAirNet:

applying

recurrent

networks for air quality prediction. Procedia Comput. Sci. 132, 1394-1403 (2018) Hong, Y.Y., Rioflorido, C.L.P.P.: A hybrid deep learning-based neural network for 24-h ahead wind power forecasting. Appl. Energy 250(April), 530-539 (2019)

30.

Kingma, D.P., Ba, J.L. : Adam:

31.

Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, p. 15 (2015) He, F., Zhou, J., Feng, Z., Liu, G., Yang, Y.: A hybrid short-term load forecasting model based on variational mode decomposition and long short-term memory networks considering relevant factors with Bayesian optimization algorithm. Appl. Energy 237, 103-116 (2019)

a method for stochastic optimization. In: 3rd International

® ‘upaates

Optimization Study of n-ZVI Oxidation for Organic Pollutants Removal from Wastewater Muhammad Raza Ul Mustafa®, Tahir Haneef, Brenda Tan Pei Jian, Khamaruzaman Wan Yusof, and Hifsa Khurshid

Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS, 32610 Seri Iskandar, Malaysia raza. mustafa@utp. edu. my

Abstract. The wastewater obtained during the exploration of oil and gas is known as produced water (PW). It contains organic and inorganic pollutants that

are generating problems for aquatic life and humans. For degradation of hazardous organic pollutants in PW, Fenton like oxidation experiments were conducted using nanoscale zerovalent iron (n-ZVI). For this study laboratory-scale batch reactor was performed. Experiments were carried out at room temperature and 200 rpm magnetic stirrer speed. Design-Expert software was used for the optimization of Fenton like oxidation process to obtain the maximum removal of COD

under optimal values of independent variable such as n-ZVI dosage, pH and contact

time.

To

obtain

suitable experimental

design,

Box-Behnken

design

(BBD) based on response surface modelling (RSM) was applied. n-ZVI concentration, pH, and contact time were coded factors as A, B and C, respectively. At

optimal values of variables such as; n-ZVI concentration 5.17 g/L, pH 2.80 and contact time 98.3 min., the maximum removal of COD Keywords:

1

Wastewater - n-ZVI - COD

- BBD

was predicted as 61.7%.

- RSM

Introduction

Suitable experimental design and appropriate response analy: an important tool for assessing an efficacious environmental technology [1]. Multiple mathematical models comprise of useful tools for designing a suitable experimental model to optimize the variables for various wastewater treatment processes. Mostly wastewater treatment techniques have more than two variables and the optimization of these variables via a conventional

method

is inefficient, time-consuming,

and

stubbom

[2]. Box

Behnken

design (BBD) and response surface modeling (RSM) are powerful tools to optimize the multiple operating variables in wastewater systems. RSM and BBD have been successfully utilized to optimize the variables for (PW) treatment using different materials/methods [3]. But there is still no reported application of RSM for n-ZVI oxidation for the effective removal of COD from PW. PW is by-product of oil and gas exploration industries; it usually e: subsurface and brought to the surface during oil and gas extraction [4]. It is the most significant offshore

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 30-38, 2021. https://doi.org/10.1007/978-981-33-631 1-3_4

Optimization Study of n-ZVI Oxidation

31

discharge connected with the production of fossil fuels [5]. PW volume can be ten times

higher than petroleum volume produced during the extraction process over the civic life of the production sector. It is typically pumped into geological rocks in many coastal regions of the world or directly injected into wetlands and coastal areas [6, 7]. PW comprises a high amount of organic and inorganic pollutants. The concentration of these pollutants depends on the type of extrication method, variety of hydrocarbons, geological location of well, life of well, and depth of the well [8, 9]. Hydrocarbons mainly present in PW include polyaromatic hydrocarbons (PAHs), organic acids, and phenols. These organic pollutants are possible to contribute in increasing PW toxicity and radioactivity. PAHs and heavy metals have been reported as potential long-term impact sources of toxicity in PW [10]. Recently in scientific community it is widely believed that the water-soluble portion of organic pollutants leads towards much more acute and chronic toxicity of PW [11]. Thus, organic pollutants are the primary concern of further analysis in this work for PW treatment before discharging into any nearby waterbody. Nanoscale

zerovalent

iron (n-ZVI)

oxidation

has gained

considerable

interest

as

one of the environmental friendly technologies of wastewater treatment systems [12]. It is iron’s nanomaterial with 100 m?/g surface area, which makes it more reactive than other metals. n-ZVI redox potential is E = —0.44V, which implies that it can efficiently reduce the concentration of organic pollutants by degrading them [13]. n-ZVI is an active reagent, which can serve as an electron donor. In additions it is non-toxic, readily available, and environmentally friendly compound. n-ZVI (Fe°) oxidizes into ferrous (Fe**) and ferric ion (Fe**), causing to release electron to become

free (active) for the

degradation of other compounds. n-ZVI oxidation was found useful for the removal of organic pollutants from municipal wastewater [14], arsenite removal from water [13], heavy metals from wastewater [15], COD and heavy metals from landfill leachate [16],

and degradation of organic pollutants in olive mill wastewater [12]. In order to analyze the performance of n-ZVI for the degradation of organic pollutants in PW, this study was conducted. BBD experimental design in RSM was used to evaluate the effect of independent parameters (such as; pH, n-ZVI dosage, and contact time) for maximum

2 2.1

removal

of COD

from PW.

Methods and Chemicals Methods

Batch experiments were carried out for n-ZVI oxidation method. used as a reactor. Calculated volume of PW sample was taken in purpose. The pH of the sample was adjusted using 1M solution of placed on a hot plate and calculated amount of n-ZVI was added

A glass beaker was reactor for oxidation HCI. The rector was to start the oxidation

reaction. The oxidation reaction was practiced at constant temperature (25 +

5 °C) and

magnetic stirrer speed of 200 rpm. After the estimated contact time, the oxidation reaction was stopped by increasing the pH of the sample up to pH 10 using 1M solution of NaOH. Then the COD of the treated sample was measured using High range COD vials and HACH

DR 2800

PW was 2213 mg/L.

spectrophotometer [17]. The initial concentration of COD

in

32 2.2.

M. R. U. Mustafa et al. Experimental Optimization Design

Box-Behnken design based on RSM was chosen to find out the optimal conditions of independent variables (A: pH, B: n-ZVI dosage, and C: Contact time) for maximum removal of COD. The parameters were designated as low, center, and high levels (—1, 0, and +1) as represented in Table

Table 1.

1.

Designated parameters for design-expert software.

Parameters

High and low values of independent variables =I 0 +1 pH; A (-) 2 3 4 n-ZVI; B (g/L) 2 6 10 Contact Time; C (mins.) 30 90 150

Depending on the combination of different values of independent variables, 17 experimental runs were generated, including five center points replicates by BoxBehnken design as shown in Table 2. The 5 replicates were used at the center of the. design to estimate pure error sum of squares. Based on Design-Expert software, a quadratic regression model (Eq. 1) [18] was used to obtain the predicted response data to fit according to BBD.

Y= Bot SO B+ Sy Bint+ ey ya Biss te

()

Here Y is representing the response, while variables are written in Xi and Xj form. The number of factors studied are indicated by K. Bo, Bj, Bij, and Bij, known as constant coefficient, linear interaction coefficient, quadratic interaction coefficient, and the 2nd order terms interaction coefficient respectively. In addition the F-test and P-values were used for the determination of significance of each factor. 2.3

Analysis of Variance

For analysis of data and interaction between dependent (responses) and independent variables, Analysis of variance ANOVA was used. R? indicates fit quality of polynomial model while F-test analyzes its statistical significance. P-value with 95% confidence level was used to analyze model terms. For removal of COD, 3D plots were obtained. In Table 2, values of responses for different independent variable values are

given.

Optimization Study of n-ZVI Oxidation Table 2. No _

Experimental runs and dependent variables for COD

Factors

COD

A

B

pH (-)

n-ZVI dosage (g/L)

removal.

removal (%)

Cc Contact time (mins)

Actual responses

Predicted responses

1

2

2

90

35.90

37.45

2

2

5

150

48.50

47.46

3 4

3 3

2 5

30 90

30.80 62.30

30.19 61.28

5

4

5

150

35.70

36.64

6

4

2

90

29.70

29.27

7

3

2

150

41.00

40.49

8

3

8

150

39.00

39.61

9

4

8

90

37.20

35.65

10 11

2 4

5 5

30 30

38.50 35.20

37.56 36.24

12

3

5

90

63.00

61.28

13

3

8

30

39.10

39.61

14

3

5

90

62.30

61.28

15

3

5

90

61.60

61.28

16

3

5

90

57.20

61.28

17

2

8

90

39.20

39.62

3

33

Results and Discussion

Optimization of independent variables was accomplished for the removal of COD in PW during oxidation process. Percent removal of COD was selected as an experimental response to analyze the results for n-ZVI application, in order to evaluate optimum conditions of independent parameter. All the independent parameters (A, B, and C) and dependent

variable

Y (COD

removal

(%))

are

given

in Table

2. A 2nd order poly-

nomial regression equation was obtainedby conducting multiple regression testing on the experimental data to explain the interaction between the independent parameters and responses. The result for n-ZVI oxidation was fitted according to the quadratic model as shown in Eq. 2.

Y = 61.28 — 3.04A + 2.14B +2.58C+ 1.05AB — 2.38AC — 2.57BC — 11.89A2 — 13.89B2 — 9.92C2

(2)

34

M. R. U. Mustafa et al.

3.1

Analysis of Variances

The variance analysis (ANOVA) in Table

3. P-values < 0.05,

for the quadratic polynomial equation is represented

show

that

the

terms

of the

model

are

significant

[2].

Values > 0.1000 suggest the terms of the model

are not significant. The

F-value of

“Lack

shows

lack

of Fit”

for n-ZVI

oxidation

(0.05 < 0.63)

non-significance

of fit

which relates to pure error, and it is good. There is 63% chance that a large value of Lack of fit might happen due to noise. The model’s R-squared values and p-values indicate that quadratic model fitted well to the experimental results. The model is significance for COD removal as it has 54.38 F-value and low probability (P-value) value. The model’s ANOVA displayed reliable confidence in estimating the removal of COD. The square correlation coefficient for the model response was determined as coefficient

of determination

(R?).

At

a confidence

level

of 95%,

it exhibited

high

significant regression. The Adjusted R* is in good agreement with predicted R* as given in Table 4. High value of R? implies that the quadratic model is satisfactorily adjusted to experimental data. Table 3.

ANOVA

Sources | Mean square Model | 249.68 A 2381 B 36.55 c 53.05 AB 441 AC 22.56 BC 26.52 rS 595.25 B 812.35 Ce 413.93 Lack of fit] 3.45

In Table 4, PRESS

for BBD

| DF | 9 | 1) 1 1 | 1 1 1 | 1 | 1 | 1 | 3. |

with significant terms.

Sum of squares |F value | P value 2247.14 54.38 | 10-7 cmésec needs to be improved soil (treatment) so that the soil is impermeable so that the sea water that is

accommodated in the pond land is not reduced/depleted because it seeps into the soil. When the entire soil reaches saturation, which is after almost 8.5 h, the infiltration discharge has decreased very significantly and the volume of water needed to reach the

saturation condition is 1.5 x 10-* m*/m? or 15 L/m?, after experiencing more infiltration discharge smaller than 4 x 10? m/sec. Acknowledgment. The authors would also like to thank to BSG soil consultant for providing all the necessary data for this study and encouragement to conduct such studies for the benefit of

science and society. References Beven, K.: R-E. Horton perceptual model of infiltration process. Hydrol. Process. 18, 3447-3460 (2004) Horton, R.E.: An approach towards physical interpretation of infiltration capacity. Proc. Soil Sci.

Soc. Am. 5, 399-417 (1940) Mgruder, K.: Halite, Guidelines for Rock Collection (2007)

Oklahoma De Fries, R., Eshleman, K.N.: Land-use change and hydrologic (2004) Beven, K., Young, P., Romanowicz, R., O'Connell, E., Ewen, J., O'Donnell, G.M., Holman, L., Posthumus, H., Morris, J., Hollis, J., et al.: Analysis of historical datasets to look for impacts

of land use and management change on flood generation; Final Report FD2120; DEFRA: London, UK (2008)

Viessman, W., Lewis, G.L., Knapp, J.W.: Introduction to Hydrology, 3rd edn., Harper & Row, New York (1989)

USDA: USDA-NRCS Soil Survey Division, National Strategy Database (1998)

®

Check updatesfor

Application of Inhibition Model to Prevent Nitrification Upset in Petrochemical Wastewater Treatment Plant Idzham Fauzi M. Ariff? Project Delivery and Technology Division, Petroliam Nasional Berhad (PETRONAS), Kuala Lumpur, Malaysia idzhamfauzi_mariff@petronas. com. my Abstract. An industrial wastewater treatment plant at a petrochemical facility in Terengganu, Malaysia was modeled using a wastewater treatment plant simulation software. The simulation was commissioned in order to develop control and operational strategies to ensure robustness of the treatment facility against plant upsets due to ingress of high concentrations of nitrification-

inhibiting substances, namely benzene and monoethanolamine (MEA). To model the inhibitory effects of the compounds to the nitrification activity, a noncompetitive inhibition model was incorporated in the activated sludge model.

Based on the simulation results, it was found that benzene concentration of 220 mg/L in the wastewater treatment plant feed significantly inhibited nitrification. The results also indicated that increasing MEA concentration will lead to excessive carbonaceous and nitrogenous load before significant nitrification inhibition is observed. Based on the findings, a set of wastewater influent parameter limits for specific wastewater influent sources were proposed to prevent nitrification upsets due to excessive ingress of these inhibitory

compounds. Keywords:

Activated sludge models - Industrial effluent treatment -

Inhibition - Nitrification - Petrochemical - Simulation

1

Introduction

Production and processing of petrochemicals produce wastewater containing a complex, blend of organic and inorganic contaminants. For example, in an olefins and ethylene oxide/ethylene glycol plant, diverse compounds such as ethylene glycol, acetaldehyde, formaldehyde, phenol, benzene, toluene, ethyl benzene, xylene, and poly-aromatic hydrocarbons have been reported

[1].

Biological treatment processes (e.g. activated sludge plant) are often used to reduce the organic and nitrifiable load of the effluent to ensure discharge specifications are met. Wastewater treatment plant simulation software packages are commercially available that incorporate the International Water Association (IWA) activated sludge models (e.g. ASM1, ASM2, etc.) to model the biological transformation kinetics and stoi-

chiometry including nitrification and denitrification processes. Although the IWA models have been developed specifically for municipal wastewater systems, the

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 81-92, 2021. https://doi.org/10.1007/978-981-33-6311-3_10

82

1. F. M. Ariff

simulator packages have been successfully used to model industrial effluents, given adequate characterisation and calibration of key parameters [2]. Model modifications may be required for industrial effluents; acute nitrification inhibition by inhibitory components is a significant modification that is often necessary [3]. In a petrochemical facility in Terengganu, Malaysia producing olefins, glycols and other olefin derivatives, the industrial effluents from the processing facilities are treated in a centralised industrial effluent treatment system consisting of primary treatment, secondary treatment, tertiary treatment and sludge handling system. The main processing equipment in the treatment plant include an equalization tank, a diversion (offspec) tank, biological aeration basins, secondary clarifiers, lamella clarifiers, sludge thickeners, belt filters and a land farm. In this facility, several compounds have been found to have historically caused upsets to the effluent treatment system, resulting in elevated levels of ammoniacal nitrogen at the clarifier effluent due to impaired nitrification, risking breach of effluent discharge limits. In extreme instances, plant production has been stopped or curtailed resulting in significant production opportunity loss in order to prevent discharging offspecification effluent. Following a nitrification upset, reinstatement of stable nitrification is also time-consuming due to the relatively low growth rate of nitrifying autotrophic bacteria. Two of these inhibitory compounds have been identified at the facility, namely benzene; produced in the olefins plant and amine (primarily monoethanolamine, or MEA)

from the ethanolamine

plant.

Benzene has been found to inhibit nitrate formation by 53% cultures

at 10 mg/L

[4]. On

the

other hand,

in biological

in batch nitrifying

ammonium

removal

from

industrial process water, 10 mM of MEA was required to decrease nitrifying bacteria activity by 50% [5]. In a related study of inhibition potential on nitrifying biofilms in a moving bed biofilm reactor, it was found that ECsq values for ammonia-oxidizing bacteria

(AOB)

and

_nitrite-oxidizing

bacteria

(NOB)

for

MEA

was

between

17—

118 mM [6]. The objective of this study is to determine the allowable limits for benzene and MEA at the effluent sources in the petrochemical facility that will prevent nitrification upsets at the wastewater treatment plant. This was achieved by characterising the influent streams, developing a model of the wastewater plant and inhibition effect, calibrating the model using available data and conducting simulation runs on the computer model. Specifically, the wastewater simulation software GPS-X (v 6.5.1) by Hydromantis was employed in the study. The Model Developer option in the software was used to incorporate non-competitive inhibition model for benzene and MEA to the autotrophic growth rate in order to model inhibition of nitrification process by these compounds. This work was conducted to ensure robustness of treatment and continuous and reliable compliance to the legal discharge limit

2

Methods

The overall methodology for the modeling and simulation can be summarized in Fig. 1 below:

Application of Inhibition Model to Prevent Nitrification Upset

Datagathering

|>] Create model layout

>|

Model calibration (normal conditions)

>|

Run model simulation to determine source limits

83

yy] Model modification (inhibition term)

>

Determine inhibition constant,

Fig. 1. Study methodology 2.1

Data Gathering

Data for operational parameters as well as influent and effluent characteristics were obtained from 2 primary sources, which are: 1. Plant Information Management System (PIMS). This system collects and integrates information about the entire production process from various sources. These include time-stamped data from online instruments, analysers and laboratory test results of samples for performance monitoring and product quality. 2. Industrial effluent characterisation study. An effluent characterisation study was commissioned which consists of a 3-consecutive day sampling program to collect samples from a total of 41 sampling points covering all the main effluent sources, wastewater treatment units and final clarifier discharge. The analytical testing of up to 31 water quality parameters was conducted by a commercial laboratory accredited under MS ISO/IEC 17025. 2.2

Model Layout Construction

The model layout used for this study was based on a previously developed model layout of the same facility described by M Ariff [7]. This layout was recalibrated for the purposes of this work. After recalibration of the existing model layout (which has a single combined influent object), new influent objects representing the multiple upstream effluent sources were added to the layout. In addition, some changes were made to the layout to simplify the model including removal of the sludge dewatering objects (i.e. dissolved air flotation (DAF) thickener, sludge digester and filter press equipment) and all objects downstream of secondary clarifiers. This was done to reduce the simulation time by removing all processes not under investigation. Figure 2 below shows the updated model layout of the effluent treatment system and influent sources in GPS-X. Olefin, EOG and Butanol streams are represented by wastewater influent objects which are used to characterise continuous streams of wastewater flow. As for the Derivatives stream, it is a combination of 4 wastewater sources in the Derivatives plant and represented in the model layout by their respective influent objects. Apart from the 4 main process influents, there are multiple other sources of effluent into the treatment plant; however, these streams only represent a

84

1. F. M. Ariff

small percentage of the total flow and contaminant load and are combined into a single influent object labelled ‘Others’.

Fig. 2. Model layout in GPS-X Supplementary carbon addition was also modeled accordingly. The aeration basins were modeled as a Completely-Mixed Tank object using the MANTIS activated sludge model, whereas the secondary clarifiers were modeled using the Circular Secondary Clarifier object using the simpleld model. 2.3,

Model Calibration

The model parameters were calibrated under dynamic conditions using plant monitoring data from Ist January to 31st March 2017. This calibration was conducted using the previous model layout with a single combined influent i.e. prior to adding the upstream effluent sources. Inhibition model was not incorporated in the model. Dynamic time-stamped data from PIMS were used as a basis to generate the GPS-X input data files for the following parameters: Influent flowrate Influent Influent Influent Influent

total Chemical Oxygen Demand Total Kjeldahl Nitrogen (TKN) ammoniacal nitrogen (NH3-N) nitrate nitrogen (NO3-N)

Influent alkalinity Splitter box split ratio

(COD)

Application of Inhibition Model to Prevent Nitrification Upset

e ¢

85

Aeration tanks dissolved oxygen Clarifier return activated sludge (RAS) flowrate.

Some of the above input parameters were calculated values e.g. TKN, which was based on pre-established ratio between organic nitrogen and COD and Splitter box split ratio, which was calculated from the measured flowrates. Calibration was done by repeatedly running the model and adjusting model parameters to optimise goodness-of-fit to key parameters. This was done by minimising the absolute root mean

square deviation

(RMSD)

between

the model

prediction

and

measured data. For this study, RMSD of less than 10% of the maximum or discharge limit was considered acceptable. The parameters optimised in the calibration were as follows: Average

NH,-N

Average COD

at clarifier overflow

at clarifier overflow

Average

total suspended solids (TSS) at clarifier overflow

Average

mixed

liquor suspended

pended solids (MLVSS)

solids (MLSS)

and mixed

liquor volatile sus-

in Aeration Tanks.

The general approach to the calibration was to firstly calibrate MLSS and MLVSS in the aeration tanks by adjusting the instantaneous sludge wasting rate. Once the MLSS and MLVSS has been calibrated, the effluent TSS and COD was then calibrated by adjusting the inlet particulate fractions, namely the particulate COD/VSS (volatile suspended solids) ratio and VSS/TSS ra while ensuring feed TSS is obtained. The effluent COD was then calibrated by adjusting the inert fraction of soluble COD. For new influent objects, the calibrated influent fractions from the earlier single combined influent object were used as the basis and the fractions were adjusted to match the corresponding feed TSS values. As the averaged 3-day parameter values from the IECS did not adequately balance in terms of mass, the ‘Others’ stream characteristics was adjusted to reconcile the combined effluent feed characteristics with the actual values taken for the simulation model basis. 2.4

Inhibition Model

Inhibition models that are frequently used to model reduction to growth rates include (i) competitive half-saturation bition, and (iii) the growth rate

inhibition, where higher inhibitor concentration reduces the substrate constant, (ii) Haldane inhibition, which describes high substrate inhinon-competitive inhibition, where high inhibitor concentration reduces [8]. The Haldane model is not suitable as it does not describe the effect

of the inhibitor. Since both MEA and benzene are not substrates for the nitrifying autotrophs, the non-competitive inhibition model is selected for this study. This is modeled by addition of an inhibition function (switching function) to the general autotrophic growth kinetic expression as shown in the following equation [9]: Ha == Ha max

x K,+58, Ki x\ Ky+S s

(1)

86

1. F. M. Ariff In the

autotrophic

above

equation,

maximum

4

is the

autotrophic

specific

growth

rate, “a max is the

specific growth rate, K; is the half saturation coefficient for

inhibitor (i.e. inhibition constant), S; is the inhibitor concentration, K, is the half saturation coefficient for substrate S (ammonia) and S is the substrate (ammonia)

concentration. In GPS-X, this was achieved by adding two industrial soluble substrate state variables to represent MEA and benzene from the Carbon — Nitrogen — Industrial Pollutant (CNIPLIB) State Variable Library using the Model Developer function. The switching functions for MEA and benzene inhibition using the industrial state variables were also added to the rate function for growth of autotrophs in the MANTIS model. To simplify the analysis, the substrates were assumed to be unconsumed by any processes in the model i.e. inert. 2.5

Determination of Inhibition Constant

The inhibition constant for benzene inhibition to nitrification was determined by running the calibrated model using plant monitoring data during a period of high benzene ingress in the combined influent with a corresponding impairment in nitrification as indicated by an increase in NH3-N concentration at the clarifier overflow. The inhibition constant was determined by parameter optimization in GPS-X (Nelder-Mead simplex method) by selecting the clarifier overflow NH3-N as the target variable. MEA was taken as the model compound for modeling amine inhibition, as it is reportedly the primary form of alkanolamine found in the wastewater effluent. The MEA inhibition constant used in the model was not calibrated from plant data, due to lack of reliable MEA and total nitrogen data during previous incidents. Instead, the inhibition constant was based on the study by Colaco [5], which found that nitrifying bacteria activity decreased by 50% in the presence of 10 mM of MEA, corresponding to 610 mg/L or 800 mg O3/L. 2.6

Determination of Source Limits

The MEA and benzene maximum limits were determined by conducting steady-state simulation runs and increasing the MEA or benzene concentration progressively until the concentration of key effluent parameters reach 80% of the discharge limit. The effluent parameters were COD, NH3-N and TSS with discharge limits of 200 mg/L, 20 mg/L and 100 mg/L respectively. For MEA, as the concentration was increased in the simulation, the corresponding increase in TKN and COD were also calculated based on stoichiometric ratio of N:MEA for TKN and reported COD ratio of 1.54 gCOD/gMEA for MEA [10]. In the simulation runs, the dissolved oxygen levels were maintained at a high level (3.1 mg/L) to ensure aeration requirement was not limiting.

Application of Inhibition Model to Prevent Nitrification Upset

3 3.1

87

Results and Discussion Calibration and Validation Results

Figure 3 shows the model-predicted MLSS and Fig. 4 shows

the COD,

TSS

and MLVSS

and NH3-N

results at the aeration tanks

results at the clarifier overflow

for the

combined influent model after model calibration compared with the actual plant performance monitoring results in the period between Ist January to 31st March 2017. The goodness-of-fit results are indicated in Table 1. Overall, the calibrated model was able to predict the actual results with a reasonable degree of accuracy.

MLSS and MLYSS for EAB

o

0

2%

© MLSS factual]

3

4 so 6 7% 8 9 Time (days) MLVSS [actual] —=MLSS [model] ====MLVSS [model]

MLSS oF MLVSS (mg/L)

MLSS and MLVSS for NAB

0

0

© MLSS [actual] Fig. 3.

2%

3

4 9 6 0 8 9 Time (days) MLVSS [actual] MLSS [model] --==MLVSS [model]

Actual and model-predicted MLSS

and MLVSS

at Aeration Basins

The calibrated parameters are shown in Table 2. All other kinetic and stoichiometric parameters were not adjusted from the default values in GPS-X. It can be seen that only 4 influent fractions were required to be calibrated in order to obtain a good match with the actual effluent parameters during normal conditions. It is noted that in

88

1. F. M. Ariff

COD (mg/L)

sse88

COD at Clarifier Overflow

°

1

2

3

4 so 6 Time (days) © COD [actual] ——COD {model]

7

8

99

TSS at Clarifier Overflow

= 50

2 4 30 20 0} ° 0

= — 1

xe

‘

8200630

iC] ‘Time (days) TSS [actual] ——TSS [model]

Ammonia-N (mg/L)

Ammonia at Clarifier Overflow

Time (days) © NH3-N [actual] | —=NH3-N [model] Fig. 4.

Actual and model-predicted results for COD, TSS and NH3-N at the clarifier overflow

this exercise, the model system can be considered to be under-specified; if influent fractions data were available, it is expected that some kinetics and stoichiometric parameters may need to be changed in order to adequately calibrate the model.

Application of Inhibition Model to Prevent Nitrification Upset Table 1.

Root

mean

square

deviation

results

between

model

prediction

89

and

parameter

at the clarifier effluent after calibrating the benzene

inhibition

measurements Parameter MLSS

RMSD (mg/L)

aeration 1

MLVSS MLSS

aeration 1 aeration 2.

MLVSS

aeration 2,

cop TSS NH3-N

Table 2.

481 506 523 573

18.1 27 14

Calibrated parameters and values

Parameter

Value | U!

Influent particulate COD/VSS ratio Influent VSS/TSS ratio

3.2. The

| 1.8 | gCOD/gVSS 0.75 gVSS/gTSS

Influent soluble fraction of total COD

0.94

Influent inert fraction of soluble COD

0.05 | —

—

Benzene Inhibition Constant results for NH3-N

constant is shown in Fig. 5. The benzene inhibition constant obtained by the parameter optimization tool is 45 mg/L with a root mean square deviation of 3.9 mg/L between simulation prediction and actual

measurement

of NH3-N.

It should

be noted that the

benzene inhibition constant obtained in this work was higher than the level for 53% nitrate production inhibition obtained by Zepeda et al. in 2008, which is 10 mg/L. In this work, benzene is assumed to inhibit autotroph growth rates without itself undergoing any transformation. This was done pragmatically to limit the number of additional model parameters and simplify the analysis. In reality, benzene undergoes biological oxidation as well as significant volatilization, which would reduce its effective concentration in the activated sludge system. Hence, this would be reflected in the higher inhibition constant that was obtained by the calibration. Acclimatisation of the local autotrophic microbial population to presence of benzene in the feed can also contribute to this effect.

90

1. F. M. Ariff

Ammonia at Clarifier Overflow

8

*

: Seal” 4

2

Fig. 5.

.

_. © 2 MW Time (days) NHB.N [actual] —=NH3-N (model) 6

§&

Ww

i

2%

Comparison between model prediction and actual NH3-N concentration at the clarifier

effluent after calibration of benzene inhibition constant. 3.3

Determination of Source Limits

Main parameter values for the base case model after the above calibrations are shown in Table 3 below:

Table 3.

|TSS

(mh)

(mg/L) | (mg/L) | (mg/L) | (mg/L)

Olefins 50.0 EOG 408 Butanol = 1.2

/61.3 | 152.2

AKM Sump 1.7 EOABGE 3.37 BABGE 17

61.2 17.0 (359

Overflow

|COD

|TKN | NH;-N

90.1 | 656.4 |15 | 1859 |116.4 349.2. | -

Derivatives 9.0 TIL Sump 2.1

Others EQtank East aeration tank North aeration tank Clarifier

Main parameter values for base case model

Flow

2006/4298 4566 | -

= =

MEA

(mg/L)

(mg/L) _| (mg/L) _| (mg/L)

fins3 ||-

0) = =

MLSS | MLVSS

[N/A [N/A [N/A

[NIA | NIA | NIA

| |-

4128 -

|N/A | N/A

| NIA N/A

| 1376 | 339.717 || 1696 |970 |700 0 ||101 — |= =

= 1100 =

[N/A |N/A N/A

| NIA | NIA | NIA

268 (03 14 126.7 | 40.82 Hass N/A NIA |NIA-

266 -

Benzene

20 34.7 [NIA

5 19.9 NIA

=

NIA

N/A

NIA

|NIA-

[NIA

NIA

| NIA

N/A

165

562/135

042

-

=

NA

NIA N/A | 5883

| NIA Iw | 4894

NA

| 5678

| 4701

=

|

Application of Inhibition Model to Prevent Nitrification Upset

91

The benzene concentration limit was found by progressively increasing benzene concentration at Olefins stream from the base case model, which yielded a limit of 570 mg/L of benzene. At this concentration, the resultant benzene feed into the biological system was 220 mg/L and the NH-N concentration at the clarifier outlet rose to 16.3 mg/L from base model concentration of 0.42 mg/L. The base case TKN loading was set at 34.7 mg/L. However, it may be the case that the benzene ingress occurs during a period of higher nitrogen load to the plant. To investigate this, the NH3-N concentration was increased (with corresponding increase in TKN) at the EOA/BGE stream to achieve a resultant high nitrogen load condition, where TKN and NH3-N at the EQ tank was 103.4 mg/L and 89.1 mg/L respectively. Alkalinity in the feed was also correspondingly increased to ensure that nitrification was not limited by alkalinity level. The results indicated that the limit of 570 mg/L of benzene

at

15.1 mg/L.

For MEA,

Olefins

was

valid,

as

the

outlet

NH3-N

concentration

at concentration of 13,800 mg/L in EOA/BGE

EQ Tank), the NH3-N

also increased to 3,867 mg/L in EOA/BGE

EQ Tank). In actuality, the increase in NH;-N

to

stream (367.1 mg/L in

level at clarifier overflow rose to 16.15 mg/L.

high concentration, TKN

increased

However,

at this

(111.8 mg/L in

level was due to alkalinity limitation at

the high TKN load to the plant. When alkalinity in the feed was increased, the NH3-N level reduced significantly. In fact, if alkalinity was ensured to be high and not limiting, even at 4% MEA concentration in EOA/BGE

stream (1,064 mg/L in EQ Tank), the NH3-N

still not exceed the set limit. At this level, the COD and TKN in the 2,593 mg/L and 271 mg/L respectively, while the COD at the clarifier 158.7 mg/L compared to 56.2 mg/L in base case model. This indicates logical system would fail primarily due to excessive nitrogenous and load and not from the inhibitory effect of the MEA on the nitrifiers.

4

level would

EQ tank was outlet rose to that the biocarbonaceous

Conclusions and Recommendations

The non-competitive inhibition model was used in an industrial effluent treatment plant to develop control and operational strategies to ensure robustness of the treatment facility against plant upsets due to ingress of high concentrations inhibiting substances. The model was applied to model inhibiting compounds to nitrification process namely benzene and MEA. This model was used in a calibrated activated sludge model using a wastewater plant simulation software for a wastewater treatment plant at a petrochemical facility producing olefins, glycols and olefin derivatives. Benzene inhibition constant was determined by model calibration using plant performance monitoring data during high benzene events while literature inhibiting concentrations were used to approximate the inhibition constant for MEA. The simulation results found that benzene inhibited nitrification at 220 mg/L at the feed from the EQ tank, which corresponds to 570 mg/L at the Olefins stream. Hence, this level should be adopted as the limiting concentration at the source streams and controls should be implemented to ensure this.

92

I. F. M. Ariff

However, for MEA, the simulation work indicates that higher concentrations in the feed resulted in an increase in TKN and COD load and would not necessarily lead to excessive inhibitory effect, as long as operational parameters critical for nitrification function are maintained, especially aeration and alkalinity levels. Nonetheles: s the inhibition constant for MEA was based purely on literature, it is recommended that bench scale nitrification treatability tests using the actual MEA wastewater be conducted to validate the findings from this simulation study.

References

v

. Bayat, M., Mehrnia, M.R., Hosseinzadeh, M., Sheikh-Sofla, R.: Petrochemical wastewater

treatment and reuse by MBR: a pilot study for ethylene oxide/ethylene glycol and olefin units. J. Ind. Eng, Chem. 25, 265-271 (2015) . Andres H.W., Kujawski D.G., Schraa 0.J., Lin, C.J., Wong, A.D.: Process optimization of a petroleum refinery wastewater treatment facility using process modeling and site specific biokinetic constants. In: WEFTEC 2011 Conference Proceedings (2011)

. Gernaey, K.V., van Loosdrecht, M.C.M., Henze, M., Lind, M., Jorgensen, $.B.: Activated sludge wastewater treatment plant modelling and simulation: state of the art. Environ. Model

Softw. 19, 763-783 (2004)

. Zepeda, A., Texier, A.C., Gomez, J.: Benzene transformation in nitrifying batch cultures. Biotechnol. Prog. 19, 789-793

(2003)

. Colago, A.B.: Biological water treatment for removal of ammonia from industrial process water. Universidade de Lisboa, Master’s dissertation (2009) . Henry, LA.: Biologi cal nitrogen removal of effluents from amine-based CO, capture plants. Norwegian University of Science and Technology, Ph.D. dissertation (2016) . Mohd Ariff, -F., Bakir, M.: Dynamic simulation of petrochemical wastewater treatment using wastewater plant simulation software. In: International Conference on Civil, Offshore & Environmental Engineering 2018 (ICCOEE 2018), MATEC Web of Conferences 203,

03005 (2018) . Rieger, L., Gillot, S., Langergraber, G., Ohtsuki, T.. Shaw, A., Takacs, I., Winkler, Guidelines for Using Activated Sludge Models. TWA Publishing, London (2013)

S.:

. Nowak, O., Svardal, K., Schweighofer, P.: The dynamic behaviour of nitrifying activated sludge systems influenced by inhibiting wastewater compounds. Wat. Sci. Tech. 31(2), 115— 124 (1995) . Monoethanolamine Material Safety Datasheet. Dow Chemical Company. http://msdssearch. dow.com/Published_iteratureDO WCOM/dh_0044/0901b80380044789.pdf?filepath=amin es/pdfs/noreg/1 11-01388.pdf& fromPage=GetDoc

®

Check updatesfor

Design Expert Application for Optimization of Ag/AgBr/TiO, Visible Light Photocatalyst Preparation Augustine Chioma Affam'?, Wong Chee Chung'?, Poh Lin Lau’, Olufemi Adebayo Johnson’, Khor Cheng Seong*, Lavania Baloo’, Bryan Wong Lee Peng', and Fung Xinru'

' University College of Technology Sarawak, 96000 Sibu, Sarawak, Malaysia [email protected],

affamskii@yahoo. com

> Centre of Research, Innovation and Sustainable Development (CRISD), Sarawak, Malaysia 5 Heriot-Watt University, Malaysia, no. 1, Jalan Vena P5/2, Precinct 5, 62200 Putrajaya, Malaysia + Universiti Teknologi PETRONAS Malaysia, Persiaran UTP, 32610 Seri Iskandar, Perak, Malaysia Abstract.

Titanium

dioxide

(TiO2)

was

activated

under

visible

light

after

modification by doping with silver/silver bromide (Ag/AgBr) through simple impregnation-precipitation-photoreduction process. The use of central composite design tool of design expert in order to optimize the preparation parameters namely silver dopant, precipitation time and photoreduction time for

the degradation of pesticide chloropyrifos was feasible. From the analysis of variance

(ANOVA),

the

most

influential

factors

affecting

the experimental

process and interaction was identified. The chemical oxygen demand (COD) removal increased with increasing the amount of Ag dopant and photoreduction time while precipitation time was not a significant factor. Optimized

condition for preparation of the Ag/AgBr/TiO, was mass ratio of Ag to TiO,: 1:1, precipitation time: 60 min, and photoreduction time: 90 min. Experimental COD removal efficiency was 77% while the predicted value was 77.4%. The percentage error was 0.51%. The close agreement between both predicted and actual COD removal shows good model.

Keywords: Visible light photocatalysis - Chlorpyrifos wastewater - CCD Optimization - Modeling - COD - Phosphate 1

Introduction

Water is the basic need for all life forms on earth but obtaining quality water has become a challenge for many countries. This is because the industrial effluent mix with natural water bodies and water reservoirs have exposed the life forms across the world face health issues [1]. About 25%

of the total water consumption

is used

in industrial

activities globally [2]. The agricultural and industrial wastewaters are classified as one

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 93-102, 2021. https://doi.org/10.1007/978-981-33-6311-3_I1

94

A.C. Affam et al.

of the significant sources of the pesticides in the environment [3]. The high toxicity of pesticides presented in water exposes human health to risk [4, 5]. Chlorpyrifos (CPF) is the most widely used broad spectrum non-systemic organophosphorus pesticide in the world since 1960s to control crop insects and house-hold pests is toxic to non-target living forms, including humans [6, 7]. CPF has long half- life within 60 to 120 d, high residual concentration (0.01-0.62 mg/kg), highly contaminates lands and aquatic ecosystems [8, 9]. Thus, the treatment or removal of CPF is important. Titanium

dioxide

(TiO)

is abundant

in

the

earth

and

quite

resistant

to

photo

corrosion, nontoxicity, chemical inertness become the attractive photocatalyst. Due to its reusability and self-regenerated properties. TiO has been selected to be used widely compared to other treatment materials or catalyst [10-12]. However, its wide bandgap makes it only harvest light in the ultra-violet light region [13, 14]. The rapid recombination of photogenerated electron-hole of TiO, leads to low quantum yields [15]. Thus, modification of TiO, is needed. Doping is able to enhance the photo catalytic activity by extending the harvested light in visible light region [16, 17]. AgBr is recognised as an essential photosensitive material and it has been applied to TiO, for the enhancement of its photocatalytic properties with varying degree of success [18, 19]. A major challenge in the production of Ag/AgBr/TiO, is its ability to degrade the CPF wastewater effectively. The most important characteristic of the Ag/AgBr/TiO3 is its photocatalytic properties which is highly effected by the preparation conditions. In order to know how the preparation conditions affect the quality of the Ag/AgBr/TiO2, the use of an adequate experiment design is important [20]. Response surface methodology (RSM) is a mathematical and statistical method that is used to design the experiments, and develop models by studying the interactions of the parameters and optimise the experimental conditions [21-23]. Optimization of the catalysts preparation has been applied in preparation of activated carbons from olive-waste cakes [24], activated

carbon

from

palm

oil fronds

[20]

and

hydroxyapatite

(HAP)

from

catfish

bones [25]. Simple impregnation-precipitation and photoreduction [26] was selected to prepare the Ag/AgBr/TiO2. From the literatures, various preparation methods and types of TiO2 were selected to produce the Ag/AgBr/TiO2. This includes, tetrabutyltitanate (Ti(OBu)4) and sol- gel method

[27], titanium (IV) isopropoxide (TTIP),

sol-hydrothermal [28] deposition-precipitation methods. Powdered TiO2 was used to prepare Ag/AgBr/TiO2 in this study [29]. The objective of this research was to optimize the preparation conditions of Ag/AgBr/TiO2 and apply it for the removal of COD and phosphate in CPF wastewater based on the RSM experimental design approach so as to validate the quantity of catalyst used.

Design Expert Application for Optimization of Ag/AgBr/TiO2. 2 2.1

95

Materials and Methods Materials

Titanium

(IV) oxide (TiO2),

was purchased

from

Merck,

silver nitrate (AgNO3)

was

purchased from Bendosen, potassium bromide (KBr) was purchased from BDH. These chemicals were used to prepare the Ag/AgBr/TiOz. The Chlorpyrifos pesticide (38.7 w/w %) was purchased from a local shop. 2.2

Preparation of Ag/AgBr/TiO,

One gram of TiO2 powder was placed in a 20 mL volume of known concentration of AgNO3 for 30 min. Thereafter, 20 mL of KBr with a concentration greater than the AgNO3 was added drop by drop to the mixture and stirred continuously between 30 to 120 min in order to induce the precipitation of deposited AgNO3 into AgBr. The reduction of AgBr was carried out via irradiation by visible light at room temperature for 30 to 120 min. Finally, the mixture was washed several times with distilled water and dried in a vacuum oven at 70 °C [30]. The prepared samples were denoted by weight ratio Ag/AgBr/TiO2. The catalyst was prepared and the mass ratio of Ag:TiO2, was calculated assuming that all of the AgBr was reduced to Ag, for simplicity. 2.3,

Design of Experiment

Central Composite Design (CCD) tool of the Response Surface Methodology (RSM) was employed for design of the experiment. The preparation conditions are shown in Table 1.

Table 1, Input Variables and ranges.

Variables

Units | Factor code | Level -10

Amount of silver |g

| XI

Precipitation time min | X2 Irradiation time | min | X3

41

T1)2)3

60 90120 30 60 90

There were 28 experimental runs generated as shown in Table 2. These experimental runs were used to study the effects of the parameters on the COD and phosphate removal since organophosphorus pesticides produce phosphate ions [31], thus it was chosen as one response factors.

96

A.C. Affam et al.

Table 2. Experimental design matrix generated by CCD. Run

Amount of silver (g)

Irradiation time (min)

60 60

30 90

3.2

90

60

4

2

90

0

5

1

120

90

6

2

90

60

7

3

120

90

8 9

1 1

60 60

90 30

10

2

90

120

ll

3

120

30

12.3

60

90

13,2

90

60

142

90

60

120 90

90 60

15 16

j1 0

172

3

Precipitation time (min)

11 2 3

90

60

18

3

60

30

19

1

120

30

20

(1

120

30

21

41

60

90

22.3

120

90

23004 242

90 30

60 60

25

(2

90

60

26

3

120

30

27

~«(\2

150

60

28

3

60

30

Results and Discussion

3.1 Response surface methodology (RSM) From the analysis, a quadratic model was suggested and thus selected as the best fit model for the COD removal while the linear model fitted the phosphate ion removal. From Fig. 1, the COD removal (%) ranged from 65.7 to 80.3% where the phosphate removal ranged from 6 to 55%. From Table 3 and Table 4, the R® values for quadratic model was 0.5879 and the adjusted R* was 0.3819. Meanwhile the R* for linear model was 0.5018 and the adjusted R? was 0.4396 which was less correlated between the observed and predicted value. This implies that the regression model does not give a sufficient explanation of the relationship between the variables and the response

[32].

Design Expert Application for Optimization of Ag/AgBr/TiO2. Table 3.

Analysis of variance (ANOVA)

for COD

97

removal.

ANOVA for Response Surface (Quadratic model) for COD removal (%) Source Sum of squares | Degree of Mean F value freedom square Model 257.51 9 28.61 2.85 A-Amount of 29.93 1 29.93 2.98 silver B-precipitation | 11.76 1 11.76 117 Cc 937 1 9.37 0.93

| P value 0.0279 0.1012 0.2931 0.3464

photoreduction

AB AC BC

13.32 23.04 23.04 143.83 15.19 3.05 Coefficient of Variance

4.25

1 1 1 1 1 1

13.32 23.04 23.04 143.83 15.19 3.05

1.33 2.30 2.30 14.34 152 0.30

R?

Adj-R?

Predicted | Adequate R? precision

0.5879

0.3819

0.2641 0.1469 0.1469 0.0013 0.2342, 0.5879

-0.1998 | 6.190

Table 4, Analysis of variance (ANOVA) for phosphate removal. ANOVA for response surface (Linear model) for phosphate removal (%) Source Sum of squares | Degree of Mean F value | P value freedom square Model 2615.15 3 871.72 | 8.06 0.0007 A-Amount of 2462.40 1 2462.40 | 22.77 < 0.0001 silver B-precipitation 26.25 1 26.25 0.24 0.6267 time

C-photoreduction

126.50

1

126.50

| 1.17

0.2902

R?

Adj-R?

| Predicted | Adequate R? precision

0.5018

0.4396

time

Coefficient of Variance

28.72

0.3239

| 10.307

Even though the R? value is not close to unity, the adequate precision ratio of the models were 6.190 and 10.307, respectively which was greater than 4 indicating that the models were significant. Both models were significant as the F-value was 2.85 where there was only a 2.79% chance that an F-value this large could occur due to noise. The p-value of the model was 0.0279 which was less than 0.05. In the phosphate removal, the F-value of the model was 8.06 which showed that only 0.07% chance that an F-value this large could occur due to noise. P-value of the model was less than 0.05 as well. The only significant model terms for quadratic model for COD removal was A?

98

A.C. Affam et al.

while the term A was the only significant one for the phosphate removal which was a linear model. The model equation for the models in terms of coded factors for the COD and phosphate removal is shown in Equations ((1) and ((2). The positive sign indicates that

the increasing factor can enhance the performance of the reaction while the negative sign indicates that the decreasing factor reverses the performance of the reaction [20]. From the equations, it can be deduced that COD and phosphate removal declined when the amount of silver increased, while the precipitation and photo-reduction time brought different impacts on COD and phosphate removal.

COD removal(%)= +77.51 — 1.12*A-0.70*B

+ 0.62*C+ 0.91*AB + 1.20*AC-

1.20 * BC - 2.37 * A? — 0.77 * B? — 0.35 2

() Phosphate removal(%) = + 36.21 — 10.13 * A + 1.05 *B-2.30*C

(2)

where A = mass of silver, B = precipitation time and C = photoreduction time Model Diagnosis Single Factor Influence The COD removal increased when the amount of silver used increased from | to 2 g and it reduced afterward. Meanwhile the phosphate removal declined when the amount of silver increased from | to 3 g. Appropriate amount of silver doped on the surface of TiO, increased the surface area of the TiO, for the photocatalytic reactions. However, when an excess amount of AgBr was doped on the surface of TiO2, clusters were formed [33]. Thus the catalyst amount should be controlled to avoid the wastage of Ag. The COD removal decreased with the increase in the precipitation time while phosphate removal increased with the elongated precipitation time. COD removal showed increasing value with elongated photoreduction time while phosphate removal showed adverse effect when the photoreduction time increased. Different factors produced different results, thus, interaction of the factors was significant as seen in Fig.

1. (a-c).

Interaction of the Factors on the Response Three-dimensional

(3D)

response

surface

with contour

plots

are

able

to reveal

the

interaction between the three parameters on COD removal. The effect of two relative input parameters on the COD removal was studied while the other parameters were kept constant. From Fig. | (a), the 3D response surface and contour representation show the interaction of Ag and precipitation time on COD removal. When 1 g of Ag dopant was used, COD removal was slightly decreased with an increased precipitation time. However, as the dosage of silver increased at the shortest precipitation time, the curved line was obtained indicating that the percentage of COD removal became lower with an increase in the amount of Ag. Fig.

1 (b) shows the interaction of silver dose and

Design Expert Application for Optimization of Ag/AgBr/TiO2

99

COD removal (mot)

COD removal (mg/L)

"5 A Amount o sve(Q)

(a)

(£00 removal (mol)

COD removal (mg/L)

(b) COD removal (mg/L)

1 precigtaton (nie)

© Fig. 1. Effect of the amount of silver, precipitation time and photoreduction time (a) photoreduction time was kept constant at 90 min; (b) precipitation time was kept constant at 60 min;

(©) Ag dose, 1 g

100

A.C. Affam et al.

photoreduction time on the COD removal. COD removal was the highest when the lowest amount of Ag was used. However, the photoreduction time did not affect the COD removal when | g of Ag dose was used. The may be due to the formation of Ag from Ago when AgBr being photo reduced under the visible light [34] and was stopped since the amount of AgBr was quite a small quantity. The highest COD removal was 78% when about 2 g of Ag was irradiated for 60 min. The 3D surface shown in Fig. 1 (c) showed that COD removal did not increase when both precipitation and photoreduction time increased. However, from the elliptical shape of 2D contour the COD removal was highest (about 77%) at 70 min precipitation time and 70 min irradiation time. This indicated that the interaction of precipitation and irradiation time was significant at shorter reaction time. As the precipitation and irradiation time increased, the contour lines were parallel. This demonstrates that no interaction could exist between the two variables. At low precipitation time (60 min) the COD removal slightly increased

as

photo-reduction

time

increased

[35].

As

the

photo-reduction

time

increased, some of the Ag+ ions on the surface of AgBr/TiO2 were converted into Ag. This was observed as grey particles derived from light yellow AgBr compound [36].

4

Conclusion

Ag/AgBr/TiO, visible light catalyst was synthetized using impregnation-precipitation photoreduction method and it was effective in COD and phosphate removal under visible light. Central composite design tool of the response surface methodology was used to optimize the parameters including the amount of Ag, precipitation time and photoreduction time. The quadratic mathematical model was suggested for COD removal whereas the linear model was suggested for phosphate removal. The optimal conditions were found to be 1 g of Ag doped on TiO with the precipitation time of 60 min and photoreduction time of 90 min. This was sufficient to remove 78.5% and 46.2% of COD and phosphate, respectively. The photo-reduction and irradiation time was significant in terms of preparation and application of the catalyst in wastewater

treatment. Acknowledgement.

The author is thankful

to the management

and authorities of University

College of Technology Sarawak, for the grant no. UCTS/RESEARCH/4/2019/09 which supported this research.

References

v

1. Bora, L.V., Mewada, R.K.: Visible/solar light active photocatalysts for organic effluent treatment: fundamentals, mechanisms and parametric review. Renew. Sustain. Energy Rev.

76, 1393-1421 (2017) . Wang, W., Serp, P., Kalck, P., Faria, J.L.: Photocatalytic degradation of phenol on MWNT and titania composite catalysts prepared by a modified sol-gel method. Appl. Catal. B Environ. 56, 305-312 (2005)

Design Expert Application for Optimization of Ag/AgBr/TiO2

‘101

. Neumann, M., Schulz, R., Schiafer, K., Miiller, W., Mannheller, W., Liess, M.: The significance of entry routes as point and non-point sources of pesticides in small streams.

Wat. Res. 198, 835-842 (2002) Yola, M.L., Atar, N., Eren, T.: Determination of amikacin in human plasma by molecular imprinted SPR nanosensor. Sens. Actuators, B Chem. 198, 70-76 (2014) . Eriksson, E., Baun, A., Mikkelsen, P.S., Ledin, A.: Risk assessment of xenobiotics in

stormwater discharged to Harrestrup A, Denmark. Desalination 215, 187-197 (2007)

. Bicker, W., Lammerhofer,

M., Genser, D., Kiss, H., Lindner, W.: A case study of acute

human chlorpyrifos poisoning: novel aspects on metabolism and toxicokinetics derived from liquid chromatography-tandem mass spectrometry analysis of urine samples. Toxicol. Lett.

159,

235-251 (2005)

. Ejaz, M., Afzal, M.B.S., Shabbir, G., Serrao, J.E., Shad, S.A., Muhammad,

W.: Laboratory

selection of chlorpyrifos resistance in an invasive pest, phenacoccus solenopsis (homoptera: pseudococcidae): cross-resistance, stability and fitness cost. Pestic. Biochem. Physiol. 137, 8-14 (2017)

. Ngan, C.K., Cheah, U.B., Abdullah, W.Y.W., Lim, K.P., Ismail, B.S.: Fate of chlorothalonil,

chlorpyrifos and profenofos in a vegetable farm in cameron highlands, Malaysia. Water Air Soil Pollut. Focus 5, 125-136 (2005)

. Nawaz, K., Hussain, K., Choudary,

N., Majeed, A., Ilyas, U., Ghani, A., Lin, F., Ali, K.,

Afghan, S., Raza, G., Lashari, M.L.: Eco-friendly role of biodegradation against agricultural pesticides hazards. Afr. J. Microbiol. Res. 5, 177-183 (2011)

. Velmurugan,

R., Swaminathan,

M.:

An

efficient nanostructured

ZnO

for dye

sensitized

degradation of reactive Red 120 dye under solar light. Sol. Energy Mater. Sol. Cells 95, 942— 950 (2011)

. Saravanan,

R., Gupta, V.K., Narayanan, V., Stephen, A.: Comparative study on photocat-

alytic activity of ZnO prepared by different methods. J. Mol. Lig. 181, 133-141 (2013)

. Saravanan, R., Gupta, V.K., Mosquera, E., Gracia,

reparation and characterization of

V20s/ZnO nanocomposite system for photocatalytic application. J. Mol. Lig. 198, 409-412 (2014)

. Fujishima, A., Honda, K.: Electrochemical photolysis of water at a semiconductor electrode. Nature 238, 37-38 (1972) .: Comparing graphene, Pan, X., Zhao, Y., Liu, S., Korzeniewski, C.L., Wang, S., Fan, TiO, nanowire and graphene-TiO2nanoparticle composite photocatalysts. ACS Appl. Mater. Interfaces. 4, 3944-3950 (2012) . Emeline, A.V., Zhang, X., Jin, M., Murakami, T., Fujishima, A.: Application of a “black body” like reactor for measurements of quantum yields of photochemical reactions in heterogeneous systems. J. Phys. Chem. B. 110, 7409-7413 (2006) . Sunada, K., Watanabe, T., Hashimoto, K.: Bactericidal activity of copper-deposited TiO2 thin film under weak UV light illumination. Environ. Sci. Technol. 37, 4785-4789 (2003) . Elahifard, M.R., Rahimnejad, S., Haghighi, S., Gholami, M.R.: Apatite-coated

Ag/AgBr/TiO, visible-light photocatalyst for destruction of bacteria. J. Am. Chem. Soc. 129, 9552-9553 (2007)

. Wang, Y., Sun, L., Fugetsu, B.: Morphology-controlled synthesis of sunlight-driven plasmonic photocatalysts Ag@AgX Chem. A. 1, 12536-12544 (2013)

(X = Cl, Br) with graphene oxide template. J. Mater.

. Fan, H., Zhu, J., Sun, J., Zhang, S., Ai, S.: Ag/AgBr/Co-Ni-NO; layered double hydroxide nanocomposites with highly adsorptive and photocatalytic properties. Chem.-A Eur. J. 19, 2523-2530 (2013)

102 20.

A.C. Affam et al.

Salman, J.M.: Optimization of preparation conditions for activated carbon from palm oil fronds using response surface methodology on removal of pesticides from aqueous solution. Arab. J. Chem. 7, 101-108 (2014) . Behbahani, M., Moghaddam, M.R.A., Arami, M.: Techno-economical evaluation of fluoride removal by electrocoagulation proce: : optimization through response surface methodology.

Desalination 271, 209-218 (2011)

. Khedmati, M., Khodaii, A., Haghshenas, H.F.: A study on moisture susceptibility of stone matrix warm mix asphalt. Constr. Build. Mater. 144, 42-49 (2017) . Haghshenas, H.F., Khodaii, A., Khedmati, M., Tapkin, S.: A mathematical

model

for

predicting stripping potential of hot mix asphalt. Constr. Build. Mater. 75, 488-495 (2015)

. Bagaoui, A., Yaacoubi, A., Dahbi, A., Bennouna, C. , Luu, R.P.T., Maldonado-Hodar,

F.J..

Rivera-Utrilla, J., Moreno-Castilla, C.: Optimization of conditions for the preparation of activated carbons from olive-waste cakes. Carbon N. Y. 39, 425-432 (2001)

. Grace, F.A., Sarafadeen, A.O., Olanrewaju, A.A., Aderonke, O., Akeem, O.0., Akanmu, A. S.: Department of science laboratory technology, osun state polytechnic, iree. Beni-Suef

Univ. J. Basic Appl., Sci (2018)

. McEvoy, J.G., Zhang, 2

Synthesis and characterization of Ag/AgBr-activated carbon composites for visible light induced photocatalytic detoxification and disinfection. J. Photochem. Photobiol. A Chem. 321, 161-170 (2016)

. Wang, D., Duan, Y., Luo, Q., Li, X., An, J., Bao, L., Shi, L.: Novel preparation method for a new visible light photocatalyst: mesoporous TiO2 supported Ag/AgBr. J. Mater. Chem. 22, 4847-4854 (2012) . Liu, X., Zhang, D., Guo, B., Qu, Y., Tian, G., Yue, H., Feng, S.: Recyclable and visible light sensitive Ag-AgBr/TiO: Surface adsorption and photodegradation of MO. Appl. Surf. Sci. 353, 913-923 (2015) . Rong, X., Qiu, F., Zhang, C., Fu, L., Wang, Y., Yang, D.: Preparation of Ag-AgBr/TiO2graphene 30.

and its visible light photocatalytic

activity enhancement

for the degradation of

polyacrylamide. J. Alloys Compd. 639, 153-161 (2015) Sui, Y., Su, C., Yang, X., Hu, J., Lin, X.: Ag-AgBr nanoparticles loaded on TiO2 nanofibers as an efficient heterostructured photocatalyst driven by visible light. J. Mol. Catal. A: Chem.

410, 226-234 (2015) . Carp, O., Huisman, C.L., Reller, A.: Photoinduced reactivity of titanium dioxide. Prog. Solid

State Chem. 32, 33-177 (2004) . Dutta, S., Ghosh, A., Moi, S.C., Saha, R.: Application of response surface methodology for optimization of reactive azo dye degradation process by fenton’ s oxidation. Int. J. Environ.

Sci. Dev. 6, 818-823 (2015) 33. Zhang, X., Lei, L.: Effect of preparation methods on the structure and catalytic performance of TiO/AC photocatalysts. J. Hazard. Mater. 153, 827-833 (2008) 34. Zhang, L., Shi, Y., Wang, L., Hu, C.: AgBr-wrapped Ag chelated on nitrogen-doped reduced

35.

36.

graphene oxide for water purification under visible light. Appl. Catal. B Environ. 220, 118125 (2018) Sohrabi, S., Akhlaghian, F.: Modeling and optimization of phenol degradation over copperdoped titanium dioxide photocatalyst using response surface methodology. Process Saf. Environ. Prot. 99, 120-128 (2016) Shi, L., Liang, L., Ma, J., Meng, Y., Zhong, S., Wang, F., Sun, J.: Highly efficient visible light-driven Ag/AgBr/ZnO composite photocatalyst for degrading Rhodamine B. Ceram. Int. 40, 3495-3502 (2014)

®

Check updatesfor

Hydrodynamics of Flow over Axonopus Compressus

(Cow

Grass) as a Flexible

Vegetation Muhammad Mujahid Muhammad", Khamaruzaman Wan Yusof?,

Muhammad Raza Ul Mustafa”, Aminuddin Ab. Ghani’,

Abdurrasheed Said Abdurrasheed?, Abdulkadir Taofeeq Sholagberu*,

Abdullahi Sule Argungu', and Umar Alfa Abubakar! ' Water Resources and Environmental Engineering, Ahmadu Bello University,

Zaria, Nigeria {mmmujahid, uaabubakar}@abu. edu. ng, abdummul@gmail. com > Universiti Teknologi Petronas, Seri Iskandar, Perak, Malaysia {khamaruzaman. yusof, raza. mustafa, abdurrashee_16000331}@utp. edu. my > Universiti Sains Malaysia, Nibong Tebal, Penang, Malaysia redacO2@usm. my + Water Resources and Environmental Engineering, University of Iorin,

Ilorin, Nigeria abdulkadir. ts@unilorin. edu. ng

Abstract.

In this experimental

work,

laboratory channel

was constructed

to

evaluate the hydrodynamics of flow through cow grass. Acoustic Doppler Velocimeter (ADV) was used in determining the magnitude of velocity across various segments of the channel due to flow vegetation— interaction. The effect of vegetative roughness on flow velocity, Manning’s resistance, and velocity

profiles were investigated. Results revealed the changes in the normal parabolic velocity profile to nearly flat in the streamwise direction. It was also observed

that the Manning’s coefficient generally decreases with rise in velocity of flow, and it increases with increase in the fractional flow depths, as the grass was

completely submerged. Thus, the relationships and n-V curve obtained could be use in proper design of grassed channels.

Keywords: Hydrodynamics - Cow grass - Vegetative roughness - Velocity profile - Streamwise 1

Introduction

Traditionally, vegetation were being eliminated in waterways to improve conveyance capacity without paying attention for its potential ecological benefits in improving the water quality received by stormwater drainage systems [1]. Thus, studying the hydrodynamics of vegetation covers in waterways such as grassed channels will reduce the cost of concrete lining which does not allow natural infiltration of rain. Also, grassed channels protect the soil from scouring forces and enhancing vegetative

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 103-110, 2021. https://doi.org/10.1007/978-981-33-6311-3_12

104

M. M. Muhammad et al.

growth, channels,

which can and

withstand

higher hydraulic

this will lead to reduction

forces

in soil loss

on slopes, stream [2].

In addition,

banks

using

and

natural

vegetation improves water quality by removing particulate matter through the action of the vegetative roughness, and results in enhancing the aesthetics of a catchment [2]. For instance, vegetation was applied in the removal of Total Suspended Solid (TSS) using Lepironia Articulata as emergent vegetation, which was served as a control of pollution into water body [3]. Nowadays, huge land areas become active construction sites by exposing earthen materials as a result of rapid infrastructural developments. This activity usually leads to stormwater pollution during the occurrence of rainfall, which enters into waterways

and reservoirs

to cause sedimentation

[4]. Thus,

this has great

impact on water quality and environmental ecology at a large. To mitigate these problems, vegetation can be applied on landscaping and urban planning, in which the vegetation acts as a filter strip and to as well alter the impact of surface runoff respectively. In the same vein, to understand the hydrodynamics of vegetation, artificial vegetation were mostly used, in this way, numerous models were developed to predict the discharge, but the models were found to overestimate the discharge when applied to a natural vegetated channel or constructed ones such as grass swales. Thus, the major objective of the present study is to employ natural grass to evaluate its flow characteristics in terms of flow resistance, velocity profile, drag coefficient and other flow parameters. Through this, an eco-friendly grassed channels that will be more erosion resistant can be properly designed.

2

Theoretical Analysis

The manifestation of flow resistance in open channel could be viewed on the actions of dynamic forces over the roughness of wetted perimeter. For vegetated open channels, vegetative flow resistance is being affected by density of vegetation, grass geometry and nature of vegetation [5]. The flow resistance is usually evaluated using Darcy— Weisbach

constant

(f), Chezy’s

factor

(C)

or Manning’s

coefficient

(n).

However,

Manning’s n is the commonly applied in open channel analysis and design [6, 7]. The n — value can be estimated using Manning’s Eg. (1) as follows:

1 Vs ERPS y1/2

(1)

It was found that vegetative roughness due to Manning’s n changes progressively with VR, where V is the average velocity and R equals the hydraulic radius (VR), for not being affected by channel slope and geometry. Further investigations revealed that varieties of vegetation having the same properties in terms of distribution, density, stiffness, grass heights and degree of submergence display same pattern of n-VR relationships [7]. Sequel to this, it was observed that the n-VR curve is alike with the n-

Re curve. This happens as the expression VR is directly related to Reynolds number Re, which is given according

to Chow

[8], as in Eq. (2):

Hydrodynamics of Flow over Axonopus Compressus (Cow Grass)

ai VR

105

Q)

Where, # is the kinematic viscosity of water. Therefore, n-value decreases with increase in magnitude of VR, as the vegetative bending and submergence increase. The Lowest n — value is produced, when the force of flow due to VR is adequate enough to cause the grasses to become flattened. But, VR decreases with decrease in the friction constant, f, when flow fluctuates between laminar and transitional, that is, when Re < 10* [9, 10]. Conversely, in very high flood levels, that is, when the depth of flow is much larger than the height of grass, which diminishes the magnitude of boundary region created by vegetative development and hence rendered the vegetative roughness to be negligible. Accordingly, the resistance coefficient remains

a constant

small-scale waterways

value

[5]. However,

for flow-vegetation

interaction

in

and floodplains in large rivers, caution need to be taken in using

a constant n — value, as it can result to incorrectness in their predictable flood level [11].

3 3.1.

Methodology Construction of Experimental Channel

Fig. 1.

0a

-Lasem | mos =s9m—|.

Lesa

Lasep

sup}

=o} ssa},

=stmp

=su

2

Laboratory studies was performed using Axonopus Compressus also known as cow grass. The cow grass was planted on a bed slope of 1:500 using a manmade concrete channel of dimension 12 m x 1.5 m x | m in the Physical Modelling Laboratory of River Engineering and Urban Drainage Research Center (REDAC), Universiti Sains Malaysia (USM) Engineering Campus. The channel slope were determined through surveying the channel bed which conforms to the existing slopes of ecological grassed swales of USM. The concrete channel has an overall length of 16 m comprising of three compartments as inlet, test section and outlet. The Plan view of the test section is shown in Fig. 1, where the Cow grass was planted across 10 m length, labelled CD (vegetated zone). The grass has an average height of 70 mm.

Plan view of test channel adapted from Muhammad et al. [12]

106

M. M. Muhammad et al.

3.2.

Experimental Procedure

After preparing the channels as Muhammad er al. [12] which experiment begins by selecting y = 0.25 m, with aspect ratios investigate the hydrodynamics Velocimeter

(ADV)

was

shown deals three of B/y of the

earlier in Figs. 1, and following the approach of with predicting vegetative flow resistance. The (3) flow depths of y = 0.15 m, y = 0.20 m, and = 10, B/y = 7.5, and B/y = 0.6 respectively, to grass in a submerged state. Acoustic Doppler

used to measure

the

velocity

of flow

along seven (7) cross

sections with distances of 3.0 m, 4.5 m, 5.5 m, 6.0 m, 6.5 m, 8.5 m, and 11.5 m away from the inlet, labelled from CH3.0 m up to CHI11.5 m as Fig. 1, respectively. The ADV measured 3D point velocities. Each cross section was divided into five locations, at every location 8 point velocities were determined with respect to the flow depths as 0.2y, 0.25y, 0.3y, 0.4y, 0.5y, 0.6y, 0.7y and 0.8y, respectively, as illustrated by Fig. 2. For the purpose of this study, the streamwise and vertical velocity profiles were considered in the analysis, as obtained using the ADV.

Water surface

Distance from the left wall (m) Fig. 2.

4 4.1

Measurement points in vertical depths

Results and Discussion Velocity Profiles

The velocity profiles along each cross section in the grassed channel was investigated in terms of the streamwise and vertical velocity profiles. Tables 1, 2 and 3 show streamwise velocity of the different flow depths, using these tables the streamwise velocity profiles were plotted as in Fig. 3. The trends show that when the flow depth was 15 cm higher velocities were observed, followed by the depth of 20 cm and 25 cm depths having the lowest velocity values. Also, maximum stream-wise velocity was observed at the center line of the channel cross section. From Table 4, the centerline vertical velocity profile along the channel length reach was drawn as in Fig. 4. Also, the same thing happens with the flow depth of 0.15 m having larger velocity range compared to the rest of the depths, which are in line with the studies of Muhammad et al. [13] on velocity distributions grassed channel, although they utilize a very mild slope of 1:1000 in their studies, that resulted in very low velocity range compared to the present study.

Hydrodynamics of Flow over Axonopus Compressus (Cow Grass) Table 1.

Distance from Left bank (m) 0.5 03 06 09 12

Mean streamwise (Vxm) velocity for flow depth of 0.15 m

CH3.0__CHiSm_CHSSm_CH60m

CH6Sm__CH8Sm_CHITSm

Vxm

Vxm

Vxm

(m/s) 029 028 043 032 036

Vxm

bank (m)

Vxm

Vxm

‘Average (wis) 027 034 052 050 027 Velocity

Mean streamwise (Vxm) velocity for flow depth of 0.2 m

CHB.0

from Left

Vxm

___(mu/s)_(amis)_(amis)_—_(amis)_—__(amis)_(ans) 0.18 021 024 0.28 049 (08 032 0.16 oa 0.46 033 052 on 0.16 0.43 035 076 = 077 0.40 052 037 082 045 0.40 0.13 049 025 036 Os 03

Table 2.

Distance

Vixm

(ows)

CHiSm

CHS.5m

CH6.0m

—CH6.5m_—

CH8.5m_—CHITSm

Vim

(vs)

Vim

(mms)

Vxm

Vxm

Vxm

(ms)

ms)

‘Average

Vxm

ms)

Velocity

s)

0.15

0.08

0.15

0.14

0.16

0.13

0.15

0.13

0.14

03

0.07

0.15

0.12

0.14

0.16

0.17

0.13

0.13

06

0.08

0.13

0.13

O14

0.18

0.16

0.13

0.14

09

0.08

0.13

Ol

O14

0.16

0.15

0.13

0.13

12

0.08

014

0.12

0.15

0.18

0.18

0.13

0.14

Table 3.

Distance fromLeft bank (m)

0.5 03 06 09 12

Mean streamwise (Vxm) velocity for flow depth of 0.25 m

CH3.0__CHiSm__CHSSm_CH60m_

CH6Sm__CHS3m_CHILSm

Vam =

Vxm_—

Vm

Vom

Vxm

‘Average Velocity

ais)

(as)

(ais)

(a's)

(as)

0.05 0.16 0.10 0.08 0.10

006 =O. 009 0.06 010 0.07 0.06 0.08 0.06 __ 0.05

5 3 3 3

(m/s)

(m/s)

(m/s)

0.04 0.09 0.19 0.17 0.05

0.06 0.16 0.06 050 02

025

+ *

z «

Vxm——

0.08 0.06 016 007 020

0.125

g>

Vam 9

0.08 0.12 0.04 0.06 0.06

1.00

os

_—

1

0.06 ot 0.10 os ol

2

Brose91 = ve

i

Fran rose Ree h 0.06

0.03 LATERAL DISTANCE FROM THE LEFT BANK

Fig. 3. Average streamwise velocity profiles

*y=0.15m my=0.20m ay=0.25m

107

108

M. M. Muhammad et al.

Table 4, Average centerline vertical velocity (V.) profiles along the channel length Measurement

y=15em

Location from the Inlet of Channel CH3.0m

y=20cm

y=25em

ol Distance (m) 3.00

‘Ve (m/s) 0.65

Ve (m/s) 0.22

Ve (m/s) 0.32

CH4.5 m

4.50

1.17

0.31

0.37

CHS.5m CH6.0m CH6.5 m CHS8.5 m CH11.5m

5.50 6.00 6.50 8.50 11.50

0.37 0.71 0.86 1.52 117

0.28 0.20 0.24 0.37 0.35

0.28 0.09 0.19 0.15 0.14

4.00

Average Centerline Velocity (m/s)

2.00 1.00

1.00

2.00

. otitis 490

é

:

50.7637 © : 92808

16.00

0.25

ons

reabene 6

0.08

Longitudinal Distance fromthe inlet

Fig. 4. Centerline velocity profiles along the channel length 4.2

Variation of Resistance Coefficient with Flow Depth and Velocity

From Table 5, it shows that all the flow depths have almost the same pattem of velocity profile. However, flow depth of 0.15 m has wider velocity range of 0.32 up to 1.67 m/s compared to the other depths. The flow depth of 0.20 m has velocity range of 0.22 to 0.32 m/s, while that of 0.25 m has velocity range of 0.06 to 0.36 m/s which is the lowest. While for the case of Manning’s, n decreases with increase in flow velocity, which

agrees

with

the results obtained

by

Ahmad

et al.

[14]

and Chen

et al.

[15].

However, the velocity profile when y = 0.20 m differs from the rest, this occurs because at this depth the aspect ratio B/y = 7.5 which create maximum velocity close to the surface of water and diminished the secondary current which was similar to the

Hydrodynamics of Flow over Axonopus Compressus (Cow Grass)

109

findings of Afzalimehr ef al. [16]. Moreover, the Manning’s n generally increases above the grass height as the fractional flow depth increases, as a result of complete submergence of the gr

Table 5. Flow depth, vertical velocity profiles (V) and Manning’s (n) yas om Points

O2y 025y O3y O4y sy

5

~~ h(m)

V (m/s)

yom 2

him)

V (mis)

yasom a

him)

V (mis)

a

0.03 167 0.001 0.04» 0.28 0.004.005 0.190.008 0.0375 (1.06 0.001.905) 032 0.005006) 0.250.007 0.045 0.670.002 0.06 = 0.290.006.0835 (0.006 0.06 164 0.001 9080.22, 0.01000 0.13002 0.075 091 = 0.002so10 0.290.010.8006 (0.053 009 0320.08. ga 0.29 NLS 030 0.013 0.105 0.590.005 og 0.260.014.0004 012048 ——0.007_—oo.ts 0.28 0.01S-— 020036 0.014

Conclusion

The flow resistance, n, due to submerged grass has greatly being affected by flow depth and velocity. The value of n show an inverse relationship with the flow velocity, while a proportional relation was generally observed with the flow depth. The velocity profiles of the various flow depths displayed similar trends especially for the streamwise and along the channel length. Among these flow depths, when y = 0.15 m has the greater velocity range than the remaining depths, with the 0.25 m having the least. However, the velocity profile and n-V curve when y = 0.20 m deviate from others, as it weaken the secondary current, which makes flow velocity to be fairly constant with tendency of producinng maximun velocity near the water surface.

Acknowledgements. The authors acknowledge Universiti Teknologi PETRONAS, Malaysia, River Engineering and Urban Drainage Research Centre (REDAC), Universiti Sains Malaysia for providing the laboratory facilities. The financial support of the Ministry of Education under HiCOE’s niche area Sustainable Urban Stormwater Management has been well appreciated (Grant No.311,PREDAC.4403901). The first author ackowledge Ahmadu Bello University, Nigeria, for granting his study leave fellowship.

110

M. M. Muhammad et al.

References

v

. Buckman, L.: Hydrodynamics of partially vegetated channels: stem drag forces and application to an in-stream wetland concept for tropical, urban drainage systems. Delft (MSc thesis report) (2013) . Shahkolahi, A., Tadman, A., Crase, J.: An economical solution for flood protection and

channel erosion control, In: Institute of Public Works Engineering Australiasia (IPWEA) Queensland Conference. Australiasia Queensland (2014) . Yahyapour, S., Golshan, A., Ghazali, A-H.B.: Removal of total suspended solids and turbidity

within experimental

vegetated

channel:

optimization

through

response

surface

methodology. J. Hydro-Environ. Res. (2013)

. Yahyapour, S., Golshan, A., Ghazali, AH] turbidity

within experimental

vegetated

channel:

Removal of total suspended solids and optimization through response surface

methodology. J. Hydro-Environ. Res. 8(3), 260-269 (2014)

. Temple, D.M., Robinson, K.M., Ahring, R.M., Davis, A.G.: Stability design of grass-lined open channels. In: Agricultural Handbook, USDA, Editor. Washington, D.C. (1987) . Wu, F.-C., Shen, H.W., Chou, Y.-C.: Variation of roughness coefficients for unsubmerged

and submerged vegetation, J. Hydraul. Eng. 125(9), 934-942 (1999) . Kirby, J.7., Durrans, S. . Pitt, R., Johnson, P.D.: Hydraulic resistance in grass swales designed for small flow conveyance. J. Hydraul. Eng. 131(1), 65-68 (2005) . Chow, V.' : Open Channel Hydraulics. McGraw-Hill, New York (1959) . Chen, C.L.: Flow resistance in broad shallow grassed channels. J. Hydraul. Division 102(3),

307-322 (1976) . Green, J.C.: Modelling flowresistance in vegetated streams: review and development of newtheory, Hydrol. Process. 19, 1245-1259 (2005) . Rhee, D.S., Woo, . Kwon, B.A., Abn, H.K.: Hydraulic resistance of some selected vegetation in open channel flows. River Res. Appl. 24(5), 673-687 (2008) . Muhammad,

M.M.,

Yusof,

K.W.,

. Muhammad,

M.M.,

Khamruzzaman,

Mustafa,

M.R.U.,

Zakaria,

N.A.,

Ab.

Ghani,

A.:

Prediction models for flow resistance in flexible vegetated channels. Int. J. River Basin Manage. 1-26 (2018) W.Y.,

Muhammad,

R.U.M.,

Aminuddin,

A.G.:

Velocity distributions in grassed channel. In: 4th Annual International Conference on Architecture and Civil Engineering (ACE 2016). Singapore, GSTF (2016)

. Ahmad, N.A., Ab. Ghani, A., Zakaria, N.A.: Hydraulic characteristic for flow in swales. In: 3rd International Conference on Managing Rivers in the 21st Century: Sustainable Solutions

for Global Crisis of Flooding, Pollution and Water Scarcity. Rivers, Penang, Malaysia, pp. 183 - 189 (2011) . Chen,

Y.-C.

Kao,

S.-P.,

Lin,

J.-Y.,

Yang,

H.-C.:

Retardance

coefficient

of vegetated

channels estimated by the Froude number. Ecol. Eng. 35(7), 1027-1035 (2009)

. Afzalimehr, H., Sui, J., Mofhbel, R.: Hydraulic parameters in channels with wall vegetation and gravel bed. Int. J. Sediment Res. 25, 81-90 (2010)

®

Check updatesfor

Greenhouse Gas Emission from Domestic Wastewater Treatment and Discharge in East Java Province — Indonesia Yatnanta Padma Devia!® and Dian Tristi Agustini? ' Department of Civil Engineering, Brawijaya University, Malang, Indonesia [email protected].

id

? Environmental Agency East Java Province, Surabaya, Indonesia dian. [email protected]

Abstract. The sewer system for collecting municipal wastewater, wastewater treatment and the discharge of wastewater treatment plants (WWTP) contribute to the emissions of greenhouse gases (GHG) in direct emissions. The potential

of global warming of GHGs emission were represented by carbon dioxide (CO,) and expressed as equivalent carbon dioxide (CO -eq). The GHG emissions in cities of developing countries, such as Indonesia, were predicted increase in the future. However, developing countries still have problems in quantifying of

GHG emission. The aim of this paper is estimating GHG emissions by calculating methane (CH4) and nitrous oxide (NO) from domestics wastewater treatment and discharge in East Java Province, Indonesia. The result show that

from domestic wastewater in East Java Province contribute GHG emissions 3,417,655 ton CO -eq in 2017. 3,625,692 ton CO -eq in 2026.

This

emission

will

increase

6.08%

up

to

Keywords: Carbon dioxide - Domestic wastewater « Greenhouse gases Methane - Nitrous-oxide

1

Introduction

The sewer system for collecting municipal wastewater, wastewater treatment and the discharge of wastewater treatment plants contribute to the direct emission of greenhouse

gases (GHG)

[1]. Indirect GHG

emissions

are produced

at WWTPs

mainly

by

electricity supply, transportation, use of chemicals and additive, and disposal/reuse of residual. The potential of global warming of GHGs emission were represented by carbon dioxide (CO) and expressed as equivalent carbon dioxide (CO -eq) [1]. According to Intergovernmental Panel on Climate Change (IPCC), CO? emission from

WWTP is not considered due to its biogenic source that attracted down from the atmosphere and turned back it to the atmosphere. Hence, CO2 emissions from WWTP are not calculated in total emissions [2]. Other GHG, such as methane (CH4) and nitrous oxide (NO), also emitted from WWTP. Anaerobic conditions in sewage and sludge treatment works are emission source of CH4 [3] which has GHG effect

approximately 28 times higher than CO, [3, 4]. Nitrification and denitrification as biological nitrogen removal from wastewater, are forming N20 [5] that has GHG effect

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 111-118, 2021. https://doi.org/10.1007/978-981-33-6311-3_13

112

Y. P. Devia

and

D. T. Agustini

265 times higher than CO, [3, 4]. Unfortunately, the comprehension of factors that caused N30 in nitrification and denitrification are extinct [6] due to a seasonal dynamic

[5]. from

The Indonesia government has ratified the Kyoto Protocol, committed to increasing 26% to 29% reduction in GHG emissions by applying Intended Nationally

Determined Contribution

(INDC). To

achieve

this target, emission reduction

through

many sectors must be attained. The GHG emissions in cities of developing countries, such as Indonesia were predicted increase in the future [7]. However, developing countries still have problems in quantifying of GHG. The lack of standardized methods and the complexity of the GHG emission quantification in full scale WWTPs make difficulties and differences of quantification of GHG [3]. Most developing countries, such as Indonesia, do not have measurement of local emission factors from WWTP due to lack of operation data/information and environmental situation [8]. The

aim of this

paper is quantifying GHG emission from domestic wastewater treatment in East Java Province Indonesia, that have linked to the in situ production CH, and N,O. East Java Province has several industrial areas such as in Pasuruan Industrial Estate Rembang (PIER),

Surabaya

Industrial

Estate

Rungkut

(PIER)

[9],

Ngoro

Industrial

Park,

Mojokerto Industrial Park, and in other cities, that equipped with off site WWTP. Otherwise, the off site treatment for domestic wastewater in East Java Province is still very rare. This study should be the preliminary research in the completed quantification GHG er ion. It also aims to identify the drawback and problem for quantifying GHG emission in developing country, in particular East Java Province Indonesia.

2

Methods

Intergovernmental Panel on Climate Change (IPCC) stated in its guidelines that wastewater treatment and discharge contribute direct GHG emissions in the atmosphere. CH, and N,O are inserted in direct GHG emission by calculating their production during wastewater treatment and at sewers. The important factor in determining the CH4 outcome of wastewater is the quantity of degradable organic material in the wastewater named Biochemical Oxygen Demand (BOD) or Chemical Oxygen Demand (COD)

[2]. A higher amount of BOD

lower amount of BOD emission

WWTP 2.1

factor when

or COD information

in national GHG

or COD

will commonly

in wastewater.

produce more CH,

IPCC recommends

is limited to estimate

CH4

and NO

than

applying default emission from

inventory.

Methane Emission

Since Indonesia is a developing country, the method for CH, emission is using Tier 1 which applied default values for the emission factor and activity parameter [2]. Tier 1 is contemplated as a right calculation for countries with insufficient data. First step of this method is to estimate total organically degradable carbon in wastewater (TOW) as Eq. | follow:

Greenhouse Gas Emission from Domestic Wastewater Treatment

113

TOW = P.BOD .(0.001). 1.365

(1)

TOW = total organics in wastewater in inventory year (kg BOD/year) P = country population in inventory year (person) BOD = country-specific per capita BOD in inventory year (g/person/day) 0.001 = conversion from grams BOD to kg BOD I = correction factor for additional industrial BOD discharge into sewers The BOD default values for Asia, Middle East and Latin America ranged in 40 g/person/day. The correction factor for collected and uncollected default is 1.25 and 1.00, respectively

[2].

Equation 2 shows calculation of emission factor that represent a function of the maximum

CH,

producing potential (B,) and the methane correction factor (MSF)

for

the wastewater treatment and discharge system. The CH4 maximum can be produced from

a given

amount

of organics

(BOD

or COD),

that expressed

in By. The

MCF

represent the extent to which the CH, producing capacity (Bo) is realised in each type of treatment dan discharge pathway and system, especially in anaerobic system.

EF; = Bo. MCF; EF; = emission factor (kg CH4/kg

(2)

BOD)

j = each treatment/discharge pathway or system B, = maximum CH, producing capacity (kg CH4/kg MCF; = methane correction factor (fraction)

BOD)

For By, 0.6 kg CH,/kg BOD as a default value can be used if country-specific data are not available. The default MCF

values for domestic water are shown in Table

| [2].

To estimate CH, emission from domestic wastewater is as Eq. 3 follows:

Table 1. Calculation of methane (CH,) emission factor for domestic wastewater in East Java Province [2]. Type of treatment or

Maximum methane

Methane correction

Emission

producing capacity (Bo)

factor for each

factor

(kg CH4/kgBOD)

treatment system

(EF)

(MCF)

(kg CHa/ kgBOD)

0.6

0.1

0.06

| 0.6

0.0

0.00

Septic system

0.6

0.5

0.30

Latrine (dry climate, ground water table

0.6

07

0.42

discharge

Untreated system Sea, river and lake

discharge Flowing sewer (open or closed) Treated system

higher than latrine)

114

Y. P. Devia

and

D. T. Agustini

CHyemissions

=

i

(U;. T,j. EF))| (TOW — S)-R

(3)

CH, emission = CH, emission in inventory year (kg CH,/year) S = organic component removed as sludge in inventory year = fraction of population in income group i in inventory year j = degree of utilisation of treatment/discharge pathway or system, j, for each income group fraction i in inventory year i = income group: rural, urban high income and urban low income each treatment/discharge pathway or system R = amount of CH, recovered in inventory year (kg CH,/year) The fraction of population (U;) of rural, urban high income, and urban low income

are 0.54, 0.12 and 0.34, respectively [2]. The degree of utilisation of treatment/discharge pathway or system, j, for each income group fraction i in inventory year are shown in Table 2.

Table 2.

Estimation of CH, emissions from domestic wastewater

Income group with type of treatment

| Degree of

Emission

Net methane emissions

or discharge pathway utilization factor Rural with fraction of population income group 0.54

(kg CHy/year)

Septic tank

53,902,727.68

0.58

0.30

Latrine

0.16

0.42

Other

0.05

0.06

Sewer

0.00

0.00

0.00

None

0.21

0.00

0.00

20,817,605.17 929,357.00

Urban high income with fraction of population income group 0.12 0.58

0.30

11,978,383.93

Latrine

Septic tank

0.16

0.42

4,626,134.48

Other

0.05

0.06

Sewer

0.00

0.00

0.00

None

0.21

0.00

0.00

206,523.86

Urban low income with fraction of population income group 0.34 Septic tank Latrine

0.58 0.16

0.30 0.42

Other

0.05

0.06

585,150.94

Sewer

0.00

0.00

0.00

None

0.21

0.00

0.00

Total (kg CH/year) Total (GgCH,/year)

33,938,754.98 13,107,381.03

140,092,018.93 140.09

Greenhouse Gas Emission from Domestic Wastewater Treatment 2.2

115

Nitrous Oxide Emission

For estimating NO emission, data of nitrogen concentration of wastewater discharge, the amount of human population and average annual per person protein consumption are needed. The equation for estimating N2O emission are presented in Eq. 4 and Eq. 5 below:

Neffluent = (P . Protein . Fypr- Fnon—con- Find-com) —

Ntudge

(4)

Nefuent = total annual amount of nitrogen in the wastewater effluent (kg N/year) P = human population Protein = annual per capita protein consumption (kg/person/year) Fypr = fraction of nitrogen in protein, default = 0.16 kg N/kg protein Fron-con = factor for non-consumed protein added to the wastewater Find-com = factor for industrial and commercial co-discharged protein into sewer Natuage = nitrogen removed with sludge (default = zero) (kg N/year)

According to Central Bureau of Statistic of East Java Province, 2017, annual per person protein generation in East Java is 61.93 g/person/day or 22.6 kg/person/year [10]. The default data for factor for non-consumed protein added to the wastewater and

factor for industrial and commercial co-discharged protein into the sewer system are 1.1 and

1.25, respectively

[2].

N2Oemissions = Neftuent- EFertuent-44/28 N2Oemissions = N2O

emissions

in inventory

(5)

year (kg N>O/year)

EF .jiuent = emission factor for N50 emissions from discharged to wastewater (kg N,O-N/kg N) 44/28 = the conversion of kg NO-N into kg N,0. Due to insufficient field data and a specific estimation regarding the process of nitrification and denitrification in rivers and in estuaries, the default IPCC emission factor for NxO emissions from domestic wastewater nitrogen effluent is 0.005 kg NO-

N/kg N is used.

2.3.

GHG Emission

The GHG emissions included CH alents (CO -eq) by global warming for CH4 and 310 for N30. The total sum of CO>-eq conversion of CH4

and N30 are converted into carbondioxide equivpotentials (GWPs) over 100 years that 1 for CO 21 direct GHG emission was calculated by arithmetic and NjO emissions.

116

3 3.1

Y. P. Devia

and

D. T. Agustini

Result and Discussion Methane Emissions

To calculate TOW in wastewater, the population should be known. The population of East Java Province that consist of 38 cities/regencies are 39,292,972 persons [11]. By using that amount of population, 40 g/person/day for BOD default values and 1.00 for correction factor for uncollected sewers, the TOW result is 573,677,391 kg BOD/year. The production of CH, depends on the degradable organic component, the temperature, and the treatment type. The rate of CH, production increases with increases in temperature. The East Java Province - Indonesia have a tropical climate with the temperature above 15 °C that CH, is likely to produce. Based on Eq. 2, the calculation result of CH,

emission factor for domestic

wastewater is shown

in Table

1.

Using Eq. 3, the estimation of CH emissions from domestic wastewater are presented in Table 2. The data of sludge separation and recovered CH, are not available. The degree of utilization are considered by expert judgement. Total of CH, emissions

from domestic

wastewater in East Java in 2017

is 140.09

Gg CH,/year. The same method and data set can be used for estimating CH, emissions from domestic wastewater in each year. The prediction of CH, emissions from domestic wastewater in East Java in 2026 is 148.62 Gg CH,/year. If the wastewater treatment types changes and sludge removal and CH, recovery records over the time periods, the changes should be included.

Table 3.

Estimation of nitrogen in effluent from domestic wastewater

Population (person) Per person protein consumption (kg/person/year) Fraction of nitrogen in protein (kg N/kg protein)

39,292,972 22.60 0.16

Fraction of non-consumption protein

1.10

Fraction of industrial and commercial co-discharged protein | 1.25

Total nitrogen in effluent (kg N/year)

Table 4.

195,364,657

Estimation of emission factor and emission of indirect N>O from wastewater

Nitrogen in effluent (kg N/year) Emission factor (kg N20-N/kg N) Conversion factor of kg N2O-N into kg N20 (44/28) Emission from wastewater plants (kg N3O-N/year) Total N3O emission (kg N2O-N/year) Total N3O emission (Gg NoO-N/year)

195,364,657 0.005 | 1.57 0.00 1,535,008 Ls4

Greenhouse Gas Emission from Domestic Wastewater Treatment 3.2.

117

Nitrous Oxide Emissions

The NO emissions can produce as direct emissions from treatment plants or from indirect emission from wastewater after effluent discharge into waterways, river or sea. Using Eq. 4 and Eq. 5, the result of estimation of N20 emissions are presented in Table 3 and Table 4. Total of NoO-N

emissions from domestic wastewater in East Java in 2017 and the

prediction in 2026 is 1.54 Gg N.O-N/year and 1.63 Gg N2O-N/year, respectively.

3.3.

GHG Emissions

Table 5.

GHGs emission from domestic wastewater in East Java Province.

Year

GHGs emission from domestic wastewater in East Java Province (ton CO>-eq)

2017

3,417,655

2018

3,439,744

2019 2020

3,462,311 3,485,067

2021

3,508,014

2022

3,531,155

2023

3,554,491

2024

3,578,025

2025

3,601,757

2026

3,625,692

The degree of greenhouse warming due to a mass of gas relative to carbondioxide, from initial assessment period to years selected planning horizon, as be known as Global ‘Warming Potential (GWP) unit. This unit is employed because it can be a reference to a given time, to overcome a different warming intensity and a different expected atmospheric residence time of each gas [12]. The calculation GHG emissions by arithmetic sum

of CO-eq

conversion

of CH4y and N20

emissions,

are presented

in Table 5.

Eventhough IPCC recommends using default factors when data is limited, these calculation can be uncertain, due to the lack of reliable data about the explanation of treatment process and specific local conditions. The accurate utilization degree for estimating CH, emissions should be investigated in an appropriate field tracer. The data of sludge removal for CH, and N,O and also CH, recovery should be recorded clearly.

4

Conclusion

Direct GHG emission from domestic wastewater in East Java Province were studied in two

direct

GHG

emission

CH,

and

N30.

Total

of CH,

emissions

from

domestic

wastewater in East Java in 2017 is 140.09 Gg CH,/year that predicted increase in 2026

118 at

Y. P. Devia 148.62

Gg

and

CH4/year.

D. T. Agustini The

calculation

of total

N,O-N

emissions

from

domestic

wastewater in East Java in 2026 will enhance from 1.53 Gg N3O-N/year in 2017 to 1.63 Gg N2O-N/year. Finally, from domestic wastewater in East Java Province contribute 3,417,655 ton CO>-eq in 2017 that rise up 6.08% to 3,625,692 ton CO3-eq in a following ten years. The limited of reliable data about treatment process and local situations should be overcomed by performing accurate tracer.

References 1. Parravicini, V., Svardal, K., Krampe, J: Greenhouse gas emissions from wastewater treatment plants. Energy Procedia 97, 246-253 (2016) 2. IPCC: IPCC

Guidelines for National

Greenhouse

Gas Inventories, Chapter 6 Wastewater

Treatment and Discharge (2006) 3. IPCC: Climate Change

2013:

The Physical

Science Basic (Cambridge

United Kingdom:

Cambridge University Press and New York USA), p. 1535 (2013) 4. Massara, T.M., Malamis, S., Guisasola, review on nitrous oxide (N2O) emissions wastewater and sludge reject water. Sci. 5. Daelman, M.R.J., Voorthuizen, E.M.V., M.C.M.V.: Methane and nitrous oxide

A., Baeza, J.A., Noutsopoulos, C., Katsou, E.: a during biological nutrient removal from municipal Total Enviro. 596-597, 106-123 (2017) Dongen, L.GJ.M.V., Volcke, E.LP., Loosdrecht, emissions from municipal wastewater treatment —

result from a long-term study. Water Sci. Technol. 67 (10), 2350 -55 (2013) 6. Law, Y.Y., Ye, L., Pan, Y.T., Yuan, Z.G.: Philosoph. Trans. Royal Soc. B-Bio.

Sci. 367

(1593), 1265-1277 (2012) 7. Friedrich, E., Trois, C.: Quantification of greenhouse gas emissions from waste management processes for municipalities - a comparative review focusing on Africa. Waste Manage. 31, 1585-1596 (2011). https://doi.org/10.1016/j.wasman.2011.02.028 8. Noyola, A., Paredes, M.G., Guereca, L.P., Molina, L.T., Zavala, M.: Methane correction factors for estimating emi ns from aerobic wastewater treatment facilities based on field data in mexico and on literature review. Sci. Total Enviro. 639, 84-91 (2018) 9. Anwar, M.R., Devia, Y.P., Rahman, A.A.: Studi evaluasi pengolahan air limbah industri

secara terpusat di kawasan industri rembang pasuruan (PIER). Rek. Sipil 2(3), 205-214 (2008) 10.

Badan

Pusat Statistik (BPS) Jawa Timur:

Konsumsi

Kalori dan

Protein Penduduk

Jawa

Timur (2017) 11.

Badan Pusat Statistik (BPS) Jawa Timur: Jumlah Penduduk dan Laju Pertumbuhan Penduduk Menurut Kabupaten/Kota di Provinsi Jawa Timur, 2010, 2016, dan 2017 (2019)

12. Pickin, J.G., Yuen, S.T.S., Hennings, H.: Waste management option to reduce ghg emissions from paper in Australia. Atmo. Enviro. 36, 741-752

(2002)

®

Check updatesfor

Removal of Cadmium from Aqueous Solution by Optimized Magnetic Biochar Using Response Surface Methodology Anwar Ameen Hezam Saeed", Noorfidza Yub Harun’, Mohamed Mahmoud Nasef', Haruna Kolawole Afolabi’, and Aiban Abdulhakim Saeed Ghaleb?

' Department of Chemical Engineering, Universiti Teknologi PETRONAS,

Bandar Seri Iskandar, Perak, Malaysia aahsanwar@gmail. com ? Department of Civil and Environmental Engineering, Universiti Teknologi

PETRONAS, 32610 Bandar Seri Iskandar, Perak, Malaysia

Abstract. Rice husk, whether in the form of raw materials ash or char is one of the best potential materials for the production of a variety of value-added products. This study intends to optimize biochar from Malaysian rice husk, as a

friendly low-cost adsorbent for removing cadmium from aqueous solution. Biochar production has been done using a tube furnace with different input parameters which are heating temperature, residence time, and impregnation ratio. The effect of pyrolysis process parameters was examined by using

response surface methodology (RSM). Responses parameters, biochar yield, and removal efficiency were investigated. Depend on the Central Composite Design

(CCD), three 2FI models were developed to compare three independent response variables. The optimum production condition for magnetic biochar was found as follows: a heating temperature of 395 °C, the residence time 120 min,

and 0.5 g/g impregnation ratio. The optimum magnetic biochar showed 44.5% of biochar yield and 94.25% removal efficiency which was in agreement with the predicted values.

Keywords: Rice husk - Adsorption - CCD - RSM - Cadmium - Magnetic biochar 1

Introduction

Most of the environmental problems and water unattended development of industries technology the most hazardous metals. Its danger lies in that it paint, batteries, and dyes, but municipal waste is neamess

to human

reach [2, 3]. Cadmium

pollution come from the rapid and [1]. Cadmium is considered one of is used in many industries, including the most complex source due to its

can be found

in the industries as a form of

cadmium nitrate, cadmium sulfate and it is more likely to leach from industrial waste and is the most toxic to living organisms [4]. Treatment of removing heavy metals is many such as ion exchange, chemical precipitation, filtration, and adsorption, but recently adsorption is considered as the best method due to its simplicity, cheapness,

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 119-126, 2021. https://doi.org/10.1007/978-981-33-6311-3_14

120

A. A. H. Saeed et al.

and the possibility of using it on a larger scale [5]. Many adsorbents have been used and consumed in various industries and yet used until now such as commercial activated carbon, activated alumina, silica gel, molecular sieve carbon, molecular sieve zeolites, and polymeric adsorbents [6]. The adsorbents require great efforts and many costs. Therefore, researchers resort to utilizing biomass waste to produce low-cost adsorbent for heavy metal removal [7, 8]. Currently, there is an alternative technology to produce adsorbents with high efficiency and distinctive properties at a reasonable cost which are biochar made from raw agricultural materials with magnets so that the magnet improves the removal of these materials and gives them a special possibility of regeneration [9]. This study aims to conduct and optimize magnetic biochar as an adsorbent for adsorbing cadmium ions from aqueous solution.

2 2.1.

Materials and Methods Chemical and Materials

Raw materials that had been treated as biochar was rice husk. Rice husk has been taken from rice mill (Bota), Perak. Rice husk material was cleaned from dirt then dried at 105

°C for 2 days. Samples were grinding, sieving to one type of size which ranges from 500-1,000 pm. It then placed in an airtight container at room temperature before pyrolysis. Ferrous chloride, malic acid, and cadmium nitrate tetrahydrate were purchased from Avantis Laboratory Supply. All the reagents and pellets were analytical grade or highest purity available and were used without further purification. 2.2

Magnetic Biochar Preparation

20 grams of ground rice husk was soaked with 500 ml of 1.0 M malic acid along with the different ratio of ferrous chloride then stir it using a magnetic stirrer for about 10 h. The deposited was separated by using a coffee filter and rinsed with distilled water several times until the pH becomes 5 then dry at the oven for 24 h. 2.3

Pyrolysis

Alumina Tube Furnace tube was prepared for performing the pyrolysis process which has performed with different input parameters which are labeled in Table 1. Purified nitrogen

(99.995%)

with a flow rate

of 100 cm?/min

was

used and

heating rate 7 °

C/min during the pyrolysis process. RSM is such software which usually carried out to widen screening mathematical and statistical methods for analyzing and improving products, which fellow up by experimental quantitative data to determine regression models and optimal operating conditions.

2.4

Batch Study

A 1000 mg/L Cd (II) stock solution was prepared with cadmium nitrate trihydrate and then diluted to 25 mg/L for batch studies. A 25 mg/L cd** stock solution was used for the adsorption study. The solution was treated with a fixed variable to obtain the best

Removal of Cadmium from Aqueous Solution

121

Table 1, Input parameter for the pyrolysis of magnetic biochar Name Units | Low | High Temperature °c [300 500 Time Min |30 120 Impregnation ratio | g/g 0.25, 1 pyrolysis process parameters for cadmium removal. The fixed parameters were the dose of the adsorbent, the pH of the adsorbent, the stirring speed, and time. Add each selected dose and pH of the adsorbent (magnetic biochar) to 250 ml ofa 25 mg/L cd** stock solution. ICP-OES (Inductively Coupled Plasma-Emission Spectroscopy) was used to determine the initial and final concentrations of the cd? * stock solution. The difference between the initial and equilibrium metal ion concentrations determines the number of metal ions absorbed by the magnetic biochar. Calculate the yield, removal efficiency, and adsorption capacity of magnetic biochar according to the following formula.

iocharYield% == BiocharYield% RemovalEffiecincy% =

Wmb

« 100

(1)1

ci-c ai i ©. 100

(2)

in the previous equations, Wmb is the magnetic biochar’s mass and Wr is rice husk’s mass. Ci and Ce are initial concentration and the final concentration.

3 3.1.

Result and Discussion Optimization of Magnetic Biochar and Adsorbent Using CCD

Response surface methodology is such common technology which widely used in materials implementation. This technology usually used for evaluating the capability effects of input parameters on the quality of the product over statistical and mathematical approaches [10]. The model usually should be fitted with the data presented. Model selection is always selected based on the sum of squares and based on the highest or lowest order model. In this analysis, the selected and recommended model was

a two-factor

interaction

model

(2FI)

for the

two

responses

(biochar

yield

and

removal efficacy). The experimental equations for a yield of magnetic biochar and removal efficiency are expressed in arranged variables and are displayed in Eq. (3) and

Eq. (4).

BiocharYield = 55.58 — 8.28A — 1.56B +2.87C — 2.47AB — 1.41AC — 1. 46BC

(3)

122

A. A. H. Saeed et al.

Removal efficiency = + 84.78 + 4.26A + 8.54B + 0.62C + 2.88 * AB +2.26AC +0.91 « BC

4

In term of regression models equation and interaction between more than variables, there are some terms need to figure out whereat the positive sign in the above equation express a synergistic effect whereas a negative sign indicates the antagonistic effect. In the above equation, A is temperature value, B is time value and C is the impregnation ratio of raw materials to the magnet materials. In this study, there were 20 runs which were generated by the Design-Expert software as labelled in Table 2. Three significant input parameters with two response parameters. Input parameters were temperature, time, and impregnation ratio while the response parameters were biochar yield and removal efficiency.

Table 2, Design matrix of experiments with results. Run

Input parameters

Responses

A:Temperature

|B: time

CIR

Biochar yield

Co)

(min)

(g/g)

(%)

1/300 2 350 3400 4 450 5 500 6 300 7 350 8 400 9 450 10 500 11300 12 350 13. 400 14450 15500 16 300 17350 18 400 19450 )

30 30 30 30 30 60 60 60 60 60 90 90 90 90 90 120 120 120 120 120

I 0.75 05 0.25 1 0.75 05 0.25 1 0.75 05 0.25 1 0.75 05 0.25 1 0.75 05 0.25

52.3 49.5 48.5 41.6 45.95 48.5 46.8 46.6 45.4 44.5 45.8 43.9 43.8 43.1 42.9 42.3 40.7 41.9 40.9 40.1

| Removal efficiency (%)

89.55 92.32 94.51 98.2 97.55 89.8 92.55 94.72 99.21 97.35 89.92 92.36 94.6 98.08 97.12 89.8 92.28 94.48 99.3 96.92

Based on the proposed model, the variance of the two responses was analysed. The recommended model is 2FI, which also creates coefficient values for R, F-value, and P-value. All these values represent the important impact of each run. Regression model analysis is a known method for comparing design models with response statistics.

Removal of Cadmium from Aqueous Solution

123

F-tests are considered an important aspect of finding comparisons and relationship between the mean square and the residual of the regression model is established by the high value of F and the low value of P. Therefore, the F test with high F value and low P-value is a more effective and accurate model

[11]. From

the experimental

analysis

found in ANOVA analysis, the values of F were found to be 349.87 and 149.52 and lower P values were (

0.7

z

0.6

é

143

2

z2 $ s

E03 3 S02

sme

S

ol

& a

——

0

(a)

05

1

1S

65mm/hr

2

25

——

3

1 Shr: $0mm/br

——— 2.0br:1,70mnvhr —— 2. Shr:1,.65mn/hr

0

0

Rainfall duration (hr)

0.28br:L1 75mm Oshrommie 1ohr1 100mm

O5

(b)

1

15

2

25

3

Rainfall duration (hr)

Fig. 4. Depth of water in StormPav structure and volume capacities at 10 yrs average recurrence interval, ARI at various rainfall events.

08

0.6

07

2”

= °°

5204

Eos

3

220 £5 o> BF 02 4 o1

—

0 (@

zo 02ou

B03

5.

o

1

2

3

ge SARE: 35mm/br

WARE-40mm/hr 4

5

JARI: 25mnvhe

aa:SARI: 3Smmvbr Som

—— 10ARI:40mmvnr

0 6

Rainfall duration (hr)

) oO

1

2

3

Fig. 5. Storage capacities of StormPay a) depth of water and b) volume average recurrence intervals, ARI within 6 hr rainfall duration

respective storage volumes from 1.995 m? to 2.565

4

5

6

Rainfall duration (hr) storage for various

m*. In [6], an indicative volume for

the stormwater component was 2m? for a standard house. Thus, the StormPav fulfils the criteria of the stormwater component. In Fig. 6a, the rainfall intensities applied was for 10 yrs rainfall events for continuous 12 h rainfall, with the maximum rainfall depth of 312 mm, similar to the StormPav’s effective storage design depth. Furthermore, the maximum rainfall volume collected was less than 0.8 m?, which was 60% less than the total volume. The rainfall peak discharge for 10 yrs ARI of residential roof gutter size

range from 7 m’Vhr to 15 m*/hr, for roof area of 60 m* to 120 m?. [25] mentioned that the tank size of 1 m* for the roof area of 100 m? was near optimum size to store 10 mm rainfall. The StormPav can provide a detention volume of less than 2.5 m> to convey an underground rainwater harvesting tanks. Besides, Fig. 6a and 6b provide rainwater harvesting tank sizes of less than 8 m* volume capacity. The duration was less than 30 min within the typical time of concentration of small urban watersheds [26, 25] at peak

discharge of 15 m*/hr.

144

N. Bateni et al.

0.8 J 06 % 04 8 202

0 (b)

45

20

95

Rainfall duration (min)

30

Volume (m*)

Volume (m’) CU hae

Rainfall duration (hr) 8 6 4 2 0 0 (c)

1549

25

Rainfall duration (min)

30

Fig. 6. StormPav Volume capacities a) at 10 to 13.5 m? car park area, RWH volume capacities collected b) in a cylindrical shape, and c) rectangular tanks shape. To investigate the performance of StormPav in LID Control in SWMM 5.1 between measured and simulated data, thesensitivityanalysiswasperformed. The calibrated model was a good fit between the simulated and measured water depth, as confirmed by its determination coefficient R” of 0.993 . Table 1 shows the recommended height of the StormPav, represented as a rain barrel in SWMM and maximum amount of water depth collected in the system. From the Kota Samarahan rainfall record, April to September received less than 10 mm of mean monthly rainfall. In wetter months (December to March) the mean monthly rainfall was 20 to 35 mm. The most torrential

monthly rainfall received was about 250 to 300 mm. From Table 1, the relationship between rainfall and StormPav depth gave a linear trendline,;whereby the recommended maximum height of StormPav should be 1.5 times the rainfall depth received. Table 1.

Recommended

maximum height of StormPav cylinder based on 24 h rainfall

Rainfall depth, mm

0-10

Recommended maximum height of the cylinder, mm

|20

4

| 11-35

| 50

| 36-84

120

| 85-140

| 200

|: 141-210

300

| 211-350

500

Conclusion

This study investigated the storage capacity of StormPav as a rainwater harvesting system. StormPav had shown unique hydrological features, which proved advantageous on a bigger space and volume of void. The experimental results concluded that the StormPav can be applied for both rainwater harvesting system and as the permeable pavement with a connected underground storage tank. Therefore, StormPav is a

Performance of Permeable Pavement with Subsurface Micro Detention Storage

145

promising sustainable green pavement solution altemative for roadwork and detention facilities for rainwater harvesting devices in a hydrology perspective.

Acknowledgments. The authors would like to express their gratitude for the financial support received by OSAKA experiments.

Grant and Universiti Malaysia Sarawak for the facilities provided for their

References 1. Haris, H., Chow, M.F., Usman, F., Sidek, L.M., Roseli, Z.A, Norlida, M.D.: Urban stormwater management model and tools for designing stormwater management of green infrastructure practices. In: IOP Conference on Series Earth Environment and Science, vol.

32 (2016) 2. Er, H., Lim, L.LP., Bong, C.H.J.: A hydrology and hydraulic case study on January 2015 flash flood in UniGarden,

Kota Samarahan,

Sarawak.

In: Proceedings of the 37th IAHR

World Congress (2017) 3. Khan, B.M., Geiss, S.: Special Section: Sewers and Drainage Systems. Storm Water Solutions, New Mexico, pp. 30-33 (2012) 4, Wong, T-H.F.: Water Sensitive Urban Design - the Journey Thus Far. Engineers Australia, Australia, p. 10, August 2007

5. Schiitte, S., Schulze, R.E.: Projected impacts of urbanisation on hydrological resource flows : a case study within the uMngeni Catchment, South Africa. J. Environ. Manage. 196, 527543 (2017) 6. Woods-Ballard, B., Kellagher, R., Martin, P., Jefferies, C., Bray, R., Shaffer, P.: The SuDS

manual, London (2011)

7. Porter-bopp, S., Brandes, O.M., Sandborn, C., Brandes, L.: Peeling back the pavement A Blueprint for Reinventing Rainwater Management in Canada’s Communities, Canada (2011) 8. Webber, J.L., Gibson, MJ., Chen, A.S., Savic, D., Fu, G., Butler, D.: Rapid assessment of surface-water flood-management options in urban catchments. Urban Water J. 15(3), 210—

217 (2018) 9. Yong, C.F., Deletic, A., Fletcher, T.D., Grace, M.R.: The clogging behaviour and treatment efficiency of a range of porous pavements. In: 11th International Conference on Urban

Drainage (2008) 10.

Scholz, M., Grabowiecki, P.: Review of permeable pavement systems. Build. Environ. 42

(11), 3830-3836 (Nov. 2007)

11. Castro-fresno, D., Andrés-valeri,

V.C., Saiiudo-fontaneda, L.A., Rodriguez-hernandez,

J.:

Sustainable drainage practices in spain, specially focused on pervious pavements. Water 2, 67-93 (2013) 12. Saraswat,

C.,

Kumar,

P., Mishra,

B.K.:

Assessment

of stormwater

runoff management

practices and governance under climate change and urbanization: an analysis of Bangkok, Hanoi and Tokyo. Environ. Sci. Policy 64, 101-117 (2016)

13. Jayasuriya, L.N.N., Kadurupokune, N., Othman, M., Jesse, K.: Contributing to the sustainable use of stormwater: the role of pervious pavements. Water Sci. Technol. 56(12),

69-75 (2007) 14. Joseph Battiata, G.H., Collins, K., Hirschman, D.: Updating the runoff reduction method. J. Contemp. Water Res. Educ. (146), 11-21 (2010) 15.

Boomsma,

W.,

Huurman,

M.:

Permeable

paving

systems

International Conference on Concrete Block Paving (2006)

with storing capacity.

In: 8th

N. Bateni et al.

. Marchioni, M., Becciu, G.: Experimental results on permeable pavements in urban areas: a synthetic review. Int. J. Sustain. Dev. Plan. 10(6), 806-817 (2015) . Bateni, N., Lai, S.H., Putuhena, F. Mah, Y.S., Mannan, A.: A rainfall simulator used for testing of hydrological performances of micro-detention permeable pavement. Int. J. Eng.

Technol. 7, 44-48 (2018) . Rossman,

L.: Storm Water Management

Model

User’s Manual, Version 5. United States

Environmental Protection Agency (US EPA), Cincinnati, Ohio (2015) . Barszcz, M.: Influence of applying infiltration and retention objects to the rainwater runoff on a plot and catchment scale — case study of stuzewiecki stream subcatchment in Warsaw. 20.

Pol. J. Environ. Stud, 24(1), 57-65 (2015) Qin, H., Li, Z., Fu, G.: The effects of low impact development on urban fl ooding under

21.

different rainfall characteristics. J. Environ. Manage. 1-9 (2013) Mah, Y.S.: Potential of road subsurface on-site stormater detention

system.

UNIMAS

Publisher, Sarawak (2016) 22.

Liow,

C.V.:

Modelling

of

StormPay

green

pavement:

Inlet

and

outlet

of integrated

permeable road and stormwater detention. vol. 10, no. 2, pp. 966-976 (2019) 23.

Szota, C., Coutts, A.M., Thom, J.K., Virahsawmy, H.K., Fletcher, T.D., Livesley, S.J.: Street tree stormwater control measures can reduce runoff but may not benefit established trees.

24.

Rose, K., Hodges, B.R.: Evaluating the Effects of Low Impact Development on Texas A &

25.

DID,

Landsc. Urban Plan. 182, 144-155 (2019) M University West Campus By, Austin (2010) 26.

Urban

Stormwater

Management

Manual

for

Malaysia,

2nd

edn.

Department

of

Irrigation and Drainage (DID) Malaysia, Kuala Lumpur, Malaysia (2012) Leming, MLL., Malcom, A., Tennis, P.D.: Hydrological Design of Pervious Concrete. Silver Spring, Maryland, USA (2007)

®

Check updatesfor

A Proposed Framework of Life Cycle Cost Analysis for Petrochemical Wastewater Treatment Plants Muhammad Ilyas), Freselam Mulubrhan Kassa, and Mohd Ridzuan Darun

Faculty of Industrial Management, University of Malaysia, Kuala Lumpur, Pahang, Malaysia ilyas@kiu. edu. pk, {freselam, mridzuand}@ump. edu. my Abstract. Stringent regulations has made it mandatory for petrochemical industries to have wastewater treatment plants for the discharge of safe and

environment friendly water. Lack of comprehensive and easy to use framework for life cycle cost analysis hinders the economic evaluation of wastewater treatment plants. This paper presents a comprehensive and easy to use framework for life cycle cost analysis of wastewater treatment plants. The framework

includes the purpose of analysis, cost categories, LCC methodology, data collection, calculation of LCC

and analysis to select the best alternative process.

A brief overview of existing frameworks, need analysis and wastewater process flow is also presented. The proposed framework will provide the foundation through which the life cycle cost of different alternatives can be estimated for

effective decision making. Keywords: Life cycle cost analysis « Cost breakdown structure - Framework Cost data - Wastewater treatment plant - Petrochemical industries

1

Introduction

Petrochemical wastewater contains various organic and inorganic components. Appropriate treatment is required for reuse, discharge, or final disposal. The composition of wastewater and environmental regulations requires a combination of different treatment (WWTP)

methods in petrochemical is a combination of various

industries. [1]. Wastewater treatment plant processes to treat wastewater and reproduce

environmentally safe water. The processes may include filtration, clarification, biodegradation, oxidation, ozonation and sludge disposal [2]. Malaysia has made it mandatory for industries to have wastewater treatment plants and made regulations for WWTPs to ensure the discharge of environment-friendly and safe water [3]. The quality of effluent from treatment plants is regulated by the Environmental Quality Act 1974 and its regulations such as the Environmental Quality (Sewage) Regulations 2009 and Environmental Quality (Industrial Effluent) Regulations 2009 [4]. Since wastewater treatment is mandatory in petrochemical industries

and priority agenda of the national government

and international organizations, their

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 147-153, 2021. https://doi.org/10.1007/978-981-33-6311-3_17

148M. Tlyas et al. economic, environmental and social sustainability should be properly assessed [5]. Recycled industrial wastewater could be a potential water source for agriculture and industrial use. The concept of “Life cycle Costing” refers to the traditional approach, devised in the 1930s, when the US General Accounting Office included operating and maintenance costs in public procurement [8-10]. The main purpose of LCC is to assess and facilitate decision making in investments requiring high initial capital [11], carried out through cost assessment during the life cycle of products, services, and technologies [12]. This study develops an easy to use, explicit and transparent LCCA framework for ‘WWTPs of petrochemical Industries. LCCA is a useful tool for assessing the cost and benefits of several alternatives to help decide which one has the lowest LCC. The selection criteria are based on the economic assessment by comparing the LCC of various alternatives. Alternatives may include membrane technology, advanced oxidation, adsorption, and Biological treatment techniques [1].

2

Methodology

The goal of this study is to develop a framework for LCCA. This framework can be used as a tool to carry out cost analysis for new and existing processes of WWT in petrochemical industries. Current practices in the field of LCC of WWTP uses a hybrid approach of combining LCA with LCC with more focus on LCA, compromising key elements of LCCA. This practice compromises methodological depth and scale, making LCC a secondary tool with little efficacy. One of the possible reasons for this practice is the lack of standards and models to perform LCCA. As a part of the current study, LCC case studies are critically analyzed. Some of the key findings reflect that more focus is on the environmental aspect as compared to the economic aspect. Further, cost estimates are based on assumptions rather than empirical evidence. The reason for this practice is not having a single LCC model which has been accepted as a standard model and employed widely. [13]. These limitation implore to develop a comprehensive framework of LCC for WWTPs. LCC was first introduced about sixty years ago by US Department of Defense and it predates LCA. However, its application in wastewater treatment infrastructure and processes is relatively new. Cole and Sterner [14] identifies gaps between LCC theory and practice and concludes that lack of formalized framework, and data availability hinders its application. They suggest a comprehensive framework for LCC application in decision making process. [15] developed LCC model of membrane technology for WWTP. Cost categories are well identified but estimation techniques are not elaborated. Heijungs, Settanni [7] proposed a methodology for the calculation of LCC by including environmental and social aspects into LCC which can be used to calculate eco-efficiency indicators for Life cycle sustainability analysis. Heijungs, Settanni [7] admits that precise and general computational structure for LCC calculation is lacking. The methodology proposed by Heijung itself is too wide in terms of scope and application. Glick and Guggemos [16] includes sustainability aspect into the calculation of LCC

and combines

it with LCA.

Bouachera,

Kishk

[17] claims that in theory

LCC methodology is quite straightforward but its application has many shortcomings.

‘A Proposed Framework of Life Cycle Cost Analysis

149

To address these shortcomings he suggest a mathematical model to deal with data uncertainties. Mustapha, Manan [3] developed a life cycle performance assessment framework for wastewater treatment with an emphasis on environmental aspects. It has been observed that the used models are too complex for practical application in industrial settings, consequently, practitioners had little motivation to perform LCCA.

3

Proposed Framework

Th six step proposed framework is presented in the figure bellow. Discussion section explains each components of the framework in detail (Fig. 1).

Purpose (Somparsive of Life CycleAnalysts) Cost Analysis

a

[

Detine Cost Categories

(_exsoten memes e166 ternuiston ) [

]

) (eens con) [ Rare Rarer)

[__Seisbueh cost estat

Sata Conection [cpmonrmeieus)

[[ensrreearaion )

[

Gaicuimuon of Lec

Anslyens to Select best ehemative Fig. 1.

4 4.1

LCC

]

[_Onmerimesen ]

[[ceccomsonente ][ roweraroncion } [ oncounmonecx ]( [

rinions

vec

—*d

|

]

framework for WWT

Discussion Purpose of LCC

LCC was originally designed to assist decision making for procurement purposes from clients point of view [9]. The purpose and nature of LCC depends on the product and process in consideration. Smit [19] mentioned that LCC

can be used in different ways

150M. Tlyas et al. and is useful in a wide range of applications such as alternative solutions and source selection, evaluating affordability of a process, managing and controlling current budget, developing profile of future expenditures, analyzing cost reduction opportunities, assessing areas of financial risks and uncertainties and improving processes of the organization. Barringer, Weber [20] classified LCC on the bases of purpose into Affordability studies, Source selection studies, Design/process tradeoff studies, Repair level analysis, Warranty & repair cost estimates and Supplier sales strategy. Greene and Shaw [21] provides several types of LCC which includes comparative analysis, cost effectiveness analysis, Maintenance concept analysis, LCC estimates for source selection, Trade studies, Cost benefit analysis, Repair level analysis and provisioning analysis. The purpose of LCC analysis in wastewater treatment is to select between different alternative techniques. Broadly, alternatives usually include Membrane treatment, Physico-chemical treatment, Biological treatment and Ion exchange treatment. 4.2

Cost Components/Categories

Cost components

can be reflected through Cost Breakdown

Structure

(CBS).

CBS

is

the comprehensive list of all cost items required for LCC analysis. It illustrates all costs attributed to each phase of the project, product or process. CBS proposed for LCC of ‘WWTP

is based on El-Haram, Marenjak [22] which divided CBS into five levels in top

down hierarchy which are project level, phase level, category level, element level and task level. At project level it will be LCC of wastewater treatment plant (WWTP). As per life cycle of WWTP, LCC is divided into following phases; Initial phase, Operation phase, Repair & Maintenance phase and finally Decommissioning & upgrading phase. Once phases are identified the CBS structure can be broken down into category level and subsequently to element and task level. 4.3

Selection of LCC Method/Model

For LCC calculation, different cost estimation methods exists and their application depands on many factors. [23, 24] introduced six different ways to estimate costs. They are estimating by Engineering Procedures, Estimating by Analogy, Parametric Estimating methods, Artificial Neural Network-based Estimation, and Fuzzy logic based Estimation. Any combination of these six can be used to estimate the LCC of a system or process. The selection of the cost estimation method depends on many factors which include the phase of the process, purpose of analysis, availability, quality, and certainty of data. There is no single technique that could be credited as the best technique therefore, the application of any technique is contextual. 44

Data Collection

Data collection in terms of resources consumed is the most critical and important part of conducting LCCA. As it is a data-driven process, it requires time and effort. The quality and ease of data collection define what method or model can be applied and what analysis can be performed.[19] As we progress through different phases of

A Proposed Framework of Life Cycle Cost Analysis

151

process development, data availability in terms of quality and quantity increases. At the initial phase of process development, comprehensive data is lacking and estimator has to rely on analogy and expert judgment whereas, in the operation phase detail engineering method could be applied which results in a more accurate estimation. It has been observed from the literature that most of LCCA in wastewater treatment is conducted in the operation phase which gives comprehensive data however estimator needs to identify cost categories and elements before collecting cost data. Once categories are identified, the data source should be identified. Once data is collected it needs to be prepared for the analysis. As data is collected from different sources therefor, there is a lack of uniformity and consistency in the data. Data preparation implies here changes and adaptations to make it applicable for LCCA. [25]. 4.5

Calculation of LCC

To calculate and the present value amounts to their convert them to

compare the LCC of each process alternative, all costs are presented in by discounting them to the base date. The LCC method escalates all future year of occurrence and discounts them back to the base date to present values. This can be accomplished using these equations:

LCC, = IC + OC + RMC + DUC

(1)

Where IC is Initial cost, OC is Operation cost, RMC is Repair and Maintenance cost and DUC is Decommissioning and Upgradation cost. Present values are calculated through discounting back the future value to bring these values to common base using present value valuation.

NPV, = PC

diva}

(2)

y=1

Equations 1-2 can be used to calculate NPV of each alternative process. The resulting values can then be compared to determine which alternative has lowest LCC and more economical in the long run. Based on the last 10 years average, suggested inflation rate is 2%, the discount rate is 4% and the analysis period for WWTP is 30 years. 4.6

Uncertainty Analysis

Any appraisal process has Uncertainty and risk embedded in input parameters. [26]. LCCA could be performed in a deterministic or probabilistic manner. In a deterministic approach, the analyst makes discrete assumptions about the uncertain parameters and calculates a point estimate of the outcome. However, it is not capable of incorporating variability and uncertainty of input parameters. Probabilistic LCC is more comprehensive as it can handle uncertainties. [26] LCC estimates of new process will certainly contain uncertainty and risk. [25]. For effective decision making, uncertainties asso-

ciated with LCC results must be accounted for by substituting deterministic input cost

152M. Tlyas et al. variables with probabilistic ones.[27] much output results are affected by performed by identifying the output and conclusions are drawn to choose

5

Sensitivity analysis is widely used to explain how a change in input values. Sensitivity analysis is and input functions. Finally results are analyzed best alternative method for wastewater treatment.

Conclusion and Future Research

A review of past studies indicates that lack of framework and methodology is the main obstacle in conducting LCCA. An attempt has been made to fill this gap by presenting a detailed framework in the context of WWTPs. This paper presents a comprehensive and easy to use framework of LCCA for WWTPs in petrochemical industries. The first step in this framework is to decide on the purpose of LCCA followed by the development of detailed cost breakdown structure, formulation of LCC method, data collection, calculation of LCC and finally conducting sensitivity analysis to select the best process alternative. This framework can be used for selection between various alternatives based on the lowest LCC. Further research can be conducted to include the environmental cost into the framework and to design a methodology for conducting Parallel LCA and LCC.

Acknowledgement. The authors would like to acknowledge Universiti Malaysia Pahang for its financial support conference.

through

RDU191802-3

TRGS

grant for this paper

to be presented

at the

References 1. Wei, X., et al.: Treatment of petrochemical wastewater and produced water from oil and gas.

Water Environ. Res. 91(10), 1025-1033 (2019) 2. Aljuboury, D., et al.: Treatment of petroleum wastewater by conventional and new technologies-a review. Glob. Nest J. 19, 439-452 (2017) 3. Mustapha, M.A., Manan, Z.A., Alwi, S.R.W.: A new quantitative overall environmental performance indicator for a wastewater treatment plant. J. Clean. Prod. 167, 815-823 (2017) 4. Mat, E.A.T., Shaari, J.. How, V.K.: Wastewater production, treatment, and use in Malaysia. In: Safe Use of Wastewater in Agriculture 5th Regional Workshop Southeast and Eastern

Asia, Bali, Indonesia (2013) 5. De Menna, F., et al.: Life cycle costing of food waste: a review of methodological approaches. Waste Manag 73, 1-13 (2018) 6. Corrado, S., et al.: Modelling of food loss within life cycle assessment: from current practice towards a systematisation. J. Clean. Prod. 140, 847-859 (2017) 7. Heijungs, R., Settanni, E., Guinée, J.: Toward a computational structure for life cycle

sustainability analysis: unifying LCA and LCC. Int. J. Life Cycle Assess. 18(9), 1722-1733 (2013)

8. Gluch, P., Baumann, H.: The life cycle costing (LCC) approach: a conceptual discussion of its usefulness for environmental decision-making. Build. Environ. 39(5), 571-580 (2004)

9. Korpi, E., Ala-Risku, T.: Life cycle costing: Auditing J. 23(3), 240-261 (2008)

a review of published case studies. Manag.

A Proposed Framework of Life Cycle Cost Analysis

153

. Woodward, D.G.: Life cycle costing—theory, information acquisition and application, Int. J. Proj. Manag. 15(6), 335-344 (1997) 11. Dhillon, B.S.: Life Cycle Costing for Engineers. CRC Press, New York (2009) 12. Swarr, T.E., et al.: Environmental life-cycle costing: a code of practice. Springer (2011) 13. Waghmode, L.Y., Sahasrabudhe, A.D.: Product life cycle cost modelling a suggested framework. In: 2008 First International Conference on Emerging Trends in Engineering and Technology. IEEE (2008) . Cole, R.J., Sterner, E.: Reconcil:

1g theory and practice of life-cycle costing. Build. Res. Inf.

28(5-6), 368-375 (2000) . Jiran, N.S., et al.z ife cycle costing model for hollow fibre membrane module: a review and further research. In: Applied Mechanics and Materials. Trans Tech Publications (2014) . Glicl S., Guggemos, A.A.: Rethinking wastewater-treatment infrastructure: case study

using life-cycle cost and life-cycle assessment to highlight sustainability considerations. J. Constr. Eng. Manag. 139(12), A5013002 (2013) . Bouachera, T., Kishk, M., Power, L.: Towards a generic framework for whole life costing in the oil industry. In: 23rd Annual ARCOM Conference (2007)

. Ghimire, N., Wang, S.: Biological treatment of petrochemical wastewater, In: Petroleum Chemicals-Recent Insight. IntechOpen (2018) . Smit, M.: NATO initiatives to improve life cycle costing. TNO Defence Security and Safety the Hague (Netherlands) (2009) 20.

Barringer,

H.P.,

Weber,

D.P.,

Westside,

M.H.:

Life-cycle

cost

tutorials.

In:

Fourth

International Conference on Process Plant Reliability. Gulf Publishing Company. Citeseer (1995) 21.

Greene, L-E., Shaw, B.L.: The steps for successful life cycle cost analysis. In: IEEE Conference on Aerospace and Electronics. IEEE

22.

23. 24.

El-Haram, collecting (2002) Fabrycky, Farr, J.V.,

(1990)

M.A., Marenjak, S., Horner, M.W.: Development of a generic framework for whole life cost data for the building industry. J. Qual. Maint. Eng. 8(2), 144-151 W.J., Blanchard, B.S.: Life-cycle cost and economic analysis (1991) Faber, I.: Engineering Economics of Life Cycle Cost Analysis. CRC Press, Boca

Raton (2018) 25.

Smit,

26.

J. Comput. Integr. Manuf. 25(4-5), 444-456 (2012) Noor, N.A.M., Eves, C., Mutalib, N.F.A.: An exploratory review of whole life cycle costing for Malaysia property development. J. Techno Soc. 4(1), 37-47 (2012)

27.

M.C.:

A

north

atlantic

treaty

organisation

framework

for life cycle costing.

Int.

AbouHamad, M., Abu-Hamd, M.: Framework for construction system selection based on life cycle cost and sustainability assessment. J. Clean. Prod. 241, 118397 (2019)

® ‘upaates

A Concise Review of Major Desalination Techniques: Features and Limitations Tijani Oladoyin Abimbola'®®, Khamaruzaman Wan Yusof'®®), Husna Takaijudin', Abdurrasheed Said Abdurrasheed'?, Ebrahim Hamid Hussein Al-Qadami', Samiat Abike Ishola*, Tunji Adetayo Owoseni*, and Suleiman Akilu”

' Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS,

32610 Bandar Seri Iskandar, Perak Darul Ridzuan, Malaysia

[email protected],

khamaruzaman. yusof@utp. edu

2 Department of Civil Engineering, Ahmadu Bello University, Zaria, Kaduna State, Nigeria

> Department of Marine Sciences, University of Lagos, Akoka, Lagos State, Nigeria

+ Department of Mechanical Engineering, The University of Nottingham, Nottingham NG72RD,

UK

5 Centre for Nanotechnology Research, Universiti Teknologi PETRONAS, 32610 Bandar Seri Iskandar, Perak Darul Ridzuan, Malaysia

Abstract. This paper provides a succinct and simplified account of desalination techniques. The techniques are categorized as membrane-based, thermal-based, combined membrane/thermal-based and miscellaneous techniques. Essential features of membrane-based, thermal-based and combined membrane/thermal-

based techniques are presented, while their associated limitations are equally highlighted. Reverse Osmosis (RO)

remains the state-of-the-art technology

in

seawater desalination with the highest installed capacity of 65% globally. Whereas, solar desalination potentially shows as a cost and energy-effective means of ensuring steady supply of potable water in off-grid arid-coastal environments, where freshwater reserves are non-existing. In general, if the world is to aim for global freshwater sustainability, seawater desalination must be considered crucial, and a good understanding of its techniques will further drive the vision closer to reality. Keywords: Desalination - Desalination techniques - Membrane-based. techniques - Thermal-based techniques - Solar-thermal desalination - Reverse osmosis - Solar still

1

Introduction

Desalination, from its root word ‘desalt’, is the removal of salt from saline water. By convention, seawater desalination is the use of specialized techniques to recover freshwater

from

seawater by eliminating

the dissolved

salt and

mineral contents

[1].

Several techniques have been adopted for the extraction of freshwater from seawater. They are generally classified under membrane-based, thermal-based and chemical-based

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 154-162, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_18

A Concise Review of Major Desalination Techniques.

technologies. Prominent among

membrane-based

technologies

155

are Reverse Osmosis

(RO), Nanofiltration (NF) and Electrodialysis (ED). Thermal-based technology involves

conventional thermal distillation, which uses heat to evaporate seawater, after which the vapour is collected as distillate. The chemical-based options—involving ion-exchange, have not been found very practical to treat water with high level of dissolved solids [2]. Overall, RO excels as the most installed and most effective desalination technology with 65% total installation capacity globally (Fig. 1). Thus, it remains the state-of-the-art in seawater desalination [3].

% Installed capacity

Categorizations from literature furbish overlapping details that make recognition of the fundamental principles on which specific desalination techniques function to be tedious. This unfolds a research gap of a summarized and simplified approach to classifying these techniques. Therefore, the aim of this review paper was to give more exhaustive synopsis of the desalination technologies from the literature and examine each of them on account of its characteristic features and associated limitations.

RO.

MSF

MED

ED

NF

Others

Fresh water extraction techniques

Fig. 1. Overall global installed desalination capacity by techniques. RO: Reverse Osmosis, MSF: Multi-Stage Flash Distillation, MED: Multi-Effect Distillation, ED: Electrodialysis, NF: Nanofiltration, Others: comprise Freezing/Hydrate-based desalination, Membrane Distillation (MD), Forward Osmosis (FO) and Solar Still Distillation (SSD) [4]. 2

Desalination Techniques

The major desalination techniques can be categorized into four broad categories: Membrane-based, Thermal-based, Combined membrane/thermal-based, and Miscellaneous techniques, as presented in Fig. 2.

156

T. O. Abimbola et al.

+ ‘Membrane-based techniques

¥ based techniques

¥ ‘Membrane/thermalbased techniques

+ Miscellaneous techniques

¥

+

¥

+

Reverse Osmosis (RO) Biectrodialysis (0) ‘Nanofitration (NF)

Mutt-stage Flash (MSF)_| | Membrane Distillation (MO) | | FreezingMyarate Formation IMutieftect Distitation (MED)| | Forward Osmosis (FO} Ton-Excnange

Fig. 2. Categorization of major desalination techniques 2.1

Membrane-Based Techniques

The membrane-based techniques are gaining attention, nowadays, with the advent of high permeability membranes. They involve the use of polymeric membranes to separate salt solutes from seawater. In practice, by using either hydraulic pressure, osmotic pressure or electrical potential, feed saline water is driven across a semi-permeable membrane. The membrane selectively permits freshwater to permeate, while retaining the solutes

[1]. The

overall effectiveness

of the membrane

techniques

is, however,

hindered by the high energy requirements to drive seawater across the membrane [3, 5]. The minimum theoretical energy required to desalinate typical seawater of 35,000 mg/L salt concentration at 50% recovery is 1.06kWh/m*, which is regardless of the type of system used, nor the energy inputs for seawater supply, pre and post treatments, and brine disposal [5]. This figure translates to the energy, in form of applied pressure, required to overcome the osmotic pressure of seawater for crossboundary flow of water through the membrane. Apparently, a higher value should be expected for waters with higher concentrations, such as found in the Arabian Gulf with 45,000 mg/L characteristic concentration. In addition to the energy demand of the membrane technology, fouling, internal concentration polarization, membrane effectiveness in the low pH range (for cellulose acetate

and cellulose

triacetate

membranes)

and

susceptibility

to chlorine

attack

(in

polyamide-polysulfone thin-film composite membranes) are known performance indices in saline water membrane applications [6, 7]. For RO, the actual energy consumption for the best-in-class of seawater, being the one with the lowest salt concentration is between 2.5-2.8 kWh/m’, translating to a cost range of about 0.5-0.8 USD/m? of produced water [8]. But for the most conventional plants, operational energy consumptions ranges between 3-5 kWh/m? [9].

A Concise Review of Major Desalination Techniques 2.2

157

Thermal-Based Techniques

Differently, the thermal-based techniques use heat energy obtained mostly from the combustion of fossil fuels to separate freshwater in the form of vapour-to-condensate from saline water. These techniques are energy intensive, except where waste heat is reused.

Multi-Stage

Flash distillation (MSF)

and Multi-Effect

Distillation (MED)

are

entirely thermal processes of seawater desalination [4]. While the thermal technology appears less cumbersome compare to the membrane-based processes, its operation suffers low thermal-to-vapour conversion efficiency and greenhouse gasses emission from the combustion of fossil fuel, except where clean renewable energy sources are employed. 2.3.

Combined Membrane-Thermal Techniques

Some other desalination technologies combine both the membrane and the thermal applications, making it a two-stage process. Membrane Distillation (MD) [10] and Forward Osmosis (FO) [11] are typical examples. In MD, thermal energy is needed to vaporize seawater, while a vapour-permeable membrane is used to separate the resultant vapour from the water bulk, which is later condensed to recover freshwater condensate. In FO, membrane process precedes the thermal application stage. A thermolytic hypersaline solution, such as ammonium carbonate (NH4),COs, is applied as a draw solution against the seawater feed solution—the former draws water molecules from the later and becomes diluted. The process is then completed by applying heat to the diluted thermolytic draw solution to disintegrate the ammonium ion (NH4*) and carbonate ion (CO3) and recover freshwater, as represented in Eq. (1). Besides the

fact that MD and FO exploitations are still vying for acceptance beyond the R&D quadrant, the overall success of the duo techniques hopes to leverage the availability of low-grade or recycled heat for its thermal process stages.

(NH4)2CO3 (aqeuos)— (heat)—>

2.4

2NH3+

CO, +

H,0

(1)

Miscellaneous Techniques

A few other desalination techniques are less popular and have, thus, received a fairly modest recognition in literature. One of them is freezing or hydrate formation, where water molecules are frozen and cleansed of salt contents, then de-frozen to regain pure water [12]. Another is ion-exchange technique, whose efficiency is only limited to lowsalinity water sources [13, 14]. However,

in all the desalination techniques mentioned,

high energy requirements, huge costs of plant installations, operations and maintenance, as well as operational technical expertise are stupendous challenges. These challenges preclude the techniques as potential solutions in remote settlements with potable water supply challenges. This further justifies the need to develop viable altematives to the much-praised RO technology.

158

T. O. Abimbola et al.

Solar-Thermal

Desalination.

Solar-thermal

desalination

(STD)

offers

a potentially

cost-effective and technologically sustainable approach to seawater desalination, especially for remote arid-coastal environments, where access to the grid is totally nonexisting or extremely inadequate [15]. STD is not only cheap, but it also benefits from the abundance of solar energy potential typical of most arid environments, to produce potable water. The energy renewability and its ecofriendly inclination puts STD ahead of other possible sustainable solutions. As a thermal-based technology, it relies on solar radiation to heat up seawater to produce vapour, which is then collected as condensate for onward usage. Thus, STD functionality is measured by the specific water productivity

(SWP)—which

pivots, as illustrated in Fig. 3, around three metrics:

1) solar

absorptivity (x), which is a measure of the percentage of solar radiation that is verted to heat; 2) thermal efficiency (1,) of the solar absorbing surface, which measure of the heat that is required to convert water to vapour via evaporation; 3) output ratio (GOR), a measure of the percentage of the resultant vapour that is densed as distilled water. Fundamentally, GOR gives an indication of how much latent heat of condensation

is reused for further distillation

[15]. Prominent

conis a gain conthe

seawater

desalination techniques that leverage direct solar radiation are solar still desalination (SSD)

[16] and

humidification-dehumidification

(HDH)

[17], both working

with

the

same principle and along a common desalination pathway.

SWP E.

SWP = —

on,GOR

L LD)

|

+a> Solar

1

Thermal

e>4 Vapor

Water

Fig. 3. Direct solar-thermal desalination pathway. Eis the solar radiation (kW/m?) and Lis the latent heat of evaporation (kWh/litre) [15]. Solar

Still Desalination

(SSD).

Basically,

solar-thermal

desalination

(STD)

can

be

categorized into direct and indirect solar desalination [18, 19]. SSD—a typical representative of the direct solar desalination techniques, is a simple assembly, comprising a basin, a transparent inclined cover, distillate collection channels and collectors, seawater inlet compartment, and a host of pipe connections. It uses direct solar radiation to heat up seawater in the basin. The water evaporates to the ambient enclosed atmosphere created between the contents of the basin and the transparent inclined cover. The water vapour condenses on the cover and rolls down the inclined plane into the distillate channel

and collected

as potable

water

[20-25].

For the indirect

STD,

on

the other

hand, solar energy is collected via a collector or a photovoltaic (PV) modules and stored as electrical energy in batteries, which can be coupled directly to a solar still to

A Concise Review of Major Desalination Techniques

159

enhance the thermal input through a heating device [26], or used as a hybrid with other forms of desalination techniques as the energy source [27]. Table | gives an overall comparison between the conventional desalination techniques discussed in the preceding sessions and solar-thermal desalination via solar still desalination. Considerations such as energy concerns, economics and costs, and technological sophistications and technical expertise favour the adoption of STD for low-income communities.

Table 1. Summary of special issues concerning conventional desalination techniques in comparison with solar-thermal desalination via solar still desalination Special issues Conventional desalination Solar-thermal desalination techniques

Energy and

+ Non-renewable and

environmental

environmentally impairing, as GHG emission is typical * Reliance on steady supply of refined energy or fossil fuel burning, if otherwise Installation requires high technological sophistications, as

concerns

Technology and technical expertise

plants components are huge

Sound technical expertise (though varies from a technique to another)

+ Renewable, eco-friendly and potentially cost-effective + Inconsistent availability and spatial variation of solar radiation

Installation requires moderate

technological sophistications * Modest technical expertise is sufficient for its operation and

maintenance

is needed for operation and maintenance Economics

and costs

+ High start-up and installation

capitals, as plant sizes are usually large + High operation and maintenance costs are inevitable, and operation

+ Low start-up/installation costs, making it suitable for low-income

communities + Almost insignificant operation and.

maintenance costs

could stop for days during repair or maintenance Productivity

and water quality

* Productivity is usually high, depending on plant capacity, while specific output depends on techniques * Good water qualities that meet. acceptable standards. However, the suitability to desalinate highsalinity water varies with technique

Development and commercial

status

+ Some techniques (such as RO, NF, MSF & MED)

are

commercially functional globally, while many others still domicile within the R&D domain

+ Low productivity, typically below 10 L per square meter per day. + Water quality meets acceptable standards. However, occasional contamination is not unlikely

because of low system sophistication. It can desalinate

high-salinity water

Itis still within the R&D domain.

But a number of functional units has been installed to serve households and small

communities

160

3

T. O. Abimbola et al.

Conclusions

Desalination techniques have been categorized in this study into membrane-based, thermal-based, combined membrane/thermal-based and miscellaneous techniques. Each of the techniques has been sufficiently entrenched in literature, but their main features and limitations were brought to focus in this review to serve as an easy and quick reference for ongoing and future studies. The membrane-based techniques are gaining rising-attentions in the desalination industry globally due to the advent of highpermeability membranes. However, their overall efficiency is hindered by fouling, concentration polarization, susceptibility to chlorine-attack and effectiveness over a low range of pH, which are distinctive characteristics of polymeric membrane and saline water interactions. The thermal-based techniques, on the other hand, are promising, as they are free from limitations associated with the membrane techniques. However, their popularity is hindered by low thermal-to-vapour conversion efficiency and the inherent green-house gas (GHG) emission when powered by fossil fuels. While the combined membrane/thermal-based techniques present great potential to desalinate seawater at fairly low costs, their success seeks to leverage the uninterrupted availability of lowgrade/waste heat for the affiliated thermal process. Some other techniques are, however, less common and use approaches different from the ones previously mentioned—they are classified as miscellaneous

techniques. Finally, solar-thermal desalination (STD) is

a thermal-based technique that benefits directly from the energy from the sun to desalinate seawater. Although the eco-friendliness and renewability of its energy source and its low installation, operation, and maintenance costs are some of the considerations that have eamed STD preference over other techniques, it would need extensive research attentions to productively compete with them for commercial

acceptance.

Acknowledgements. The authors are grateful for the supports provided by Universiti Teknologi PETRONAS for this study under the YUTP-FRG grant, cost-center: 015LC0-215. References 1. Kucera, J.: Desalination: Water From Water. Wiley, Hoboken (2019)

v

. Younos, T., Tulou, K.E.: Overview of desalination techniques. J. Contemp. Water Res. Edu. 132, 3-10 (2005) 3. Gary, A., Noreddine, G., Zhenyu, L., Lijo, F., Valladares, L.R., Thomas, M., Sabine, L.: Membrane-based seawater desalination: present and future prospects. Desalination 401, 16— 21 (2017). https://doi.org/10.1016/j.desal.2016.10.002

4. Li, Z., Afreen, S., Diaz, A-L., Venkatesh, N.: Towards sustainability in water-energy nexus: Ocean energy for seawater desalination. Renew. Sustain. Energy Rev. 82, 3833-3847 (2018). https://doi.org/10.101 6/.rser.2017.10.087 5. Elimelech, M., Phillip, W.A.: The future of seawater desalination: energy, technology, and the environment. Science 333(6043), 712-717 (2011). https://doi.org/10.1126/science. 1200488

A Concise Review of Major Desalination Techniques

161

. Singh, L.B., Victor, K., Raed, H., Nidal, H.: A review on membrane fabrication: structure,

properties and performance relationship. Desalination 326, 77-95 (2013). https://doi.org/10. 1016/j.desal.2013.06.016 . Lau, W. ., Lai, G.-S., Li, J., Gray, S., Yunxia, H., Misdan, N., Goh, P.-S., Matsuura, T., Azelee, LW., Ismail, A.F.: Development of microporous substrates of polyamide thin film composite membranes for pressure-driven and osmotically-driven membrane processes: a review. J. Ind. Eng. Chem. 77, 25-59 (2019). hitps://doi.org/10.1016/j,jiec.2019.05.010 . Nikolay, V.: Energy use for membrane seawater desalination — current status and trends. Desalination 431, 2-14 (2018). https://doi.org/10.1016/j.desal.2017.10.033 . Tufaa, R.A., Noviello, Y., Di Profio, G., Macedonio, F., Ali, A., Drioli, E., Fontananova, E.,

Bouzek, K., Curcio, E.: Integrated membrane distillation-reverse electrodialysis system for energy-efficient seawater desalination. Appl. Energy 253 (2019). https://doi.org/10.1016/). apenergy.2019.113551

. Deshmukh, A., Boo, C., Karanikola, V., Lin, S., Straub, A.P., Tot T. , Warsinger, D.M., Elimelech, M.: Membrane distillation at the water-energy nexus: limits, opportunities and

challenges. Energy Environ. Sci. 11, 1177-1196 (2018). https://doi.org/10.1039/c8ee00291f . Imane, C., Souad, A., Sébastien, V., Tijani, B.: Water desalination by forward osmosis: draw

solutes and recovery methods — review. Environ. Technol. Rev. 8(1), 25-46 (2019). https:// doi.org/10,1080/21622515.2019. 1623324

. Ponnivalavan, B., Abhishek, N., Tianbiao, H., Karimi Iftekhar, A., Dong, L.J., Peter, E., Praveen, L.: A review of clathrate hydrate based desalination to strengthen energy-water

nexus. ACS Sustain. Chem. acssuschemeng.8b01616

Eng. 6(7), 8093-8107

(2018).

https://doi.org/10.1021/

. Subban, C.V., Gadgil, A.J.: Electrically regenerated ion-exchange technology for desalination of low-salinity water sources, Desalination 465, 38-43 (2019). https: loi.org/10.1016/j. desal.2019.04.019 .. Tian, S., Niu, J., Kong, W., Lin, J., Hao, X., Guan, G.: An electrochemically switched ion exchange process with self-electrical-energy recuperation for desalination, In: Separation and Purification Technology, vol. 239 (2020). https://doi.org/10.1016/j.seppur. 2020.116521 . Wang, Z., Horseman, T., Straub, A.P., Yip, N.Y., Li, D., Elimelech, M., Lin, S.: Pathways and challenges for efficient solar-thermal desalination. Sci. Adv. 5 (2019) . Alkaisi, A., Mossad, R., Sharifian-Barforoush, A.: A review of the water desalination

systems integrated with renewable energy. Energy Procedia 110, 268-274 (2017). hitps:// doi.org/10.1016/.egypro.2017.03.138

. Dehghani, S., Date, A., Akbarzadeh, A.: Humidification-dehumidification desalination cycle.

In: Emerging Technologies for Sustainable Desalination Handbook, pp. 227-254 (2018) . Omar, B.: Direct and indirect solar-powered desalination processes loaded with nanoparticles: a review. Sustain, Energy Technol. Assess. 37 (2020). https://doi.org/10.1016/j.seta. 2019.100597 . Chandrashekara,

20.

21.

M., Avadhesh, Y.: Water desalination system using solar heat: a review.

Renew. Sustain. Energy Rev. 67, 1308-1330 (2017). https://doi.org/10.1016/}.rser.2016.08. 058 Riahi, A., Yusof, K.W., Sapari, N., Singh, B.S., Hashim, A.M.: Novel configurations of solar distillation system for potable water production. In: IOP Conference Series: Earth and Environmental Science, vol. 16 (2013) Ali, R., Emmanuel, O., Khamaruzaman, W.Y., Nasiman, S., Balbir, S.M.S., Mohamed, H.L-: The performance investigation of triangular solar stills having different heat storage materials. Int. J. Energy Environ. Eng. (2015). https://doi.org/10.1007/s40095-015-0185-x

162

T. O. Abimbola et al.

22.

Riahi, A., Yusof, K.W., Isa, M.H., Singh, B.S.M., Mustafa, Z., Ahsan, A., Ul Mustafa, M.

23.

R., Sapari, N., Zahari, N.A.M.: Potable water production using two solar stills having different cover materials and fabrication costs. Environ. Progr. Sustain. Energy, (2017). https://doi.org/10.1002/ep.1271810.1002/ep Dhivagar, R., Sundararaj, S.: A review on methods of productivity improvement in solar desalination. Appl. Mech. scientific.net/ AMM.877.414

24.

Mater.

877,

414-429

(2018).

https://doi.org/10.4028/www.

Indra, M., Satyapal, Y., Hitesh, P., Shivani, B.: A review on solar simple desalination technology to obtain potable water. Int. J. Amb. Energy 40(3), 335-342 (2017). https://doi. org/10.1080/01430750.2017. 1393776,

25.

26.

27.

Sapari, N., Ahmadan, N.A.M., Riahi, A., Uka, O.K.: The performance of trapezoidal glass cover solar still during monsoon period of tropical environment. Appl. Mech. Mater. 567, 161-166. (2014). https://doi.org/10.4028/www.scientific.net/AMM.567.161 Ali, R., Khamaruzaman, W.Y., Balbir, S.M.S., Mohamed, H.I., Emmanuel, O., Noor, A.M. Z.: Sustainable potable water production using a solar still with photovoltaic modules-AC heater. Desalinat. Water Trea. (2015). https://doi.org/10.1080/19443994.2015.1070285

Kasra, M., Mohammad, S., Kevin, E., Kody, P.: Hybrid concentrated solar power (CSP)-

desalination systems: 2019.114083

a review.

Desalination

468

(2019).

https://doi.org/10.1016/j.desal.

®

Check updatesfor

Impact of Treating Ammonia-Nitrogen Contamination from Chemical Fertilizer Plant Using Extended Aeration Activated Sludge System Mohammad Fakhuma Ubaidillah Bin Md Hafiz, Shamsul Rahman Bin Mohamed Kutty, and Shekhah Norafizah Binti Shekh Imaduddin Hakmi

Universiti Teknologi PETRONAS, 32610 Perak, Malaysia {mohammad_19001703, shamsulrahman, shekhah_22245}@utp. edu. my Abstract. A chemical fertilizer plant in Kedah, Malaysia produced high Ammonia-nitrogen content in the discharge outlet, exceeding the allowable standard limit (Standard B). The standard is based on Environment Quality Act (EQA) 1974, to ensure the integrity of water is maintained. At the plant, current wastewater treatment used is reed bed system, which is incapable to treat the excessive amount of Ammonia-nitrogen. Feasibility of extended aeration activated sludge process (ASP) was then discovered in this study. Objective of this study is to determine the impact of treating wastewater generated from the chemical fertilizer plant using bench scale extended aeration activated sludge system at 20% of contamination, diluted with domestic wastewater of an

average of 500 mg/L chemical oxygen demand concentration. Bench scale ASP was set up using 5000 mg/L initial biomass, a 20 L’ influent tank with a heavyduty mixer, connected to a pump via 10 mm tube at 0.1 rpm which is equivalent to 5 L of water infused into the aeration tank for 24 h consistently, and finally discharged into an effluent tank. Influent and effluent samples were monitored

for Total Suspended Solids (TSS), Chemical Oxygen Demand (COD), Nitrate (NO;"), Ammonia (NH), and Total Phosphorus (PO,*) while the mixed

liquor sample was monitored for Mixed Liquor Suspended Solids (MLSS) and Mixed Liquor Volatile Suspended Solids (MLVSS). Keywords:

1

Ammonia-nitrogen - Extended aeration - Activated sludge process

Introduction

A petrochemical plant in Kedah, Malaysia produced urea fertilizer that includes Ammonia-nitrogen by-products. Ammonia-nitrogen was known as one of pollutants in wastewater that carries toxic and hazardous substances. It was discovered that the limit of Ammonia-nitrogen produced at the plant exceeded the regulatory requirement. Environmental Quality Act (EQA) 1974, a Malaysian regulatory requirement has aligned two standards for effluent discharge: Standard A for discharge upstream of any

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 163-173, 2021. https://doi.org/10.1007/978-981-33-6311-3_19

164

M. F. U. B. Md Hafiz et al.

raw water intake, and Standard B for discharge downstream of any raw water intake (Malaysian Sewerage Industry Guideline 2009). The treatment system implemented at the plant was Reed Bed system that used bioremediation method. It has two acres Reed Bed plot, designed to treat 10.83 m*/hr of wastewater, equivalent to 260 m*/d inclusive of rainwater. The plot was divided into two stages with total residence time of seven days. Stage one was vertical feeding with residence time of four days and stage two was horizontal feeding with residence time of three days. It received effluent wastewater from three separate locations that was pumped to reed bed system namely, collecting pit at wastewater stripping unit (primary source), guard ponds

(secondary

source) and storm drain (secondary

source).

The treated water from reed bed was discharged to storm drain at final discharge point and guard ponds and flowed into Sungai Bongkok, Gurun Kedah. Bioremediation was initiated by soil microorganisms, physical and chemical properties of the soil, sand and gravel, and the plant themselves. Bioremediation process began when wastewater passed through the root zone of reeds. Treatment was achieved through naturally occurring soil microbes that broke down contaminants in the wastewater. Dissolved organic materials were taken up by the roots of the plants, while oxygen and food were supplied to the soil and underwater microorganisms through the same root system. Bacterial

oxidation

of

NH3

to

NO;

through

nitrification

relied

on

aeration

as

biodegradable organics. NO; was then removed by bacterial process of denitrification. Anaerobic conditions were needed, which could be found in the reed beds away from the root zone. Initially, wastewater stripping unit (WSU) was installed to remove Ammonia, urea and methanol in the wastewater using steam stripping. However, the WSU was not designed to meet the Clean Air Regulations 2014 and Industrial Effluent Regulations 2009 from Department of Environment (DOE). The unit imposed high operating cost and had significant design issues that causes ineffective treatment operation. Thus, the ‘WSU ceased operation. Presently, the plant employs pilot project to remove Ammonianitrogen by performing Performance Test Run (PTR). However, due to lack of working instrumentation associated with the PTR, elements of uncertainty on the system’s efficiency have been discovered. The volumes of effluent delivered to the system were inaccurate due to pumping logs and flow meters, the assumed capacity for the pump must be used, with no account for any loss in capacity, excessive pumping beyond the agreed timings, or delayed operations to the pump in any way affecting the theoretical pumped volume. In addition, weather has affected the volumes captured by the system. Minor rainfalls during the progress of the PTR have been discounted. However, rainfalls toward the end of the PTR exceeded the initial degradation limit, whereby rainfall intensity was almost equal to the volume of effluent received in the system. Based upon an extrapolated volume of 260 m*/d, with a target inflow of 850 mg/L Ammonia-nitrogen discharged from the system were in the range of 50 mg/L to 120 mg/L. The allowable discharge limit was 20 mg/L (Standard B — AMN effluent discharge to streams), thus the plant had discharged 3.5% to 14.11% higher than the allowable discharge limit. This shows that the reed bed system was ineffective in removing Ammonia-nitrogen accordingly. Thus, the feasibility of extended aeration

Impact of Treating Ammonia-Nitrogen Contamination Activated

Sludge

Process

(ASP)

to remove

Ammonia-nitrogen

in

165

wastewater

was

proposed in this study. Commonly, wastewater treatments that would be introduced in current markets were primary treatment, secondary treatment and tertiary treatment. Primary treatment removed settleable organic and inorganic solids from raw wastewater by sedimentation [1]. However,

the

wastewater

treated

in this stage

were

polluted

with dissolved

and

colloidal particles, unsatisfactory to be discharged into water bodies. These particles were micro-sized with poor settling ability. Therefore, they must be attached to each other to form sizable particles for separation process. Thus, secondary treatment was initiated and known as biological treatment. Several design considerations of ASP includes characteristics of influent wastewater and effluent wastewater, types of reactors, volumetric and organic loading for reactors, preferred oxygen quantity and aeration system to assist mixing process, microorganism nutrients requirement, environmental conditions of treatment plants and the necessary quantity of sludge for additional operations and disposal [2]. On top of that, some elements to be established prior to the design of ASP includes dimensions of reactor or typically known as aeration tank, power requirements for aeration process, volume and dimensions of secondary settling tank (clarifier) that includes hopper bottom for sludge collection. ASP was one of the most generic secondary treatment systems in the industry [3]. The activated sludge required microbial development in suspension mode to achieve biodegradation of organic materials in oxygenated environment. Thus, biological floc for solid separation would form in the clarifier tanks. The oxygen diffuser installed in the system was responsible to control aerobic condition in the reactor. It was a versatile system for nitrogen removal to enhance nitrification and denitrification processes. Parameters that impose significant impacts towards nitrification were rate of sludge wasted, BOD loading, MLSS and retention time. Extended period of sludge age became critical when this scenario takes place to encourage the conversion of Ammonia to Nitrate [4]. Nitrification would take place at sludge age greater than 96 h [5].

This system developed microorganisms to feed organic matter in the wastewater which

then

produced

nontoxic

run-offs,

to be

released

to water

bodies

[6].

These

microorganisms would grow through time and reached a chunking stage. During this stage, the chunks which were called floc increased in density and settled at the base of the reactor. Thus, when these particles settled, the upper section of water became clear with reduced contamination of organic matter and suspended solids.

2 2.1

Methodology Experimental Setup

The setup comprised 20 L aerated influent tank that was connected to aeration tank via 10 mm tube powered by the up-flow pump, and the overflow was discharged into an effluent

tank

(Fig.

1). One

aeration

tank

(reactor)

was

designed

named

as modified

reactor. Modified reactor was injected with raw wastewater from domestic sewage treatment plant (STP) on sampling day 1 until sampling day 17, and addition of 20% urea fertilizer wastewater on sampling day 18 until sampling day 24.

166

2.2

M. F. U. B. Md Hafiz et al.

Preparation of Synthetic Wastewater

Synthetic wastewater was used as the baseline wastewater to the reactor to manage the consistency of influent wastewater quality and its organic loadings. A high protein dog meal was used to stimulate raw municipal wastewater. The dog meal was grinded thoroughly until it reached powder consistency. Then, 13 g of dog meal was mixed with one liter of distilled water. The solution was poured with 20 L of influent wastewater from domestic STP and kept suspended using pneumatic mixer (Stir-Pak heavy duty mixer model 50007-05). Several tests were conducted to verify the characteristics of synthetic wastewater. 2.3

Experimental Methodology

The designed reactor was fabricated with 5 mm thick acrylic glass, and six tube diffusers were installed to ensure minimum D.O. of 2 mg/L. During the operation stage, the reactor was set up in the laboratory using biological biomass from the domestic STP as the starting biomass. Influent wastewater was pumped into the reactor continuously at the rate of 5 L/day by using peristaltic pump silicone tubing by Longer Pump. The reactor was operated at extended aeration (SRT = 40 days, design MLSS = 3000 mg/L). Acclimatization was achieved after 17 sampling days to stabilize the biomass, and the sludge age was controlled through daily sludge recycling and wasting. Influent and effluent samples were collected daily at a pre-determined time and the performance of reactor was monitored for every two days continuously.

Reactor Peristaltic Pump

Ht

co Raw wastewater (20% urea fertilizer was ‘added on sampling day 18 onwards)

Effluent from Reactor

Fig. 1. Experimental setup of bench-scale activated sludge process

Impact of Treating Ammonia-Nitrogen Contamination

3

167

Result and Discussion

This

section

Mixed

is dedicated

to present

Liquor Suspended

(MLVSS),

the

Solids (MLSS),

Chemical Oxygen Demand

results

Mixed

for Total

Solids

(TSS),

Liquor Volatile Suspended

Suspended

Solids

(COD), Nitrate (NO;—), Ammonia

(NH3_), and

Total Phosphorus (PO,*).

TSS concentration (mg/L)

Based on Fig. 2, the raw wastewater shows greater amount of TSS in comparison to effluent from reactor throughout all sampling days. This shows some portion of suspended solids were removed in the reactor, despite the contamination of 20% urea fertilizer wastewater included at sampling day 18 until sampling day 24. The highest percentage of TSS removal was recorded on sampling day 14, at 88.9% while the lowest TSS removal was recorded at sampling days 2, 9, and 18 at the range of 13.0% to 15% removal. Despite the reduction of TSS values in effluent from reactor, it could be analyzed that greater TSS removals took place at sampling days | until 17. It was observed only several sampling days recorded low reduction of TSS during this period which were sampling day 2 and 9. The difference of TSS removal before adding 20% urea fertilizer wastewater and after may be induced by the suspended solids that were contained in the urea fertilizer wastewater.

193 —*— Fig. 2.

5

7

Raw wastewater

9

MW 13 1S Sampling Days ~~

17

19 2 23

Effluent from reactor

Comparison of TSS concentrations between raw wastewater and effluent from reactor

According to Fig. 3, the average MLSS concentration for sampling day 1 until sampling day 17 was 3042 mg/L while the average MLSS concentration for sampling day 18 until sampling day 24 was 2667 mg/L. For sampling day 1 until 17, raw wastewater from domestic STP was fed into the reactor while for sampling day 18 until sampling day 24, 20% of urea fertilizer was included in the raw wastewater and was fed into the reactor. The average MLSS concentration between raw wastewater feeding (sampling day 1 until 17) and raw wastewater with 20% urea fertilizer wastewater feeding (sampling day 18 until 24) was reduced by 12.3%.

168

M. F. U. B. Md Hafiz et al.

Evidently, the addition of 20% urea fertilizer wastewater was detrimental for the microbes in the mixed liquor solution. Interestingly, the MLSS concentrations from sampling day 18 until sampling day 19 were increasing from 2667 mg/L to 4667 mg/L. which was equivalent of 42.9%, and then it proceeded to decrease on sampling day 20 until sampling day 24 from 4000 mg/L to 1000 mg/L which was 75% of concentration reduction. The average concentration of volatile suspended solids recorded for sampling day 1 until sampling day 17 was 2313 mg/L while for sampling day 18 until sampling day 24 was 1292 mg/L. The average MLVSS concentration between raw wastewater feeding and raw wastewater with 20% urea fertilizer wastewater feeding was declined by 55.9%. It was observed the lowest volatile suspended solids recorded was on sampling day 24 at only 333 mg/L while the highest concentration recorded was 5000 mg/L on sampling day 8. The significant reduction by 93.3% proves that the microbes could not sustain the 20% urea fertilizer contamination in raw wastewater.

6000

&& 8Ss

Concentration, mg/L

5000

3000 2000 1000

13°95

7

—=— MLSS

9

MW 13 IS 17 19 21 23 Sampling Days ‘a

MLVSS

Fig. 3. Comparison between MLSS and MLVSS concentrations COD concentrations in mg/L of raw wastewater from domestic STP were recorded from sampling day | until sampling day 17, with an average of 507.9 mg/L. 20% of urea fertilizer wastewater was added into raw wastewater from sampling day 18 until sampling day 24 (Fig. 4). On sampling day 18, when 20% of urea fertilizer wastewater was added into raw wastewater, it has resulted in COD concentration inclination by 62.3%, from 488.2 mg/L to 1294.0 mg/L. During the entire phase of urea fertilizer wastewater addition into raw wastewater, the average of effluent COD concentration was 1326.54 mg/L. Meanwhile, the average of COD concentration for effluent from reactor for sampling day | until sampling day 17 was 37.9 mg/L. The recorded effluent concentration met Standard B limit which was 200 mg/L, signified that the activated sludge process was effective to treat raw wastewater from domestic STP.

Impact of Treating Ammonia-Nitrogen Contamination

169

COD Concentr: ion (mg/L)

However, when 20% of urea fertilizer wastewater was added into raw wastewater from sampling day 18 until sampling day 24, the effluent from reactor recorded an average of 886.8 mg/L of COD concentration. This has shown that treatment process that supposedly occurred in the reactor was failing. Specifically, the oxidation of organic matter, nitrification and endogenous respiration were all depleting when the raw wastewater was contaminated with 20% of urea fertilizer wastewater. Evidently, 20% of urea fertilizer wastewater was extremely excessive for activated sludge process to take place potently.

1800 1600 1400 1200 1000 800 600 400 200 0

—e— Fig. 4.

Comparison of COD

ca fries ‘dition

2

4

6

Raw wastewater

8 10 12 14 16 18 20 22 24 Sampling Days —----@--- Effluent from reactor

concentrations between raw wastewater and effluent from reactor

Nitrate concentrations of raw wastewater from domestic STP for sampling day 1 until sampling day 17 were stabilized at an average of 57.9 mg/L because nitrifiers were growing, and induced nitrification when MLSS concentration exceeded 1000 mg/L since sampling day 1. The Nitrate limit for the treatment plant was 50 mg/L. Evidently, from sampling day | until sampling day 17, Fig. 5 shows that the Nitrate concentrations were relatively low, but it did not meet the limit by 15.8%, equivalent to 7.9 mg/L in surplus. Before the addition of 20% urea fertilizer wastewater, nitrification was occurring because Nitrate was consistently produced as shown in Fig. 4. On sampling day 18, when the addition of 20% urea fertilizer wastewater took place, it was observed that the Ammonia limit had failed. This could be justified by depletion of nitrification, that has caused Ammonia to be untreated and consequently, Nitrate was untreated as well because raw wastewater of domestic STP reported high concentration of Nitrate throughout sampling day 18 until sampling day 24. Seasonal nitrification occurred on some sampling days because Nitrate levels in effluent from reactor were higher than raw wastewater. For example, on sampling day 19, 22, 23 and 24. Possibly, there might be some anoxic condition in the reactor because some sludge was not aerated or idle in the tank. Even though nitrification took place on these sampling days, it was insignificant as Nitrate and Ammonia limits were not achieved accordingly.

170

M. F. U. B. Md Hafiz et al.

Nitrate Concentration (mg/L)

12000 10000 8000 6000 4000 2000

13°5

7

—e— Raw wastewater Fig. 5.

9

Wt 13 15 17 19 21 23 Sampling Days ----@--- Effluent from reactor

Nitrate concentrations between raw wastewater and effluent from reactor

The Ammonia limit was 20 mg/L, whereby both domestic STP raw wastewater and effluent from reactor from sampling day 1 until sampling day 17 met the Standard B limit at 14.4 mg/L and 6.3 mg/L. Meanwhile, the concentrations of Ammonia with addition of 20% of urea fertilizer wastewater into raw wastewater from sampling day 18 until sampling day 24, collected from raw wastewater and effluent from reactor recorded 2414.2 mg/L and 5176.0 mg/L. In average, the extreme increment of Ammonia concentration with 20% urea fertilizer contamination were 99.4% and 99.9% higher than raw wastewater without urea fertilizer wastewater contamination. Impact of Ammonia in effluent was observed on sampling day 18, whereby it began to spike at 205 mg/L. The result has proven that the Ammonia rate exceeded the limit. However, Fig. 6 shows that Ammonia is higher in effluent than the influent for sampling day 18 onwards. This is justified by the composition of urea fertilizer wastewater that contains urea. Ureases would undergo enzymatic hydrolysis whereby it breaks down

into carbon dioxide and Ammonia

(1) and (2) [7]. Conclusively, extended aer-

ation activated sludge process could not remove contamination of urea fertilizer and Ammonia from wastewater but further escalated Ammonia contamination, at maximum level of 9280.7 mg/L on sampling day 24.

HN

© | “NN

NH:

© |) HO

wN~

NH

©

1,0 ——+ urease,

xn +

——

ww +00.

Ho’

'Ni

a @)

Impact of Treating Ammonia-Nitrogen Contamination

2S o 2S a &
WPA Partners Sdn. Bhd., No. 31-4, Block C2 Dataran Prima, Jalan PJU 1/39, Petaling Jaya, Malaysia weijing. lee@hotmail. com

> Fundamental and Applied Sciences Department, Universiti Teknologi PETRONAS, Seri Iskandar, Perak Darul Ridzuan, Malaysia beh. hoeguan@utp. edu. my

+ Ethnobiology Research Group, Research Center for Biology, Indonesian Institute of Sciences (LIPI), Bogor, Indonesia wawan. [email protected] 5 M. Kumarasamy College of Engineering, Thalavapalayam, Karur, India hariharanchenu@gmail. com, balamuruganp. civil@mkce. ac. in

Abstract.

Coagulation-flocculation

is the key

process of water

treatment

in

water turbidity removal. However, the usage of inorganic coagulants has raised

awareness of researchers due to the threat possess on human health and environment. Therefore, studies in natural coagulant and modification of natural coagulant have raised various interest of researchers. Natural coagulants are widely available and non-toxic that can be obtained from animal, fungi, bacteria,

and plant. Preliminary studies of plant-based natural coagulant aid, Hylocereus undatus in synthetic turbid water was also conducted. Remarkably, the usage of A. undatus with the mixture of ferric chloride has achieved a turbidity removal of 93.03%.

Keywords: Biocoagulant - Coagulation - Flocculation - Polysaccharide 1

Introduction

‘Water is an essential need for human beings and other life forms on earth. However, in third world countries as well as developing countries, around 80% of the illness occurred are closely linked to the consumption of contaminated water [1]. Therefore, it

has raised awareness of researchers to provide a safe and affordable water which is on par with the Sustainable

Development

Goals

(SDG)

indicator 6.1.

To provide clean water that are pathogenic free, non-toxic, odourless, tasteless and colourless [2], which fulfil the basic drinking quality, water treatment is required.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 187-194, 2021. https://doi.org/10.1007/978-98 1-33-63 1 1-3_22

188

J.-S. Lau et al.

Coagulation-Flocculation is one of the most important processes in water treatment for the production of clean water.

2 2.1.

Coagulation-Flocculation Coagulation-Flocculation Process

Coagulation-flocculation process is used to separate the suspended solid particles from water. By removing the suspended solid particles and turbidity of water, there will be a significant increase in water quality as it is able to remove the harmful contaminants absorbed on the suspended solid particles [3]. Coagulant is required during the coagulation-flocculation process by acting as a mechanism to neutralise the negatively charged particles in the turbid water to form microflocs and flocs. Aluminium Sulphate (alum) and Iron (III) Chloride are conventional inorganic coagulant that are commonly

used in the industry due to its high efficiency and low cost [4]. However, various shortcomings have been identified in recent studies of inorganic coagulant usage such e

Alum

usage

Alzheimer’s

e e 2.2

of coagulant Disease

has

been

categorised

as

a contributing

factor

for

[5].

Alum generated sludge are considered as scheduled waste where special treatment and proper disposal method are required which leads to a higher cost [6]. Production of inorganic coagulant as contributing factor to ozone depletion [7]. Plant-Based Coagulant

Natural coagulant alternatives are highly anticipated as they are mostly non-toxic and widely available in the world. Natural coagulant can be obtained from various sources, which include bacteria, fungi, animals and plants. Based on recent studies, plant-based coagulant has shown positive results and are usually able to obtain from the seeds [1] or the leaves [8] of the plant through various extraction methods such as solvent extraction

and precipitation [9] as well as dry and ground process [10]. Mucilage From Plants. Mucilage is a type of gelatinous liquid that contains high polysaccharides and can be extracted from plants or animals which are widely available and harmless to both human and environment. Mucilage is also commonly used in the food industry as they have unique properties such as structuring, texturizing, emulsifying and thickening as well as great dietary aspect [11]. There are various studies of mucilage from plants for water treatment as well, includes okra seeds [12], garden cress [13] and cactus (Opuntia ficus-indica)

[14].

Pectin from Fruit Peels. Pectin is polysaccharides called polyacids or anionic polymers that are able to obtain from fruit peels [15]. Usage of fruit peels are highly beneficial as they are widely available, cost-effective, biodegradable, non-toxic as well as eco-friendly. Studies have also shown various positive results obtained from fruit peels such as peels or pith from Orange

[15], Mango

[16] and Banana

[17] (Table

1).

Emerging Coagulant in Water Treatment

189

Table 1, Summarised example of various plant-based coagulant Natural

Part

coagulant Okra

Test

Optimum results

Seeds.

Raw

water

Turbidity

(Mucilage) Garden

17NTU —

3.8NTU

[12]

removal Seeds.

Cress

Turbid

Turbidity

water

(Mucilage)

Opuntia (ficus-

Reference

extracted — sample

>92%

[13]

93.33%

[14]

removal

(Kaolin)

Cactus pad

Raw tailing

indica

Turbidity Removal

pond water

(Mucilage) Sesbania Seed Gum

Seed

Raw river water

Turbidity removal

Orange (Pectin)

Piths and peels

Turbid water (Bentonite)

Turbidity removal

Mango

Peels

Raw water

| Turbidity

98.3%

[18]

(Mucilage)

(Pectin) Banana

(Pectin)

2.3.

SOONTU

—

1.89NTU

92.7%

[15]

[16]

removal Piths

Raw

river

water

Turbidity

279NTU

— 4NTU

[17]

removal

Modification on Plant-Based Coagulant

Grafting. Polymer grafting is to improve the flocculation capacity of the natural coagulant [19]. Various grafting methods can be adopted to graft natural coagulants which includes high energy radiation method, microwave-based method chemical free radical

initiator and UV-radiation

based method

[20]. Microwave

grafting method

is

commonly used as it has maximum reproducibility, easy operation, less time consuming and able to precisely control the percentage of grafting requirement [19]. Microwave grafting method will first dissolve both the desired material into water and mix them well prior to inserting them in the microwave. After heating the mixture in the microwave, the solution is dried with an oven and will then be pulverised into powder form [21]. According to several researchers’ results, grafted natural coagulant with a higher percentage of grafting efficiency will eventually lead to better flocculation results [19, 21] (Table 2).

Crosslinking. Polymer crosslinking has great potential in the use of coagulationflocculation process [25]. Usage of crosslinking modification in natural coagulant aims to reduce the cost as well as improve the coagulation-flocculation efficiency of an existing coagulant in water treatment process [26]. Existing bonds between the molecules are expected to be reinforced through the crosslinking modification of natural coagulant [27]. The most optimum condition of crosslinking results are

190

J.-S. Lau et al.

Table 2, Summarised example of various grafted natural coagulant Grafted natural coagulant

Polyacrylamide grafted Carboxymethy! guar

Test sample

| Municipal sewage

Polymethylmethacrylate

(21)

| 58%

(221

| S8NTU > 14NTU

(20)

Turbidity removal

| 187NTU — 64NTU

(23)

| Turbid

Turbidity

| 800NTU ~ = 9NTU

[24]

water

removal

wastewater

| Turbid

Turbidity

water

removal

PMMA)

Polyacrylamide grafted starch (St-g-PAM)

| Turbidity removal

Reference

| 64NTU — 9NTU

gum (CMG-g-PAM) grafted guar (GG-g-

Optimum results

(Kaolin)

| Turbid

Turbidity

water

removal

(Kaolin)

Polymethylmethacrylate | Turbid grafted psyllium (Psy-g- | water PMMA)

2-methacryloyloxyethyl trimethyl ammonium

(Kaolin)

chloride grafted lentil

extract (LE-g-DMC)

(Kaolin)

governed by different parameters such as dose ratio of the materials, dose ratio of crosslinker, reactive temperature and reaction time during the crosslinking process [26]. There are various types of cross linking agents also known as crosslinkers include borox

3 3.1.

[25], epichlorohydrin

[27] and glutaraldehyde

[28].

Future Perspective Environmental Impact

Natural coagulant in water treatment process is highly beneficial to both human health and environment as it does not possess any threat. Usage of natural coagulant is able to replace the usage of conventional inorganic coagulant. With that, it will significantly reduce the carbon footprint to the environment during the production of inorganic coagulant. Moreover, sludge produced from natural coagulant are able to dispose through normal disposal as sludge produced as they are biodegradable and harmless to the environment. 3.2.

Cost

Marketing of natural coagulant into the industry highly dependent on the feasibility study of cost and economics of the product. Usage of natural coagulant has also shown, significantly reduce in cost of coagulant as well as cost of sludge disposal [29]. Cost of natural coagulant is usually lower as they are widely available on earth or may be

Emerging Coagulant in Water Treatment

191

obtained from food waste. Moreover, sludge produced from natural coagulant can also be disposed into the landfill or acts as fertilizer which reduces the cost as well.

4

Preliminary Study on Kaolin Suspension with Hylocereus Undatus

Hylocereus undatus or better known as pitahaya fruit peel is obtained and used as source of natural coagulant in the coagulation-flocculation water treatment process. the preliminary study, simple extraction, characterisation, and optimisation of natural coagulant H. undatus is conducted. For the extraction of mucilage from undatus peel, the aqueous extraction method

Dry pectin an oven and grind them to powder Fig. 1.

Boilthe powder in water for extraction

is conducted

Use muslin cloth to filter the solution

[13] (Fig.

Precipriate the filtered water with Ethanol

the In the H.

1).

Dry the precipitate in the oven overnight

Extraction process for H. undatus peel through aqueous extraction method.

Preliminary Study in Synthetic Turbid Water. Coagulation-flocculation process performance is dependant and controlled by several factors and parameters which include type of coagulant, dosage of coagulant, pH, mixing speed and settling time. In the preliminary study, the screening factors of H. undatus as natural coagulant were considered

in terms

of pH,

dosage

of ferric chloride

(FeCl;)

and

dosage

of natural

coagulant which are the variable chemical parameter for the coagulation-flocculation process. From the result obtained, the turbidity removal has a higher efficiency in acidic condition as compare

to alkaline condition

Table 3.

Run — pH__|

Turbidty removal for turbid water (Kaolin)

Concentration of Fe**

(mg/L)

(Table 3).

| Concentration of natural

coagulant (mg/L)

Turbidity

removal

1

9

15

0.01

39.68

2 3

9 3

15 15

0.1 0.01

38.52 84.83

4

3

15

0.1

93.03

192

5

J.-S. Lau et al.

Conclusion and the Way Forward

Plant-based natural coagulant as an alternative for conventional coagulant is relatively feasible in coagulation-flocculation process. Various modification technologies can also be practiced to improve the efficiency of natural coagulant and shorten the time taken for coagulation-flocculation process. Plant-based coagulant has shown positive results which includes the results obtained from Hylocereus undatus peel as a preliminary study for natural coagulant in this study. Furthermore, the study of natural plant-based coagulant still has room for studies as limited work has been conducted in terms of experimenting the natural coagulant in a large scale as well as its optimisation in large amount. Therefore, industrial scale testing can also be conducted to understand the efficiency of the natural coagulant once the results in laboratory testing is justified. In addition, cost consideration for large-scale production of natural coagulant should also be taken into account for the analysation between the conventional coagulants. All in all, natural coagulant benefits both human health and environment as well as improve the coagulation-flocculation efficiency and provide a cost-effective alternative for coagulation-flocculation process in the water treatment industry. Exploration of natural coagulant in various wastewater treatment such as agriculture wastewater and textile dye will also be anticipated.

Acknowledgement. This research was funded by PETRONAS through YUTP grant (015LCO169). The authors would love to appreciate and acknowledge technical assistance of Mdm. Norhayama Bt Ramli as well as Universiti Teknologi PETRONAS for providing required facilities for experiment. Also, author Jia-Shen Lau gratefully acknowledge Siong-Chin Chua for his guidance and proofreading of manuscript. References

v

1. Ernest, E., Onyeka, O., David, N., Blessing, O.: Effects of pH, dosage, temperature and mixing speed on the efficiency of water melon seed in removing the turbidity and colour of Atabong River, Awka-Ibom State, Nigeria. Int. J. Adv. Eng. Manag. Sci. 3, 427-434 (2017) . Yongabi, K.A.: Bio-coagulants for water and waste water purification. Int. Rev. Chem. Eng.

2 (2010)

3. Mulligan, C.N., Davarpanah, N., Fukue, M., Inoue, T.: Filtration of contaminated suspended solids for the treatment of surface water. Chemosphere 74, 779-786 (2009) 4. Malik, Q.H.: Performance of alum and assorted coagulants in turbidity removal of muddy

water. Appl. Water Sci. 8 (2018) 5. D’Haese,

P.C., Douglas, G., Verhulst, A., Neven, E., Behets, G.J., Vervaet, B.A., et al.:

Human health risk associated with the management of phosphorus in freshwaters using lanthanum and aluminium. Chemosphere 220, 286-299 (2019)

6. Environmental Quality (Scheduled Wastes) Regulations (2005) 7. Vince, F., Aoustin, E., Bréant, P., Marechal, F.: LCA tool for the environmental evaluation of potable water production. Desalination 220, 37-56 (2008) 8. Benalia, A., Derbal, K., Panico, A., Pirozzi, F.: Use of acorn leaves as a natural coagulant in a drinking water treatment plant. Water 11, 57 (2018)

Emerging Coagulant in Water Treatment

. Zaid, A.Q., Ghazali, S.B.: Dataset on phys

193

chemical properties of particle-sized Moringa

oleifera seed cake and its application as bio-coagulants in water treatment application. Chem.

Data Collect. 24 (2019) ). de Jesus,

Victor Cruz, P., Adair Pacifico, J., Santos Silva, A.: Removal

of turbidity,

suspended solids and ions of Fe from aqueous solution using okra powder by coagulationflocculation process. Am. J. Water Resour. 1, 20-24 (2014)

. Soukoulis, C., Gaiani, C., Hoffmann, L.: Plant seed mucilage as emerging biopolymer in food industry applications, Curr. Opin. Food Sci. 22, 28-42 (2018) . Mishra, S., Singh, S., Srivastava, R.: Okra seeds: an efficient coagulant. Int. J. Res. Appl. Sci Eng. Technol. (IRASET) 5 (2017) . Lim, B.-C., Lim, J.-W. . Ho, Y.-C: Garden cress mucilage as a potential emerging biopolymer for improving turbidity removal in water treatment. Process Saf. Environ. Prot.

119, 233-241 (2018) . Wan, J., Chakraborty, T., Xu, C., Ray, M.B.: Treatment train for tailings pond water using Opuntia ficus-indica as coagulant. Sep. Purif. Technol. 211, 448-455 (2019)

. Kebaili, M., Djellali, S., Radjai, M., Drouiche, N., Lounici, H: Valorization of orange industry residues to form a natural coagulant and adsorbent. J. Ind. Eng. Chem. 64, 292-299 (2018)

. Zaidi, N.S., Muda, K., Loan, L.W., Sgawi, M.S., Abdul Rahman, M.A.: Potential of fruit peels in becoming natural coagulant for water treatment. Int. J. Integr. Eng. 11 (2019)

. Kakoi, B., Kaluli, J.W., Ndiba, P., Thiong’o, G.: Banana pith as a natural coagulant for

polluted river water. Ecol. Eng. 95, 699-705 (2016) . Chua, S.-C., Chong, F.-K., Malek, M., Ul Mustafa, M.R., Ismail, N., Sujarwo, W., et al.: Optimized use of ferric chloride and Sesbania Seed Gum (SSG) as sustainable coagulant aid for turbidity reduction in drinking water treatment. Sustainability 12, 2273 (2020)

. Nandi, G., Changder, A., Ghosh, L.K.: Graft-copolymer of polyacrylamide-tamarind seed gum: .

Synthesis,

characterization

and

evaluation

of

flocculating

potential

in

peroral

paracetamol suspension. Carbohydr. Polym. 215, 213-225 (2019) Mishra, S., Mukul, A., Sen, G., Jha, U.: Microwave a: ted synthesis of polyacrylamide grafted starch (St-g-PAM) and its applicability as flocculant for water treatment. Int. J. Biol.

Macromol. 48, 106-111 (2011) . Pal, S., Ghorai,

S., Dash,

M.K.,

Ghosh,

S., Udayabhanu,

G.:

Flocculation properties of

polyacrylamide grafted carboxymethyl guar gum (CMG-g-PAM) synthesised by conventional and microwave assisted method. J. Hazard Mater. 192, 1580-1588

(2011)

. Mishra, S., Sen, G.: Microwave initiated synthesis of polymethylmethacrylate grafted guar (GG-g-PMMA), characterizations and applications. Int. J. Biol. Macromol. 48, 688-694 (2011) . Mishra, S., Sinha, S., Dey, K.P., Sen, G.: Synthesis, characterization and applications of polymethylmethacrylate grafted psyllium as flocculant. Carbohydr. Polym. 99, 462-468 (2014) . Chua, S.C., Chong, F.K., Mustafa, M.R.U., Kutty, $.R.M., Sujarwo, W., Malek, M.A., et al.: Microwave radiation-induced grafting of 2-methacryloyloxyethyl trimethyl ammonium chloride onto lentil extract (LE-g-DMC) as an emerging high-performance plant-based grafted coagulant. Sci. Rep. 10, 1-3 (2020) . Thombare, N., Jha, U., Mishra, S., Siddiqui, M-Z.: Borax cross-linked guar gum hydrogels as potential adsorbents for water purification. Carbohydr. Polym. 168, 274-281 (2017) . You, L., Lu, F., Li, D., Qiao, Z., Yin, Y.: Preparation and flocculation properties of cationic starch/chitosan crosslinking-copolymer. J. Hazard Mater. 172, 38-45 (2009)

194 27.

J.-S. Lau et al. Yusoff, M.S., Aziz, H.A., Zamri, M., Suja, F., Abdullah, A.Z., Basti, N.E.A.: Floc behavior and removal mechanisms of cross-linked Durio zibethinus seed starch as a natural flocculant for landfill leachate coagulation-flocculation treatment. Waste Manag. 74, 362-372 (2018)

28. Lu, Y., Wang, Z., Ouyang, X.K., Ji, C., Liu, Y., Huang, F., et al.: Fabrication of cross-linked chitosan beads grafted by polyethylenimine for efficient adsorption of diclofenac sodium from water. Int. J. Biol. Macromol. 145, 1180-1188 (2020) 29. Chua, M., Chong, S., Ho: Red lentil (lens culinaris) extract as a novel natural coagulant for turbidity reduction: an evaluation, characterization and performance optimization study.

Water 11 (2019)

®

Check updatesfor

Eco-composite Porous Concrete Drainage Systems: An Alternative Mitigation for Urban Flood Management Feroz Hanif Mohamed Ahmad'**), Mohamad Hidayat Jamal’, Abdul Rahman Mohd. Sam, and Nuryazmeen Farhan Haron*

' Faculty of Civil Engineering, Universiti Teknologi Malaysia, Johor Bahru, Malaysia ferozhanifahmad@gmail. com

> Center for Coastal and Ocean Engineering (COED, Faculty of Civil Engineering, Universiti Teknologi Malaysia, Johor Bahru, Malaysia mhidayat@utm. my

> Construction Research Centre (CRD), Faculty of Civil Engineering, Universiti Teknologi Malaysia, Johor Bahru, Malaysia abdrahman@utm. my + Water Resources and Environmental System (WRES), Faculty of Civil

Engineering, Universiti Teknologi MARA, Shah Alam, Malaysia neemzay@yahoo. com

Abstract. Nowadays, there are several problems cause by excessive of stormwater runoff, such as problem on drainage systems, flash floods, polluted water increase, and the environmental issue. Thus, this paper presents on eco-

composite porous concrete using rice husk ash (RHA) and effective microorganisms (EM) as a cement replacement, as green technology

development in urban drainage systems. For example, and EM-CSD (Effective Microorganisms Composite technology of drain covers have been implemented Alam, Selangor. It’s able to reduce flash floods, and

and sustainable

drainage cover, EM drain, Scupper Drain). This new in several areas in Shah also reducing the drainage

vandalism. Eco- composites, which are obtained from recyclable waste materials, such as RHA, egg shells, and many more; combination with EM in porous

concrete are designed, for ecological sustainability development purposes, mainly as part of urban stormwater managements. Thus, the eco-composites porous concrete quality is good in terms of concrete strength, workability, porosity and so on.

Keywords: Eco-composite - Rice husk ash (RHA) - Effective microorganisms (EM) - Porous concrete - Drainage systems 1

Introduction

The increasing of urbanization rapidly has influenced, specifically to cities, and also public, including researchers and engineers in order to discover in several ways to cut the number of impervious surfaces and to involve with stormwater management in an

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 195-201, 2021. https://doi.org/10.1007/978-98 1-33-63 11-3_23

196

F. H. M. Ahmad et al.

environment friendly way and sustainable elements. Regarding on this issue, several researches related to the urban storm water management using pervious concrete either for drainage system or in parking areas have been conducted in many countries around the world with variety of purposes, such as to recognize the utilisation of pervious concrete in order to decrease the excessive stormwater runoff and to overcome water pollution, to develop a new techniques in using pervious concrete for paving roads, or any non-structural components, to apply and enhance the sustainable ways to recharging groundwater. The porous concrete in the drainage system, specifically for urban storm-water management, mainly contribute to solve the flash flood. It is designed with the mix of concrete and a cement replacement; which have the element waste material that being recycled, for example, rice husk ash (RHA), egg shells, wood waste ash, and many more. There are some benefits of RHA, like strength improvement, minimise the material costs due to reduction on cement usage, durability characteristics, environmental benefits to the removal of RHA waste where the wastes have been identified to use in concrete mixture. Meanwhile, the effective microorganisms (EM) that was founded in the early

1980s

by Dr. Teruo Higa, a Horticultural Professor from Japan are added as admixtures in porous concrete drainage system as the EM have many benefits, such as improved the strength and the workability of the concrete, as investigated by several researchers [1— 4]. EM is a set of co-existing microorganisms in a culture solution, where it is naturally available, not genetically modified, environmentally friendly, and display as coexistence and co-prosperity. However, there is lack of significant research using eco-composite (RHA) and EM in porous concrete for drainage system. This type of waste materials are being used as construction and building materials such as RHA usage by [5-7], egg shells by [7-9], sawdust ash (SDA)

[10], and many

more.

This paper presents the preliminary study on eco-composite using RHA and EM in porous concrete for drainage system, focusing on drainage cover, EM drain and EMCSD as an alternative for urban flood management systems.

2

Problem Statement

In order to overcome a few issues involving drainage system around Shah Alam, Selangor (i.e. flash flood issue, stagnant water in the drain leads to the fertilization mosquitos, and vandalism of the cover slab, which cause a safety issue for public), the pervious surfaces need to be installed at the certain areas. Rainfall will infiltrate more in the pervious surfaces as compared to the impervious surfaces. The improper drainage system, such as, clogging, sedimentation, etc., is one of the reasons that the flash flood occurs. Hence, the eco-composite, such as, RHA, egg shells, fly ash, saw dust ash, etc., will be used as the porous media in the concrete drainage system as part of best management practices in order to manage the urban stormwater. Meanwhile, as referring to the proposed study area in the small catchment at Seksyen 28, Shah Alam, the flash flood have been occurred so many times as the numerous of the impervious surfaces at parking lots, roads, drainage systems, and also

Eco-composite Porous Concrete Drainage Systems along the pedestrian areas. Besides, improper drainage conditions floods. Figure 1 shows the catchment of the study area.

197

also lead to the

Fig. 1. Small catchment at Seksyen 28, Shah Alam, Selangor as the proposed study area (Source: Google Earth) Figure 2 presents the current condition of drainage system some part at Seksyen 28, Shah Alam, Selangor indicates that, there are stagnant water conditions of certain part of the walkway along the public and facilities area due to a number of impervious pavement leads to reduce on infiltration rate. Other than that, rubbish and also sedimentation in drainage system trigger to water clogging and water stagnant, thus leads to increase in breeding place of Aedes mosquito.

Fig. 2. Current condition of drainage system in Seksyen 28, Shah Alam 3

Eco-composite Porous Concrete Drainage Systems

There are several aspects that being considered in order to get the optimum design of this composite porous concrete, besides the reason on choosing the RHA and EM in this drainage systems. For example, from Table | below shows the comparison in terms of the cost between using existing drain design and using eco-composite porous concrete. From Table |clearly indicates the cost saving when using the eco-composite porous concrete, specifically for EM-CSD.

198

F. H. M. Ahmad et al.

Table 1. Cost comparison between the existing drainage systems and the EM-CSD Type of drain | Cost of labour Cost of machinery Existing 4 persons x RM80.00/day x 2/month = RM | 1 x RMS550.00/day x 2/month= RM drainage system | 640.00 1,100.00 EM-CSD 1 person x RMB80.00/day x 2/month=RM 0 x RM80.00/day x 2/month = RM 160.00 0.00 Cost saving | RM 480.00 RM 1,100.00 using EM-CSD Besides, several laboratory testings have been conducted in order to define in terms of the strength, and the workability of the mixture as shown in Fig. 3 to Fig. 5 below. From all the tests, the results show the samples with EMC (Effective Microorganisms Concrete) is better as referring to the strength and the workability as compared to the control concrete samples based on the slump test from Fig. 3 and Fig. 4 for compressive strength. While, the temperature of the EMC sample is lower and cold as compared to the control concrete sample (Fig. 5).

(a) Control concrete

‘Slump (mm)

150

(b) EMC

“CONC =EMC

100 60 $0

Y

°

We ‘Types of Samples

(c) Result from slump test for both samples Fig. 3. The slump test for both samples, (a) control concrete, (b) EMC, and (c) the result of the slump test

Eco-composite Porous Concrete Drainage Systems

199

gé

Es

é3 eo

a

2

0 4

Mend

10

20

30

AGE (Days)

(a) The samples after the testing

(b) The results for both samples

Fig. 4. The testing of the samples compressive strength . a

s

z

40 30

4.27 RT

(4+)

(4) 2

Oxy * Oui) Sa Oe; CPE

6)5

In other words, the above equations can be written as Rate of

Transport

change | +] of kore

Transport

of koreby | =| convection

Rate of

ofkoreby | +| diffusion

production of kore

Rate of

— | destruction

(6)

of kore Here k and ¢ are turbulent kinetic energy and rate of viscous turbulence dissipation respectively. vr is defined as below:

R

vr = Cue Here

we

have

five

adjustable

constants,

o%, 62, Cy,

(7) Cie,

Coe

and

the

values

in

present analysis are takes as 1.00, 1.30, 0.09, 1.44 and 1.92 respectively [19]. 2.2.

Solver

OpenFOAM uses implicit solvers to solve the Navier stokes (N-S) equations. However, for a give timestep, the Pressure-Implicit with Splitting of Operators (PISOFoam) solver used in this analysis does

not include

the recirculation of N-S

equations

[20].

This results the use of low Courant number to maintain the numerical stability. The Courant number is defined in Eq. (8). where At is the timestep and Ax is the minimum

size of the cell width. As a result, the time step significantly decreases to have fine mesh for higher Re and hence the simulation time upsurges disproportionality. By default, this solver does not include Courant number-controlled time stepping. The convergence criteria for the OpenFoam solver include residual calculations for solution variables such as velocity, pressure and turbulence. The simulations are solved for 600 non-dimensional time units.

Numerical Assessment of Flow Around Circular Cylinder

UAt

Cr=5 The hydrodynamic

293

(8)

force coefficients such as drag (Cp) and the lift coefficient (C,) are

calculated using the formulas presented in Eqs. (9) and (10) respectively. The Strouhal number (St) is calculated by using the Eq. (11). Equation

pressure coefficient Cp. The data from the OpenFoam

(12) is used to calculate the

was post-processed by using paraview.

Fp Cp =?

?

05pU2D FL

a= 0.5pU2D

(D 1=G

9

) (10)

St=—

IL (11)

c, = PoP ?0.5p,.U2,

(12)

Where Fp and Fy, are drag and lift forces induced by fluid flow on the cylinder respectively. p represents the density of the fluid flow, p is the static pressure and the Poo: Uses Poo are pressure, velocity and density

of the fluid flow taken at 0.1 units of

distances in x-direction from the inlet and center line of the computational domain in y-

axis. 2.3.

Computational Grid

2.3.1 Boundary Conditions Inlet A flow with uniform speed is specified at inlet. The Re is defined as in Eq (13) based on the fluid flow speed of the fluid U, where D is diameter of the cylinder and v is kinematic viscosity.

Re =——

(13)

Outlet The outlet of the computational domain is adequately at downstream such that the vortices formed by the flow won’t get affected by the boundary wall. In this case, the velocity and pressure outlets are capable of showing identical results. Boundary conditions for velocity and pressure are taken as zero gradients.

294

M. R. Lekkala et al.

Cylinder Walls The cylinder walls are set to no-slip condition, meaning that the velocities and the pressure are set to set to zero (U, = Uy = 0) and zero gradients respectively.

Periodic Periodic faces are performs velocity

Faces faces are referred as front and back sides of the flow domain. The periodic considered in the form of empty boundary conditions, meaning that the solver effectively 2D simulations in transverse and span-wise directions. Here the and pressure gradients are assumed as equal on both sides.

Free Slip Walls The top and bottom faces of the domain are assigned as free-slip conditions as shown. in Fig.

1, which sets the velocity normal to wall and shear stress are zero (Uy = 0 and

Twatt = 0) and velocity parallel to wall is calculated (U,).

Y

4,

—

Lo.

Cylinder

nd

51 51,

101

Transverse direction

es

Streamwise direction Fig. 1.

Definition of flow domain around a circular flow.

Turbulence Properties To solve the RANS equations using k-e as discussed in above section, the values for turbulent kinetic energy (k) and turbulent dissipation rate (¢) are required to be defined. Both the parameters are set using the Eqs. (14) and (15).

k=

3

5 (Use)

.

0.164k15 ¢=—

2

(14)

(15)

Numerical Assessment of Flow Around Circular Cylinder

295

Where I is defined as turbulence intensity which is ratio of magnitude of turbulent

fluctuations to the magnitude of the characteristic mean velocity J = 0.16Re~'/8. 1 is characteristic turbulent length (=0.07D).

2.3.2

Meshing

Figure | shows the rectangular computational domain with circular cylinder center at the origin of the Cartesian coordinate system. Coordinates x, y, z represents streamwise, cross-stream and span-wise directions respectively. The structured meshes are created by ICEM. The grid topology details are shown in Fig. 2. The computational domain is 3D rectangular box defined in Cartesian coordinate system (x, y, z), cylinder center is at (0, 0) as shown in the Fig. 1.

Fig. 2. Example of mesh generated for computation flow around circular cylinder. The grid was stretched in radial direction from the center of the cylinder and the grid was gradually merged into a quadratic region in four sectors as shown in Fig. 2. The mesh in 3D can be defined into three parameters Ni, Nj, Nk, which indicate the

number of grid points in circumferential, radial and axial directions respectively. For 2D cylinder cases, the mesh can be defined in only two parameters in circumferential and radial directions i.e. Ni and Nj.

3 3.1 3.1.1

Numerical Results and Discussion Convergence Studies Domain Dependence Study

In this section, the Streamwise and transverse wise direction lengths are selected based

on the domain independence tests. Initially, Streamwise domain will remain constant

296

M. R. Lekkala et al.

and the transverse direction length will be varied to ensure that the domain size in this direction won’t show any effect on the results. Similarly, the Streamwise direction domain size will be selected. Transverse Domain At the start, the Streamwise direction length of the domain is kept at 13], and the transverse direction length is varied. From here, length in later direction can be selected by ensuring that it won’t show any effect on numerical results. The transverse lengths used are 3],, 51, and 71, as shown in Table 2. When the transverse domain length changes from 5], to 7], the change of numerical results less than 1%. Therefore, the

transverse domain size is selected as Sle for subsequent numerical analysis. Table 2.

Change of Cp and C,, for transverse computational domain sizes

Transverse length | Grid size 3k. 320 x 216 Sle 320 x 254 Tle 320 x 298

Streamwise Domain The above selected transverse domain Streamwise direction. The Streamwise the effect of domain length on results. the effect of size decreases on the Cp

[Cp | 1.114] 1.090 | | 1.081 |

[Ci 0.992 0.978 0.972

length is selected and fixed for simulations in direction lengths are 10], 121, 14], to analyze It is observed that as when the length increases, and C,. When the length changes from 121, to

141, the change of numerical results is less than

1%. Hence

121, is fixed in streamwise

direction for succeeding simulations.

Table 3.

Change of Cp and C,, for streamwise computational domain sizes Streamwise length

Grid

Cp

(Cy

101,

264 x 254

size

1.106

0.990

121,

296 x 254

1.094 | 0.979

141.

324 x 254

1.091

0.976

Therefore, the optimal computational domain size is selected as ~31, 1.5 g, the F,, value increases from SA to SC, then it decreases from SC to SE. Another case with the site coefficient F,, value. As shown in Table 2, the value of F,, is always amplified from SA to SE, and tends to decrease in magnitude as the value of S\ increases.

From this observation and refer to studies by Sutjipto [15-17], it is indicated that the cause of the design response spectrum anomaly is the site coefficient F, which has de-amplified properties at high spectral acceleration (earthquake prone areas or near fault areas).

It should be noted that the de-amplification case of the response spectrum in soft soils in areas with high spectral acceleration has been revealed by Seed (1976), Mohraz and Elghadamsi

(1989), Idriss (1990).

UBC,

IBC

and ASCE

7 have

already adopted

since long time ago. Unfortunately, it was not adopted in SNI 1726-2002 even though this code used the 1997 UBC as a reference.

3

Design Response Spectra of 17 Major Cities in Indonesia

For the purpose to observe the behavior of the design response spectra, 17 major cities which are spread throughout Indonesia were selected. Selection of the cities are based on that they are the capital of provinces in Indonesia; they have a significant increment in economic development and construction industry, and they have a quite high population and vulnerability to earthquakes. The parameter values used for developing the design response spectra of those 17 selected cities, are shown in Table 3. The values are obtain from separate studies of Prof. Sengara, Dr. Asrurifak and Dr. Partono, and had been verified by Prof. Irsyam.

378

S, Sutjipto and I. Sumeru

Table 3. Spectral response acceleration parameters at short periods (Ss) and 1-second period (S\) of 17 major cities in Indonesia. No.

cy

1 2 3 4 s 6 1 8

‘Ambon Balikpapan Banda Aceh Bandung Denpasar Jakarta Joyapura Kupang Makassar Manado Manowar Medan Padang Palembang Pontianak Semarang Surabaya

9. 0 u 2, 1B 4 Is, 16, 17

Latitude “36584 “12379 55483 69175 “86705 “6.1781 2.5916 “101772 “S477 Lams, 08615 35982 -asari 29761 00263 -7.081 12518

erin

Longitude 1281908 1168829 983238 or6191 is2126 106.8680 40.6000 123.6070 94307 eager 1340020 oxen oosi72 1047784 1093425 ho436i 27821

Ss TaRs 0128 sto Lite 0984 om 1500 1049 022 Lose 1500 0682 Last 0292 ons 0890 710

0398 0.088 0.600 osto 0397 039 0.600 0380 one 0470 0.600 03a 0.600 0248 087 039 31s

The design response spectra for the site class categories of SC (hard soil), SD (medium) and SE (soft soil) of the 17 major cities in Indonesia at short periods (0.2 s), Sps, and at a 1-second period, Sp, are tabulated in Table 4 and Table 5, respectively.

Table 4. Design response spectra at short periods (Sps) of 17 major cities in Indonesia. Ne. p 2 3 4 5 a 1 8 9. 0 " 2, 8 4 Is. 16, V7.

cus ‘Ambon Balikpapan Banda Acch Bandung Denpasar Jakarta Jayapura Kuping Makassar Manado Manokwar Medan Padang Palembang Pontianak Semarang Surabaya

Hard Soil(SC) O68 0107 1208 oat 0787 0625 1200 089 0192 oss 1200 0539 Liss 0253 0.098 0704 0516

Sas Medium Soil SD) o7m ont Loo? 0807 07% 0617 1000 0786 0237 0757 000 0596 0.987 030s ont 067s 0583

Soft Soil(SE)__ 0746 0197 80s 0782 0730 0.668 0.800 0742 0385 742 0.800 0633 0797 oa 1st 702 0.646

‘Anomaly 1 Anomaly? Anomaly’3 SD Department of Civil and Construction Engineering, Faculty of Engineering and Science, Curtin University, CDT250, 98009 Miri, Sarawak, Malaysia Abstract. Excessive usage of sand in construction industries has generated many

environmental issues. Silica aerogel, is able to minimise environmental

issues while providing thermal resistance for building materials. Silica aerogel has properties such as lightweight, nano-porous and very low thermal conductivity compared to sand in the mortar matrix. This paper studied the compression

and flexural strengths of mortar with silica aerogel as a sand replacement. Cement to sand ratio of 1:3 was used and sand was replaced with silica aerogel in the volumes of 15%, 20% and 25%. 15% volume of sand replacement with silica aerogel powder was the optimum ratio as it possessed the highest strength

during experimental work. All the specimens were able to achieve the minimum strength for Type N non load bearing wall with the optimum ratio of 15% volume silica aerogel powder. In conclusion, silica aerogel mortar achieved the minimum

strength of type N mortar.

Keywords: Sand - Silica aerogel - Compression strength - Flexural strength + Mortar

1

Introduction

Excessive usage of sand in the construction industry has caused some environmental problems like river bank erosion, river bed degradation, river buffer zone encroachment and deterioration of river water quality (DID 2009). Depletion of natural resources is the current main issue in maintaining sustainability. Artificial aggregate may serve as one of the solutions for this environmental issue. There are some identified artificial lightweight coarse aggregates being applied in the construction industry, namely expanded clay, expanded shale, and processed volcanic rocks (Mohd et al. 2001). Waste pebble is also used to produce artificial sand and applied into concrete. Mortar is a mixture of cement, sand and water with the purpose of binding masonry units into a single unit. Mortar also serves as plaster, multi-purpose repair and floor levelling. It is established that sand is an important construction material in the construction industry. However, due to the massive usage of sand, it has been depleted over the years. India is taking the necessary steps to combat this problem by restricting sand excavation due to the exhaustion of natural sand (Klangvijit and Sookramoon 2018.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 493-500, 2021. https://doi.org/10.1007/978-98 1-33-63 11-3_57

494. T. Tay etal. Moreover, indoor thermal comfort is gaining the attention of scholars, architects, contractors, governments and residents recently. A building envelope is known as the physical separator between the conditioned and unconditioned environments of a building. There is a close relationship between building envelope and indoor thermal comfort. Factors such as absorptivity of external surfaces, thermal capacity, and thermal conductivity of the building envelope hugely affect the internal environment and consequently the energy consumption inside the building in order to maintain thermal comfort (Hwaish 2015). According to Gratia and De Herde (2007), more than 50%

of

the embodied energy distribution in major elements in residential buildings come from building envelopes, contributing approximately 50%-60% of their total heat gain (Uihlein

and

Eder 2010).

In addition,

Utama

and

Gheewala

(2009)

discovered

that

building envelope materials are able to greatly affect the cooling load in high-rise buildings while contributing to their energy life cycle. This is because roofs and walls tend to absorb heat and release long wavelength, which release the heat and causing UHI (Mansouri et al. 2017). The texture and materials used for facades and roofs also. affect the inside and outside thermal ambiance (Mansouri et al. 2017).

Construction materials such as brick and concrete have high thermal mass and thermal conductivity. Thermal conductivities for concrete and brick are 0.9 W/m-k (Antwi-boasiako et al. 2018) and 0.6 W/m-k

(Visser and Yeretzian 2013) respectively.

High thermal mass materials can be referred to as high-density items as more heat is required to increase their temperature. High thermal mass materials will store heat during daytime and release the heat at night. Therefore, an easier and practical way to maintain indoor thermal comfort level is to use thermal insulation materials to control the heat transfer. Nevertheless, conventional thermal insulation materials are either non-environmentally friendly or non-biodegradable. Polyurethane has low thermal conductivity, but it will release toxic gas, HCN which is harmful to humans (Jelle et al. 2012). Expanded polystyrene (EPS) and extruded polystyrene (XPS) are polystyrenes used to provide thermal insulation due to their low thermal conductivity; however, polystyrene is non-degradable and EPS is one of the top items of debris recovered from shorelines and beaches worldwide (Garrity and Levings 1993). Mineral wool is widely used as thermal insulation but it is vulnerable to moisture. Meanwhile, the negative impact of cellulose is that it is made up of recycled newsprints and people may have allergies to the dust of newspapers (Jeon et al. 2017). Replacing natural sand with artificial fine aggregate is an alternative to solve both environmental issues and thermal resistivity of building materials. In order to produce similar strength with normal concrete, high content of silica dioxide in sand should exist in the artificial aggregate. Silica aerogel possesses the properties of being lightweight, with high specific surface area, high porosity, low density and high thermal insulation value, and could potentially replace sand in concrete matrix. Therefore, this paper presents a study of sand replacement with silica aerogel powder for mortar. The replacements were performed in different levels, namely 15%, 20% and 25% of sand volume for non-bearing mortar type N. The results were compared with ASTM specifications.

Compressive and Flexural Strengths of Mortar with Silica Aerogel Powder

2 2.1

495

Experimental Procedure Materials

Silica aerogel, as shown in Fig. 1, was bought from Changsa Eachem Co. Limited with the properties given by the manufacturer, as displayed in Table 1.

Table 1, Properties of silica aerogel powder. Properties Value Thermal conductivity (25 °C W/m-K) | 0.015 Particle size (um) 90

Fig. 1. Silica aerogel powder. Sand 42.5 N)

that passed through was

used

for

the

600 ym sieve and Ordinary Portland cement (CEM

mixing.

Silica

fume

and

superplasticiser

(RF-611)

I

were

needed for the mixing. 2.2

Mix Design

Silica aerogel powder was used to replace the sand in mortar design mix. Mix ratio for cement to sand was 1:3. Mortar type N with sand replacements of 15%, 20% and 25% of silica aerogel powder in volume were casted and tested for compression and flexural strength based on ASTM C1329 and ASTM C 348 respectively.

496

L.T. Tay etal. Specimen size for compression testing was 50 mm cube. Meanwhile, specimen size

for flexural testing was 40 mm

(width)

x 40 mm

(thickness)

x

160 mm

(length). All

specimens were tested for 7 and 28 days of concrete ages. Table 2 shows design for various sand replacements with silica aerogel powder.

the mix

Table 2. Mix design for silica aerogel mortar. Sample

WiC

name

ratio

ratio

Cement:Sand

(%)

(%)

1

0.83

1:3

0

0

0

2 3

O81 0.83

1:3 1:3

10.8 10.8

5 5

15 20

4

0.85

1:3

10.8

5

25

2.3.

Mixing Procedure

Ordinary Portland cement (CEM 600 um sieve), water and silica mixing. Due to the hydrophobic well with mortar. Therefore, silica to enhance cohesion and produce the cementitious

materials

Silica fume

| RF-611

Silica aerogel

percentage (%)

I 42.5 N), oven-dried fine aggregate (passed through aerogel powder were weighted accordingly before property of silica aerogel powder, they did not mix fume and superplasticiser were added during mixing a well-mixed mortar. Silica fume in 10.8% weight of

(Gao et al. 2014)

and

5%

weight

of superplasticiser (RF-

611) were prepared and then OPC, fine aggregate, silica fume and silica aerogel powder were mixed together. Diluted superplasticiser was then introduced to the mortar.

3 3.1.

Results and Discussion Compression Strength

Compression strength test was conducted according to ASTM C 109/C 109 M — 02. Loading was applied to the specimen faces that were in contact with the true plane surfaces of the mould at the rate of 1 KN/s with a pace rate of 10 mm/min.

Compressive and Flexural Strengths of Mortar with Silica Aerogel Powder

497

18 w

a

16

214 s

% 12 S

3

—

.

10

Compression

strength at 7th

8

3

6

S §

4 2

days (MPa)

——Compression strength at 28th days (MPa)

0 0%

15%

20%

25%

Percentage of silica aerogel

Fig. 2.

Compression strength results.

Based on Fig. 2, all the groups of samples achieved the minimum compression strength for type N mortar as per ASTM C 1329, which were 3.5 N/mm” and 6.2 N/mm’ at 7 days and 28 days. It was observed that compression strengths at 7 and 28 days decreased from 15% to 25% replacement by silica aerogel and 15% achieved the highest compression strength. The largest differences for 7 days and 28 days were 3.54 MPa and 8.83 MPa respectively. It was because silica aerogel is nano-porous and has low mechanical strength. Besides that, from the study by Gao et al. (2014), it is confirmed that silica aerogel is quite stable and chemically inert due to its hydrophobic property. It repels water and thus prevented a reaction with the surrounding cement paste and formation of gaps along the silica aerogel. This can also be explained by decreasing in terms of the amount of binder in mortar. Binder serves to tie the materials or particles inside the matrix. Due to the gaps presented in the matrix, bonds cannot completely hold cement matrix with sand. Therefore, a higher volume of sand being replaced with aerogel caused the mortar to become more fragile. However, normal mortar had lower compression strength than 15% sand replacement with silica aerogel powder, owing to the fact that silica fume and superplasticiser were added in to the mix together with silica aerogel. Meanwhile, there were only sand and cement present in normal mortar. Silica fume and superplasticiser (RF-611) help in developing strength and cohesion. Both materials have proved that they are able to increase their concrete strength. Silica fume improved aggregate-paste bond and microstructure due to its fineness (Barbhuiya and Qureshi 2016; Raveendran et al. 2015; Xiao and Jiang 2016).

It is able to fill in the gaps inside the mortar and induce additional C-S-H bond; thus, it is denser with the presence of silica fume. Furthermore, it improved the cohesion of the mortar mix which tackled the problem caused by the hydrophobic property of silica aerogel. Therefore, silica fume was able to

498

L.T. Tay etal.

create a well-mixed condition. Moreover, superplasticiser (RF-61 1) is able to accelerate

high early strength in mortar. In short, the presence of silica fume and superplasticiser induced the high strength of 15% sand replacement with silica aerogel. Another observation was found that all percentages of silica aerogel increased steadily from 7 days to 28 days with almost 3 KN; however, 15% replacement raised steeply at 7.65 KN. This can be explained by the lowest water to cement ratio with a value of 0.81 to get a flow rate of 110 + 5. High water to cement ratio will result in low compression due to excess evaporation of water and lead to the formation of pores. In short, 15% sand replacement of silica aerogel powder achieved the highest compressive strength at 7 days and considered as the optimum mix in this series of tests. 3.2

Flexural Strength

Flexural or bending test was carried out based on ASTM C 348 - 02. Load was applied at the rate of 0.05 kN/s with a pace rate of 10 mm/min. Figure 3 shows the results of flexural tests.

Flexural strength (MPa)

3.5

3

————

25

15

——Flexural strength at 7th days (MPa)

1 0.5

0%

15%

20%

25%

Percentage of silica aerogel

Fig. 3.

Flexural strength results.

Based on the results, it can be confirmed that silica aerogel mortar with the volume of 15% achieved the highest flexural strength than the other percentages. This trend was similar with compression strength. Since silica aerogel is nano-porous and very weak in mechanical strength; thus, a higher volume of silica aerogel will make mortar become more fragile. In the meantime, 0% of silica aerogel mortar had a lower flexural strength than 15% volume silica aerogel mortar due to the presence of silica fume and superplasticiser (RF-611) in 15% volume silica aerogel mortar. Both silica fume and

Compressive and Flexural Strengths of Mortar with Silica Aerogel Powder superplasticiser (RF-611) confirmed

by

Ahmed

flexural strength. Hence, lower bending strength.

4

helped in developing

et al. (2016)

who

high early strength. This

stated that silica fume

mortar without adding

was

499

is further

able to increase

these two materials tend to have a

Conclusion

In conclusion, silica aerogel is a promising thermal insulator with a lower thermal conductivity compared to existing thermal insulator materials. Silica aerogel possesses the properties of being lightweight, with high specific surface area, and high porosity and low density to potentially replace sand in concrete and mortar matrix. To ensure its suitability as a thermal insulation material, silica aerogel powder was used to replace the sand partially to become silica aerogel mortar without compromising the strength required by mortar. Based on the results, silica aerogel mortar achieved the minimum strength for Type N non-load bearing wall with an optimum ratio of 15% volume silica aerogel powder. The characteristic of silica aerogel is known to reduce heat transfer, but there is a lack of research to investigate the thermal conductivity and thermal performance when silica aerogel is incorporated in mortar. Thus, it is recommended to study the thermal properties and thermal performance of silica aerogel mortar. Besides that, silica aerogel mortar can be recommended to apply as plastering and/or introduced as a new type of concrete brick and wall panel to act as a thermal insulation material.

Acknowledgement. The authors would like to acknowledge Ministry of Education and Universiti Malaysia Sarawak for their financial support through FRGS project of F02/FRGS/1884/2019. References Mohd, S., Wah, C.K., Lim, P.Y.: Development of artificial lightweight aggregates. In: Structural Engineering, Mechanics and Computation, pp. 1399-1406. Elsevier Science (2001) DID. River Sand Mining Management Guideline, Ministry of Natural Resources and Environment

Department

and Drainage (2009)

of Irrigation

and

Drainage

Malaysia.

Department

of Irrigation

Klangvijit, W., Sookramoon, K.: Study of the mix cement properties of mortar cement used in.

masonry and plaster from the waste biscuit firing of ceramic. In; MATEC Web of Conferences, vol. 187, p. 02005. EDP Sciences (2018) Hwaish, A.N.A.: Impact of heat exchange on building envelope in the hot climates. Int. J. Emerg. Technol. Adv. Eng. 5(2) (2015) Gratia, E., De Herde, A.: Guidelines for improving natural daytime ventilation in an office building with a double-skin facade. Sol. Energy 81(4), 435-448 (2007)

Uihlein, A., Eder, P.: Policy options towards an energy efficient residential building stock in the EU-27. Energy Buildings 42(6), 791-798 (2010) Utama, A., Gheewala, S.H.: Indonesian residential high rise buildings: a life cycle energy assessment. Energy Buildings, 41(1 1), 1263-1268.-27 (2009) Mansouri, O., Belarbi, R., Bourbia,

F.: Albedo effect of external surfaces on the energy loads and

thermal comfort in buildings. Energy Procedia 139, 571-577 (2017)

500. T. Tay etal. Antwi-boasiako, C., Boadu, B.K.: Thermal conductivity, resistance and specific heat capacity of chemically-treated, widely-used timber for building-envelope (2018) Visser, F., Yeretzian, A.: Energy Efficient Building Guidelines. European Union, Heliopolis (2013) Jelle, B.P., Gustavsen, A., Baetens, R.: Innovative high performance thermal building insulation materials-todays

state-of-the-art

and

beyond

tomorrow.

In: Proceedings

of the Building

Enclosure Science & Technology (BEST 3 - 2012), Atlanta, Georgia, U.S.A., pp. 1-2, 2-4 April 2012 Garrity, S.D., Levings, S.C.: Marine debris along the Caribbean coast of Panama. Mar. Pollut.

Bull. 26(6), 317-324 (1993) Jeon, C.K., Lee, J.S., Chung, H., Kim, J.H.,

Park, J.P.: A study on insulation characteristics of

glass wool and mineral wool coated with a Polysiloxane Agent. Advances in Materials Science and Engineering 2017 (2017) Gao, T., Jelle, B.P., Gustavsen, A., Jacobsen, S.: Aerogel-incorporated concrete: an experimental study. Constr. Build. Mater. 52, 130-136 (2014)

Barbhuiya, S., Qureshi, M.: Effects of silica fume on the strength and durability properties of concrete. Age 3(7), 28 (2016) Raveendran, K.G., Rameshkumar, V., Saravanan, M., Kanmani,

P., Sudhakar, S.: Performance

of silica fume on strength and durability of concrete. Int. J. Inn. Res. Sci. Eng. Technol. 4, 10162-10166 (2015) Xiao, L.G., Jiang, B.: Silica fume and fly ash admixed can help to improve the PRC durability combine microscopic analysis. In: MATEC Web of Conferences, vol. 67, p. 06033. EDP Sciences (2016) Ahmed, M., Mallick, J., Hasan, M.A.: A study of factors affecting the flexural tensile strength of concrete. J. King Saud Univ. Eng. Sci. 28(2), 147-156 (2016)

®

Check updatesfor

Study of C-S-H Formation of Cemented Sediment Brick L. W. Ean", M. A. Malek', B. S. Mohammed?, Chao-Wei Tang’, and C. Y. Ng*

' Institute of Sustainable Energy, Jalan IKRAM-UNITEN, Universiti Tenaga Nasional, 43000 Kajang, Selangor, Malaysia {Leewoen, Marlinda}@uniten. edy. my > Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS, 32610 Perak, Malaysia {bashar. mohammed, chengyee.ng}@utp. edy. my > Department of Civil Engineering and Geomatics, Cheng Shiu University, No. 840, Chengching Road, Niaosong District, Kaohsiung 83347, Taiwan tangew@geloud. csu. edu. tw Abstract.

This

paper

presents

an experimental

study of chemical

structure

formation in newly developed sediment bricks. Morphology analysis was conducted by Scanning Electron Microscope (SEM) images to observe morphological structure and amorphous material while mineralogy analysis through Energy Dispersive Spectroscopy (EDS) and X-Ray Diffraction (XRD) to determine amount of the amorphous materials. C-S-H formation that formed a dense material was observed in Mix 4, which made up from 20% weight of sediment-sand, 70% weight of sediment-silt and stabilized by 10% weight of cement content, which was identified to be the best mix in earlier publication [1]. Higher content of calcite in Mix 4 shows existance of carbonates that indicates lower C-S-H formation compared to Mix 1, which has higher cement content. Meanwhile, larger amount of SiO in Mix 6 does not indicate higher

production of C-H-S. This is verified by SEM images whereby unreacted sediments were observed in Mix 6. This study showed that increment of sediment content disturbed the production of hydration products. In addition, C-S-H

formation is not solely verified by amount of SiOz, it has to also verify through SEM

images since SiO, can also be the unreacted raw materials.

Keywords:

1

C-S-H

- Sediment brick - Structure formation

Introduction

Full utilization of dredged sediment in cemented bricks is one of the remediation taken to prevent secondary pollution when the sediments accumulated in landfill areas. Several initiatives have been taken to utilize the sediment into bricks by different method such as by pressing and sintering [2-4], vibration and compaction

[5], as well as firing [6, 7].

Myrin et al. [8] used dredged sediment from Paranagua Port with mineralogical composition of quartz SiO, microcline KALSi;Ox and halite NaCl to produce sediment bricks. SEM images observed amorphous material of halite which appeared in classic

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 501-509, 2021. https://doi.org/10.1007/978-981-33-6311-3_58

502

L. W. Ean et al.

hex-octahedron forms and morphological structure shows rounded and no sharp edges grain of sand. In higher magnification, abrasion including peeling was noticed on the grains as a result of friction. Myrin et al. [8] found that 30% weight of lime production waste is sufficient to neutralize acid salts in 50% weight sediment and presented an increase in the amount of new growths in pore space with 20% weight of construction and demolition debris. Wang et al. [9] illustrated SEM image that shows irregular and rough crystal aggregates which suggested the presence of high magnesium calcite on the sample’s surface and agglomerated particles with low magnesium calcite that depended on the concentration of soluble Ca?* and Mg** during the precipitation process. This is to show

that magnesium

oxide cement

(MOC)

in the binary cement

used in the mixture

with contaminated sediment from kwun Tong Typhonn Shelter in Hong Kong facilitated and accelerated the carbonation process that resulted in the formation of magnesium calcite and magnesium carbonate hydrate. Wang et al. [10] illustrated SEM images of C-S-H gel formation for sediment bricks using sediment with pretreatment at 400 °C which contained lesser impurities of organic matter and other unstable materials. The C-S-H gel formation is corroborated by EDS analysis with the presence of Ca, Si and O. This research shows the study of structure formation of the newly developed sediment bricks. This is conducted by evaluating the formation of amorphous materials in the mixtures.

2

Experiments

The sediments used in sediment bricks were characterized and classified in two (2) groups, namely sediment-silt and sediment-sand based on the physical and chemical properties. The mix design developed with water content and sediment-sand content that fixed at 10% and 20% of total weight respectively; sediment-silt content that varies from 60-90% and cement content varies from 5-20%. Table 1 shows the mix proportions

of the

sediment

bricks

[1]. The

sediment

bricks

were

compressed

with

a

pressing load of 220 KN [1]. The properties of sediment bricks were tested accordance with the requirements of the ASTM C140 and ASTM C67, whereby the recomended mix design was published

in [1].

Microstructure and chemical structure formation studies of the sediment bricks were conducted by identifying amorphous minerals of calcium silicate hydrate (C-S-H) using SEM images (morphology analysis) and further clarified through Energy Dispersive Spectroscopy (EDS) and X-Ray Diffraction (XRD) analyses (mineralogy analysis). The micro chemical analysis was determined by method of EDS using INCA Energy system that attached to SEM machine (JSM-6010Plus/LV) and Panalytical X’Pert Powder for XRD. Diffractogram patterns of XRD were used to show amounts of the amorphous materials; SEM images to observe amorphous material; and EDS analyse the crystalline bodies to reaffirm results from XRD. In addition, SEM images also validated the type of grain by observing the morphological structure. The sediment brick samples were grind to fine powder for XRD analyses while EDS and SEM analysis were using the brick sample that cut into 1.5 cm x 1.5 cm cube. XRD analysis was conducted with 2-Theta scaled at 5°-80°.

Study of C-S-H Formation of Cemented Sediment Brick

Table 1, Mixing proportions [1]. Mix No. | Percentage by Weight (%)

503

‘Average compressive

Sediment Silt | Sediment Sand | Cement | Total _| strength (MPa)

1 2 3 4 5 6 3 3.1

60 65 68 70 2 75

20 20 20 20 20 20

20 15 12 10 8 5

100 100 100 100 100 100

| 21.1 | 12.8 67 63 52 27

Results and Discussions Microstructure of Sediment Bricks

This section presents results on morphology analysis and mineralogy analysis of the mixes. Figure 1 shows SEM images for sediment-sand and sediment-silt. Results on Scanning Electron Microscope (SEM)

for Mix

| to Mix 6 are shown

in

Fig. 2. It is observed that the particles are perfectly bonded to each other with little pores for Mix 1, with 20% of cement content. Scattered patches of small noncontinuous and plate-like particles that grow outwards from the clay particles are the precipitation of calcium silicate hydrate (C-S-H) gel. The insoluble hydration product (C-S-H) is formed when cement reacts with water while the soluble hydration product

(calcium hydroxide) precipitated as discrete crystals. According to Oti et al. [11], lesser amount of free calcium hydroxide will produce a more chemically stabled system. C-SH gel is clearly visible as the cement content increased that is showed in Fig. 2(a) at 20% cement content, compared to Fig. 2(f) that shows lesser C-S-H gel in 5% cement content. The hydrated compound is beneficial to enhance the compressive strength of the brick. The compacted and continuous microstructure indicated that the brick has the potential resistance in heat and sound resistance. The SEM image at 50-times magnification of Mix 1 in Fig. 2(a) shows denser structure on the surface area of the bricks compared to Mix 6 in Fig. 2(f). The sediment-sand particles in Fig. | are visible in Fig. 2(f) while Fig. 2(a) shows evidence

Fig. 1. (a) Sediment-sand x S0magnification; (b) Sediment-silts x350magnification

504

L. W. Ean et al.

se

(f) ISkV, WD 10mm.

Fig. 2. SEM images of the sediment bricks for mix proportion: (a) Mix 1 (b) Mix 2 (c) Mix 3 (d) Mix 4 (e) Mix 5 and (f) Mix 6 at magnification x50, x 1000, and x3000.

Study of C-S-H Formation of Cemented Sediment Brick

505

of new formation covering the sediment sand particles. The new formation of C-S-H are observed at 3000-times magnification with denser surface structure that link the sediment particles together. Pore chains were observed under 1000-times magnification for all mixes and they are more visible as cement content decreases. This is possibly due to lesser C-S-H formation as binder of the sediment particles and remains of organic materials that promote development of cracks which significantly reduce mechanical properties of the bricks. 3.2

Chemical Structure Formation of Sediment Bricks

Mineralogy of crystalline phase for Mix 1, 4 and 6 were analysed by XRD method and presented in Fig. 3. The mixes were chosen based on the physical and mechanical properties tested in [1], whereby mix 4 was found to be the best mix in the said paper. Despite the increasing amount of compressive strength with the increased cement content, the usage of cement content was controlled in order to minimize carbon footprint in the production as well as uncertainties such as early usage before the bricks matured. Since mix 4 has higher compressive strength than requirement in ASTM C129 of 4.2 MPa at 7 days, therefore mix 4 has been selected as the optimum mixture [1]. Mix 4 is compared to Mix

1 and Mix 6 which are the lowest cement content for the

former and the highest cement content for the latter. Dissolved amorphous SiO. precipitated as calcium silicate hydrates (C-S-H) in existence of large amounts of free CaO in cement, and the C-S-H become binder between the sediments particles as similarly found by Mymrin

et al. [8].

Q

Q-Quartz

i

K- Kaolinite

M-Microcline [- [Ilite

C- Calcite g

Bork Py Kile K

6

A

ia

P- Portlandite

M P

Q

9 aMice

K

6

oo

-

[ara

NL.

[

:

meee

A

es

*

2Theta (Coupled TwoTheta/Theta) WL=1.541

Fig. 3. X-ray Diffractogram pattern of XRD for Mix 1, Mix 4 and Mix 6. These amorphous SiO, were observed in large amount in Mix | with 50.873% weight compared to Mix 4 with 27.4% weight as shown in Table 2. Meanwhile, the higher peak of calcite in mix 4 surmise the existence of carbonates which indicated low activity of C-S-H formation. This has validated the large amount of C-S-H as amorphous SiO, in Mix 1. The amorphous substances of carbonates and C-S-H formation

506

L. W. Ean et al.

contribute to the gain of mechanical properties of the material as found by Mymrin et al. [8]. In contrary, Mix 6 shows larger amount of SiO2, which is 55.008% compared

to Mix 1. However, this does not indicate high amount of C-S-H as in Mix 1. High content of SiO, mineral in Mix 6 shows unreacted sediment. This is be verified from SEM images in Fig. 2(f) that shows lesser existence of C-S-H compared to Mix | in Fig. 2(a). In addition Mix 4 and Mix Table 2.

6 show

low amount of unreacted portlandite.

Mineral phase of Mix 1, Mix 4 and Mix 6 from XRD.

Mineral phase

Percentage weight (%) Mix 1 | Mix 4| Mix 6 Quartz (SiO2) 50.873 27.4 | 55.008 Kaolinite (AlSix0;(OH),) 6.752) 17.3 | 13.234 Calcite (CaCO) 10.632 19.9 12.030 Microcline (KAISi;Ox) 2047 23.0 9.677 Ilite ((K,H30)Al2(SisAI)Oio(OH)>-H30) 11.272) 9.7 8.125 Portlandite (CaO-H20) 27 1.926 Chemical compositions measured through EDS (Table 3) of the sediment bricks in total areas (Fig. 4) in 500-times magnification for all mixes confirmed the results of

SEM and XRD on the presence of amorphous minerals. EDS analysis shows the presence of Ca, Si and O that forming C-S-H gel. Less Ca content in Mix 6 has verified low C-S-H formation activity and that SiO. in Mix 6 resulted from XRD analysis is the unreacted sediments rather than C-S-H

amorphous

mineral. On the other hand, carbon

content in all mixes indicated the present of organic matters in the sediments platinum, Pt is the coating material used in SEM-EDS analysis.

and

Fig. 4. SEM images of the sediment bricks for EDS analysi : (a) Mix 1 (b)Mix 2 (c) Mix 3 (d) Mix 4 (e) Mix 5 and (f) Mix 6 at magnification «500.

Study of C-S-H Formation of Cemented Sediment Brick

Table 3, Chemical analyses of total area (labeled in Fig. 4) by proportions with 500-times magnification. Spectrum Values of elements, wt% c io [Mg lat {si (s |[K Total Area 1 (Mix 1)/4.84 35.70/0.42/3.10/13.47— |- | Total Area 2 (Mix 2) | 7.18 40.29/0.60/2.71/ 8.30 |- | Total Area 3 (Mix 3) | 8.36 36.25/2.18 3.30 0.70|Total Area 4 (Mix 4) | 5.04 47.08 | 0.48 | 7.76 | 11.17 0.38| 0.92 Total Area 5 (Mix 5) | 4,64 52.90 | 0.38 |5.87/ 13.81 0.76 Total Area 6 (Mix 6) | 4.22 49.16 | 0.59 8.46 | 12.25 _|0.83, Other OPC aluminate

hydrates

sulphate

such as calcium

hydrate

(C-A-S-H)

EDS method for all mix [Ca [Fe 13.42/— 7.38/ 4,91 17.54/9.03 1.61, 6.36, 0.91, 4.60/ 1.95.

aluminate hydrate (C-A-H)

and calcium

507

hydroxide

(CH)

[Pt _29.05| 28.64/ 31.68 | 16.53 | 14.38 | 17.94

|Total 100 100 100 100 100 100

gels, calcium which

enhance

strength development of bricks were not detected in XRD analysis (Fig. 3) although the element of Al, Ca and S were detected in EDS analysis. This is in line with finding by L. Wang et al. [10] that large amount of sediment in the mixture significantly hindered the formation of CH due to calcium complexation by organic materials and interferences by heavy metals, also by salt content in the sediments. EDS analysis by selecting total area (Fig. 4), partial area and points (Fig. 5) are significantly different due to complexity in mineral composition at micro level. It is also significantly different from that obtained from bulk analysis such as XRD. Homogenous composition is impossible to be obtained in microlevel and chemical composition of area near C-S-H gel are significantly different from bulk method as explained by Mynmnrin et al. [8]. Figure 5 show EDS results on partial areas and points for Mix | and

Spectrum Areal Point! Point2 Point4

Values of elements, wt % CO Mg Al Si_ SK Ca 8.79 3621 - 4.56 675 - 11.46 492 3429 0.58 4.83 9.52 1.50 9.17 7.49 39.52 - 3.08 5.48 0.67__- 10.99 7.07 30.22 - 1.99 5.98 0.66 - 15.09 10.98 38.29 11.52 1.61 1230 ‘1.84 5.22 = 29.11

Pt_ 32.25 35.19 32.77_ 39.00 30.33 49.92

Total 100 100 100 100 100

(a) Mix! Spectrum

CO ‘Areal 9.55 34.10 Area2 19.25 43.84 Area3 13.79 37.18

Values of elements, wt % Al Si K Ca Fe Pt Total 8.38 13.12 0.66 3.03 1.72 29.43 100 0.31 6.82 9.74 0.66 0.97 0.82 17.58 100 0.52 8.39 11.36 3.25135 24.16 100

Point1

0.80

8.20 14.59

46.10 38.88

8.17

42.34 55.06 50.70

7.43

Fig. 5.

(b) Mix6

Mg

7.10 7.10

11.30 8.83

2.18

0.62

6.00_ 11.50

15.31 13.42

0.46

7.38_

11.98

3.69

1.05

19.57 30.60

100 100

2.88

24.12 18.60

100 100

0.94

18.74

100

EDS analysis of areas and points of 5000-times magnification for (a) Mix 1 and (b) Mix 6.

508

L. W. Ean et al.

Mix 6. It is proved that amount of Ca content decrease as cement content decrease from Mix | to Mix 6 and Si content increase as the Ca content decrease which indicates lesser C-S-H formation. Carbon content in partial area and points is averagely at 10% wt which indicate existence of organic matters in the sediments.

4

Conclusions

This study shows the microstructure and chemical structure formation for the proposed mix proportion to enhance the properties studies of the sediment bricks in previous publication

by L. W.

Ean et al. [1,

12,

13]

that recomended

Mix

4 as the best mix

design. Mix 4 is made up from 20% weight of sediment-sand, 70% weight of sedimentsilt and stabilized by 10% weight of cement content using high pressure cementing method (without firing). Although increment of sediment content disturbed production of hydration products, the sediment bricks have developed sufficient hydration product in the form of C-S-H to produce a dense material that satisfied the requirement for nonload bearing bricks accordance with ASTM C129. In fact, different mix proportions of the sediment brick can be used depending on the requirements of the product. Higher compressive strength can be obtained by increasing the cement content of up to 20%.

Acknowledgements. The authors would like to acknowledge financial support from Bold 2020 research grant of University Tenaga Nasional, Malaysia. Grant code: RJO10517844/079. References 1.

Ooi, J.L., Ean, L.W.,

Mohammed,

B.S., Malek, M.A., Wong,

L.S., Tang, C.-W., Chua,

H.

v

Q: Study on the properties of compressed bricks using cameron highlands reservoir sediment as primary material. Appl. Mech. Mater. 710, 25-29 (2015). http://doi.org/10. 4028/www. scientific.net AMM.710.25 . Liang, H.H., Li, J.L.: The influence of hydration and swelling properties of gypsum on the preparation of lightweight brick using water supply reservoir sediment. Constr. Build. Mater. 94(September), 691-700 (2015). https://doi.org/10.1016/j.conbuildmat.2015.07.111 3, Tang, C.W., Chen, H.J., Wang, S.Y., Spaulding, J: Production of synthetic lightweight aggregate using reservoir sediments for concrete and masonry. Cem. Concr. Compos. 33(2), 292-300 (2011). https://doi.org/10.1016/). cemconcomp.2010. 10.008 4. Chiang, K.Y., Chien, K.L., Hwang, $.J.: Study on the characteristics of building bricks produced from reservoir sediment. J. Hazard. Mater. 159(2-3), 499-504 (2008). hitps:/doi. org/10.1016/j.jhazmat.2008.02.046 5. Said, L, Missaoui, A., Lafhaj, Z.: Reuse of Tunisian marine sediments in paving blocks: factory scale experiment. J. Clean. Prod. 102, 66-77 (2015). https://doi.org/ 10.1016). jclepro.2015.04,138 6. Samara, M., Lafhaj, Z., Chapiseau, C.: Valorization of stabilized river sediments in fired clay bricks: Factory scale experiment, J. Hazard. Mater. 163(2-3), 701-710 (2009). hitps://doi. org/10.1016/j.jhazmat,2008.07.153 7. El Fgaier, F., Lafhaj, Z., Chapiseau, C.: Use of clay bricks incorporating treated river sediments in a demonstrative building: Case study. Constr. Build. Mater. 48, 160-165 (2013). https://doi.org/10.1016/j.conbuildmat.2013.06.030

Study of C-S-H Formation of Cemented Sediment Brick

509

. Mymrin, V., et al.: Utilization of Sediments Dredged from Marine Ports as a Principal Component of Composite Material. J. Clean. Prod. 142, 4041-4049 (2017). https://doi.org/ 10.101 6/j,jclepro.2016.10.035 . Wang, L., Yeung, T.L.K., Lau, A.Y.T., Tsang, D.C.W., Poon, C.S.: Recycling contaminated sediment into eco-friendly paving blocks by a combination of binary cement and carbon dioxide curing. J. Clean. Prod. 164, 1279-1288 (2017). https://doi.org/10.1016/j.jclepro. 2017.07.070 . Wang, L., Kwok, J.S.H., Tsang, D.C.W., Poon, C.-S.: Mixture design and treatment methods

.

for recycling contaminated sediment. J. Hazard, Mater. 283, 623-632 (2015). https://doi.org/ 10.101 6/j,jhazmat.2014.09.056 Compressive strength and microstructural analysis of unfired Oti, LE. Kinuthia, J., Bai, clay masonry bricks. Eng. Geol. 109(3-4), 230-240 (2009). https://doi.org/10.1016/. enggeo.2009.08.010

. Woen, E.L., Malek, M.A., Mohammed,

B.S., Chao-Wei, T., Tamunif, M.T.: Experimental

study on compressive strength of sediment brick masonry. AIP Conf. Proc. 1930(1), 020017 (2018). https://doi.org/10.1063/1.5022911

. Ean, L.W., Malek, M.A., Mohammed, B.S., Tang, C.-W., Tamunif, M.T.: Flexural and shear bond strength of sediment brick masonry. Int. J. Recent Technol. Eng. 8(4), 6288-6294 (2019). https://doi.org/10.35940/ijrte.D5 109.1 18419

® ‘upaates

Flexural Behaviour of Glass Fiber Reinforced Polymer

(GFRP)

Tubes Subjected

to Static Load Agusril Syamsir'?, Abdulrahman Alhayek”, Audrey Yeow Yee Keng”, Daud Mohamad’, Mohamad Zakir Abd Rashid*, and Shuhairy Norhisham? ' Institute of Energy Infrastructure (IED, Universiti Tenaga Nasional,

JIn IKRAM-UNITEN, 43000 Kajang, Selangor, Malaysia Agusril@uniten. edu. my > Civil Engineering Department, College of Engineering, Universiti Tenaga Nasional, JIn IKRAM-UNITEN, 43000 Kajang, Selangor, Malaysia rahman. [email protected], [email protected], {Daud, shuhairy}@uniten. edu. my > Grid Solution Expertise, Tenaga Nasional Berhad, Bangunan Dua Sentral, no.8, Jalan Tun Sambathan, 50470 Kuala Lumpur, Malaysia Zakir. rashid@tnb. com. my Abstract. This paper presents the flexural behaviour of glass fiber reinforced polymer tubes subjected to static load. Two samples of GFRP crossarms produced by Wagner were tested under static four-point bending in order to determine the flexural behaviour. The results show that, both samples were able to withstand the same load and moment with a maximum load of 75 kN and

19.69 kN.m respectively. Wagner I had a lower mid-span deflection and a higher flexural modulus as compared to Wagner 2. However, Wagner 2 had a higher flexural toughness due to the contribution from the side of tubes to sustain the load until failure, unlike Wagner 1 where the initial crack occured at the edges of the tube so that the tube failed at 11 mm mid-span deflection. The different behaviour between these two samples is because the samples were taken from different location of transmission tower and already exposed to the

environment and load from the transmisson tower’s conductor.

Keywords: Flexural behaviour - GFRP - Four-point bending 1

Introduction

Composite materials are typically made of two components, a reinforcement phase dispersed in a matrix phase [1]. These two components however are insoluble to each other and as a result form a composite

[2]. The

matrix constituent is made

from either

polymer, metallic or ceramic and it acts as a protection layer to the reinforcement by enclosing it at the desired location besides transferring the load between fibres. Reinforcement however is made from fibres, particles or filaments and functions as the main load bearing element. Matrix is a homogeneous and monolithic material besides

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 510-516, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_59

Flexural Behaviour of Glass Fiber Reinforced Polymer (GFRP)

S11

being completely continuous while the reinforcement is the discontinuous or dispersed phase of the composite which provides high strength and stiffness. The matrix transfers the load to the fibres after the fibre reinforced composite is loaded [1, 3, 4]. In addition,

the matrix provides finish, texture, colour, durability and functionality [4]. Therefore, combining these two composites together produces better mechanical properties. In the case of GFRP, glass fibre acts as reinforcing phase, also known as hard phase while thermosetting epoxy acts as matrix phase, also known as soft phase [3]. FRP has an additional protective layer of insulation and is non-flammable which enables the material to be suitably used as a construction material for the cross arm of transmission towers [5]. The transmission tower is an important component

to the transmission line.

This is because the transmission tower is responsible to hold the transmission conductor at a certain height from the ground. The main structure of the transmission tower is the cross arm which holds the responsibility to support the transmission conductor [6]. FRP can be considered a suitable material to be used in transmission tower due to

the low electric conductivity, light weight and the ease of assembly which due to most fabricators producing GFRP pole products similar to the wooden counterpart. When comparing the type of section which was suitable to be used in transmission tower, the FRP pultruded circular-section with 200 m span between frames can be considered the optimum solution amongst the circular-section, rectangular-section and I-section as it was the most cost effective [7]. In this study, GFRP

crossarm was subjected toa

static

load at a single direction in order to investigate the flexural behavior by setting up the GFRP at a self-reaction test frame with two supports in the laboratory.

2

Material and Methodology

In this study, 6.3 mm)

the

hollow

were taken from

pultruded

GFRP

the main member

square

tubes

(127 mm

of a transmision

x

127mm

x

tower’s cross arm. The

tubes are made up of vinyl ester resin and E-glass fibre reinforcement. The tubes are made of 9 plies of E-glass fiber manufactured by Wagner’s Composite Fibre Technologies, Australia in this arrangement —[0°/+45°/0°/-45°/0°/-45°/0°/+45°/0°]. Two

GFRP

square tubes

labelled as Wagner

1 (W1)

and Wagner

2 (W2)

were taken

from different location of cross arm has been tested in this study. The samples are identical and were prepared by cutting the composite into a length of 2000 mm for each sample. A static four-point bending test as per ASTM D7250 [8] was conducted by applying static loading to the sample with loading increment of3 mm/min as shown in Fig. 1. In order to observe the flexural behaviour of the sample, two unidirectional strain gauges were attached at the top and bottom surfaces of the GFRP cross arm at the mid-span. Two strain gauges, a linear variable differential transducer (LVDT) and portable data logger were used to record the strains, applied load and the deflection at the mid-span of tubes, respectively.

512A. Syamsir et all.

Lt= 1350 mm

Fig. 1. Setup of crossarm member in self-reaction test frame 3

Results and Discussion

This section describes the results of the flexural testing on the samples by measuring the strains and deflection of the tubes. Flexural toughness, modulus and failure mode will also be discussed in ths section. 3.1

Flexural Testing Results

Flexural test result for these two sample were presented in Table |. The tubes have the same failure load and different strain at top and bottom part of the cross arm. W2 shows more deflection and also higher strain at the bottom and top part of the tube. Table 1.

Flexural testing result.

Sample|d L/L (mm)/a | Failure | Failure (mm) (mm) load Moment Wi W2

(kN)

[127 | 2000/1350 525/75 | 127 2000/1350 525 | 75

(kN.m)

19.69 19.69

Deflection (mm) 1 18,

|Top | Bottom strain | strain (us)

-317 -225

(ud)

| 532.8 | 718.2

Figure 2 shows that as the load exerted increases, the deflection of the GFRP cross arm increases until the occurrence of failure. By comparing the samples, sample W2 has the highest mid span deflection which is 18 mm at the loading of 75 KN. These figures imply that the load deformation behaviour is more or less linear elastic to failure for both tubes. This result is in line with the results found by Majid Muttashar et al. [9],

the load-deflection curve is linearly elastic until failure. Figure 3 shows the load and strain (at mid span at the topmost and bottom most section) relationship of the hollow tubes which is linear elastic to failure. It can be seen

that the W1

tube failed at a compressive and tensile strain of 317 microstrains and

513

Ss

Failure Load (KN)

Flexural Behaviour of Glass Fiber Reinforced Polymer (GFRP)

—a— Wagner 1 0

5

—— Wagner 2 10

15

20

Mid-span deflection (mm) Fig. 2. Load vs deflection graph of the samples 532.8 microstrains, respectively, while sample W2 failed at a compressive and tensile strains of 225 microstrains and 718.2 microstrains, respectively. With increasing load, however, the measured strain tend to become positive indicating the initiation of cracks at the maximum moments locations where the strain gauges were sitting and as a result they started to get detached from the top surfaces. This finding in line with Guades, E et al. [5], where the load-strain curve for GFRP

is linearly elastic for the bottom strain

which is in tension until the occurrence of failure happen. —*— Wagner

80

1-

top strain

Load (KN)

—— Wagner 1bottom strain —— Wagner 2top strain we

Wagner 2-

bottom

-500

Oo.

500

Strain (microstrains)

1000

train

Fig. 3. Load vs strain graph for sample Wagner I and 2 3.2

Flexural Toughness of GFRP

Sample

The flexural thoughness as shown in Table 2 was obtained from the calculated area under the load vs deflection curve of each sample from Fig. 2.

514A. Syamsir et all. Table 2.

Flexural Toughness Result.

Sample _ | Flexural Toughness (kN.mm) Wagner1 | 357.5 Wagner2 756.0

Based on the flexural toughness table, the sample Wagner 2 has higher flexural toughness as compared to Wagner 1. The approximate difference between these two samples is 7%. It is shown W2 is tougher as compared to W1. This result is in line with the finding from Fig. 2, which proves that W2 still can support the load until it achieves 18 mm deflection. 3.3.

Flexural Modulus of GFRP

Sample

To determine the flexural modulus, a stress vs strain graph was plotted and the slope of the curve was measured. The slope taken is the linear curve at the beginning of the curve. The top strain and bottom strain plotted were extracted from the data logger while the stress of the GFRP cross arm was determined by the following formula:

oa

(1)

Whereby;

o = Stress (N/mm?) M = Moment

(N.mm)

y = Neutral Axis (mm) = % h I=

Area moment of inertia (mm*) = bh/12

Table 3.

Flexural Modulus.

Sample | Flexural stress (N/mm’) | Flexural strain (1) | Flexural Modulus (N/mm?)

Wl W2

56.13 | 46.13

468 522

119933 88377

Table 3 shows the flexural modulus of sample W1 and W2. It is shown that flexural modulus for W1 is higher as compared to W2. The higher flexural modulus, the higher stifness of the tubes and the lower the mid-span deflection. It can be observed in Table 1 where sample W1 has a lower deflection of 11 mm as compared to W2. 3.4

Failure Mode of Crossarm

Figure 4 shows crack pattern of Wagner | where the cracks were formed as dents at the top surface of the cross arm situated at loadings. Slight vertical cracks were extended

Flexural Behaviour of Glass Fiber Reinforced Polymer (GFRP)

515

Fig. 4. Deform of sample Wagner 1 top and side view inwards at the top and slight horizontal cracks were formed at the sides of the cross arm. It was observed the failure of this tube due to the crack innitiated at the edges of tube.

Fig. 5. Deform of sample Wagner 2 top and side view It can be observed in Fig. 5 the transverse crack occured at the top flange of the W2 tube, it indicates the tube still can sustain the load until the mid-span deflection reached 18 mm as indicated by the presence of crack at the top flange of the tube and continued by cracrking the edges. The behaviour is a little bit different with W1, where the tube failed at a deflection of 11 mm due to the crack at the edges of the tube. It can be concluded that the edges of the tube has functioned as a support to the top flange of tube during loading.

516 4

A. Syamsir et all. Conclusions

This paper investigated the flexural behaviour of glass fiber reinforced polymer (GFRP) corssarms subjected to static load. It can be concluded that, Wagner | has lesser midspan deflection and higher flexural modulus as compared to the Wagner 2. However, eventhough Wagner 2 has higher mid-span deflection, this sample has a higher flexural toughness due to the contribution from the side that acting as support to top part of tube to sustain the load until failure. The different behaviour between these two samples is because the samples were taken from different transmission tower and also already exposed to the environment and the load from the transmisson tower’s conductor.

Acknowledgements. The authors acknowledge Tenaga Nasional Berhad (TNB), UNITEN R&D, and Institute of Energy Infrastructure (IED) for the lab facilities and financial support (TNB

Seeding Fund: U-TS-RD-19-03). Special thanks to those who contributed to this project directly or indirectly. References

v

1. Karatas, M.A., Gokkaya, H.: A review on machinability of carbon fiber reinforced polymer (CFRP) and glass fiber reinforced polymer (GFRP) composite materials. Defence Technol. 14(4), 318-326 (2018) . Syamsir, A., Ishak, Z.A.M., Yusof, Z.M., Salwi, N., Nadhirah, A.: Durability control of UV

radiation in glass fiber reinforced polymer (GFRP) - A review. In: AIP Conference Proceedings. vol. 2031, pp. 020033 (2018) 3. Chaudhary, S.K., Singh, K.K., Venugopal, R.: Experimental and numerical analysis of flexural test of unfilled glass fiber reinforced polymer composite laminate. Mater. Today Proc. 5, 184-192 (2018) 4. Syamsir, A., Mohamad,

D., Beddu, S., Itam, Z., Sadon, S.N.: An examination on durability

and degradation of glass fiber reinforced polymer structures. Test Eng. Manage. 81, 33793388 (2019) 5. Guades, E., Aravithan, T., Mainul Islam, M.: Characterisation of the mechanical properties of

pultruded fibre reiforced polymer tube. Mater. Design 63, 305-315 (2014) 6. Nadhirah, A., Beddu, S., Mohamad, D., Zainoodin, M., Nabihah, S., Zahari, N.M., Itam, Mansor, M.H., Kamal, N.L.M., Alam, M.A., Muda, Z.C.: Properties of fiberglass crossarm in

transmission tower - a review. Int. J. Appl. Eng. Res. 12(24), 15228-15233 (2017) 7. Godat, A., Légeron, F., Gagné, V., Marmion, B.: Use of FRP pultruded members for electricity transmission towers. Comp. Struct. 105, 408-421 (2013) 8. ASTM D7250. ASTM D7250/D7250M-06 : Standard practice for determine sandwich beam

flexural and shear stiffeness. West Conshohocken, (PA): ASTM International (2006) 9. Muttashar, M., Manalo, A., Karunasena, W., Lokuge, W.: Influence of infill concrete strength on the flexural behaviour of pultruded GFRP square beams. Comp. Struct. 145, 58-67 (2016)

®

Check updatesfor

Investigation on Behavior of Concrete Slab Due to Low Velocity Impact Using Numerical Modeling Agusril Syamsir'®, Syahidatul Islamiah”, Shuhairy Nothisham?, Nur Liyana Mohd Kamal’, Norazman Mohamad Nor*, and Vivi Anggraini* ' Institute of Energy Infrastructure (IED, Universiti Tenaga Nasional,

Jin IKRAM-UNITEN, 43000 Kajang, Selangor, Malaysia Agusril@uniten. edu. my > Civil Engineering Department, College of Engineering, Universiti Tenaga Nasional, JIn IKRAM-UNITEN, 43000 Kajang, Selangor, Malaysia syahidatulislamiah@gmail. com, {shuhairy, Yana_Kamal}@uniten. edu. my > Faculty of Engineering, Universiti Pertahanan Nasional Malaysia (UPNM), 57000 Kuala Lumpur, Malaysia azman@upnm. edu. my + School of Engineering, Monash University Malaysia, Jalan Lagoon Selatan, Bandar Sunway, 47500 Subang Jaya, Selangor, Malaysia vivi.anggraini@monash. edu Abstract. This paper presents on the low velocity impact loading on plain slab concrete. There are three different thicknesses of concrete slabs; namely 20 mm, 30 mm, and 40 mm, all of which have been subjected to 1 blow of impact from

1.25 kg steel ball per slab. The resulting deflections and stresses on the concrete midspan was then observed. It was found that the highest deflection occurs for the concrete slab with the 20 mm thickness with a value of 9.965 mm, while the lowest deflection was found to be at 4.157 mm for the slab thickness of 40 mm.

The highest mid-span stress occured at 20 mm slab for 500 mm drop height. During impact it was observed the slab is still in elastic linear region.

Keywords: Low velocity impact - Concrete slab 1

Introduction

Reinforced concrete slabs are among the most common structural elements used in construction throughout the world. As part of the structural elements, slabs have always been subjected to static and even dynamic loadings such as impact from falling objects onto the slab. However, the requirement for the impact resistance of slab is not properly adressed in the design code. Some experimental studies have been conducted pertaining to drop weight impact onto the slab by measuring its energy absorption as well as the crack pattern of the slab. In this study, the authors presented the different crack patterns and energy absorption for various slab thickness due to impact loading [1-6].

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 517-523, 2021. https://doi.org/10.1007/978-98 1-33-63 1 1-3_60

518

A. Syamsir et al.

Since conducting numerous repetitive experiments for various parameters inquire a higher cost, it was decided that a numerical study is a better suited solution to solve this issue cost-wise. Prediction on the behaviour of structure due to static and dynamic loads has shown good agreement with experimental results as reported by previous researchers [7-10]. Therefore a numerical model has been developed in this study to predict deflection and energy absorption of slab due to various velocities of impact loads.

2

Methodology

Table 1.

Material properties for numerical simulation.

No| Material | Density | Young’s modulus | Poisson ratio (kg/m3) (N/m2) 1 [Concrete [2400 | 4100 0.21 2. | Steel Ball 7850 | 2.1 x 109 0.33 In this study, all the materials were modelled by the using of 3D solid elements in ABAQUS analysis software. The material properties are outlined in Table 1. In this study, unreinforced

slab with various thickness of 20 mm,

tested in this simulation.

The

numerical

test is simulated

30 mm,

and 40 mm

for these

were

three varying

thicknesses with the same scale of velocity at the drop heights of 20 cm, 30 cm, 40 cm,

and 50 cm. The schematic diagram of this simulation is shown in Fig. 1. In this model the unreinforced concrete slab was clamped by steel clamps.

Impact Mass

—* [|

|

}

Drop Height (h)

Concrete Slab

Fig. 1. Schematic diagram of the concrete slab setup

Investigation on Behavior of Concrete Slab

519

Fig. 2. Geometrical model with boundary In the numerical model, the slab was pinned of slab as shown in Fig. 2. A steel ball weighing various heights ranging from 0.2 m to 0.5 m that from 1.98 m/s to 3.13 m/s which are calculated

v=

supported at the edges of bottom part 1.25 kg is dropped vertically from a corresponding to velocities that range by using Eq. |.

V2gh

(1)

where v is the velocity of the steel ball (m/s), g is the acceleration due to gravity (m/s) and h is the drop height of the steel ball (m). The potential energy which are released by the steel ball from three different drop heights can be calculated using Eq. 2.

Ep = mgh

(2)

where Ep is the potential energy of steel ball (Nm), m is the mass of steel ball (kg), g is the acceleration due to gravity (m/s) and h is the drop height (m).

3

Results and Discussion

in this section, velocity of steel ball, as well as its potential energy, mid-span deflection and stress of slab is presented. Three thicknesses of samples were used for the simulation, namely: concrete slab 20 mm,

30 mm,

blow of impact are presented in Table 2.

and 40 mm

thick. The deflections for 1

520A. Syamsir et all. Table 2.

Simulation analysis results for each slab thickness.

Thickness (mm) | Drop height | Velocity (m/s) | Potential energy (Nm) | Deflection | Stress (mm)

20

200 300

30

40

(mm)

[1.98 (2.53

[6.833 8.164

9.225 9.965 3.754

49.53 57.81 | 28.97

4301 [4.845 (5.014 (3.302 3.754

| 33.53 | 38.12 | 39.9 | 10.4 | 12.5

400 500 200

2.8 3.13 1.98

491 6.13 2.45

300 400 500 200 300

(2.53 (28 (3.13 1.98 (2.53

3.68 (4.91 (6.13 | 2.45 3.68

400 500

28 3.13

(N/mm?)

[2.45 (3.68

491 6.13

3.895 4.157

34.5 | 42.9

14.01 15.75

It can be observed in Table 2, that the higher velocity of steel ball will consequent in higher values of potential energy and deflection located at the middle of slab. The maximum deflection for the 20 mm, 30 mm, and 40 mm slabs are found to be are 9.965 mm, 6.590 mm, 4.157 mm, respectively. For the 20 mm slab, the drop height of 0.5 m resulted in a 46% higher deflection as compared to the 200 mm drop height. Meanwhile, the 30 mm and 50 mm slabs show the drop height of 500 mm resulting in a 34% and 26% higher deflection as compared to the 200 mm drop height. It could also be observed that the slab with 20 mm thickness has the highest sensitivity to the velocity of the steel ball. This could be due to the fact that lower thicknesses tend to produce a lower second moment of inertia for the slab, resulting in higher deflection for that particular as compared to 30 mm and 40 mm thickness. Another observation made from Fig. 3 was that the correlation between the velocity of the steel ball and the resulting deflection of slab has a R* value that is higher than 0.95. it is found that during impact loading, the slab is still in the elastically linear range. Table 3 shows the relationship between two parameters, which are velocity versus deflection and stress at mid-span of slab. Y represents deflection and stress, while X represents the velocity of the steel ball. Those equations expressed that the velocity of the steel ball plays a big role in in deflection and stress of slab as well. The linearity of the relationship between velocity and deflection and stress can be observed in these equations.

Table 3.

Empirical formulation of velocity versus mid-span stress and deflection

Thickness (mm) | Deflection Stress 20 Y =2.78X + 1.28 20.15 X — 6.4 30 Y=LISX+15 |/Y=9.94X+9.18 40

Y = 0.73 X +

1.86

Y = 4.65 X + 1.03

Investigation on Behavior of Concrete Slab

521

Cy s

a

y =1.1535x + 1.468 R? = 0.9661

N

Mid-span deflection (mm)

»8

b5B

y = 2.7848x + 1.2785 R? = 0.9909

y = 0.7342x + 1.8608 R?=0,9951 30mm = @40mm

420mm 15

2

25

3

3.5

Velocity (m/s) Fig. 3.

Velocity versus deflection for various slab thickness of 20 mm, 30 mm, and 40

Correlation between velocity of steel ball versus mid-spa stress of the slab can be observed in Fig. 4. It can be observed that 20 mm slab produce the highest stress as compared to 30 mm and 40 mm slab. Mid-span stress of slab has linear correlation velocity of steel ball which reveal that the slabs still in linear elastic region during

impact.

Mid-span stress (N/mm?)

70 y = 20.153x - 6.4154 R? = 0.9817 y

50

=9.9427x + 9.1797 R? = 0.9719

x 30

a y = 4.6491x + 1.031 R? = 0.9908

© °

15

420mm

2

830mm

25

eo @40mm

3

3.5

Velocity (m/s) Fig. 4. Velocity 40 mm

versus

mid-span

stress for various

slab thickness

of 20 mm,

30 mm,

and

522

A. Syamsir et al.

Figure 5 shows the effect of slab thickness on the stress of the slab with the highest stress (red color) occurred

at the mid-span

of slab. It can be seen

20 mm thickness shows the greater area with high stress as compared 40 mm slab thickness. As presented in Table 2, where the mid-span 57.81, 39.9, and 15.75 N/mm? for 20 mm, 30mm, and 40 mm respectively. This due to the highest deflection occurred at 20 mm lowest of second moment of inertia (J) for slab 20 mm that producing

that the slab with

to the 30 mm and stress of slabs is slab thickness, slab and also the the highest stress.

Fig. 5. Stress pattern at failure load for various thicknesses of slab 20 mm, 57.81 N/mm? (a), 30 mm, 39.9 N/mm? (b), 40 mm, 15.75 N/mm? (c) for drop height of 0.5 m.

4

Conclusions

It can be concluded that thickness of slab has effect on deflection of slab at the middle span. The thinnest slab produces the highest deflection and stress due to the lowest of second moment of inertia (J) of slab. Steel ball velocity also has effect on the deflection

of slab, the higher velocity the higher the deflection. It is concluded that finite element software is able to predict and simulate the behavior of concrete slab subjected to low velocity impact load.

Acknowledgments, The authors express their gratitude to Universiti Tenaga Nasional (UNITEN), Malaysia for supporting this research. Special thanks to those who are contributed to this, project directly or indirectly. References

1, Zineddin, M., Krauthammer, T.: Experimental study of reinforced concrete slabs subjected to impact loading, In: Fourth International Conference on Concrete Under Severe Conditions, Environment and Loading. pp. 27-30, Seoul, Korea (2004) 2. Nguyen, M.Q., Jacombs, $.S., Thomas, R.S., Hacnhenberg, D., Scott, M.L.: Simulation of impact on sandwich structures. Comp. Struct. 67(2), 217-227 (2005) 3, Wang, Z., Li, Y.C., Shen, R.F., Wang, J.G.: Numerical study on craters and penetration of concrete slab by ogive-nose steel projectile. Comput. Geotech. 34, 1-9 (2007) 4, Faham, T.: Numerical Modelling of Reinforced Concrete Slabs Subjected to Impact Loading. University of Wollongong (2008)

Investigation on Behavior of Concrete Slab

523

. Muda, Z.C., Beddu, S.: Impact resistance behaviour of light weight rice husk concrete with bamboo reinforcement. IOP Conference Series: Earth and Environmental Science (2016) . Muda, Z., Mustafa, K.N.: Impact resistance performance of banana fiber reinforced slabs. IOP Conference Series: Earth and Environmental Science (2016) . Zahari, N.M., Zawawi, M.H., Sidek, L.M.: Introduction of discrete phase model (DPM) in fluid flow: A review. AIP Conference Procedings. vol. 2030 (2018) . Zawawi, M.H., Saleha, A.H., Salwa, A., Hasan. N.H.: A review: Fundamentals of

computational fluid dynamics (CFD). AIP Conference Proceedings. vol. 2030 (2018) . Norazman, M.: Static analysis and design of sandwiched composite long-span portable beam. Int. J. Phys. Sci. (IIPS) 6, 6323-6328 (2011) . Norazman, M.: Dynamic analysis of sandwiched composite foldable structure under heavy vehicle load. Appl. Mech. Mater. (AMM) 110-116, 2331-2336 (2011)

® ‘upaates

Investigating an Optimum Mixing Method to Produce Foam Concrete Fulfilling the Workability, Density, Shrinkage, Strength and Total Volume M. I. Safawi!®, SN. L. Taib!, L. P. Hua’, and A. Rashidi! ' Department of Civil Engineering, Faculty of Engineering, Universiti Malaysia

Sarawak, 94300 Kota Samarahan, Sarawak, Malaysia mzmisafawi@unimas. my

? Public Works Department, Wisma Saberkas, Kuching, Malaysia

Abstract. Production of foamed concrete may encounter segregation, shrinkage and poor strength. The nature of mix is flowable but must maintain

homogeneity of both foam and sand particles. One of the reasons for these problems is the difficulty in mixing method. Two mixing techniques, wet and dry mixing methods, were investigated and resulting properties recorded. Five main properties characterized a foamed concrete mix and these are density, workability, shrinkage, strength and total volume. The first mixing method, wet mixing method, tend to have segregation and shrinkage problems. The target density was difficult to achieve. The second mixing method, dry mixing method,

proved to achieve target density easily and lesser shrinkage problem. The dry mixing method proved to produce foamed concrete with specific target density

and workability.

Keywords: Foamed concrete « Production « Shrinkage - Segregation » Dry mixing method

1

Introduction

Today’s construction industry is constantly in need of new materials for use. One of the most common materials is concrete. One special concrete that are needed by the industry but not easily available is lightweight foamed concrete. This lightweight concrete is foamed concrete by introducing stable foam in mortar. Foam concrete is produced

from

four basic

materials,

viz cement,

water,

sand

and

foam

(Fig.

1). The

foam used in this process is a category of stable white bubbles flowing from a compressed tank. According to Van Deijk (1992), the air-pores are initiated by agitating air with a foaming agent diluted with water then the foam was carefully mixes together with cement slurry to form foam concrete. Aldridge (2005) concluded that the S:C ratio

of mix have a proportional relationship to strength. The higher sand content in a mix will produce a stronger concrete. However, higher solid content like sand will cause the mix less flowable and posed higher risk to segregation (Aslam et al. (2017)).

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 524-533, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_61

Investigating an Optimum Mixing Method to Produce Foam Concrete

=

we

Fine Sand

™

Fig. 1.

cement

water

“ Foamed

a

525

>

concrete

|

Stable

~~

¥

foam

showing the four ingredients that produce foam concrete

in the production of foamed concrete, there are two phases viz fresh and hardened phase. The thre important properties measured in the fresh phase are workability, density and volume. Upon hardening, the two properties measured are shrinkage and compressive strength. Fresh foam concrete is a self-flowing slurry mix. The density of concrete depends primarily on the amount of foam added to the mix. The more foam added the lower the density. Hitherto, there are no standard method of determining a density measurement that reflect on the homogeneity of mix (BSI 1983). During the mixing process, its volume increases as more foams are added in the mix. Addition of foam must be strictly done without segregating the mix. Once the mix is segregated, it has to be disposed and impossible to remedy (Kunhanandan et al. 2008). With respect to hardened foam concrete, the most important aspect to mitigate is shrinkage. In this study, shrinkage is measured as percentage to original dimension of mould. After casting the mould or formwork, lighter materials like water or foam tend to move upwards to the surface. Due to relative humidity or evaporation the materials vaporized and cause the concrete to shrink when dried (Yuan 2012). Strength property of foamed concrete depended on the volume of solid content in the mix proportion (Ramamurthy

et al. 2009).

Table | summarizes some common problems with foamed concrete with respect to the five properties mentioned above. The mixing of foamed concrete is not as simple as normal weight concrete. Any errors in mixing will cause problems to the five properties mentioned above. It is pertinent that the right mixing technique be identified in order to produce foamed concrete as per designed. Table 1.

Problems encountered in failed mixing technique

Phase —_| Properties Typical difficulties Fresh | 1 | Workability | Too much water can cause segregation 2|Density | Wrongly added foam in mix affects target density 3|

Volume

Too little mix volume means uneconomical mix.

Hardened |4 | Shrinkage | Segregated mix will cause shrinkage problem 5 | Strength

Percentage of solid content determine strength

526

M. I. Safawi et al.

There are two main objectives of this study. The first objective was to compare the production of foam concrete using the conventional wet and newly introduced dry methods. The second objective was to measure five main properties in foam concrete production and investigate its relationship with respect to production methods. The outcome of this study shall allow simpler method of mixing foam concrete with specific

property targets.

2

Experimental Procedure

2.1

Materials

In this study, foamed concrete were made from basic equipment readily available in the market. One of the objectives of this research is to identify a practical mixing method of producing foamed concrete. With the right mixing method that is simple and practical, the production of foamed concrete should be easily done like any ordinary concrete. Figure | and 2 show the compressed cylindrical tank and mini slump cone used to produce the foam and measure slump spread, respectively.

100,70mm dia x 60mm ht

Fig. 2. Ordinary 70 1 compressed tank

Fig. 3. Standard mini slump cone

Ordinary compressed water tank and standard mini slump cone (as shown Fig. | and 2 respectively) were used to produce foam bubbles and measure slump spread. Workability of fresh mix was measured using the mini slump cone. Other ingredients used for the experiments

are shown

Table 2. Ingredient Sand Blended cement Foaming agent Superplasticizer

in Table 2 (Fig. 3).

Properties of mix ingredients Value Dry, passing 600 micron 80% OPC, 20% limestone Kao Foam Agent Softclean 27AE | RF-611

Investigating an Optimum Mixing Method to Produce Foam Concrete

2.2

527

Mixing by Wet and Dry Methods

There are two mixing methods used in this study. Table 3 shows the two methods and differentiate the mixing procedures used. The main difference is in determining the slump spread and density values. The conventional mixing procedure, labelled here as wet mix, require mixing with water before adding the foam bubbles. Workability is measured first before density. By designing a new mixing procedure, termed as dry mix, the measurement of density preceded the workability. The mix is dry initially until the time water is added, thus, become flowable.

Table 3. Mixing procedures for wet and dry mix Wet Mix Time __| Ingredients

Measured

Dry mix Time __| Ingredients

Measured

(s)

properties

(s)

properties

0

Start

120

Sand + cement

240 360

Water + Sp Foam

Slump Density

0

Start

120

Sand + cement

420 720

Foam Water

840

End

Density Slump

End

The mix proportion for the two methods are given in Table 4 below. Specimen size for compression testing was 150 mm cube and air cured. All specimens were tested for 28-days compressive strength.

Table 4.

Mix proportion for the two methods

Method |C:$_| w/e

Wet Dry

3 3.1.

| SP (%C) | Foam

| 1:1.3 0.29 13 | 1:2 | Until flowable state | NA

Until target density Until target density

Results and Discussion Concrete Samples Using Conventional Wet Mixing Method

A series of 10 experiments were done in the laboratory using the conventional wet mixing methods. Initially sand and cement were added. Then water and superplasticizer. The mix was workable and flowable. Using the mini slump cone, the slump spread was measured. By taking two perpendicular length measurements and calculate the average value. Foam bubbles were added to reduce the density. By taking 1-L sample in a fixed volume container, the mix density was measured using a balance. Once the target

528

M. L Safawi et al.

density is achieved, the mixes were cast into 30 numbers of 150 x 150 x 150 mm moulds. The following observations were noted. a)

b)

Volume: The total mix volume per batch was around

100-120

L. At this rate it is

comparatively too small a quantity if the concrete were to be produced at commercial rate. With too small volume the foamed concrete will not be economically viable. The conversion to commercial rate procurement of foamed concrete is expected to be around RM300-350/m*. As a comparison, an average rate for Grade 30 normal weight concrete is RM250-300/m°. Density: The target density was set at 1000 kg/m’. Based on the mixing results, it was difficult to achieve the target value. The density obtained from the 10 samples

ranged from 851 to 1152 kg/m*. A gap of 301 kg/m? was practically too big a c)

d

e)

margin between the smallest and largest density values. Workability: For all 10 samples the average slump spread was 200 mm. This means all the mixes were deemed flowable but having too high risk of segregation. It is evident that most of the flowable mix had segregated. The average shrinkage percentage of the hardened samples was 0.9%. In all 30 samples in each batch, more than half of the exhibited significant dimensional change. Figure 4 and 5 shows some typical samples with shrinkage and brittle surface condition, respectively. Strength: For all the hardened samples the concrete strength varied from 010 N/mm”. Since the samples experienced segregation, the strength value was rejected and did not portray the expected strength. The height of tested samples was lesser than 150 mm. Shrinkage: Fig. 6 shows the shrinkage results in percentage of hardened foamed concrete ranging from all 10 samples (density 852-1152 kg/m’). The shrinkage value is measured 24 h after casting the samples and the samples hardened. The results showed that most of the samples changed in dimension between 0-2.5%. A shrinkage of 2.5% is translated as a depression of 4 mm, which is physically visible. After casting the mixes in the moulds lighter materials like water and foam tend to move upwards to the surface. As it reached the top surface the liquid slurry slowly evaporated due to temperature difference or relative humidity. As it hardened, the loss in water and foam contributed to the dimensional change.

Fig. 4. Shrinkage upon hardening

Fig. 5. Poor hardening at lower density

Investigating an Optimum Mixing Method to Produce Foam Concrete

529

Shrinkage

YD SS

arg

SP Sv

WR

SF GO DPDOD DP PP L LWSOLO HPWH NO a LLM LL NL LN

Sample density between 852-1151 kg/m?

Fig. 6. Percentage shrinkage of samples produced using wet mix method Discussion: In the conventional wet mixing technique the initial target was to achieve target flowability. The mix must be flowable by adding water and superplasticizer. The subsequent target was to achieve the target density by adding foam in the mix. A slump spread value was measured before the process of adding the foam. Adding water, sand, water and superplasticizer produced a mix that can fulfil the flowability criteria. However, the mix viscosity was too high such that the sand particles easily segregate from the liquid materials. The process of adding foam into the mix tend to further aggravate the risk of segregation. As the foam were added, the mix volume increased. This increase in volume was due to the foam embedded in body of water. As the concrete hardened, the foam and water were evaporated and the concrete shrank. Based on the experimental results, it was evident that the conventional wet mixing technique has a problem with concrete shrinkage as it hardens. 3.2.

Concrete Samples Using Dry Mixing Method

The next stage was to mix foamed concrete using the dry method. This method was contrived based on the researchers’ innovation and creativity. In this dry method, sand and cement were added first. Then foam was sprayed on the mix until target density was achieved. Subsequently, water was added until it reached the target workability as given by a slump spread measurement of above 140 mm In the ensuing experiment, the target density was again fixed at 1000 kg/m’, similar to the earlier ones. After sand and cement were mixed thoroughly, foam bubbles were added to the mix until it reached a density of 1000 kg/m*. For a mix with C:S ratio of 1:1 the amount of foam mixed into the mixer was about 12.0 + 2 L. By now the target density can be achieved by controlling the amount of foam added into the mix. The subsequent process was to improve its workability by adding water. In order to observe workability change towards becoming a flowable mix, a preliminary experiment was conducted whereby | L of water was added step by step while taking the slump spread.

530

M. L Safawi et al.

Figure 7 shows the change in slump spread as | L of water added step by step. It became obvious that addition of water in the mix improves the workability without affecting target density. The photos of the slump spread at initial and final phase of adding water showed the change from a rigid to flowable state. After the above preliminary test, another series of 10 experiments were done in the laboratory using the new dry mixing methods. In each batch of mixing, the mix was cast into 150 x 150 x 150 mm mould. The following observations were noted.

b)

c)

d)

Volume: The total mix volume per batch was around 160-200 L per mix. It was possible to cast 40 number of moulds per batch mixing. This is a comparatively abundant. For the mix proportion used in this experiment, the market estimated cost of foamed concrete can range between RM150-200/m*. At his rate, foamed concrete will be a competitive option to industry players that intend to use it in construction works. Density: The same target density was set at 1000 kg/m? as in the earlier experiment using wet mixing method. Out of 10 batches the range of density was between 900 and 1114 kg/m*. The difference between the maximum and minimum value was 214 kg/m*. This dry mixing method is better than the wet mixing method with respect to achieving the target density. Workability: The slump spread was taken at the end of the experiment. The amount of water needed to achieve flowable state was in the range 10.0 + 2 L. Precautionary steps were taken so that segregation do not occur. The flowable state was obtained by slump value 140 mm and above. It is important to obtain a good workability to easily cast the concrete into the moulds. A flowable mix will be selfcompacting and fill all the spaces. Strength: All samples recorded compressive strength less than 1 N/mm/. It was evident that the mix with S:C ratio of 1:1 is inadequate to produce a strong concrete

Slump (Cm)

a)

0

2 4 6 Water added per litre each time (L)

8

Fig. 7. Change in slump spread as water was added per litre step by step

Investigating an Optimum Mixing Method to Produce Foam Concrete

531

Shrinkage

° ww SELLE SSSoS S 2

7

2

>

Sample density between 900-1114 kg/m? Fig. 8. Variation of percentage shrinkage using wet mixing method

Fig. 9. A piece of hardened foamed concrete showing little shrinkage

e)

Fig. 10. A batch of foamed concrete produced using dry mixing method

above 1 N/mm?. There is a need to do more research in this aspect to determine the factors affecting strength of foamed concrete. Shrinkage: The dry mixing method had a better performance with respect to shrinkage values. Figure 8 shows the variation of percentage shrinkage for all the samples. It was clear that shrinkage reduces with increased density. The average shrinkage percentage was 1.05% which is physically translated as insignificant depression of 1.5 mm. Figure 9 and 10 shows the condition of hardened cube samples that are dimensionally neat and tidy with little or no depression.

Discussion: The innovative wet mixing technique had shown a very encouraging results in terms of mitigating shrinkage of foamed concrete upon drying. For samples mixed at density 1000 kg/m? there was little shrinkage for all 10 batches of mixing.

532

M. L Safawi et al.

However, the workability of the mixes was limited to a slump spread of 140 mm were segregation be avoided. Although the mixes were flowable and self-compactable when cast in the moulds, a higher slump spread value would perform better. The innovative wet mixing method introduced the foam at an early stage of mixing. The foam adhered onto the surfaces of sand and increased the volume of mix. Thus, the mix density reduced. It became easier to measure the density and achieve the target by controlling the amount of foam added into the mix. There was little indicator that the cement particles reacted hydraulically with the foam bubbles because setting of mix did not occur. When water was introduced into the mix, the workability increased without affecting the density value. The presence of water separated the sand particles to increase its flowability. This outcome was very encouraging and allowed both density and workabililty be directly linked to the volume of foam and water, respectively.

4

Conclusion

Both mixing methods were able to produce foamed concrete as per density designed. However, it is easier to achieve the target density by producing foamed concrete using the innovative dry mixing method rather than the conventional wet mixing method. The former method successfully mitigated shrinkage problem than the latter. There was little difference in term of compressive strength of hardened concrete. Based on the volume produced per mixing batch, the dry method proved more superior than the wet method. Finally, the workability of both methods produced flowable mix though differed in average slump spread value. The wet mixing method had a high tendency to encounter segregation than the dry mixing method. In conclusion, the innovative dry mixing method is an optimum method of producing foamed concrete with specific target density and workability.

References American Society for Testing and Materials. ASTM C1611-C1611M: Standard Test Method for Slump Flow of Self-Consolidating Concrete (2015) Aldridge, D.: Introduction to Foamed Concrete What, Why, and How? Use of Foamed Concrete

in Construction, pp. 1-14 (2005) Aslam, M., Shafigh, P., Nomeli, M.A.: Blended coarse lightweight aggregates. J. Build. Eng. (2017) British Standard Institution (BSI). BS

1881

Part 107: Method

for determination of density of

compacted fresh concrete (1983) Kunhanandan, N., Ramamurthy, K.: Fresh state characteristics of foam concrete. J. Mater. Civil

Eng. (2008). https://doi.org/10.1061/(ASCE)0899-156 1 (2018)20.0(111) Yuan, L.O.: Engineering properties of lightweight foamed concrete, a thesis submitted in partial fulfilment of the requirements for the award of BSc (Hons) Civil Engineering, UNIMAS

(2012) Jones, M.R.,

McCarthy,

A.: Behavior and Assessment of Foamed Concrete for Construction

Applications, pp. 61-88 (2005)

Investigating an Optimum Mixing Method to Produce Foam Concrete

533

Lai, P.H.: An Experimental Study on Foamed Concrete Foundation as Peat Soil Replacement in Sarawak, a thesis submitted in partial fulfilment of the requirements for the award of MSc

Civil Engineering, UNIMAS (2011)

Nambiar, E.K.K., Ramamurthy, K.: Shrinkage behavior of foam concrete. J. Mater. Civil Eng. (2009) Ramamurthy, K., Nambir, E.K.K., Ranjani, G.LS.: A Classification of Studies on Properties of Foam Concrete, vol. 31, pp. 388-396 (2009) Van Deijk, S.: Foamed Concrete: A Dutch View, pp. 2-8 (1992)

® ‘upaates

Experimental Study of Two Stages on the Use of Local Rubber as Base Isolator

for Dwelling Houses

Civil Engineering Department, Krida Wacana Christian University, Jakarta, Indonesia {usman. wijaya, elly. kusumawati}@ukrida.ac. id Abstract.

Base isolators have not been extensively used in Indonesia because

they have to be primarily imported from abroad and thus, they need high costs. One of the efforts to reduce the costs of base isolators is the use of local rubber that is totally processed within the country, from harvesting the rubber latex

until manufacturing elastomeric seismic bearing. Before using local rubber as base

isolators, the prediction of vertical

and horizontal

isolator is required through a grade 40 uniaxial test;

stiffness of the base

50; and 60 shore A in IRHD

standard. The uniaxial test indicates that grade 50 samples have the best modulus so that the base isolator test only uses grade 50 rubber. From the test results,

the damping value for energy dissipation is 12%. This value gives hope for the future development of low-cost base isolator technology in Indonesia.

Keywords: Rubber - Base isolators « Uniaxial test « Horizontal stiffness + Damping 1

Introduction

Indonesia has a high earthquake potential. Throughout history, strong earthquakes with a large magnitude frequently hit many areas in Indonesia. The most recent highmagnitude earthquake in 2018 was the earthquake in Palu and Donggala, which killed thousands

of people

[1].

To prevent fatalities and damage due to earthquakes, a structure needs a tool that can control the earthquake force from the ground. Rubber is an alternative that can be used for developing a damping tool to protect a building from earthquake vibration from the ground. The use of rubber as an earthquake damper tool has a bright prospect because Indonesia is the second-largest rubber producing country in the world. In addition, the government through the Ministry of Trade intensively promotes the use of natural rubber for domestic needs and develops the natural rubber processing industry, especially the industry that can support national infrastructure development. Base isolators are very effective in reducing earthquake force through a separation of structure from ground motion [2]. In principle, base isolators extend the fundamental period of the structure to decrease the earthquake load that enters into the building structure

[3].

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 534-542, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_62

Experimental Study of Two Stages on The Use of Local Rubber

535

The high price of imported base isolators (ranging from $500 to $10,000 per unit) is partly caused by such complexity in the manufacturing process that this technology cannot be widely applied in the country. In most cases, buildings using base isolators will cost 0% to 50% higher than an ordinary building without base isolators [4]. A number of studies have been done to find low-cost earthquake absorbers, one of which is the use of replacement materials with steel plates. However, this will still produce an expensive earthquake absorber and the tool cannot be produced domestically. For this reason, innovation in the composition of the earthquake dampers is needed to produce domestically-manufactured and low-cost earthquake dampers that people can benefit from it. Rubber has not been widely used for earthquake absorbers, because so far, it is often used in the automotive industry so that the technology of mixtures for earthquake absorbers has not grown rapidly. Based on the most famous base isolator research [5], the value of shear modulus is limited to 0.7 MPa because of

limited engineering property data. To date, even in developed countries, rubber specifications are submitted directly to fabrication manufacturers that generally process the rubber for automotive purposes, so the rubber may have different behavior. Thus, the limited use of rubber as a basic damper has been incomplete works for scientists [6].

The results of this study are expected to have wider implications, not only for people who have a high income but can also be used for people who have a low income because safety belongs to everyone. The target of this research is public housing. In the future, it is expected that residential buildings in Indonesia will use a base isolator to minimize fatalities in case an earthquake occurs. Furthermore, this study is expected to encourage the development of base isolator technology in Indonesia, considering that Indonesia is one of the most earthquake-prone countries.

2

Method and Materials

This study was conducted with two stages of experimental testing. The first was the uniaxial tension test and hardness test to obtain mechanical properties of the rubber as shown in Fig.

1; the second, axial and double shear tests to determine vertical stiffness

and horizontal stiffness and to get a hysteretic loop to find the damping value ofa base isolator as shown in Fig. 2.

536

U, Wijaya and E, Kusumawati

(a) Uniaxial tension

(b) Sample of the rubber after testing

Fig. 1. Testing mechanical Properties

Fig, 2. Base Isolator testing 2.1

Dissipated Damping Energy

Dissipated damping energy values can be seen from the hysteretic loop graph. From the hysteretic loop graph, effective stiffness (Keff) can be calculated using Eq. | [2].

(Fimax = Fi

Key = inax — Armin

(1)

From the effective stiffness (Keff) damping value was subsequently calculated using Eq. 2 and 3 [2].

be

Wi (4nW,)

®

Experimental Study of Two Stages on The Use of Local Rubber

537

The value of Wad involves the area of hysteretic loop from the double shear test (horizontal

tes) of base

isolator

which

can

be

solved

using

the Takeda

model

[8].

Meanwhile, the value of Ws can be completed using Eq. 3 [2].

We = Kat (Om Amax is the average value completed using Eq. 4 [2].

of positive and

@) negative displacement which can be

Ona _* =O +07) OD 2.2

(4)

Geometry Properties

Rubber testing was conducted to determine the mechanical properties of rubber prior to base isolators (BI) testing. Testing of uniaxial tension and hardness followed the ISO37 [9] standard with grade hardness 40; 50 dan 60 durometer scale shore A IRHD [10]. Meanwhile, testing of BI was based on BS EN 15129 [11] with specifications

according to Table 1.

Table 1.

Specification of cross-section BI.

Vertical test | Horizontal test

Diameter (mm) Thickness of BI (mm)

200

200

150

150

Total layer of rubber

14

Total layer of fiber

13

Thickness of fiber (mm)

14 13

0.77

0.77

Thickness of rubber layer (mm) | 10 Total thickness (mm) 150

10 150

The samples of rubber testing according to ISO-37 standard can be seen in Fig. 3, while those of BI testing according to BS EN 15129 can be seen in Fig. 4.

Fig. 3.

Sample of the uniaxial tension rubber test

538

—-U, Wijaya and E, Kusumawati Rubber

Bas

6200 mm

Rubber, 150

t=

10 mm

Fiber Reinforcing Layer

jm Fig. 4. Sample of base isolator test 2.3

Procedure of Vertical and Horizontal Test of BI

As for the method of the vertical testing load, the specimens were examined

in the

vertical

Three

direction

with

load

control

in accordance

with

BS

EN

15129

[11].

specimens were loaded with a vertical force of up to 66 KN. In this test, the specimen was examined the maximum axial load has to be applied on specimens and released before the measurement. Following the first loading, pre-determined maximal axial load of 66 KN should be applied progressively with a minimum of five level at a 5 + 0.5 MPa/min or 2.64 KN/s. Then specimen is loaded up to 66 KN vertical force and four fully reversed cycles. Deflection must be recorded at 1/3 of the maximum load and at the maximum load. Horizontal test set-up was conducted under constant axial load and horizontal displacement control. At first, the specimen was subjected to a constant vertical load of up to 66KN. At the same time, the vertical load remained constant. Subsequently, the specimen changed shape in cyclic shear with five fully reversed cycles at a given strain rate of 10%, 20%, 50%, 75% and 100% (based on 140 mm tubber thickness)

3 3.1

[11]:

Result and Discussion Experimental Testing of First Stage

The first stage of testing involves nine samples, examining the mechanical properties of three types of rubber (each grade consisting of three samples) with grade 40; 50 and 60 tested durometer shore A yielding in hardness values that are close to the predicted

Experimental Study of Two Stages on The Use of Local Rubber

539

value of each hardness category. For grade 40, each sample produces hardness of 41, 43, and 41; grade 50 produces hardness of 50, 52, 52; and grade 60 produces hardness of 61, 61 and 62. The three hardness tests show stable results. According to Gent (2012) deviation of no more

than five levels belongs to the ‘stable’ category

[12] and

the categorization of grade can be continued to the next stage, namely uniaxial tension testing as shown in Fig. 5.

Uniaxial Tensile Test

—140 —H50

o

100

200

Fig. 5.

300

Strain (%)

400

500

600

The result of uniaxial tension

The results uniaxial tension testing of three rubber grades, shown in Fig. 5, indicates that the H50 rubber grade has the highest stress and strain values compared to grades H40 and H60. The strain energy function in the Hyperelastic model of the uniaxial rubber test which already uses H50 grade is analyzed using the constitutive model and compared to experimental test data. The results can be seen in Fig. 6. The most frequently-used constitutive models are Mooney Rivlin model, Neo Hookean model, Ogden model, Arruda - Boyce model, and Yeoh model. From the comparison, Yeoh Model becomes the most appropriate for predicting the hysteresis behavior of rubber compositions because of its ability to match experimental data points at small and large strain values below the strain value of 200% to above 500%. Overall, of the five constitutive models, the Yeoh model provides the best and most accurate adjustments for various test data despite a limited amount of test data, for example, only having simple tensile stress-strain test data.

540

U, Wijaya and E, Kusumawati

— Mette | Adee J me

a

ro} Fig. 6.

0

Constitutive model fit of experimental data

From the result of uniaxial tension and hardness testing, mechanical properties of the rubber H50 which is used to produce base isolators can be seen on Table 2.

Table 2.

Mechanical properties of the rubber used for base isolator.

Vertical test | Horizontal test

Specimen| Hardness (shore A) Elongation at break

BI 50 >500%

Elastomer shear modulus (G) | 0.7 MPa

3.2

BI 50 >500% 0.7 MPa

Experimental Testing of Second Stage

As for the second stage of experimental testing, there are three samples used for the base isolator (BI) test with H50 grade. It is assumed that BI will withstand the load in a

simple type-T36 (36 m2 building) single-story houses located in an earthquake-prone area of Padang, West Sumatra. The damping effect occurs through energy dissipated through partially constrained BI rubber movements, which usually results from inelastic deformation or friction. The amount of energy dissipated from a BI rubber system is determined from the area of the hysteretic curve obtained from the experimental double shear test results in Fig. 7, and the area of the hysteretic loop can be solved using the Takeda equation.

Experimental Study of Two Stages on The Use of Local Rubber

541

Lateral tes BI 40000 20000 —H50-1

-40000 60000

Fig. 7.

Displacement (mm)

Hysteretic loop of lateral test of BI

Figure 7 shows that the BI displacement is obtained when it was exposed to lateral loads ranging from 91 mm to 110 mm. The results of this test indicate that the deviation of each sample is not so large. This indicates that H50 grade rubber has quite stable and consistent behavior with the hardness and uniaxial tension testing at the first test stage. Therefore, the use of H50 sample for BI is a right choice. The lateral test results shown in Fig. 7 also indicate the absence of curve jump between force vs displacement, which can reflect the presence of lateral bulging. From Eq. (1) to the value of vertical stiffness and horizontal stiffness, as well as the

damping ratio of the hysteretic loop in Fig. 7, can be determined based on the area of hysteretic loop obtained from the test results. Then, the effective viscous damping ratio can be calculated, which is shown in Table 3.

Table 3.

The result of lateral test of base isolator

Sample

Damping (%)

KH (kN/m) Keff (kN/m) H50-1 | 39.44 521.71 H50-2 | 38.75 511.98 H50-3 | 38.13 505.29 Vertical stiffness (Kv): 4067 kN/m

12.20 12.17 12.01

From the results of testing the H50-1 sample, H50-2, and H-50-3, based on the vertical test, the value of the vertical stiffness is 4067 kN/m. A base isolator must result in high vertical stiffness and low horizontal stiffness. The behavior of the H50 specimens tested has met the requirement, where the horizontal stiffness value exceeds 1/100 of the vertical stiffness value. The test results of these specimens prove that the behavior of inexpensive base isolators has been compatible with the concept of lowcost public housing buildings. Then the damping value around 12% is ideal for public houses with one and two floors.

542 4

U, Wijaya and E, Kusumawati Conclusion

From the two-stage testing of base isolators using local rubber, it can be concluded that properly processed rubber can serve as a good base isolator and is appropriate for earthquake absorbers, especially for simple residential buildings for public housing, which are usually built without following an engineering design process that misses ductile detail, especially for the earthquake-prone areas. Local rubber has some mechanical properties that can fit the constitutive model that has been developed so far. This means that in the future, if Indonesia wants to develop a local base isolator, the technology to predict the mechanical properties of rubber to the base isolator can be directly adopted. Local rubber is very promising for use in the construction industry in Indonesia so that in the future Indonesia does not need to import base isolators from abroad for production of earthquake absorbers, giving the fact that Indonesia is the second largest rubber producing country in the world after Thailand, and almost all regions of Indonesia are at high risk of earthquakes.

References 1. BMKG,

Press

release

No.

UM.505/9/D3/X/2018

untuk

Gempa

Palu

dan

Donggala

v

Sulawesi Tengah, Badan Meteorologi Klimatologi dan Geofisika, no. September, pp. 160164 (2018) . Calabrese, A., Losanno, D., Spizzuoco, M., Strano, S., Terzo, M.: Recycled rubber fiber reinforced bearings (RR-FRBs) as base isolators for residential buildings in developing countries: the demonstration building of Pasir Badak, Indonesia. Eng. Struct. 192(September

2018), 126-144 (2019) 3. Tajammolian, H., Khoshnoudian, F.: Reliability of symmetric and asymmetric structures mounted

on TCFP

base isolators subjected to near-field earthquakes.

J. Perform. Constr.

Facil. 32(4) (2018) 4. Habieb,

A.B.,

Milani,

G.T., Milani, F.: Low

cost rubber

seismic

isolators for masonry

housing in developing countries. In: AIP Conference Proceedings, vol. 1906, no. 5, pp. 647— 653 (2017) 5. Kelly, J.M., Konstantinidis, D.: Seismic isolation for housing, schools and hospitals in the

urban environment. In: Second International Conference on Earthquake Engineering Disaster Mitigation, no. July, pp. 19-20 (2011) 6. Wijaya, U., Soegiarso, R., Tavio, T.: Mechanical properties of Indonesian rubber for lowcost base isolation. In:

MATEC

Web Conference, vol. 276, no. 01017 (2019)

7. Van Engelen, N.C.: Fiber-reinforced elastomeric isolators: a review. Soil Dyn. Earthq. Eng. vol. 125, no. January (2019) 8. Dong, P.: Effect of Varying Hysteresis Models and Damage Models on Damage Assessment of Ric Structures Under Standard Design Level Earthquakes Obtained Using a New Scaling Method (2003) 9. Bureau of Indian Standards, ISO37: Method of test for vulcanized rubber (2012) 10. ASTM International, ASTM D412-15 Standard Test Methods for Vulcanized Rubber and

Thermoplastic Elastomers, Annu. B. ASTM Stand., no. Reapproved 2002, pp. 1- 14 (2016) 11.

BS-EN

15129, Anti-seismic devices, UNI EN

151292009, vol. 1, no. 1 (2014)

12. Gent, A.: Engineering With Rubber (How to design rubber components). Hanser Publishers, Munich (2012)

Advanced Construction and Building Information Modelling

®

Check updatesfor

Prospects of a Sustainable EOL - Carbon Footprint Assessment of a Tropical Housing Habitat Syed Shujaa Safdar Gardezi' and Nasir Shafiq?

' Department of Civil Engineering, Capital University of Science and Technology (CUST), Islamabad Expressway, Kahuta Road, Zone-V, Islamabad 44000, Pakistan dr. shujaasafdar@cust. edu. pk

> Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS, 32610, Seri Iskandar, Perak Darul Ridzuan, Malaysia nasirshafiq@utp. edu. my

Abstract. End of Life (EOL) is a vital phase in environmental life cycle impact assessment. However, most of studies either do not consider this phase or make its assessment based on certain assumptions, thus limiting the actual contributions. A proper pre-assessment of such environmental impacts during planning can achieve an optimum sustainable design inception at an early stage. The current work evaluates the carbon footprint potential of the EOL phase of

conventional housings in a tropical climate of Malaysia. Conventional units with varying areas, height and type of construction have been analyzed. The life cycle

inventory was achieved by developing the virtual prototypes of selected units

using Building information Modeling (BIM). Partial life cycle assessment (LCA) methodology was used to obtain carbon emissions. The study highlighted a contribution of 2.5 to 3.0 tons-CO, with average intensity of 1.00 kg-CO. per

unit area. The

dismantling operation dominated the hauling operations by 50%.

Concrete and bricks were the top two materials dominating the hauling activity. Statistical technique, regression, highlighted a significant relationship between

dependent (carbon footprint) and independent (area) variables. Study, being one of the few addressing conventional low cost housing sector in tropical climate, is

expected to act as mile stone and guideline baesed upon actual data of facilities for a realistic environmental conscious and optimum sustainable decision.

Keywords: Carbon footprint - End of Life (EOL) - Life cycle assessment Building information modeling - Tropical housing 1

Introduction

The environmental impacts of a building’s life cycle should cater all the phases including design, raw material extraction, processing, construction, use, dismantling and disposal of construction wastes [1]. Life Cycle Assessment

(LCA)

has been a key

methodology adopted by different researchers in previous studies to evaluate the environmental impact of buildings. The entire life cycle may include materials

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 545-554, 2021. https://doi.org/10.1007/978-98 1-33-63 1 1-3_63

546

S. S. S. Gardezi

and

N. Shafiq

production, construction, operation, and demolition whereas demolition may include operations of dismantling, waste sorting and hauling etc. Howver, any specific phase assessment remains under the partial life cycle application. Dismantling and dispose-off is the last and vital part of every life cycle assessments

[2]. Wu, Yuan

[3] emphasized

that the magnitude of the greenhouse gas emissions should include contributions from this phase. Like other stages of life cycle, each building consumes energy during its demolition [4]. The energy consumed during these operations causing carbon footprint emissions. The consumption from this phase has been reported mainly due to the diesel fuel used for transportation and demolition machinery. The magnitude of emissions concentrations depend on different factors [3]. However, due to lack of data, either this phase has not been included in most of the studies [5] or based upon certain assumptions [6]. Franzoni [7] highlighted that some assessment tools also do not

address the demolition and disposal. Similarly, most of the studies report lack of contributions from high rise building and housing sector assessment in this regard. Therefore, in achieving a standard yardstick to assess the sustainable housing construction practices, it is important to evaluate EOL phase on actual data of facilities to come up with a realistic environmental impact. The current work aims to assess the conventional housing projects in a tropical climate to observe an actual data based environmental impacts. Since, in this work, only phase of dismantling and dispose off have been considered, so Partial life cycle assessment, in compliance with ISO14001 standard, has been adopted for this purpose. 1.1

Literature Review

Many researchers have observed the importance of the last phase [8] i.e. dismantling and dispose-off, of life cycle environmental impact assessments studies conducted for building sector. Cole and Kernan [9] reported that in current demolition practices, intensive application of energy along with transportation make them an important contributor to overall emissions during life cycle. Hernandez and Kenny [10] agreed to a potentially high impact influence. Blengini and Di Carlo [2] emphasized this phase as an essential part of every LCA study. Table | details the CO, contributions of this phase of life cycle reported by some of the researchers in their studies.

Table 1, Previous research work findings regarding last phase of LCA Researcher Cuéllar-Franca and

Findings 1% from the end-of-life waste management

Azapagic [11]

Crawford [6] Wu, Yuan [3]

1% of the life cycle energy requirements of the house for the demolition and disposal of the house (assumption) The demolition stage activities emit about 2,177.44 kg/m? CO»

Monahan and Powell

An estimated 4.9 tCO, resulted from waste, equating to

[12]

109 kgCO, per m?

Prospects of a Sustainable EOL - Carbon Footprint Assessment

2

547

Methodology

The research follows:

‘singe

Goats

methodology

t

adopted

in the study

has been presented

by

Fig. 1 as

Critical Literature review of carbon ernissions from Dismantling and Dispore-off

Defe

Scoring

Review of Housing z Sector of Malaysia : along with different design options for

Deterrnine the project scope of study and objectives ofthe wudy Stage -22 rote pee

Selection of cave tudy Grea, height, type)

J

3D Virtual Prototyping

1 Life Cycle Inventory (LCI) Stage -3 Impact assessment

q

Identification of the main sources with potential CO; emissions

Dismantling

Dispose-off ‘Results and analysis Carbon footprint potential

Stage -4 Interpretations Fig. 1.

2.1

Conclusions and Recommendations Research framework based upon ISO 14000 standard guidelines

Case Study

Five housing units were selected. These comprise of detached, semi-detached and terraced type construction usually being adopted within Malaysia with a single and double storey height. The details of housing units have been highlighted in Table 2. In case of residential building, the semi-mechanical or combination of manual and mechanical mode of dismantling was being adopted. The reason for such adoption supports the fact that some of the materials can be reused or recycled and require

548

S. S. S. Gardezi

and

N. Shafiq Table 2.

removal etc. The tributed loading

Details of housing units

Model ID

Area (sft)

H-O1 H-02

1625 2925

Single Single

No. of storey

Type of house | Type of structure Semi-Detached | RCC Frame Detached RCC Frame

H - 03

2800

Single

Detached

RCC Frame

H-04

| 4400

Double

Terraced

RCC Frame

H-05

3940

Double

Semi-Detached | RCC Frame

with least damage e.g. aluminum, mild steel, roof tiles, roof truss, PVC doors use of mechanical equipment in dismantling and dispose-off operations conin emissions. The total carbon footprint is the summation of dismantling, and transporting.

S5 €O2(d&d) = S> CO>(dismantling) + S~ CO2(transport)

(1)

However, the quantum of such emissions from this phase depends upon the type of equipment, duration of their operation, type of fuel and distance for disposal. The disposal de: ion may be a recycling unit for some of materials or waste pit. As, no specific data was available for such dismantling operation in Malaysia, the study was based on standard equipment locally available in construction works. Table 3 details the adopted equipment for different activities.

Table 3, Equipment involved in dismantling and dispose-off operations Type of equipment Excavator / Loader (M318)

Fuel type | Fuel consumption rate Diesel

Dump truck ( tons load carrying capacity) Diesel

3.5-5 gallons/h

6 km/I

“1 UK gallon = 4.54609 L The data pertinent to the duration for dismantling process was based upon the feedback from the construction industry experts and professional organizations involved in rehabilitation / reconstruction operations of housing stocks. The quantities of dismantled materials were quantified from the detailed drawings of the housing units. For accuracy, these were compared with the bill of quantities and, further, with the data extracted from 3D virtual models developed for each of the housing unit using Building Information Modeling (BIM), Fig. 2

Prospects of a Sustainable EOL - Carbon Footprint Assessment

Fig. 2.

3

549

BIM based 3D virtual model of selcted case study (H# 04)

Results and Discussions

Based upon the type of equipment involved and quantities of materials, Table 4 details the contributions of carbon footprint from EOL phase.

Table 4, Carbon footprint (Kg-CO>) from dismantling and dispose-off phase Description [H#01 (H#02 |H#03 |H#04 |H #05

Dismantling Dispose-off

1,642.60 | 1,642.60 | 1,642.60 | 2,463.90 | 2,463.90 666.74 | 1,106.81 | 771.48 | 1,385.19 | 961.57

Total carbon footprint | 2,309.34 | 2,749.41 | 2,414.09 | 3,849.09

3,425.48

The contribution varied from 2.3 tons-CO, to 3.8 tons-CO, with a variatoion of 67% for the selecetd housing units depending upon their design features. On average, Fig. 3, highlighted an average contribution of 3 tons-COz from this phase of life cycle. As the area of housing increased, the impact was also proportionate. The fuel consumed by equipment was the source for carbon footprint from dismantling operations. The dismantling dominated the contributions with an almost of 60-70% share in each of case. The rest of 30-40% was contributed by dispose-off operations. Similarly, the fuel consumed for transportation caused the emissions. However, the emission were observed to be dependent upon the type of vehicle, distance of disposal and operating time. A distance of 50 kms was adopted for disposal. The transportation of materials to the disposal point ranged from 0.65 tons-CO, to 1.1 tons-CO}.The share by different materials have been presented in Table 5.

S. S. S. Gardezi

and N. Shafiq

H#o1

a

| —

$2823 8

: i

550

H#02

H #03

H#O8

HOS

Fig. 3. Total carbon footprint contributions Table 5. Breakdown of carbon footprint - Dispose-off operation Description of Material | Carbon footprint (Kg-CO.) H#01|H#02 |H#03/H#04 H#05 ‘Aluminum 0.13) 0.35) 0.29 0.40 0.56 Bricks 281.26 | 478.59) 307.44 545.87 369.91 Concrete 279.44 | 498.28 338.43 688.59 471.39 Ceramic Tiles 8.67) 13.59] 12.46 37.14 14.15 False Ceiling 5.74) 5.39] 7.16 10.84 25.99 Glass 0.60) 131) 3.06 4.81 1.28 Mild Steel 13.38; 1.74) 16.94 3.89 141 Paint 077) Ll) 0.78 1.45 131 Plaster 38.19 | 55.22] 38.68 71.86 44.06 PVC Doors Panels 0.07) 0.07) 0.07 0.15 0.10 Roof Tiles 30.24) 35.17 37.57 17.09 Steel Rebar 643 15.12; 6.09 18.58 13.62 Wood 1.82) 0.85] 2.50 1.61 0.69 Total 666.74 | 1,106.81 | 771.48 1,385.19 961.57

Among

materials,

concrete

concrete generated 0.50 ton-COz

and bricks (46%)

are the top two

materials.

On

whereas bricks resulted 0.40 ton-COz

The two materials shared more than 85% in almost every case, Fig. 4.

average, (40%).

Prospects of a Sustainable EOL - Carbon Footprint Assessment

551

HOS

Fig. 4. Percentage contribution of materials in housing units

The contribution trend, Fig. 5, for each of the materials during the dispose-off activity depicted that all materials except form concrete and bricks followed almost a linear trend in their contributions.

552

S. S. S. Gardezi

and

Sa Sines 00 | ——=Rost Ties

N. Shafiq

Bids Milde Sed Re

Comets Pact — Wood

Cwranic Ties Piste

Fis Gating PVC DoceePenels

widen fot

8

KgCO})

om |

__————————— Hew H#G #04

HAO

Fig. 5.

The emission observe positive

= & = &8

#05

Carbon footprint contributions and trend of materials

area of housing units observed to be significant factor regarding the carbon content. Statistical technique of simple regression analysis was adopted to relationship between the area and carbon footprint. The results highlighted a significant relationship between the two variables (R? > 0.70), Fig. 6

4.500 4,000 3,500 3.000 2.500 2000

carbon footprint

Linear (carbon footprint)

.

+

=

R= 08842

1,500

1,000 500 1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500.

5,000

Area (sft)

Fig. 6. Statistical relationship between dependent and independent variable However, an inverse relationship was achieved for emissions per unit area, Table 6. In other words, the per unit contributions observed a downward trend with increse in

area.

Table 6. Per unit area carbon footprint contributions (Kg-CO4/sft) Description H#01 H#02 H #03 H#04 H#05 Carbon footprint / sft 142 0.94 0.86 0.87 0.87

Prospects of a Sustainable EOL - Carbon Footprint Assessment

4

553

Conclusions

The current study aimed to assess the carbon footprint potential of dismantling and dispose-off phase of project life cycle from the conventional construction in Malaysian housing sector. The assessment was based on the inventory extracted from construction drawings and virtually developed 3D models using BIM methodology alongwith feedback from industry experts. The study concluded that: The demolition and hauling of construction materials resulted in environmental impact of this phase. The contributions ranged from 2300 to 3500 kg of CO3. On overall, the dismantling activities dominated the share in carbon emissions. Concrete and bricks were the top two materials dominating the dispose-off part. An average intensity of almost 1.00 kg-COz /sft was observed. Carbon footprint and area of housing units observed a significant positive trend. A 66%

increase in CO;

resulted with an area increase from

1625

sft to 4400 sft.

Acknowledgements. The authors acknowledge the support of Capital University of Science and Technology (CUST), Islamabad Pakistan and Universiti Teknologi PETRONAS (UTP), Malaysia for this collobrative research study.

v

References

. Bilec, M., et al: Example of a hybrid life-cycle assessment of construction processes. J. Infrastruct, Syst. 12(4), 207-215 (2006) . Blens . G.A., Di Carlo, T.: The changing role of life cycle phases, subsystems and materials in the LCA of low energy buildings. Energy Build. 42(6), 869-880 (2010) . Wu, H.J., et al.: Life cycle energy consumption and CO2 emission of an office building in China. Int. J. Life Cycle Assess. 17(2), 105-118 (2012) . Dixit, MK., et al.: Need for an embodied energy measurement protocol for buildings: A review paper, Renew. Sustain. Energy Rev. 16(6), 3730-3743 (2012) . Wallhagen, M., Glaumann, M., Malmqvist, T.: Basic building life cycle calculations to decrease contribution to climate change-Case study on an office building in Sweden. Build. Environ. 46(10), 1863-1871 (2011) . Crawford, R.H.: Towards a comprehensive approach to zero-emissions housing. Archit. Sci. Rev, 54(4), 277-284 (2011) . Franzoni, E.: Materials selection for green buildings: which tools for engineers and architects? Procedia Eng. 21, 883-890 (2011) . Winistorfer, P., et al.: Energy consumption and greenhouse gas emissions related to the use, maintenance, and disposal of a residential structure. Wood ‘iber Sci. 37, 128-139 (2005)

. Cole, R.J., Kernan, P.C.: Life-cycle energy use in office buildings. Build. Environ. 31(4), 307-317 (1996) . Hernandez, P., Kenny, P.: From net energy to zero energy building lefining life cycle zero energy buildings (LC-ZEB). Energy Build. 42(6), 815-821 (2010)

554

S. S. S. Gardezi

and

N. Shafiq

11. Cuéllar-Franca, R.M., Azapagic, A.: Environmental impacts of the UK residential sector: life cycle assessment of houses. Build. Environ. 54, 86-99 (2012) 12. Monahan, J., Powell, J.C.: An embodied carbon and energy analysis of modern methods of construction in housing: a case study using a lifecycle assessment framework. Energy Build. 43(1), 179-188 (2011)

®

Check updatesfor

Analytical Investigation of Failure Behavior of Beam-Column Knee Joint with External Steel Plates Anchorage Using 3D RBSM Liyanto Eddy", Kohei Nagai, and Punyawut Jiradilok? ' Department of Civil Engineering, Parahyangan Catholic University,

Bandung, Indonesia liyanto. eddy@unpar. ac. id

? Institute of Industrial Science, The University of Tokyo, Bunkyo City, Japan

{nagai325, punyawut}@iis. u-tokyo.ac. jp

Abstract. External steel plates anchorage is one way to reduce the reinforcement congestion inside the beam-column joint because the reinforcing bars of the beam and column are anchored using steel plates located outside the joint. In this study, the failure behavior of the joint is investigated through the study of

internal stresses using a discrete analysis method called 3D Rigid Body Spring Model. Simulation results are compared with experimental results. The simulation results show that in BCJ-Hook (conventional hook bar anchorage), the capacity of the beam-column joint in open case is significantly lower than that in close case. Meanwhile, in BCJ-Plates (external steel plates anchorage), the maximum load in open case is roughly the same as that in close case with only

6-14% reduction in maximum load. Higher capacity in BCJ-Plates is caused by the confinement from the steel plates. Keywords:

Beam-column knee joint - External steel plates anchorage - Meso-

scale analysis - 3D RBSM 1

Introduction

To address the issue of reinforcement congestion in the beam-column joint, the reinforcing bars of the beam and column could be anchored outside the joint using the external steel plates as shown in Fig. 1. Using this type of anchorage, the reinforcing bars in beam and column are extended to outside the beam column joint. Steel plates with holes having the same location as the steel bars is attached. Washers and nuts are placed and tightened. In this study, the behavior of the beam-column knee joint with external steel plates anchorages is investigated through the comparison of the experimental observations and simulation results. The strain-stress state inside the joint is very complex and difficult to be investigated using experimental results because many steel bars are anchored in the joint. Simulation results could provide the changes in intemal stresses and cracks inside the beam-column joint at each loading step which is beneficial to figure out the failure process at the macroscopic level. A numerical investigation of the failure behavior of the beam-column knee joint with external steel plates anchorage is conducted using a discrete analysis method, specifically the Three-dimensional Rigid

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): ICCOEE2020, LNCE 132, pp. 555-563, 2021. https://doi.org/10.1007/978-98 1-33-63 1 1-3_64

556

L. Eddy et al.

Body Spring Model (3D RBSM). This analysis method is very suitable to investigate the failure behavior of reinforced concrete structures. It is a 3D simulation where the reinforcement arrangement is accurately represented by modelling the 3D shape of reinforcement including its ribs. 3D RBSM is a meso-scale simulation. It is well known that in the reinforced concrete structures, cracking occurs because of the discontinuous of concrete and the interaction of concrete and steel bars at meso-scale. Cracks are presented directly when the tensile failure occurs between two elements. In our research group, utilizing the advantages of 3D RBSM, the failure behaviors of reinforced concrete members

under mechanical

and environmental

loadings were figured out [1-4].

The ultimate aim of this research is to investigate the failure behavior of the beamcolumn knee joint with external steel plates anchorages through the study of the experimental and numerical results.

2

Numerical Method and Constitutive Models

The beam-column knee joint models are simulated using 3D RBSM proposed by Kawai [5]. A three dimensional beam-column joint model is subdivided into rigid bodies in the

form of Voronoi cells. Two neighboring rigid bodies are linked with three springs which are two shear springs and one normal spring at the contact area of those two bodies as illustrated in Fig. 2. At a computational point within the interior ofa rigid body, which is the centroid of the rigid body, six degrees of freedom consisting of three translational degrees of freedom and three rotational degrees of freedom are defined. To model the beam-column joint in 3D, concrete and steel rigid bodies are defined. The size of each tigid body is approximately 10 x 10 x 10 ~ 20 x 20 x 20 mm’, which is similar to the aggregate size. For concrete element meshing, Voronoi diagram associated with a set of random points defined in the simulation system is used. The purpose is to allow the cracks propagate in an arbitrary direction. The rigid bodies for steel are constructed manually in order to replicate the 3D shape of the reinforcing bar including its ribs as shown in Fig. 3a. The geometry of the reinforcing bar is modeled accurately, with full 3D modeling of the reinforcing bar arrangement, in order to properly account for the interlocking between the reinforcement and concrete. Fig. 3b shows a cross section to illustrate the mesh arrangement for concrete and steel used in this study. The simulation system adopted in this study is originally developed by Nagai et al. [6]. The constitutive models for concrete elements shown in Fig. 4 and steel elements used in this study are adopted from previous

studies [1, 2].

-

eT oc

Fig. 1.

External Steel Plates Anchorage

3 SPRINGS

DOF

DOF

Fig. 2.

3D computational RBSM

model

Analytical Investigation of Failure Behavior of Beam-Column Knee Joint

5

557

a

(a) 3D model reinforcing bar

_—_(b) Cross section

Fig. 3. Mesh arrangement for concrete and reinforcing bar 3 3.1

Detail of Numerical Simulations Numerical Models

Two numerical models listed in Table 1 are based on the experiments of beam-column knee joints with different anchorage systems done by Kunii et al. [7]. The numerical models are signified by BCJ-Hook and BCJ-Plates BCJ-Hook means the reinforcing bars in the beam and column are anchored using the conventional hooked bars in the joint which is designed according to Japanese Seismic Code and used as the control specimen. BCJ-Plates is basically the same as BCJ-Hook, but the hooks anchored down in the column is replaced with external steel plates. 3.2.

Geometry of Numerical Models

Figure 5 shows the geometries of the numerical. For comparison, details of the original experimental specimens are shown in Fig. 6. The locations, details, and number of the steel bars in the numerical models are exactly the same as those in the experimental specimens. Deformed bars of 16 mm and 22 mm are used as the main column and beam reinforcement, respectively. To reduce the computational time, the stirrups are modeled as plain bars. Plain bars of 16 mm and 13 mm are used as the stirrups of beam and column, respectively. The material properties of the steel bars are given in Table 2.

ratio

se 92]

Compression [Tension (a) Normal spring

w 0.03

(b) Shear spring Fig. 4.

(c) tmac criterion

Constitutive models of concrete

03)

(Wax) (d) Shear reduction

mm

558

3.3.

L. Eddy et al.

Boundary Conditions

The boundary conditions of the numerical models are included in Fig. 5. For comparison, experimental setup is shown in Fig. 7. Rigid steel plates are modeled at the beam and column ends to prevent the deformation of the plates. Pins are placed inside the steel plates representing the hinge condition where the normal springs between the pins and plates only transfer compressive stresses from the pins to the steel plates. No tensile and shear stresses are transferred from the pins to the steel plates. Cyclic loading was applied in the experiment while in the simulation monotonic-displacement loading is applied to reduce the computational time. The monotonic-displacement loading increasing by 0.1 mm at each loading step is applied to the pin located at the end of the beam,

while the pin located at the end of the column

is fixed. The simulation of each

model is stopped at step 1000. Table 1.

Case

BCJ-Hook BCJ-Hook BCFPlates BC/Plates

Detail of numerical models

Parameter

Material properties of | Number | Maximum concrete of Load ft. fi E._| elements Exp. | Ana (MPa) | (MPa) | (MPa)

(Close) | Conventional Hook (Open) | Bars Anchorage | 11 | 147 | 22,000 | 908160 (Close) | External Stee! (Open) | Plates Anchorage |! | 146 | 22,000 | 850.548

(kN) | ()

276.3 | [i397 350.0 | P3599

Table 2, Material properties of reinforcing bars Reinforcing bars

D22 DI6 DIG DB

(a) BCJ-Hook

Function Main reinforcement of beam | Main reinforcement of column Stirrups of beam, Stirrups of column

Yield Strength (MPa)

525 554 377, 375

(b) BCJ-Plates

Fig. 5. Numerical models (units: mm)

Modulus of

Elasticity (MPa)

202300 201160 187750 185370

2543 [i043 251.5 [>146

Analytical Investigation of Failure Behavior of Beam-Column Knee Joint

4 4.1

559

Results and Discussion Load-Displacement Relationships

The load-displacement relationships for all cases are shown in Fig. 8. Maximum loads for all simulated models are also listed in Table 1. At early stage loading, simulation and experimental results show almost the same initial stiffness in BCJ-Plates and BCJHook. However, the simulation results are underestimated the capacity by 8% and 34% when BCJ-Hook

(conventional hook bar anchorage) is loaded by a moment

that tends

to close and open the beam-column joint, respectively. In BCJ-Plates (external steel plates anchorage), the simulation results are underestimated the capacity by 28% and 42% when it is loaded by a moment that tends to close and open the joint, respectively. The possible cause of the discrepancy is the loading condition. In the experiment, as the lateral load from the actuator increases, the steel column near to the beam tilts. The load might be applied diagonally to the pin located at the end of the beam,while in the simulation only horizontal load is applied. The load-displacement relationships given by the simulation results have roughly the same tendency as those observed in the experiment. Simulation and experimental results show that in BCJ-Hook, the capacity in open case is significantly lower than that in close case. In BCJ-Plates, the maximum load in open case is roughly the same as that in close case with only 6-14% reduction in maximum load. It indicates that extemal steel plates increase the capacity of the beam-column knee joint when the beam-column. joint is loaded by a moment that tends to open the joint. Both experimental and simulation results also show that in close case, the stiffness in BCJ-Hook is higher than that in BCJ-Plates prior to the maximum load, while in the open case, the stiffness in BCJ-Hook is almost the same as than that in BCJ-Plates prior to the maximum load. In BCJ-Hook, the top reinforcing bars in the beam are hooked in the joint, and anchored down in the column. It causes higher stiffness when BCJ-Hook is loaded by a moment that tends to close the joint. In BCJ-Plates, all reinforcing bars in the beam are anchored outside the joint and there is no reinforcing bar hooked and anchored in the column. The role of steel plates in increasing the joint capacity in open case will be investigated through the study of simulated stresses and cracks.

(a) BCI-Hook

(b) BCJ-Plates

Fig. 6. Experimental specimens (units: mm)

Fig. 7. Experimental setup

560

—_L. Eddy et al.

Dope

Fig. 8. Load-displacement relationships 4.2

Surface Cracks

Surface cracks in 3D for all cases after failure are shown in Fig. 9. The surface cracks of simulated BCJ-Plates are compared to those of observed BCJ-Plates. In close case, crushing of concrete is predicted in the joint (1) and flexural cracks outside the steel plates (2) are predicted in BCJ-Plates. The same locations and types of cracks are also observed in the experimental specimens. In BCJ-Hook, concrete spalling (3) and diagonal cracks, propagating from the bending portion to the re-entrant corner of the joint, occurs (4). In open case, crushing of concrete is predicted in BCJ-Plates, indicated by

the distributed cracks

in the joint (5), while in BCJ-Hook,

cracks

are con-

centrated in the bending portion of the achorage (6). 4.3

Internal Stresses and Cracks

Figure 10 shows the internal stress distributions at a displacement of 20 mm. The internal stresses are shown at a cross section where the longitudinal reinforcing bars in the beam are present. When the displacement is relatively small, stresses are trasnferred from the reinforcing bars to surrounding concrete in the beam-column joint by means of the bond stresses. As the displacement increases, different internal stress distributions and crack patterns are predicted inside the joint due to the different anchorage systems. In close case, diagonal compressive stresses are concentrated inside the bending portion of the anchorage in BCJ-Hook, while in BCJ-Plates, diagonal compressive stresses are distributed in the joint. Flexural cracks are predicted outside the bending portion of the anchorage in the joint of BCJ-Hook, while in BCJ-Plates,

Close case (a) BCJ-Plates Fig. 9.

Open case

Close case (b) BCJ-Hook

Open case

Surface cracks in 3D after failure (deformation x 3)

Analytical Investigation of Failure Behavior of Beam-Column Knee Joint

=

f i

Close

BCJ-Hook

‘Open

Close

BCJ-Plates

561

‘Open

+

aa 10 MPs Tension ‘Comp. (Deformationx 10)

Fig. 10. Internal stresses at a displacement of 20 mm

Close

‘Comp. (Deformationx 10) Fig. 11.

BCJ-Hook

‘Open

Close

BCJ-Plates,

‘Open

:

Internal stresses at a displacement of 70 mm

flexural cracks are predicted outside the anchorage plates (1). In open case, cracks occur inside the joint stress state occurs because of the complex reinforcement arrangement in the joint of BCJ-Hook. Tensile stresses and cracks may occur between the reinforcing bars in the joint. In BCJ-Plates, the reinforcement arrangement in the joint becomes simpler. The steel plates confine the joint indicated by the diagonal compressive stresses formed between the steel plates (3). Consequently, the capacity of BCJ-Plates in open case is higher than that of BCJ-Hook. Figure 11 shows the internal stress distributions at a displacement of 70 mm. In close case, concrete spalling (4) and cracks inside the bending portion of the anchorage cause the significant drop in load after the maximum load in BCJ-Hook, while concrete crushing in joint is predicted in BCJ-Plates. However, in BCJ-Plates, compressive stresses still occur in the joint because of the confinement from the steel plates (5). The

load does not decrease dramatically beyond the maximum load. It indicates that the steel plates could confine the concrete in the joint after concrete crushing occurs in the joint. As a result, compressive stresses still occur in the beam-column joint. Meanwhile, because there is no confinement in BCJ-Hook, it is hard for compressive stresses formed in the joint. In open case, the cracks in BCJ-Hook open wider (6), while in BCJ-Plates, compressive stresses in the joint still occur because of the confinement from the steel plates (7).

562 5

—_L. Eddy et al. Conclusions

Based on the numerical study using 3D RBSM and experimental observations of the behavior of the beam-column knee joint with external steel plate anchorages, the following conclusions can be drawn. 1. By changing the hooked bars with steel plates, the capacity in close case of BCJPlates is the same as that of BCJ-Hook, while in open case, BCJ-Plates shows higher capacity than BCJ-Hook. Simulation and experimental results show that the capacity of BCJ-Hook (conventional hook bar anchorage) in open case is significantly lower than that in close case. In BCJ-Plates (extemal steel plates anchorage), the capacity in open case is roughly the same as that in close case with only 6-14% reduction in capacity.

2. The simulated failure behaviors in BCJ-Plates match the observed crack patterns well. In close case, crushing of concrete and flexural cracks outside the steel plates are predicted in BCJ-Plates, while in BCJ-Hook concrete spalling and diagonal cracks, propagating from the bending portion to the re-entrant corner of the joint, occurs. In open case, crushing of concrete is predicted in BCJ-Plates, indicated by the distributed cracks in the joint, while in BCJ-Hook, cracks are concentrated in the bending portion of the achorage. 3. 3D RBSM shows that the changes in the internal stress and cracks beyond the maximum load between BCJ-Plates and BCJ-Hook are different. In close case, the load in BCJ-Plates does not decrease dramatically beyond the maximum load. Although concrete crushing is predicted, compressive stresses still occur in the joint due to the confinement by the steel plates. In BCJ-Hook, concrete spalling, and cracks inside the bending portion of the anchorage cause the significant drop in load after the maximum load. In open case, cracks occur inside the joint of BCJ-Hook. As the displacement increases, cracks open wider. In BCJ-Plates, the steel plates confine the joint indicated by the diagonal compressive stresses formed between steel plates. Consequently, the capacity of BCJ-Plates in open case is higher than that in BCJ-Hook. 4. The external steel plates improves the performance of the beam-column knee joint. In addition, the reinforcement arrangement becomes simpler.

References

1. Eddy, L., Nagai, K.: Numerical simulation of beam-column knee joints with mechanical anchorages by 3D rigid body spring model. Eng. Struct, 126, 547-558 (2016) 2. Eddy, L., Matsumoto, K., Nagai, K.: Effect of perpendicular beams on failure of beamcolumn knee joints with mechanical anchorages by 3D RBSM. J. Asian Concr. Fed. 2(1), 56 66 (2016)

w

Analytical Investigation of Failure Behavior of Beam-Column Knee Joint

563

. Jiradilok, P., Wang, Y., Nagai, K., Matsumoto, K.: Development of Discrete Meso-scale Bond Model for Corrosion Damaget at Steel-concrete Interface Based on Tests with/without Concrete damage. Construction and Building Materials 236. (2020)

. Jiradilok, P., Nagai, K, Matsumoto, K.: Meso-scale modeling of non-uniformly corroded reinforced concrete using 3D discrete analysis Eng. Struct. 197 (2019) . Kawai, T.: New discrete models and their application to seismic response analysis of structures. Nucl. Eng. Des. 48(1), 207-229 (1978) . Nagai, K., Sato, Y., Ueda, T.: Mesoscopic simulation of failure of mortar and concrete by 3D RBSM. J. Adv. Coner. Technol. 3(3), 385-402 (2005) . Kuni, M., Tsukishima, D., Kuraoka, M.: Experimental investigation of behavior of L-shape beam-column knee joint with steel plate anchorage. Proc. JCI 41(2), 295-300 (2019). (in Japanese)

® ‘upaates

Dominant Success Factors of Managing Subcontractors by Main Contractors in Sustainable Development Project Bambang Endro Yuwono®*), Yuhana, and Raflis

Department of Civil Engineering, Universitas Trisakti, Jakarta, Indonesia bambang. [email protected]. id Abstract. Challenges faced by businesses, including a construction service sector, are involved. The complexity in a construction business stimulates construction project practitioners to cooperate and apply POAC (Planning, Organizing, Actuating, Controlling) management function well. However, there

always problems during the construction process, like problems between contractor-subcontractor.

A contractor plays roles as one who executes the plan,

and a subcontractor works under the contactor's commands—the contractorsubcontractor relationship based on a legal contract. There is no research on the problems conducted on the base of POAC management perspectives yet. The samples of this research are contractor-subcontractor who was working on a skyscraper project in Jakarta. The research instrument conducted four dependent

variables and 40 independent variables. Data analyzed with confirmatory factor analysis in 2 stages; each stage results in dominant numbers derived from the respondents’ perception. The research finds that the most commonly occurring problem between them is actuating Keywords:

Management function - Sustainable development - Contractor -

Subcontractor

1

Introduction

The activities of construction project implementation are considered unique and complex. The main contractors as executors of these construction projects attempt to provide the best services to the employers or project owners in compliance with the given quality standards, within a period and on an agreed budget. Although the work on the field is assigned to the main contractors, most of the work is often assigned to specific specialist subcontractors, especially building projects, it is found for 80-90% of the work to be performed

by subcontractors

[1].

The main contractor is obliged to be able to manage all the subcontractors involved in the implementation of construction projects to achieve specific project goals; grade quality based on specifications, on time as planned, and minimum cost as much as possible have become the responsibilities of the main contractor to the project owner. The accurate of shared information between contractor and subcontractor is slightly weak

due

and

often failed

[2] they make

to unresponsiveness

subcontractors’ dejection

of their need

for timely

and

with main contractors

correct

information.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 564-572, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_65

Many

Dominant Success Factors of Managing Subcontractors

565

problems often occur during the implementation of construction projects, such as the lack of trust, delay,

and lack of communication

[3].

Various studies on contractor and subcontractor management have been conducted, but the studies that focus on subcontractor management's dominant success factors are less to be found especially in sustainable development project. According to [4] researched factors that influence a construction project manager's performance but have not managed to include subcontractor management topics. [5] studied the measurement instruments of innovative performances of contracting companies but is yet to include correlations

with subcontractor

management.

[6] researched

contractor risks, but still

have not included relations to subcontractor management risks. Thus, studies need to be continued regarding subcontractor management by the main contractor. [7] studied the factors that influence the main contractor's selection of specialist subcontractors but still has to focus on the dominant success factors of subcontractor management by the main contractor. [8] have studied subcontractor selection criteria by the main contractor

and have not focused on the dominant success factors of subcontractor management by the main contractor. [9] has studied the behaviors of the main contractor in arranging a

construction project subcontract but is yet to focus on the dominant success factors of subcontractor

management

by the main contractor.

[10] has studied risks of subcon-

tractors’ work on a construction project but has put relations to dominant success factors of subcontractor management by the main contractor. According to [11] did study on the critical success factors of work relations between the main contractor and subcontractor but is not considered quite assertive to be used as dominant success factors of subcontractor management by the main contractor. Based on the studies of past researches, therefore raises a question, “What are the dominant success factors of subcontractor management by the main contractor in sustainable development project?".

2

Method of Study

This study uses questioners as the primary method to collect data from projects. The targeted respondents are project managers (PM), site managers (SM), and site engineers (SE) from leading contracting companies of multiple-story building projects who have experience in sustainable development project in academia in Jakarta. The questioner includes variables that are part of the dominant success factors of subcontractor management by the main contractor. The measurement scale used is the Likert scale (whereas | = uninfluential to 5 = very influential). The original data set of 46 samples. The samples was split into two subsamples: Sample 1 (6) and Sample 2 (40). Subsequently, factor analysis was performed on Sample | using principal component analysis and The principal component analysis revealed 1 factors unvalid. Next, confirmatory factor analysis (CFA) was performed on Sample 2 to confirm the factor

structure.

566

B. E. Yuwono et al. Table 1.

No | Sub-Variables

1

| Planning and Scheduling Factors

2

Variables and Sub-Variables

Variables

Code

‘Arranged materials and equipment schedules by provided by each subcontractor in accordance to the

XI

master schedule by the main contractor Distribution of site plans to every subcontractor both at | X2

offices and warehouses that are to be used 3 4

5 1

Subcontractors are able to carry out short project implementation times Submission of sample materials by subcontractor at least 1 month prior to installation

X3

Subcontractors making mockups for repetitive work

| X5

Budget and

The value of work subcontracted by the main

Contract Factors

contractor to the subcontractors

2

x4

x6

Obligations and rights of the subcontractor in relations with the main contractor must be regulated in the

| X7

contract 3

The role of the main contracting that is very dominant to the subcontractors and is not supported with adequate subcontract design* planning

4

The main contractor provides subcontractors with detailed and clear specifications and drawings at the time of submission of prices

1

Resource Factors

| X8

x9

The length of work/experience of an individual in the relevant subcontracting company

| X10

2

On-time mobilization of subcontractor resources (materials, tools, labor)

Xi

3

Expertise

X12

and skills as well as high work motivation

for direct workers subcontractors in the field 4

Sufficient number of workers / in accordance with existing work activities by subcontractors

X13

5

The availability of sufficient materials / as needed by

X14

subcontractors

6 1

2

Availability of working tools / equipment that are Managerial Factors

X15

sufficient / in accordance with needs by subcontractors Support from top management of the main contractor to parties involved in the project both human. resources, budget, methods, and implementation time

Mutual trust between subcontractors and main

| X16

X17

contractors 3

The long-term commitment between subcontractors and main contractors is not just one project

X18

4

Effective communication between subcontractors and main contractors

X19 (continued)

Dominant Success Factors of Managing Subcontractors

567

Table 1. (continued) No

Sub-Variables

5

Variables Productive conflict resolutions from the main

Code X20

contractor to its subcontractors 6

Giving power* to the main contractor project manager — X21 in carrying out the management of its subcontractors because it is still controlled by the head office

7

The efficient coordination system in the project from

| X22

the main contractor to its subcontractors 8 9

Regular control of the work of subcontractors by a team appointed by the main contractor

X23

Need of special markings on each subcontractor

X24

equipment 10

The documentation system in the project is neat by the

= X25

main contractor for the subcontractors from the beginning to the end of the project 11

Supervision of work from subcontractors

X26

12

Placement of supervision / supervisor of the main

X27

contractor in accordance with their abilities 13

Services and responsibilities during the maintenance period of the subcontractor

14

Proper construction implementation method according

X28

| X29

the plain of the subcontractor

15

Reports and regular meetings between the main

X30

contractors with the subcontractors.

1

Technical Factors

The experience of subcontracting companies handling

X31

the same type of work and project size 2

There is initial explanation before the subcontractor joins both the contract and project implementation

X32

3

The subcontractor must provide the main contractor with detailed and clear specifications and drawings before implementation

X33

4

Competent technical and managerial quality of

X34

personnel in the main contracting work organization

1

| Work Safety Factor

2

Procedure for handling work accidents by

X35

subcontractors, The safety policy of the main contractor to its

X36

subcontractor 3 4

Participation in subcontractor labor insurance Periodic safety talk by the main contractor to the

subcontractors

Sources: [4, 6, 12-19]

X37 X38

568

B. E. Yuwono et al.

The stages of this study are as follows: . First stage Conduct interviews and give questionnaires to experts to validate variables from previous studies and bibliography and other references that have been compiled. Categories as experts are people who have expertise in academia and practitioners with at least ten years of work experience in the field of construction. The Variables of library research for studying the main contractor success factors in the subcontractor's management in the implementation of construction were 39 (thirty-nine), whereas after being validated by experts who had long been involved in the main contractor at the time of the pilot survey as many as 38 (thirty-eight) are shown in Table

| (one).

. Second stage Distribution of questionnaires (PM),

site

managers

(SM),

to respondents. and

site

engineers

Respondents (SE)

of the

are project managers leading

contracting

company building projects in Jakarta who have experince in sustainable development project. Namely PT. MULTIKON, PT. TOTAL BANGUN PERSADA Tbk, PT. PP (PERSERO) Tbk, PT. NUSA RAYA CIPTA, PT. TOTALINDO EKA PERSADA, PT. TATA MULIA NUSANTARA INDAH, PT. PULAU INTAN, PT. WASKITA KARYA (PERSERO) Tbk, PT. JATIKARYA MEGAH LAKSANA. Of the 50 questionnaires send out, 40 were returned with varying degress of completeness. . Third stage Validating the results of the second stage to be included in the SPSS program, then create analysis and conclusion. The steps are as follows: . Tabulate results of questioner data to verify which sub-variables have been fulfilled and then can be processed. . Validation and reliability tests to identify factors that fulfill the sub-variables or not. If the data does not fulfill, therefore, the sub-variable is to be excluded. Fulfilling data are continued in the process. . Correlation analysis aims to identify the pattern and closeness relationship between two or more variables. The value of r is judged. The value of r is to be tested with a probability value of< 0,05 and value of t that is if the counted value of t < t of the table, therefore the decision is the value of correlation analysis r is not significant, and if the counted value of t, then the value of correlation analysis r is significant . Analysis factors with calculations: Calculate the value of Kaiser-Meyer-Olikin (KMO test) and Barlet test and probability (sig. = p). If KMO > 0,5 and probability < 0,05, then the sub-variables can be factored, take anti-image correlation, or Loading Faktor (A) test. if 4 > 0,5, then the sub-variables are considered valid to be factored; if initial eigenvalue > 1, then the sub-variables is considered valid to be

factored, the next step is Communalities.

Dominant Success Factors of Managing Subcontractors

569

5. Conclusion Includes results from dominant success factors of subcontractor management by the main contractor in the implementation of construction projects with the confirmatory analysis factor retrieved from the roles of its variables.

3

Results and Discussions

3.1

Validity and Reliability

From the validation test, we can conclude that Corrected item-total Correlations are higher than the value of 0.3. Thus, the independent variable, x, can be considered valid, and if the value is less than 0.3, then invalid. Data of x that are not valid are X3, X5, X7, and X20. Those, as mentioned earlier, will not be included in the next process. From the reliability test, it is shown that the value of Cronbach's Alpha is 0.922, Cronbach's Alpha Based on Standardized Items is 0.927 with N of items is 38. Which describes the level of reliability of being very reliable, 0.80 to 1.00. 3.2.

Correlation Analysis Results

From the validation and reliability tests, as many as 34 variables were used. The correlation analysis results showed that 12 variables have significant relationships with the main contractor's success in managing subcontractors in the implementation of construction. and

Those were (1) X2 Distribution of site plans to every subcontractor both at offices warehouses that were to be used (Correlation value = 0.689); (2) X16 Support

from top management of the main contractor to parties involved in the project both human resources, budget, methods, and implementation time (Correlation value = 0.622); (3) X17 Mutual trust between subcontractors and main contractors (Correlation value = 0.685); (4) X23 Regular control of the work of subcontractors by a team appointed by the main contractor (Correlation value = 0.668); (5) X25 The docu-

mentation system in the project was neat by the main contractor for the subcontractors from

the

beginning

to the end

of the

project

(Correlation

value = 0.622);

(6) X27

Placement of supervision / supervisor of the main contractor in accordance with their abilities (Correlation value = 0.413); (7) X29 Proper construction implementation method according the plain of the subcontractor (Correlation value = 0.605); (8) X30

Reports

and regular meetings

(Correlation

value = 0.658);

between the main contractors with the subcontractors (9) X32

There

is initial explanation

before

the subcon-

tractor joins both the contract and project implementation (Correlation value = 0.616); (10) X36 The safety policy of the main contractor to its subcontractor (Correlation value = 0,659) (11) X37 Participation in subcontractor labor insurance (Correlation value = 0.643); (12) X38 Periodic safety talk by the main contractor to the subcontractors (Correlation value = 0,675).

570

3.3

B. E. Yuwono et al.

Factors Analysis Results

From the results of the correlations analysis, the variables were tested with confirmatory factors analysis. The results obtained were sorted from high to low, which are as follows: (1) There was a preliminary explanation to the subcontractor before the contract or project implementation process with a contribution of 73.9% (2) Routine meetings and reports between the main contractor and subcontractors with a contribution of 59.9% (3) Work safety policy provided to the subcontractors by the main contractor with

a contribution

of 54,9%

(4) Regular control

of subcontractors’

work

results that were conducted by a team assigned by the main with a contribution of 52.9% (5) Appropriate implementation methods for subcontractors with a contribution of

52.2%

(6)

Placement

of

supervisors

or

supervision

of the

main

contractor

in

accordance to their abilities with a contribution of 29.6%.

4

Discussion

Based on the results of the analysis of data processing with statistic method (correlation analysis and confirmatory factor analysis), the top six ranked factors that require extra attention from the main contractor in managing subcontractors on construction projects implementation in Jakarta as follows: 1. There is an initial explanation before the subcontractor joins both the contract and project implementation. Explanation of the contract at the beginning must have clear and detailed contract clauses, and the method of implementation needs to be explained at the beginning to get a description of the method of work to be carried out both the availability of materials and tools to be used to minimize the risks that will occur and to facilitate monitoring of work. 2. Periodic reports and meetings between the main contractor and subcontractors. Hopefully, these periodic reports and meetings will be able to decide on existing sues and provide a positive solution or impact on the project's progress. 3. The safety policy of the main contractor to the subcontractor. The management team of the main contractor must support and work on programs that can guarantee that no accidents occur or minimize work accidents or work accident prevention measures to the subcontractors. 4. Control the work of subcontractors regularly by a team appointed by the main contractor. The designated team must understand and understand the work of their subcontractors, to minimize the delays in the work of their subcontractors and minimize complaints from project owners. 5. Proper construction implementation method according to the plan of the subcontractor is an essential factor for the main contractor to assess its subcontractors’ performance and commitment in helping accelerate the implementation of work on the project. 6. Placement of supervision/supervisor of the main contractor following their abilities. Supervision/supervisor to coordinate and control all subcontractors.

Dominant Success Factors of Managing Subcontractors

5

S71

Conclusion

The topic is needed to be restudied by adding other variables that is yet to be included in this research paper, such as the need of the main contractor to draw sequence from the steps of work that has to be done including the critical path, Design coordination for every work should be discussed earlier and more detailed, subcontractors are obliged to provide its organizational structure that is clear and only focuses on the assigned work in the related projects, Due to the fact that subcontractors vary from one another, future researches and studies should be more specific on certain subcontractors for their work assigned such as substructure, structure, architecture and MEP and Plumbing).

6

(Mechanical, Electrical

Suggestion

Recommendations for this particular subject of field would be the urgent need of further and more researches, respons and reviews from participating subcontractors to the main contractor that makes sure the subcontractors succeed at maintaining good relationship with main contractors in a construction project, therefore future researches can provide more depth in the study.

References

v

1. Wong, F., So, L.: Restriction of the multi-layers subcontracting practice in Hong Kong — is it an effective tool to improve safety performance of the construction industry (2001) . Hola, B., Sawicki, M.: Knowledge assets about construction enterprise collected in the

knowledge map, Technical Transactions. Krakow (2014) 3. Huang, R.Y., Huang, C.T., Lin, H., Ku, W.H.: Factor analysis of interface problems among construction parties a case study of MRT. J. Marine (2008)

4. Sahadi, M., Agung, W.: Faktor-faktor berpengaruh terhadap kinerja manajer_proyek konstruksi dengan Structural Equation Modeling. J. Media Komunikasi Teknik Sipil (2009) 5. Bernathius, J.: Instrumen Pengukuran Kinerja Inovasi Perusahaan Kontraktor di Indonesia. Jurnal Media Komunikasi Teknik Sipil (2014) 6. Fauziyah, S., Wibowo, M.: Analisis Perbandingan Kontrak Tradisional dan Kontrak

Berbasis Kinerja (KBK) Berdasarkan Risiko Persepsi Kontraktor dengan Metode. Analytical Hierarchy Process (AHP). Jurnal Media Komunikasi Teknik Sipil (2016) 7. Simanjuntak, M.R.A., Widjajakusuma, J., Tantri, N.: Analisis Faktor yang Mempengaruhi Keputusan Kontraktor dalam Pemilihan Kontraktor Spesialis terhadap Peningkatan Kinerja Procurement pada Proyek Jalan Lokal di Kalimantan Timur. Konferensi Nasional Teknik

Sipil 3 (2009) 8. Messah,

Y.,

Pono,

R..,

Krisnayanti,

S.: Kajian

Kriteria

Pemilihan

Subkontraktor

oleh

Kontaktor Utama Menggunakan Metode Analytic Hierarchy Process (AHP) (2012) 9. Henrico, Soekiman, A.; Analsis Perilaku Kontraktor Utama dalam Melakukan Subkontraktor Konstruksi Bangunan Gedung di Indonesia. Jurnal Kontruksia (2013)

10. Yonas, E.V.: Resiko Pekerjaan Subkontraktor pada Proyek Kosntruksi di Kota Bandung ditinjau dari Sisi

Sub Kontraktor.

Universitas Katolik Parahyangan, Skripsi (2018)

572 11.

B. E. Yuwono et al. Aanval, Y., Noer, B.A.: Critical Factors of Success of Work Relationship between Main Contractor and Subcontractor in X Ltd. Steel Fabrication Company. IOSR Journal of

Engineering (IOSRJEN) 08(7), 21-27 (2018) 12.

Lendra and Andri: Tingkat Kepercayaan Dalam Hubungan Kemitraan antara Kontraktor dan

Subkontraktor di Surabaya (2006) 13.

Nugroho, R.H., Reini, W.: Kajian hubungan

kontraktual

antara kontraktor utama dengan

subkontraktor pada proyek konstruksi (2010) 14.

Girsang,

D.S., Yohanes, L., Adianto,

D., Andreas,

W.: Analisis Faktor-Faktor Penyebab

Keterlambatan Pelaksanaan Proyek-Proyek Pemerintahan (2009) 15. Budiman, P.: Keterlambatan Waktu Pelaksanaan Proyek (1999) 16.

Rohi,

D.R.P.:

Kajian

Kriteria

Pemilihan

Subkontraktor

oleh

Kontraktor

Utama

dengan

Menggunakan Metode Analytic Hyerarchy Process (AHP) (2013) 17.

Rahman, S.H.: The importance of collaboration in construction industry from contractors’ perspectives. In: International Conference on Innovation, Management and Technology Research, Malaysia, 22-23 September 2013 (2013)

18. Othman, M.R.: Forging Main and Subcontractor Relationship for Successful Projects (2012) 19.

Mbachu,

J.: Conceptual

framework

for the assessment of subcontractors’

eligibility

and

performance in the construction industry. Construct. Manage. Econ. 26, 471-484 (2008)

®

Check updatesfor

Effect of Fire Flame Exposure on Basalt and Carbon Fiber-Reinforced Concrete Siti Nooriza Abd Razak'”, Laurent Guillaumat!, and Nasir Shafiq? ' Laboratoire Angevin de Mécanique, Procédés et InnovAtion, Arts et Metiers,

Angers, France {siti_nooriza. abd_razak, Laurent. GUILLAUMAT}@ensam. eu 2 Department Civil and Environmental Engineering, Universiti Teknologi Petronas, Perak, Malaysia nasirshafiq@utp. edu. my Abstract. Damage of concrete in fire varies according to the nature of fire, mix proportions and constituents of concrete. When concrete are caught in fire, it can

suffer consequential damage. Lots of research has been conducted so far to assess the effect of fire towards concrete. Usage of fiber becomes one of the interest in this evaluation of fire performance because fiber not only enhance the mechanical properties such as, compressive and tensile strength, but fiber can

provide additional durability which is preventing cracks. In this study, chopped basalt and carbon fiber are used as reinforcement in concrete and its performance are compared with normal concrete. This paper reported the performance of fiber-reinforced concrete of standard grade 20 and 40 when subjected to fire flame at 28 days. The effect of fire on fiber reinforced concrete covers changes taking place in cement paste, aggregates, fibers, as well as their interaction that

result in changes of physical and mechanical properties of concrete specimens. A direct fire exposure test was developed to imitate real fire event. Concrete specimens were burnt at 1000 °C temperature for 90 min continuously. After burning, the specimens were cooled at ambient temperature before further testing. From the findings, it is found that G20 OPC specimens obtained the

highest residual compressive strength. Likewise, G40 OPC + CF specimens also obtained high residual compressive strength. Apart from that, the occur-

rence of spalling and cracking is observed during the duration of the fire exposure and after. The study showed adding carbon fiber in concrete improved its properties and damages observed after fire exposure were minor as compared to basalt fiber. Keywords:

1

Fire flame - Carbon fiber - Basalt fiber - Strength reduction

Introduction

Concrete has been used worldwide as the main construction material because it possesses good strength, low cost and easy production process. Even so, concrete is at risk having its properties deteriorated when it is exposed to fire. Damage of concrete in fire varies according to the nature of fire, mix proportions and constituents of concrete. Exposure to fire resulting the concrete to gradually lose its strength over 200°C

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 573-579, 2021. https://doi.org/10.1007/978-98 1-33-63 1 1-3_66

574

S. N. A. Razak et al.

temperature, which may lead to structural collapse at instance or at long hours duration. Having concrete with good fire resistivity may allows more time for rescue and firefighting before the eventual collapse of a structure in an incident. Alternatively, addition of fiber like polypropylene, steel and nylon fibers in concrete production has improved the fire resistivity of concrete. These fiber are the most common material used to resist fire as well as preventing spalling and cracking in concrete

matrix

fibers improves

while

enhanced

the compressive

and

tensile strength

the tensile strength of concrete by providing

interfacial transition zone (ITZ) in the body

[1]. Moreover,

additional

tension of

of concrete.

Basalt fiber, which possesses high stiffness, high strength, high resistance to corrosion and chemical attacks has recently been a subject of interest to use as reinforcement

in concrete

[2, 3].

In recent

year,

few

studies

have

reported

basalt

fibre

reinforced concrete tested for compressive strength showed a maximum 14% increase in strength compared to normal concrete [4]. Since basalt fiber possess good properties as composite reinforcement material, it able to control the damage due to fire. Therefore, Novakova

[5] studies reported that concrete with basalt fiber reinforcement

having 58% reduction of compressive strength from its initial compressive strength. Another option for fibre reinforcement includes the addition of carbon fibres. It possesses high stiffness, high tensile strength, high chemical resistance and can resist heat [1]. Limited studies against high temperature have been reported that addition of carbon fiber in concrete obtained high residual compressive strength. Studies by Tanyildiz

[6] showed

that

1%

carbon

fiber content

in concrete mixture

obtained sig-

nificant increment of strength, which is 37% from its initial strength when exposed to fire at 800 °C for one hour. Similarly, Chen and Liu [7] reported high residual compressive strength compared to the reference specimen when subjected to fire at 800 °C for three hours, which is 68%. Both studies mentioned that carbon fiber are able to resist cracking and slow down the occurrence of spalling due to high modulus elasticity [6, 7]. This paper examined the resistance of basalt and carbon fiber reinforced concrete in term of their mechanical properties when exposed to direct fire as imitation to real fire

event.

2 2.1

Experimental Program Materials and Mix Proportions

For this study, cement of CEM type 1, laboratory clean tap water, crushed coarse aggregate of size 20 mm and river sand with fineness modulus 2.6, which obtained from local supplier were used. Coarse aggregates were washed prior to mixing activity. Chopped filament type carbon and basalt fiber with 40mm length were used for this study. The manufacturer provided the properties details of each fiber as shown in Table 1. Figure | displayed the physical appearance of carbon and basalt fiber. Total two series of standard grade 20 and 40 manufactured in this study. Each series has 3 batches consisted normal OPC as control, OPC with basalt fiber and OPC with carbon fiber. Trial mixes has been done prior to selection of the final mix for this study to

Effect of Fire Flame Exposure on Basalt and Carbon Fiber-Reinforced Concrete

575

obtain the optimum fiber content in each mixture. Details of mix proportion are shown in Table 2 and it has been designed

according to BS

8500- 2: 2002

Type | Specific of fiber | gravity

Table 1, Properties of fibers [1] | Tensile Elastic Monofilament strength modulus diameter (1m) (MPa) (GPa) 4800 652 205 3950 652 205

Basalt |2.7 Carbon | 1.77

[8] (Fig. 2).

Elongation at break (%) 5.13

Table 2. Mixture proportions (kg.m™*) [1] Mix ID

OPC

Sand | Water | Basalt Fiber | Carbon Fiber

G20 OPC

342 | 1211

652

G20 OPC + BF G20 OPC + CF

342 | 1211 342 | 1211

652205 652 205

G40 OPC

405 | 1193

642

190 | -

G40 OPC + BF

405 | 1193

642

190

G40 OPC + CF

405 | 1193

642

190 | -

Fig. 1.

2.2.

Coarse Agg

Basalt fiber

205

:

:

| 5.13 -

: 3.42 -

(8.1

Fig. 2.

8.1

Carbon fiber

Mixing and Casting

100 x 100 x 100 mm cubes were casted for each concrete mix. The mixing was carried out in VMI mixer. Firstly, cement and aggregates were dry mixed for one and a half minutes. Then, half of the water was added and continued to mix for one minute. Finally, fibers were added gradually into the mixture together with the remaining water until uniform dispersion of fibers were observed in the mix. The cubes were then filled with fresh concretes and compacted using tamping rod. The cubes were demoulded after 24 h and wet cured in curing tank filled with water up to 28 days for further

testing.

576

2.3

S. N. A. Razak et al.

Testing Program

The concrete specimens were burnt with direct fire exposure from natural gas fire setup at temperature 1000 °C. The burning process was maintained for 90 min upon reaching 1000 °C temperature. The temperature was recorded using thermocouple glued to the concrete surface meant for fire exposure. The specimens were allowed to cool naturally to room temperature after exposure to fire. Then, the mass loss and residual compressive strength of the test specimens were determined.

3 3.1

Results and Discussion Loss of Mass/Density

Loss in mass of concrete specimens after exposure to fire were measured and is shown in Fig. 3. Several studies have reported that, loss in mass of concrete increases with temperature increases. For this study, the concrete specimens were exposed to 1000 °C fire, which is high temperature and high loss of mass were expected due to water evaporation process as well as occurrence of spalling. The graph displayed that G40 concrete specimens suffered high mass reduction compared to G20 concrete specimens. Moreover, the mass reduction of G40 concrete specimens were significant due to spalling with reduced concrete cross-section. G40 OPC having average 7.87% of mass reduced out of the 3 specimens’ groups. Effect of spalling from internal thermal stress due to less porosity within concrete body have contributed to high mass reduction. For this concrete series, G40 OPC + CF loss of mass is lesser compared to G40 OPC + BF and G40 OPC. Although G40 series of concrete experienced spalling but damage due to spalling for G40 OPC + CF are minor, only the edge of the concrete specimens is tear off. From this results, carbon fiber was found out to improve the tensile strength properties and provides good concrete constituents (cement, aggregate and fiber) adhesion compared to basalt fiber. For G20 concrete specimens’ series, no spalling was observed throughout fire exposure test duration. G20 specimens only suffered surface cracking up to the edge of specimen for 90 min exposure. Having high porosity matrix has reduced the effect of spalling occurrence for G20 concrete specimens with regard to fiber addition. From this study, G20 concrete specimens experienced loss in mass because water was forced to leave from the concrete body through evaporation process.

Effect of Fire Flame Exposure on Basalt and Carbon Fiber-Reinforced Concrete

577

Mass reduction (%) of G20 and G40 concrete samples at 1000 °C 8 5S 7 i 8S ge oss g 5 2 45 Bo4 2 3s 3 i 25 2 ct

os 0

Gao

G20 opc

OPC+BF

OPC+CF

Fig. 3. Mass loss (%) at 1000 °C fire exposure 3.2

Residual Compressive Strength

Compressive strength of concrete at high temperature is the prime concern in fire resistance design. Concrete suffers gradual strength loss as the temperature increase in fire incident. This paper reported the residual compressive strength of fiber-reinforced concrete of standard grade 20 and 40 when subjected to fire flame at 28 days. Figure 4 illustrated the average compressive strength for two series of concrete specimen G20 and G40 when subjected to 1000 °C temperature. From the graph, it displayed that the compressive strength reduced for all concrete specimens except G20 OPC specimen. G20 OPC specimen displayed increment of 7.8% in strength compared to the specimen prior to fire exposure. The result obtained display contradictory behaviour of concrete as reported in literature. This showed that there is possibility of development in strength during exposure to fire may have occurred during the transfer of heat across the body the concrete. G20 OPC + BF specimen experienced reduction of 52.2% of strength while G20 OPC + CF specimen experienced reduction of 18.4% of strength. The differences in reduction of strength for both fibers are due to the differences in thermal properties. Basalt fiber degrade earlier than carbon fiber at temperature higher than 700 °C, therefore the strength afer exposure to 1000 °C decrease more than 52%. For G40 specimens, G40 OPC suffered higher reduction of strength of 53.8% of its initial strength. Comparatively, the G40 OPC + CF has the highest residual compressive strength at 58% of its initial strength with G40 OPC + BF with 53.8% of its initial strength. Results showed that adding carbon fiber into concrete enhance the bonding within concrete matrix, thus, retain more strength after being exposed to fire at 1000 °C compared

to basalt and control concrete specimens

[9].

578

S. N. A. Razak et al.

Compressive strength of G20 and G40 concrete samples 50,00 _, 45:00 &

40.00,

—_—_—___

= 2 5 a =

35.00 30.00 25.00 — 20.00 15.00

—

S

1000 —

—

fp

—

—_————ps—

—

—

—

5.00 0.00

Before Fire

After Fire

Before Fire

G20

mOPC BOPC+BF

After Fire G40

mOPC+CF

Fig. 4. Comparison of compressive strength (MPa) before and after fire at 1000 °C 4

Conclusions

Based on the objective of this study, the following conclusion are made: I. Cracking and spalling are occurred due to water evaporation and internal thermal stress within the concrete body after exposure to 1000 °C temperature. G20 specimens did not experienced spalling like G40 specimens and G40 OPC + CF specimens suffered minor damages from spalling. IL. For residual compressive strength, all specimens experienced strength loss more than 50% except for G20 OPC which recovers 7.8% of strength, which is contrary with the literature. G40 OPC + CF obtained high residual strength with significant difference of 11.4% from reference specimens. This support the advantageous properties of carbon fiber where the damage due to fire exposure was lower compared to reference and basalt fiber.

Acknowledgement. The authors gratefully acknowledge the technical support provided by MIT (Matériaux pour infrastructures de transport) IFSTTAR France in this study.

Effect of Fire Flame Exposure on Basalt and Carbon Fiber-Reinforced Concrete

579

References 1. Abd Razak, S.N., Shafiq, N., Azmi, Y.M., Guillaumat, L., Farhan, S.A., Ayesha, W.F., Grampeix, G.: Effect of fire flame exposure on basalt and carbon fiber reinforcedconcrete

(2020) 2. Singha, K.: A short review on basalt fiber. Int. J. Text. Sci. 1(4), 19-28 (2012) 3. Bhat, T., Chevali, V., Liu, X., Feih,

S., Mouritz, A.P.: Fire structural resistance of basalt fibre

composite. Compos. A Appl. Sci. Manuf. 71, 107-115 (2015)

4. Irine, F.:

Strength aspects of basalt fiber reinforced concrete. Int. J. Innov. Res. Adv. Eng. 1

(8), 192-198 (2014) 5. Novakova, IL, Thorhallsson, E., Bodnarova, L.: Behaviour of Basalt Fibre Reinforced Concrete Exposed to Elevated Temperatures. 5""International Workshop on Concrete Spalling

due to FireExposure (2017) 6. Tanyildizi,

H.: Effect of temperature,

carbon

fibers, and

silica fume

on

the mechanical

properties of lightweight concretes. New Carbon Mater. 23(4), 339-344 (2008) 7. Chen,

B.,

Wu,

K.,

Yao,

W.:

Conductivity

of

carbon

fiber

reinforced

cement-based

composites. Cement Concer. Compos. 26(4), 291-297 (2004) 8. BS 8500-2: 2002- Concrete - Complementary British Standard to BS EN 206-I- Part-2 Specifications for constituent materials and concrete: BSI, 389 Chiswick High RoadLondon 9. Larson, B.K., Drzal, L.T., Sorousian, P.: Carbon fibre-cement adhesion in carbon fibre

reinforced cement composites. Composites 21(3), 205-215 (1990)

® ‘upaates

Delay and Cost Overrun of Palm Oil Refinery Construction Projects: Artificial Neural Network

(ANN)

Model

Muhammad Sani Abdullah, Wesam Salah Alaloul®®, M. S. Liew, and Muhammad

Ali Musarat

Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS,

Bandar Seri Iskandar,

32610 Tronoh,

Perak, Malaysia

wesam. alaloul@utp. edu. my

Abstract.

In spite of the development

and

innovation

in the construction

technologies, still, delay and cost overrun are the most crucial challenge of the construction industry in both developed and the developing countries. This research aims to develop a prediction model using Artificial Neural Network (ANN). The prediction model consists of the most impactful causes of delays and costs overruns during the construction of Palm Oil Refinery projects which were ranked based on importance, severity and frequency. A series of 39 questions were developed from the questionnaire survey causing delays and cost overruns during construction of palm oil refinery projects. Artificial Bee Colony (ABC) algorithm was used to develop the prediction model for palm oil construction projects. Keywords: model

1

Delay - Cost overrun - Artificial Neural Network - Prediction

Introduction

One major contribution to the development growth of socio-economic is the construction industry. The contribution from the construction industry towards the nation’s economy is the reason why improvements in the efficiency in proper costing and timelines will be a saving to the whole country

[1-3]. Several advancements

have been

brought to the construction industry but still delay and cost overrun are the major distress for the stakeholders which affect the success of the project and ending in affecting the country’s economy [4—7]. Delays and cost overruns reduce the available economic opportunities and set limitations to the growth potential upon the entire country’s economy [8]. Usually, when projects get delayed, the extension is given or accelerated and thus, incurring extra cost. The usual practices are to provide a small reasonable percentage from the project’s cost budgeted as contingency funds within the contract amount and this estimation is normally based on assumption judgment [9]. The project can be considered successful upon the project meeting the technical specifications, also should there be an increased level of satisfaction on the project’s results

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 580-589, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_67

Delay and Cost Overrun of Palm Oil Refinery Construction Projects

581

amongst the clients, completed within planned cost or budget and commissioned the project on time

[10-15].

Malaysia is the second biggest palm oil producers in the world which accounts for approximately 41% of global production and 47% of world exports. The basic of palm oil is briefly explained as palm trees bearing palm fruits that are harvested from the plantations and sent to the palm oil mills for extraction of crude palm oil (CPO) from the Fresh Fruit Bunch (FFB). An extraction percentage of approximately 20% to 22% CPO. per ton of FFB, depending on the quality of the FFB produce from the plantation. Palm oil in Malaysia contributes to a total of RM44.8 billion or 3.8% of the GDP [16, 17]. Malaysia is currently one of the biggest producers of the palm oil production by accounting 28% of oil production and 33% of exports to the world and plays an important role in increasing global need of palm oil [18] which shows the importance of palm oil to Malaysia; hence, this research focus on the management of palm oil refinery construction projects. The construction industry exposed to higher risk value and uncertainty compared to other industries where 90% of large construction projects are facing delays and cost overruns [19-21]. Malaysia is constantly building refineries but has not developed a comprehensive and practical system that can be continuously used to improve the management of the projects. Delay and cost overrun in palm oil refinery construction projects are still rampant and not well organized. In many cases, the company responsible for construction plays a major part in resolving and avoiding problems. Some experienced contractors take advantage of certain problems arising to make a contractual claim for additional monies and apply for an extension of time which is also a problem to the client and the project management

team

[22].

This study intends to develop a model to predict the cost and time overrun through the understanding of problems during the construction stage of the palm oil refinery and to find out the major causes of delay and costs overrun of the project. The study provides the future palm oil refinery construction with a baseline to predict their projects so that they can prepare to face such circumstances when similar problems occur. The understanding of problems faced by the builders of the palm oil refineries during construction stage is critical as it is the balance that may result to a successful project or the one that fails to engage the fundamentals of good project management in the palm oil refinery projects.

2

Methodology

This study

is an extensive

version of research concluded

by Abdullah,

et al. [23] on

delays and cost overruns causes during the construction of palm oil refinery projects. The collected data was analyzed using the Relative Importance Index (RII) to rank the causes. Artificial Neural Network

(ANN)

running on Artificial Bee Colony

(ABC)

al-

gorithm was used to develop the prediction model. The flowchart of this research is provided in Fig. 1.

582

M. S. Abdullah et al.

Start

wes

Soran

‘Artificial Neural "Network (ANN)

‘Antificial Bee Colony Algorithm (asc) Model Validation

Fig. 1.

2.1

Research Flowchart

Artificial Bee Colony (ABC)

Artificial bee colony (ABC) algorithm is an optimization technique introduced by Karaboga [24]. The model of the algorithm simulates the intelligent foraging behaviour of honey bee, which was proposed by [25]. A set or a group of honey bees, which effectively accomplish tasks through social collaboration, is known as swarm. The ABC algorithm has three types of bees: employed bees, onlooker bees, and scout bees. The employed bees search for food and share the information with the onlooker bees. The onlooker bees select good food sources based on quality. The scout bees are responsible for exploration for a new food source to replace old one [26, 27]. For swarm size = {xj,1,%)2,..-..-.---,Xip}, the following are few derived equations: For initialization purposes:

vig = mig +iy-(ig — Mey)

()

where D is the dimension size, j is a random dimension index selected from the set {1, 2,......D}, X, is randomly selected food source, and V; is a new source for food and @ is a random number within [—1, 1].

Delay and Cost Overrun of Palm Oil Refinery Construction Projects The Eq. 2:

probability selection of onlooker bee for a new

583

food source is described in

Sit; SN

-it;

(2)

where fiti is the fitness value of the i solution in the swarm. As seen, the better the solution i, the higher the probability of the i" food source selected. 2.2

Develop Prediction Model Using ANN

A prediction model using Artificial Neural Network (ANN) was developed and designed using MATLAB version R2015B. ANN develops learning algorithm as it is more knowledgeable and aware of each process of training and learning. Learning process structure is not bounded by any parameters as it has the ability to learn and model non-linear and complex relationships. ANN can improve model data with high volatility and non-constant variance. Results of the survey questionnaire were incorporated into the ANN algorithm which got through a series of tests to determine its accuracy in the prediction model. Once good prediction results have achieved, the model was tested and validated. 2.3,

ANN

Prediction Model Result

The results from the ANN prediction model work on two stages. The first stage is the data that has been incorporated into the ANN model. These data are taken from the survey questionnaire. It is incorporated into the ANN model to learn using the selected algorithm. The second stage is the actual user, providing the data for the prediction model to process. The ANN model will display its prediction results for delays and cost overruns of the project in percentage. 2.4

Validation of the ANN

Prediction Model

Validation of the ANN prediction model goes through a 2-tier process. The to test the prediction model internally. The model was tested using data from projects to determine the accuracy of the prediction results. For the second tier validation, the ANN prediction model was sent experts in the field for their validation and to do field test on real projects to the assessment model being able to attain its objectives as an accurate determine the causes of time delays and costs overruns.

first tier is the actual to several determine model to

584

3

M. S. Abdullah et al.

Results and Discussion

A series of 39 questions were developed as the inputs for the project evaluation. From a total

179 causes of delays

and costs overruns, risk that were ranked

(with RII above

0.45) were selected as database for the ANN prediction model. Reason for shortlisting the causes is to keep the research focus on critical causes that has high risk factors. Any causes that have lesser RII below 0.45 are considered less significant and the risk can still be manageable. The prediction model start displays a selection of 3 types of projects. The selection of project size is needed as the ranking for these projects varies and therefore the user can have more accurate assessment. 3.1.

Model Training and Testing

ANN is a heuristic in nature. It absorbs the data inputs and use the layers of neurons. It processes the inputs and adjust the weights of the outputs. Afterwards it measures its outputs against the desired outputs and go conduct back propagation to adjust its weights and keeps this cycle until it has reached the nearest desired results. As such this model also needs to go through its training process. Figure 2 shows the test results of the model.

Lage

Medium

3

3

Por

s Sort

sarole tage

Small 3

*

* Soy

sample Medium

oa i Bost

oa wat 30, 0 tases sonole

sample Small

f o Oe

—

as

+ 02 iBos °

i somo

42

Fig. 2. Prediction model testing 3.2.

Mean

Square Error

Mean square error (MSE) shows the level of accuracy of the prediction model. Measures the average of the squares of the errors, i.e. the average squared difference between the estimated values and what is estimated. MSE can be calculated by using

Eq. 3.

(3)

Delay and Cost Overrun of Palm Oil Refinery Construction Projects

585

MSE tries to minimize the average squared error between the network's output, and the target value over all the example pairs. Table | and 2 show favorable MSE results of the prediction model during training and testing. Table 1.

Results of training

- MSE

Training — MSE Large project

| Medium project | Small project

Cost Overrun | 0.002916822003 | 0.000992635967 | 0.066996300789 Delay 0005327824828 | 0.016000113486 | 0.011927485853

Table 2. Results of testing - MSE Testing - MSE

Large project

| Medium project | Small project

Cost Overrun | 0.000779952629 | 0.000000001414 | 0.000142470600 Delay 0.000991252498 | 0.000000007848 | 0.000010461657

3.3

Selection of Model Project Size

The assessment model has 3 types of project size to provide the user with a more accurate assessment. The size of the project is based on the monetary value of the project as follows: 1) Large Project - RMS0 million and above, 2) Medium Project — RM30 million up to RMSO million and 3) Small Project - RM30 million below. The data taken from the survey questionnaire was ranked based on the severity and frequency. The relative importance index was applied to allow the ranking to be done based on project size. The result of the prediction model for time delays comes as 10.6215% and for cost overruns as 4.9256%, based on the user’s earlier inputs on the 39 questions. The prediction is based on likelihood occurrence based on the measurement of risk input by the user. 3.4

Model Validation — Field Assessment

To ensure that the risk prediction model able to handle actual cases in the real world, the model was sent to 6 practicing professionals to test, validate and asses the model. Data collected from the survey questionnaire makes more sense with the validation of causes assessment. As most often, the key personnel make their own assessment of causes of delays and costs overruns based on their own experiences and make their inputs into the timeline and cost. The professionals were given a short assessment to answer after using this model for their own projects. The results are discussed in Table 3.

586

M. S. Abdullah et al. Table 3.

No

1 2

| Assessment Questions

|edcng Accuracy of

| Overall assessment

Model validation summary

Totally not | Slightly

achieved _| achieved 0 1

|[°

| Achieved

0

|| 33

Greatly

RIL

achieved 2

0.79

3

0.88

assessment model

3 Objective of user

‘0

0

|2

4

0.92

4

| Time savings / delay

I°

|2

|2

2

0.75

10

ie

|2

3

0.83

| Recommendations of

|0

|4

2

0.83

5 6

avoided |Cost saving

the prediction model

Based on the RII result, the most significant result of validation is ‘objective of user’. The purpose of the objective is met and thus the rating for this is highest. The second most significant RII is ‘accuracy of risk assessment model’. These comments from the industry practitioners are important to point out the nearest risk assessment to their own inputs. The third and fourth most significant RII is ‘costs savings’ and ‘recommendations to risk’ with both having RII of 0.83. Meaning that the user has selected cost savings and risk recommendations are high. The fifth most significant RII of 0.79 is ‘overall risk assessment modeling’. The result of this shows that there is still room for improvement of the risk assessment model which is a positive response. The sixth significant RII of 0.75 is ‘time savings/delay avoided’. This is interpreted as there is an opportunity for time savings and avoiding delays is potential to other users the benefits of having this risk assessment model in practice. 3.5

Model Validation

The prediction model goes through an accuracy test by comparing it with the previous projects to measure the prediction model’s performance. Referring to Table 4, the model’s 39 questions were answered and based on 3 types of project sizes to see which of the prediction answered was nearest to the actual project percentages of delays and cost overruns. This is to determine if the prediction model does predict more accurately based on project size.

Delay and Cost Overrun of Palm Oil Refinery Construction Projects

587

Table 4, Prediction model Vs Actual project No

Project

Project | Time/Cost | Actual | Prediction model

size

problems

1 | Refinery project in | Large | Cost Kuching, Sarawak overrun Delay 2

3

Refinery Bulking Project in Kuantan,

Pahang Palm Oil Bio-Diesel | Large

Refinery Project, Kuantan, Pahang

4

Medium

Cost overrun Delay | Cost

Overrun Delay

Accuracy

Large | Medium | Small

40% | 38.34 /34.25 | 24.45 | 95.85% 16% 10% 23% 40%

15%

| 15.32 | 33.02 8.21

18.97 | 95.75%

13.48

52.63 | 96.52%

| 13.90 | 20.05 | 33.16 | 24.14

33.66 | 97.05% 37.28 | 93.16%

| 10.11

11.16

41.27 | 95.11%

Conclusion

The developed model able to predict the delays and costs overruns for 3 types of project sizes namely large projects, medium size projects and small size projects. However, if the causes are not properly managed, can cause the projects to suffer delays and cost overruns. After the development of the prediction model, it was tested and verified using mean square error (MSE) which showed favorable results. The prediction model went through two validation tests. One is done against a case study project and the other is sent to field professional to gauge their value on this prediction model. Finding of the above, for the field validation, it is found that the users being new to this prediction model, needs more time to be comfortable to fully use it in practice as it will take some time for them to assimilate their projects into the model system which is very probable way forward as a tool for enhanced project management whereby project management team can utilize it in many variations of projects. On the validation using case study, it was found that small projects show higher value to delays and cost overruns than large and medium projects on 2 out of3 projects. The highest prediction for cost overrun is 95.85% and lowest at 93.16% accuracy against actual. For the prediction of delays, the prediction result with the highest accuracy is 97.05% and lowest is 95.11%.

Acknowledgements. The authors would like to thank Universiti Teknologi PETRONAS (UTP) for the support provided for this research.

588

M. S. Abdullah et al.

References

. Aibinu, A., Jagboro, G.: The effects of construction delays on project delivery in Nigerian construction industry. Int. J. Proj. Manage. 20(8), 593-599 (2002) . Tayeh, A. Al Hallag, K., Alaloul, W.S., Kuhail, A.R.: Factors affecting the success of construction projects

in Gaza Strip. Open Civ. Eng. J. 12(1), 301-315 (2018)

. Alaloul, W.S., Liew, M.S., Wan Zawawi, N.A., Mohammed, B. , Adamu, M.: An Artificial neural networks (ANN) model for evaluating construction project performance based on

coordination factors. Cogent Eng. 5(1), 1507657 (2018) . Tayeh, A., Al Hallaq, K., Al Faqawi, A.H., Alaloul, W.S., Kim, $.Y.: Success factors and barriers of last planner system implementation in the gaza strip construction industry. Open

Constr, Buil. Technol. J. 12(1) (2018) . Alaloul,

W.S.,

Liew,

M.S.,

Zawawi,

N.A.W.A.:

Identification

of coordination

affecting building projects performance. Alexandria Eng. J. 55(3), 2689-2698 (2016)

factors

. Alaloul, W.S., Liew, M., Zawawi, N.A.W.A., Kennedy, I.B.: Industrial revolution 4.0 in the

construction industry: challenges and opportunities for stakeholders. Ain Shams Eng. J. (2019)

. Alaloul, W.S., Musarat, M.A., Liew, M., Zawawi, N.A.W.A.: Influential safety performance and as ment in construction project review. In: AWAM International Conference on

Civil Engineering, pp. 719-728. Springer (2019)

. Singh, R.: Delays and cost overruns in infrastructure projects: extent, causes and remedies. Economic and Political Weekly 43-54 (2010) . Sambasivan, M., Soon, Y.W.: Causes and effects of delays in Malaysian construction

industry. Int. J. Project Manage. 25(5), 517-526 (2007) . Othman, A., Torrance, J.V., Hamid, M.

Factors influencing the construction time of civil

engineering projects in Malaysia. Eng. Constr. Archit. Manage. 13(5), 481-501 (2006)

. Abedi, M., Fathi, M., Mohammad, M.: Effects of construction delays on construction project objectives. In: The First Iranian Students Scientific Conference in Malaysia, vol. 9, pp. 1-8 (2011)

. Musarrat, M.A., Inderyas, O., Khan, $., Shah, A.A.: Causes of delay in the execution phase of construction projects in Khyber Pukhtoonkhwa Pakistan. Sarhad Univ. Int. J. Basic Appl. Sci. 4(1), 62-70 (2017) . Musarat, M.A., Ahad, M.Z.: Factors affecting the success of construction projects in Khyber

Pakhtunkhwa, Pakistan. J. Constr. Eng. Proj. Manage. 6(4), 1-6 (2016) . Lichtenberg, S.: How to avoid overruns and delays successfully: nine basic rules and an associated operable procedure, vol. 406, (2005). Dostopno na: hitps://wwwicoste.org/ roundup . Altaf, M., Musarat, M.A., Khan, A., Shoukat, Z., Salahuddin, U.: Change order impact on construction industry of Pakistan. In: AWAM International Conference on Civil Engineer-

ing, pp. 391-402. Springer (2019)

. Sue, T-T., Pantzaris, T.: Pocketbook of palm oil uses. Malaysian palm oil board, Kuala Lumpur, Malaysia (2009) . S. 18. 19.

Malaysia.

(2019).

Selected

Agricultural

Indicators

2019.

https://www.statista.com/

statistics/952996/malaysia-palm-oil-share-of-gdp/ MPOC. Malaysian Palm Oil Industry. https://mpoc.org.my/malaysian-palm-oil-industry/ Endut, R., Akintoye, A., Kelly, J.: Cost and time overruns of projects in Malaysia. 21, 243252 (2009). Accessed August

Delay and Cost Overrun of Palm Oil Refinery Construction Projects 20.

589

Memon, H., Rahman, I.A., Abdullah, M.R., Azis, A.A.A.: Factors affecting construction cost in Mara large construction project: perspective of project management consultant.

International Journal of Sustainable Construction Engineering and Technology 1(2), 4154 (2010) 21.

Alaloul, N., Zawawi, A., Liew, M.: Coordination process in Malaysian construction industry-critical review. In: 2nd World Thought Conference on Islam & Civilization, Perak,

22.

Malaysia (2014) Patah, L.A., de Carvalho, M.M.: The value of project management methodology: a case

23.

Abdullah,

study. Prod. Manage. Dev. 7(2}, 177-182 (2017)

24. 25. 26. 27.

M.S., Alaloul, W.S., Liew, M.S., Mohammed,

B.S.: Delays and cost overruns

causes during construction of palm oil refinery projects. In; MATEC Web of Conferences, vol. 203, p. 02004. EDP Sciences (2018) Karaboga: An idea based on honey bee swarm for numerical optimization. Technical report106, Erciyes university, engineering faculty, computer ...2005 Tereshko, V., Loengarov, A.: Collective decision making in honey-bee foraging dynamics. Comput. Inf, Syst. 9(3), 1 (2005) Zhu, G., Kwong, S.: Gbest-guided artificial bee colony algorithm for numerical function optimization. Appl. Math. Comput. 217(7), 3166-3173 (2010) Shah, H., Tairan, N., Garg, H., Ghazali, R.: Global gbest guided-artificial bee colony algorithm for numerical function optimization. Computers 7(4), 69 (2018)

® ‘upaates

Human Factor Engineering in Oil and Gas Construction Works - A Case Study to Mitigate Safety Risk Mat Saaud Nor Arinee’, Othman Idris’, and Ir Baharuddin A. Rahim?

' Department of Civil and Environmental Engineering, University Technology PETRONAS, 32610 Seri Iskandar, Perak, Malaysia {nor_18003122, idris_othman}@utp. edu. my > Custodian Construction, PETROLIAM Nasional Berhad, Tower 3 KLCC, Kuala Lumpur, Malaysia arahimbahar@pe tronas. com. my Abstract. The oil and gas construction industry has become one of the major high risk industry in Malaysia. With the high volume and cost of project, the risk of safety also become higher. There are significant numbers causes the accident is the human factor. Thus, the requirements of human factor engineering in the process safety management has become compulsory to the industry. However, due to the cost constraints and schedule, management may find way to do shortcut during execution. This study will somehow categorize the human factor engineering causation and portrays its factors that need further research on the subject. The purpose of the case study is to find the factor relate to Human Factor Engineering in Oil and Gas Construction works. The recommendation from the study is to further study the awareness of the Human Factor Engineering implementation in Oil & Gas Construction Works in order to further mitigate the safety risk at projects. Keywords: Human factor - Safety - Risk - Process safety management Worksite accident - Oil & gas construction - Risk

1

-

Introduction

Safety at the workplace has become the main focus in construction industries nowadays, this is as the result from the many accidents happened especially on the heavy industry construction including oil and gas has led to the catastrophic and fatality at the workplace [1]. The Department

Of Safety

and

Health

(DOSH)

was

established

in

1994

and

is

under the Ministry of Human Resources. It is accountable for enforcing laws on occupational, safety, and health. A health and safety policy is a written statement of principles and objectives that embodies the commitment of an employer to occupational health and safety [2-4]. Table

1 and Chart

| tabulated the numbers of accident at

the worksite.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 590-601, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_68

Human

Factor Engineering in Oil and Gas Construction Works

591

Table 1, Number of Accident at construction site from 2014- 2018 (source from Construction Plus Malaysia) (10] [CONSTRUCTION ACCIDENT RATE IN MALAYSIA

Bh terme sy

‘Occupational Accidents By Sector Until October 2018 (Investigated) Mining and Quarrying Construction Agricuture, Forestya Uuities Electricty, Ga, Transport, Storage and

Chart 1.

Occupational Accidents Statis

ics by Sector until October 2018 [10]

As human factor was found to be the among the main focus area that shall be look into and manage properly. Human Factors discovers and applies information about human behavior, abilities, limitations, and other characteristics to the design of tools, machines, systems, tasks, jobs, and environments for productive, safe, comfortable, and effective human

use [5, 6].

The focus area of this study is on the construction of oil and gas, this is due to the oil and gas construction sector is among the most hazardous industry due to the working environment which exposes workers to high risk of accidents. Statistics by the Malaysian Ministry of Human Resource had shown that the number of mortality and disability cases involving construction workers were the highest among the other

592

M. S.N. Arinee et al.

sectors

[7, 8]. Table

| below

showing the statistics of Construction

Accident

Rate in

Malaysia, strengthen the statement above. Due to the schedule driven and the needs of completing the project in shorter time the human factor in the decision making process has been neglected and is the main cause of incidents, even involuntary, directly connected with the use of the product itself. If the accident rate is to be decreased, human factor must be better understood and the knowledge more broadly applied [9]. Based on history, many preventable accident and incident happened in the industries are due to lack of awareness and enforcement of proper Human Factor Engineering to be implemented in the facilities design. Human Factor Engineering (HFE) has become one of the important study elements in the Process Safety Engineering study in Oil & Gas construction and heavy industry construction. Thus the understanding on the implementation shall be fully understood by all parties related to the project. The reasons for this study is to manage the project risk in safety. Safety elements as one of the important risk factor in planning and execute the project and also branch of Project Management is the Project Risk Management. By mitigate the risk in the project, the project goals and objectives will be achieved.

2

Literature Review

This section presents HFE overview by explaining its brief history, definition and principles. It also will briefly have explained on the previous records with regards to the HFE and the importance of this elements in managing the safety risk at the workplace. 2.1

Overview

The study of the HFE in the construction industry has been conducted worldwide due to the importance of the HFE in the design to ensure the safety at the workplace to be managed properly. However, there are not as much as the study on other factors lead to accident such as machinery factor and procedure factors. The system of reporting data about internal working and safety is also minimal. The manpower driven industry is facing regular accidents in daily working, which cause heavy losses in terms of men, money

and time

[11,

12].

There are few numbers of standards are available for the construction practitioners. The standards could require, where appropriate, to take into account the physical and cognitive ergonomic assessments of the operator tasks, the equipment they will use to complete

those

tasks,

and

the environment

in which

the

tasks

occur

[13].

the

main

standard is The ISO: 6385— Ergonomic Principles in the Design of Work Systems [14]. There also another major standard to be refer to that is ISO 11064 - Ergonomic Design of Control Center [15]. This standard offers nine principles for the ergonomic design of control center and guidance on specific aspects of control room design, including layout, workstation design, controls and displays, and environmental requirements. Another cross reference that is not mentioned in the ISO 6385 is the one to the standard ISO 12100 — Safety of Machinery [16] which suggests a five steps

Human

Factor E1

neering in Oil and Gas Construction Works

593

methodology to perform risk assessment at design stage and the overall strategy to take into account safety of machinery in the life cycle, considering usability, maintainability and cost efficiency. Previous practice, the design of the facilities is less considering the human factor where the man and machine interface is required to be more tailor suite to the capability and the ergonomic or human. the requirement become more important now to ensure the accident due to the human factor is minimum. The importance of the human centered design [17] has been growing rapidly day by day. That is why the engineering of human factor in process safety management has become important in project delivery. 2.2

HFE Benefit and Importance

The implementation of HFE in the construction design has become more important nowadays due to the benefit from it not just the client, the contractor and the subcontractor as well. The construction industry suffers from debilitating and costly occupational injuries primarily to workers backs, necks, shoulders, hands and arms. These types of injuries or traumas are commonly called repetitive motion injuries (RMIs) and are caused by activities that are repeated on a regular basis [4, 18, 19], these are among the cause of the incomprehensive of HFE implementation but this issue shall be mitigated via the proper implementation of HFE in PSM. The aim for the HFE is to provide a system that is formulated which prevents and manages the accident in advance, and formulated safety management system, for that assessment

of risk is important

in construction sector [20, 21].

All construction works must be carried out in compliance with the requirements of operational safety, economic efficiency, the environment and in certain cases national security [22, 23]. Safety is one of the main basic requirements in the construction works. Thus the standard of HFE is most beneficial to support the project goals. In many cases of the accident investigation reported that not many documents like Job Hazard Analysis (JHA), work procedure and work instruction usually do not identify process safety issue and concerns about the human factor [24]. With the gaps in the assessment, many accidents occur supposed to be avoidable at the first place if all workers involved aware on the importance of the HFE even during the project execution. More in depth study on the action and interventions that could be taken to minimize the occurrence of human error in the construction site thus minimize the occurrence of accidents, injuries and even mortality [25, 26]. With the recommendation from previous study, thus this study is to be conducted in order to manage the safety risk. Unawareness of the importance of HFE may lead to many behavioral action by the workers to perform their work under the negligence and disregard towards safety and accident preventions. The workers shall be behaving differently if they know and understand the importance of the design done is inclusive the human factors to minimize the safety risk.

594

3

M. S.N. Arinee et al.

Case Study from Reported Accident

3.1

Result Study on Human Factors Affecting Causation from the Accident Reporting

Tabulation of the key factor contribute to the accident in construction site from 2012— 2019 extracted from accident and incident reporting as officially reported to the

management team.

Factor/ | Regulatory | Organizational| Supervisory svn Year | factor | Process | Violations |“ ” 2 | 4 it ay 2 203 | 4 9 8 0 204 [1 3 6 0 2s | 4 5 4 1 26 | 4 10 2 3 207 | 0 3 5 1 208 | 2 it 7 1 29 [1 3 5 1

Skil | sitBased |Valaton} Toosand | [Undeutiiation} — Eror s Equipments 8 2 2 it 4 7 7 B 2 5 3 6 3 3 4 7 6 8 il 10 5 8 5 5 5 5 5 5 3 3 2 8

9 6 5 4 6 B 7 4

Line chart for the Accident Causation Factor relate to the Human Factor contribute to the accident in construction site from 2012-2019 extracted from real reporting.

Accident Factor

—pesien

Bar chart for the Accident Cau: tion Factor relate to the Human Factor contribute to the accident in construction site from 2012-2019 extracted from accident and incident reporting as officially reported to the management team.

Human Factor Engineering in Oil and Gas Construction Works

2012

2013

design tools and equipments, violations skill based error skill underutilization adverse spiritualstate supervisory violations organizational process regulatory factor

design tools and equipments violations skillbased error skill underutiization adversespiritual state supervisory violations ‘organizational process regulatory factor o

5

2014

10

15

2015

design tools and equipments, violations skill based error skillunderutiization adverse spiritual state supervisoryviolations ‘organizational process regulatory factor

design tools and equipments violations: skillbased error skillunderutilization adversespiritual state supervisory violations ‘organizational process regulatory factor o

2

2016

esi took and equipments violations skillbased error skilunderutilization adversespiritual state supervisory violations ‘organizational process regulatoryfactor o

4

6

w

820

2017

design tools and equipments violations: skillbased error skillunderutiization adverse spiritual state supervisoryviolations ‘organizational process regulatory factor 5

10

—— — = = =m o

15

5

2019

2018

design tools and equipments, violations skill based error skillunderutiization adverse spiritual state supervisoryviolations ‘organizational process regulatory factor

595

co tools and equipments violations skillbased error skilunderutiization adversespiritualstate SLENisor Vicktor organizational process regulatoryfactor

— = — — =m — ==

8

596

M. S.N. Arinee et al.

3.2

Analysis of the Result

The case study has been conducted to identify and also to validate the theoretical assessment conducted via literature review from previous research conducted by others that lead to the causation of the accident where the key human factor identified are regulatory factor, organizational process, supervisory violations, adverse spiritual state, skill underutilization, skill based error, violations, tools and equipment’s and design [27]. The analysis conducted by mapping the accident report into the category as mentioned above. Each accident may have more than 1| factor for each accident. the mapping is based on the recommendation and root cause analysis from the reported

cases. The study conducted on 158 numbers of reported accidents varies from year 2012 to 2019. The accidents cover for all construction activities for Offshore and Onshore facilities including the road accidents on the worksite. The accident reporting form operation sage also taken into consideration as the operationability of the facilities shall also be the main factor or consideration during design and construction stage. From the analysis, it can be concluded that the top 5 key human factor contribute to the causation for the accident to happen are tools & equipment’s, organizational process, supervisory violations, design, and skill based error. These elements playing the most significant cause and it close relate to the awareness of human factor engineering importance into design and during the execution of the project themselves. Thus, it is very important to recommend for further study to be conducted in doing the analysis on the workers and stakeholders involved in the project aware and fully educate on the importance in managing the safety risk in the project. Safety management and risk prevention is a common thread throughout every workplace, yet keeping employee Safety and health knowledge current is a continual challenge for all employers [28].

4

Previous Research

This section presents the various field from previous study on the human factor engineering or ergonomics that has been conducted by previous researchers. The reasons of conducting the study on previous research is to find the gaps and identify the way forward and recommendation on this study in order to have more meaning full study in mitigate the project risk especially on the safety management. Among the many researches and case study conducted on human factor engineering, below listed the most significant reference to find the gaps for this study. The details and gaps of the researches as below:

Human

Factor Engineering in Oil and Gas Construction Works

REFERENCE Tnflucnce of the Human Factor on the Risk of Work on Scaffolding, Katarzyna Szaniawskal, Krzysztof Czamocki2, Zbigniew Wisniewskil(&), and Malgorzata Wisniewskal, 2018

DETAIL To determine safety offhuman factors, we find it very important to find out where are points of visual concentration human factorand its lack of concentration on the ‘work area affect directly on the increase of risk during work that correlate with increaseof visual concentration on scaffold area is workload further research should be cartied in the direction of founding solutions to increase visual concentration ‘on scaffolding and decreasing a workload

The Sailport Project: A Trilateral Approach to the Improvement of Workers’ Safety ‘and Health in Ports, ‘Alessandro Pilippeschil, Mauro Pellicci2, Federico Vannil, Giulia Forte2, Giulia Bassanil , Lorenzo Landolfil, 22019 Robois and Human Interaction in a Fumiture Manufacturing Industry Risk Assessment , Ana Coliml(&), Susana Costal, André Cardoso2, 2018

The factors considered are musele effort level and its duration, the frequency of the action performed, hand and wrist posture, the velocity of the execution and the duration in the overall shift ‘The deterministic approach is based on the definition of a heuristic which associates the movement and position of objects to a risk level. Collisions are one of the main causes of dockers” injuries. Although there exist safety rules to reduce the collision risk, the number of accidents is still high. the tasks performed have been described as heavy and repetitive, involving frequent lifting, pushing and pulling of heavy loads, and the adoption of awkward static postures, like bending and twisting Since automated manufacturing systems are perceived to be efficient, automation is often viewed asa tool that can potentially enhance manufacturing competitiveness Topics covered included the following factors: workspace; physical activity in general; handling loads; postures and movements; risk of accident; repetitiveness of work; decision-making; lighting; thermal environment and; noise it is important to do a systematic risk assessment in order to anticipate all the factors that can harm the workers

How Ergonomics Is Contributing to Overall Equipment Effectiveness: ‘A Case Study, Mariana Rodrigues|(&), Isabel Loureiro2, Celina Pinto Leio2, and Nelson Costa2, 2018

Ergonomic workplace analysis was conducted comprising a generalist ergonomic study allowing the identification of the workstation that presented the ‘worst ergonomic situation, This research will contribute to raise awareness to the importance of the ergonomic aspects when designing. and organizing workplaces in order to contribute to the economic and social objectives of the organization. {greater exposure to critical situationsat an ergonomic level for workers who essentially perform their activities as factory workers interacting continuously ‘with machines Ergonomic improvements can contribute to the well‘being of workers. Increasing their motivation and consequently the productive performance. 6 changes were made: (#1) maximum weight to be handled in a box of raw material to supply the machine was established; (#2) a machine-operated button was centralized; (#3; #4) procedures were established to restriet the placement of boxes of raw material on the floor of two workstations and, (#5; #6) illuminance levels were changed in two ‘workstations ‘ergonomic improvements can enbance the improvement ofthe key performance indicators (KPI) as also the satisfaction of workers with the workplace.

597

GAPS research only focus on the human factor visually po relation with the procedure ‘and work method statement the study does not relate to the tools and equipment’s used for the work ‘The study only covers work at the building construction only rot the plant / oil & gas fac construction works No relation of the accident due to tools and equipment Study focus on the monitoring system to prevent the accident ‘The study has no relation to oil and gas construction

research only focus on the human interaction with robotic po relation with the procedure ‘and work method statement ‘The study only covers work at the building construction only not the plant /oil & gas facilities construction works

research only focus on the human factor with no factor on management intervention and inhouse procedure no relation with the procedure ‘and work method statement ‘The study only covers work at the building construction only not the plant / oil & gas facilities construction works

598

M. S.N. Arinee et al.

Role of Human Safety Intervention on the Impact of Safety Climate on Workers Safety Beha viours in Construction Projects: A Conceptual Model , Emmanuel B. Boateng] (&), Peter Davis2, and Manikam Pillay! , 2018

Workplace Safety Culture Model [WSCM] Presentation and Validation , Patricia Amélia Tomei, Giuseppe Maria Russo, 2019

Human Factors Engineering , James R. Lewis , 2011

© the rise in studies relating to safety behavior in the construction industry demonstrates its importance to construction safety management « safety climate has been used to mould the perceptions workers form about their organisations « the construction industry’ chronic level of fatalities, severe injuries, and ill health seems resilient to change © mechanisms required to considerably improve safety behaviourof construction workers have not been well captured, ‘There are three main safety interventions; management safety intervention denotes the top management strategies and safety managerial actions; technical safety intervention refers to any method that guarantees a safe working climate; and human safety intervention suggests methods to change ‘human understanding and reasoning, in view of safety practices that directly affect the employee « this study develops a conceptual model for measuring construction workers’ safety behaviours by incorporating human safety intervention practices, ‘group safety climate dimensions, and types of safety behaviors © Leadership Commitment, Pressure at Work, Infrastructure, Learning, Efficiency, Management System, Feedback, Responsibility, and ‘Communication * identified two distinct perspectives: the engineering approach, which focuses mainly on the formal aspects that influence business security (procedures, managerial systems, controls and policies), and a psychological approach, which focuses on the perceptions, feelings and attitudes of employees shortcoming of understanding the value of safety and its priority within the workplace, then unsafe behavior that leads to 80-90% of accidents # demonstrated that even employees with technical knowledge of WS sometimes show behaviors that are inconsistent with the safety standards required by ‘companies « that are considered natural, premises that govern the actions, behavior and reasons for the acts of the members of the company. * the way people think or behave in relation to shared values, attitudes, perceptions and beliefs with regard to safety and reflect a view whereby safety culture is something that characterizes a company, rather than something that it possesses basic assumptions are manifested in the shared * focusing on five common dimensions: management, safety systems, risk, work pressure, and competence beliefs of the membersof the company concerning ‘what is and what is not safe and acceptable risk behavior Do not require users to hold more information in ‘working memory than necessary, and then only for as short a period of time as possible * Parse longer alphanumeric strings into three- or four item units to enhance chunking of information Do not require users to perform several actions before allowing them to dump the contents of working memory * Do not require users to update working memory too Rapidly © body awareness senses such as the sense of balance ‘and awareness of limb location is also important. there are two way’ for the observer to be right: when the stimulus is present and the observer says it s (ie. ahit) and when the stimulus

research only focus on the human factor and intervention of the management strategies in the process and managing the safety no relation with the procedure ‘and work method statement The study only covers work at the building construction only not the plant / oil & gas facilities construction works

‘Only further studies can determine the conclusive stability of the WSCM, bearing in mind the academic support of diverse authors regarding the importance of certain dimensions, such as Leadership and Commitment. samples diversified by region in multicultural countries with lange geographic dimensions The study only covers work at the building construction only not the plant / oil & gas facilities construction works

research only focus on the human factor and intervention of the management strategies in the process and managing the safety no relation with the procedure and work method statement The study only covers work at the building construction only not the plant / oil & gas facilities construction works

Human

Factor Engineering in Oil and Gas Construction Works

599

* the most common typesof input devices are keyboards, buttons, mice, touchpads, touchscreens, pointing sticks, handwriting recognition, and speech recognition important to understand human capabilities of vision, touch, proprioception, motor control, and hearing, * Anthropometry is the branch of anthropology that deals with comparative measurements of the human body, used in engineering to achieve the maximum benefit and capability of products intended for human © Thus, human factors engineering draws upon such diverse disciplines as anthropometrics, biomechanics, industrial engineering, experimental psychology, cognitive psychology, social psychology, and organizational psychology, and applies them to the physical # design of equipment, such as input and output devices, and to interaction design

5

Conclusions

This case study gauge the factor that become the major cause of accident in oil & gas industry. The HFE in PSM among workers who involved directly and indirectly in the construction industries inclusive heavy industries has contribute significantly to the accident as reported. The case study has identify the factors based on ranking of the cause and factor for the accident. from the case study, the 5 key human factor contribute to the causation for the accident to happen are tools & equipment’s, organizational process, supervisory violations, design, and skill based error. It can be concluded that in the span of 7 years, the factor is the same and this has come to the same points that the awareness of the workers on the Human Factor Engineering importance is questionable. There is highly recommended that the detail evaluation on the factors. Thus further study to investigate the awareness of human factor engineering implementation among the workers involves is required. With the study, we will be able to identify the gaps among the workers involves and can provide gap closure programme such as training, awareness talks, framework and periodic assessment on work site in order to minimize the safety risk at the workplace. Acknowledgement. porting this study.

The

author would

like to thank

Universiti Teknologi

Petronas

for sup-

References 1. Othman, I., Majid, R., Mohamad, H., Shafiq, N., Napiah, M.: Variety of accident causes in

construction industry. In: MATEC Web of Conferences. vol. 203, p. 02006, EDP Sciences (2018) 2. Ayob, A., Shaari, A.A., Zaki, M.F.M., Munaaim, M.A.C.: Fatal occupational injuries in the Malaysian construction sector—causes and accidental agents. In: JOP Conference Series:

Earth and Environmental Science. vol. 140, p. 12095 (2018)

600

M. S.N. Arinee et al.

. Othman,

I, Ibrahim,

M.F.H.,

. Othman,

I., Mohamad,

Shafig, N., Mohamad,

H., Kamil,

M.S.: HSE

management

system for hot work operation at high elevation in shipbuilding project. MATEC Web Conf. 203, 02005 (2018) H.,

Sapari,

N., Shafig,

N.,

Ibrahim,

F., Kamil,

S., 2018.

Hse

management system at high elevation in shipbuilding project. Int. J. Eng. Technol. Manage. Res. 5(11), 117-127 (2018) . Sanders, MC.: Human Factor and Engineering design (1993) . Othman, 1, Mohamad, H., Napiah, M., Hashim, Z., Cai, Z.: The framework for effective safety control and implementation

at construction

project. Int. J. Eng. Technol.

Manage.

Res. 5(12), 28-42 (2018) . Kamal, LS

, Ahmad, I.N., Ma’arof, M.LN.: Review on accidents related to human factors

at construction site. In: Advanced Engineering Forum Publications Ltd (2013)

. Othman,

I,

Napiah,

M.,

Nuruddin,

M.F.,

Klufallah,

vol. 10, pp. 154-159 Trans Tech M.M.A.:

Effectiveness

management in oil and gas project. Appl. Mech. Mater. 815, 429-433 (2015)

of safety

. Fabrizio, G., Andrea, C., Vladimiro, C.: The Human Factor in Process Safety Management

(2012)

). http://www.constructionplusas .com/towards-world-cl: . Shabin, S1., Ramesh Babu, T.: A Study of Human Construction Industry (2014)

afety-standards/ (2019) Factors

and

Risk

Related

To

the

. Othman, I, Ghani, S.N., Mohamad, H., Alalou, W., Shafiq, N.: Early warning signs of project failure. In: MATEC

Web of Conferences. vol. 203, p. 02008 (2018)

. Leva, M.C., Naghdali, F., Alunni, C.C.: Human factors engineering in system design: a roadmap for improvement, Procedia CIRP, 38, 94-99 (2015) . ISO 6385, 2004, Ergonomic principles in the design of work systems, The International Standards Organisation, Geneva, Switzerland (2004) . ISO 11064, 2006, Ergonomic design of control centres,

The

International

Standards

Organisation, Geneva, Switzerland (2006) . ISO 12100, 2010, Safety of machinery - General principles for design - Risk assessment and risk reductio, The International Standards Organisation, Geneva, Switzerland..] . Maguire, M.: Methods to support human-centred design.International J. Hum.Comput. Stud.

55(4), 587-634 (2001) 18. 19. 20. 21.

TOSR Journal of Mechanical and Civil Engineering (IOSR-JMCE Othman, I., Shafig, N., Nuruddin, M.F.: Effective safety management in construction project.

IOP Conference Series: Materials Science and Engineering 291(1), 012018 (2017) International Conference on Recent Trends in Engineering and Management, Indra Ganesan College of Engineering Othman, I., Harahap, M.LP., Mohamad, H., Shafiq, N., Napiah, M.: Development of BIMbased safety management model focusing on safety rule violations. MATEC Web of Conferences 203, 02007 (2018)

22. Terezie, V., Vera, V., Vladimir, N.: The human factor as a cause of failures in building structures (2016) 23. Othman, I, Napiah, M., Nuruddin, M.F., Klufallah, M.M.A.: Effectiveness of preventive safety management in construction. engineering challenges for sustainable future. 24. 25.

155 (2016) Gambetti, F., Casalli, A., Chisari, V.: The huan factor in process safety management. Chem. Eng. Trans. 26, 279-284 (2012) Advance Engineering Forum. vol. 10 (2013)

Human Factor Engineering in Oil and Gas Construction Works 26.

Othman,

L:

Safety

management

practices

at

construction

site.

In:

Proceeding

601 3rd

International Conference on Environment (2010) 27.

Improved HFACS

on Human

Factors of Construction Accidents: A China Perspective Gui

Ye, (2018) 28.

Advances in Human Factors and Ergonomics. In: 10th International Conference on Applied Human Factors and Ergonomics and the Affiliated Conferences (2019)

® ‘upaates

Causes of Construction Accidents and the Provisions of Safety Regulations in Construction Industry in Malaysia Ku Adenan Ku Ismail’ and Idris Othman?°

' Maju Perak Holdings Berhad, No. 1-a, Jalan Meru Casuarina, 30020 Ipoh, Perak, Malaysia kuadenan@majuperak. com. my

2 University Teknologi PETRONAS, 32610, Seri Iskandar, Perak, Malaysia idrisothman@utp. edu. my

Abstract. In order to improve safety performance, Malaysian government has gazetted an occupational safety and health policies to provide guidelines to be followed in dealing with occupational safety and health activities at site. Meanwhile, Factories and Machinery Act provide regulation with respect to the health, health and welfare of person, machinery and for matters connected therewith for the control of factories. This paper aims to identify cause of accidents and the provisions of safety regulations in construction projects in

Malaysia. Methodology is by analysing the DOSH report. Statistic from DOSH shown that from 2015 to 2019, about 116% increase in accident cases. The main causes of accidents are work at high, unsafe work practice or conditions and failure of structure. Under Section 15 of OSHA 1994, contractors and

employers can be charged under this section if fail to provide a healthy and safe working environment for their employees. Keywords:

1

Construction - Accidents - Safety - Regulations

Introduction

Accidents are still high in the Malaysian construction industry. Department of DOSH indicated cases of construction accidents in Malaysia is a total of 1143 during the period of 2015 to 2019 as shown in Fig. 1. The cases of accidents include cases of permanent disability, death and non-permanent disability [1-3]. According to Rahim et al. [4], construction industry requires a huge overhaul from the aspect of site safety practic Rosli Ahmad [5] mentioned that safety programs can increase productivity, minimize construction costs, reducing injuries and could save lives of workers. Anmadon Bakri et al., [6], proposed that reducing the hazards at construction sites can improve the image of the construction industry. Providing a healthy and safe workplace is the most effective strategies for holding down the cost of the construction [7]. Accidents directly and indirectly incur additional and also cause delays in delivering projects [8, 9].

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 602-607, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_69

Causes of Construction Accidents and the Provisions of Safety Regulations

603

Numberof Accidents in Construction Projects in Malaysia 300

aE

250

237

oe

232

€

g 700

a7

2S 150 3s = 100 25 50 °

2015

2016

2017 Years

2018

2019

Fig. 1. Number of accidents in construction projects in Malaysia [1] According to Tan Chin Keng and Nadeera Abdul Razak (2014) [10], safety practices

at construction site can minimize the accidents. Among the practices are education, training, safety policy, safety auditing, site safety inspection, personal protective equipment, safety measuring devices and fall protective system [11, 12]. The main problem of safety practices is the attitude of the workers. Due to ignorance, Personal Protective Equipment (PPE) is not used. Stern actions again errant workers, contractors, employers and should be carried out continuously in order to minimize problems in safety practices [13].

According to Zhou et al. [14], penalties should be imposed on person in charge or the organization. Penalties can be in form of warning, confiscation of the income, charges of correction, charges to temporary detainment or to stop production. In order to improve safety and health at construction site, I Othman [15] recommended that safety impact assessment should be implemented systematically as part of safety management practices[15]. Implementation of laws under OSHA 1994 and FMA 1967 are approaches undertaken by government to address the safety issues in construction sites.

2

Review on OSHA

1994 and FMA

1967

OSHA 1994 and FMA 1967 are the current fundamental legislation which regulate construction activities in Malaysia. The main objective of OSHA 1994 and FMA 1967 are to ensure the welfare, health and safety for all employees and visitors and protection for the disable. In terms of securing the health, safety and welfare among the entire Malaysian workforces, OSHA 1994 can be considered as act that provides the legislative framework

for construction

site [16].

604

K. A. K. Ismail

and

I. Othman.

The self-regulation scheme that had been designed to coordinate with the particular organization or industrial firm are the basis of the establishment of this act. The effective health and safety performance is the main aims of this act.

3

Methodology on Analysing DOSH Accidents

Reports Construction

Figure 2 shows causes of construction accidents from 2010 to 2019 that have been reported by DOSH.

Causes of Construction Accidents (DOSH)

Not Wearing

Incorrect or No Working Procedure

Unsafe Work Practice or Conditions

Failure of

Structure

Working at Height Fig, 2. Causes of construction accidents [1]

The analysis shows that the biggest causes of construction accident (40%) is working at height. Then, followed by unsafe working conditions or practice (27%), failure of structure (18%), no work procedure (10%) and not wearing personal protective equipment (PPE) (5%). Therefore, the main causes of fatality accidents on

construction sites in Malaysia are employees falling from height or from the same level and accidents cause by fallen objects. The appropriate safety procedure such as evaluate the hazard by comprehensive tisk assessment, provide adequate training for all employees working at height, using appropriate personal protective equipment (PPE) and implementing safe working practice are actions can implemented to prevent fall from height or from the same level and fallen objects accidents cases

[1, 17].

Causes of Construction Accidents and the Provisions of Safety Regulations

4

Analysis on Legal Provisions Under OSHA 1967

605

1994 and FMA

Table | below shows the causes of construction accidents and legal provisions under OSHA 1994 and FMA 1967.

Table 1.

Causes of construction accidents and legal provisions

Cause of accident Working at high

Legal provisions Section 15 of OSHA 1994-Contractors and employers can be charged under Sect. 15 of OSHA

1994 if fail to provide

a healthy and safe working environment for their employees [18] FMA Regulations 1970 (Safety, Health and Welfare) [19] the employers to follow and adhere to Guidelines for the

Prevention of Falls, under Regulation 12 practicable means that the safety of person must be ensured for working more than 10 feet high [20] Unsafe work practice or conditions

Section 15(2) of OSHA 1994 - General duties of selfemployed persons and employers. The section stated the

duties of every employer and every self-employed persons [18]. Section 19 of OSHA 1994 stated that a person is liable to a fine if contravenes the provisions of this section

21) Failure of structure

Section 20(1) of OSHA 1994—General Duties of Suppliers, Manufacturers and Designers. The Section stated that any

person who supplies, designs, import or manufactures any plant for use at work, to ensure it is safe and without to health [22] A person who contravenes the provisions of this section shall be guilty and liable to a fine [23]

Incorrect or no work procedure _| Section 15(1) of OSHA 1994 - General duties of selfemployed persons and employers are to ensure the health, safety and welfare at work of all his employees [18] Section 19 of OSHA 1994 stated that a person is liable to a fine is contravenes the provisions of this section [21]

Not wearing personal

Section 24(1) of OSHA 1994-Every employee while at

protective equipment (PPE)

work to use or wear any protective equipment provided by the employer at all times in order to prevent risks to his health and safety [24]. A person is liable to a fine if contravenes to the provisions of this section [24]

606

5

K. A. K. Ismail

and

I. Othman.

Conclusion

The cases of accidents include death, permanent disability and non-permanent disability. The top three causes accidents are working at high, unsafe work practice or conditions and failure of structure. There are legal provisions under the OSHA 1994 and FMA 1967 Act that a safe and healthy working environment for employees are responsibilities of employers. Employers are subjected to a fine if fail to provide a safe and healthy working environment for their employees. Since the biggest cause of accidents is related to working at high, monitoring of projects involving workers working at high need to be further enhanced. Various efforts carried by government agencies, employers, contractors and consultants should incorporated by continuing safety improvement through enforcement of legislations, promotion, education and training, incentive, development and technology, occupational safety and health research.

References 1. DOSH: Statistic on occupational accidents (2020). http://www.dosh.gov.my/index.php/ statistic-v/occupational-accident-statistics-v

2. Othman, I., Majid, R., Mohamad, H., Shafiq, N., Napiah, M.: Variety of accident causes in construction industry. In: MATEC Web of Conferences, vol. 203, p. 02006 (2018) 3. Othman, [., Ibrahim, M.F.H., Shafiq, N., Mohamad, H., Kamil, M.S.: HSE management

system for hot work operation at high elevation in shipbuilding project. In: MATEC Web of Conferences, vol. 203, p. 02005 (2018) 4. Rahim, A., Hamid, A., Zaimi, M., Majid, A., Singh, B.: Causes of accidents at construction sites. Malaysian J. Civ. Eng. 20, 242-259 (2008) 5. Ahmad, A.: Best practices in safety management for conventional civil construction industry

in Malaysia. Master Thesis of Science Construction Management, Malaysia. Universiti Teknologi Malaysia (2008) 6. Bakri, A., Zin, R.M., Misnan, M.S., Mohammed, A.H.: occupational safety and health management system: towards development of safety and health culture. In: Proceeding of 6th Asia-Pacific Structural Engineering and Construction Conference (APSEC 2006), Kuala

Lumpur, Malaysia, pp. 19-28 (2006)

7. Othman, I., Mohamad, H., Napiah, M., Hashim, Z., Cai, Z.: The framework for effective safety control and implementation at construction project. Int. J. Eng. Technol. Manage. Res 5, 28-42 (2018). http://www. ijetmr.com

8. Othman, I, Mohamad, H., Sapari, N., Shafiq, N., Ibrahim, F., Kamil,S.: HSE management system at high elevation in shipbuilding project. Int. J. Eng. Technol. Manage. Res 5, 117— 127 (2018). http://www. ijetmr.com 9. Othman, L, Ghani, S.N., Mohamad, H., Alalou, W., Shafiq, N.: Early warning signs of 10.

project failure. In: MATEC Web of Conferences, vol. 203, p. 02008 (2018) Keng, T.C., Razak, N.A.: Case studies on the safety management at construction site. Journal of Sustainability Science and Management, vol. 9, no. 2 December 2014. Department of Quantity Surveying, Kulliyah of Architecture & Environmental Design,

International Islamic University Malaysia, Jalan Gombak, 53100 Kuala Lumpur, Malaysia (2014)

Causes of Construction Accidents and the Provisions of Safety Regulations

607

. Othman, L., Shafiq, N., Nuruddin, M-F.: Effective safety management in construction project. IOP Conf. Ser. Mater. Sci. Eng. 291(1), 012018 (2017) . Othman, I., Klufallah, M.M.A., Napiah, M., Nuruddin, M.F. Effectiveness of preventive safety management in construction. In: Engineering Challenges for Sustainable Future, vol.

155 (2016) . Othman,

1,

Napiah,

M.,

Nuruddin,

M.F.,

Klufallah,

M.M.A.:

Effectiveness

of safety

management in oil and gas project. Appl. Mech. Mater, 815, 429-433 (2015) . Zhou, Q., Fang, D., Mohamed, S.: Safety climate improvement: case study in a chinese construction company. J. Constr. Eng. Manage. 137, 86-95 (2011) . Othman, I.: Safety management practices at construction site. Proceeding 3rd International

Conference on Environment (2010) . DOSH: Occupational Safety and Health Act 1994. DOSH Malaysia, Kuala Lumpur (1994) . Othman, L., Harahap, M.LP., Mohamad, H., Shafig, N., Napiah, : Development of BIMbased safety management model focusing on safety rule violations. In: MATEC Web of Conferences, vol. 203, p. 02007 (2018) . Section 15 of OSHA 1994 Act 514 Occupational Safety and Health Act 1994 Incorporating all amendments up to 1, p. 16, January 2006 . DOSH 20. 21. 22. 23. 24.

1983 Factories and Machinery (safety, health and welfare) regulations 1970 (revised

1983) (Kuala Lumpur: DOSH Malaysia) p. 25 DOSH 2007 Guidelines for the Prevention of Falls at Workplaces Malaysia) p. 68 Section 19 of OSHA 1994 Act 514 Occupational Safety and Health all amendments up to 1, p. 18, January 2006 Section 20 of OSHA 1994 Act 514 Occupational Safety and Health all amendments up to 1, p. 18, January 2006 Section 23 of OSHA 1994 Act 514 Occupational Safety and Health all amendments up to 1, p. 21, January 2006 Section 24 of OSHA 1994 Act 514 Occupational Safety and Health all amendments up to 1, p. 21, January 2006

(Kuala Lumpur: DOSH Act 1994 Incorporating Act 1994 Incorporating Act 1994 Incorporating Act 1994 Incorporating

® ‘upaates

Development of Framework for BIM-Based Tools to Minimize the Causes of Accidents in Construction Idris Othman', Aminu Darda’uRafindadi'*°, Madzlan Napiah’, MiljanMiki¢?, Hayroman Ahmad’, Nura Shehu AliyuYaro', Balarabe Wada Isah', Ahmed Farouk Kineber', and Muhanad Kamil Buniya!

| Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS, Seri Iskandar, Perak DarulRidzuan, Malaysia {idris_othman, aminu_18002765, madzlan_napiah, nura_19001733, balarabe_16005697, ahmed_17008588, mohanad_18000491}@utp. edu. my ? Department of Engineering Management, Faculty of Engineering,

University of Leeds, Leeds, UK m.mikic@leeds. ac. uk

* Department of Civil Engineering, Bayero University, Kano, Nigeria

adrafindadi. civ@buk. edu. ng + Universiti Teknologi Mara, Bandar Baru, Seri Iskandar, Perak, Malaysia hayro724@uitm. edu. my Abstract. Despite increased efforts given to the field of safety management, the number of accidents is continuing to rise in the construction industry. Now with

the help of Building Information Modeling (BIM) technology, incident rates may dramatically decrease. However, there is little or no information available to be utilized by the safety officers and designers in the BIM-based tools for the

training and design works respectively in Malaysia. The main aim of this paper is to develop a framework to aid in giving comprehensive safety training to workers and to gather all the necessary information that can be used in BIMbased tools for the selected types of accidents at the design stage. The research

will be carried out by using the triangulation method. The findings of the research will guide safety officers in giving comprehensive safety training in a 3D virtual environment.

Keywords:

Building Information Modeling (BIM) « Malaysian construction

industry - Major types and causes of accidents - Safety training

1 11

Introduction Background of the Study

No construction project is accident-free. Accident-free construction sites help projects to be completed within the designed time frame, budget and with standard quality. Construction accident causes many human tragedies, contribute to a lack of motivation

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 608-619, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_70

Development of Framework for BIM-Based Tools

609

in construction workers, interrupt construction processes, delay development, and adversely change the cost, efficiency, and prestige of the industry [1]. Malaysia is a developing country with fast economic development. The risks of serious injuries to construction workers can be compelling reasons to expend investments on safety management to improve safety efficiency. The construction industry plays a significant part in satisfying people’s expectations and in addressing their basic needs [2]. While more focus has been given to safety management over the last decade, the number of injuries in the construction industry continues to increase relative to other industries [3]. In the construction industry, the probability of mortality is five times higher than in the manufacturing industry, while the likelihood of severe injury is two and a half times higher [4].

In Malaysia from 2010 to 2018 (see Table 1), it was reported that approximately every two working days there was a fatal accident. A lot of researches associated with construction safety declared that majority accidents on site can be reduced or prevented through proper and reliable planning approaches for safety management, workers’ training, and site inspection. Despite recent attempts to make the construction industry safer, the number of injuries in the construction sector keeps increasing [5]. Based on the experiences and observations of safety officers and project managers, possible safety risks are typically defined and then relay the message either pictorially or verbally to the workers through safety planning and training [6]. Safety planning is essentially coming together of relevant professionals who pinpoint potential risks by imagining construction processes using 2D drawings or pictures, schedules, safety rules and a vast array of experiences, but lack of an instinctive technique to depict the construction process [7-10]. It has been acknowledged that the construction industry needs to tackle the shortfalls of conventional manual safety procedures now in use [11].

The BIM-based tools can determine the virtual construction of a building or structure before its real physical development to reduce uncertainty, ensure safety, identify issues, simulate and analyze potential hazards [12]. However, the restricted implementation and insufficient safety benefits of BIM was due to the inadequate safety information contained in current information models

[13].

The main aim of this research is to develop a framework to aid in giving comprehensive safety training to construction workers. In view of the above-stated problems, a study is proposed to fill the gap by gathering or collecting empirical data from the relevant professionals in order to possibly bring everlasting solutions using whatever means possible and available. A safety management framework will be developed for the selected BIM-based tools using the most frequent types of accidents with the intention of mitigating their effects at the design stage through safety planning and safety training. The findings of the study will be tested utilizing real case studies of projects that are accessible and available in Malaysia and through semi-structured interviews. The intent of creating a framework is to drive efforts from research, improve relations with common understanding, and incorporate related ideas into a descriptive or predictive model

[14].

610

1. Othman et al.

Table 1. Reported number of fatal construction accidents in Malaysia (2010-2018) [15] Year | Non-fatal 2010)

54

2011)

54

2012)

55

2013) 2014)

83 94

2015 | 138 2016 | 135 2017 | 123 2018 | 106

1.2

Building Information Modeling (BIM) Adoption and Implementation in Malaysia

Latest developments

in technology

have allowed the Architecture, Engineering, and

Construction (AEC) sectors to keep up with the multifaceted world today [16]. The BIM concept was developed by Professor Charles M. Eastman in 1970 [13, 17, 18]; and AEC

industries began implementing it in the mid-millennium ades, BIM can be found everywhere

[17-20]. In the last two dec-

in the field of design and construction

[21]. The

United States of America became the first nation in the world to introduce BIM in the construction sector [22]. It was in 1997 that the U.S. industries became fully aware of the

importance

of BIM

technology

in construction,

with

the first version

of Industry

Foundation Classes (IFC) files [18]. The idea of putting BIM into practice in Malaysia

was brought in 2009 by the Director of Public Works Departments (PWD), Datuk Seri Prof. Judin Abdul Karim, who urged construction companies to adopt new technology to improve productivity and efficiency [23]. The introduction and deployment of BIM technologies is seen as the potential answers to the industry’s problems [24, 25]. Among these problems include the safety issues usually encounter at the design and construction stage. It was also adopted to make the construction industry of the country into a worldclass,

innovative

and

knowledgeable

sector of global

solutions

[26,

27].

It can

also

provide an efficient, effective, flexible and innovative system while providing national productivity towards contributing to the economic growth. However, the BIM implementation in Malaysia is still at the planning and design stage; and far behind compared to the developed countries [27, 28]. Since then, it has been gaining popularity in the sector. The government of Malaysia is the biggest property holder among the country and has mandated to implement BIM for all their projects by the year 2016 [29, 30]. Looking into the previous researches, the level of BIM adoption and implementation in the construction industry is very low most especially in developing countries [23,

24,

31]; considerable

attention

and effort is needed

to change

that. In a survey

conducted by Ibrahim, et al. [32], they found out that only 25.7% of the respondents were involved in BIM project while the remaining had none experience in Malaysia. The awareness of the existence of BIM technology amongst Malaysia’s contractors is moderately high, but only a minority of contractors had BIM experience [33]. The pace

Development of Framework for BIM-Based Tools

of adoption and implementation of BIM capacity,

user-friendliness

and

business

6ll

is still slow due to legal concerns, technology structure

[13,

34]. The

effects,

barriers

and

challenges of implementing BIM in Malaysian construction projects are similar to those of other countries

[24, 35]. Table 2 gives an overall idea of BIM

implementation and

adoption in Malaysia. BIM is generally adopted in the central region (i.e. Kuala Lumpur, Selangor, Negeri Sembilan, Putrajaya & Melaka) having 78%. This is due to the fast development and large scale construction projects going on in the area.

Table 2. Adoption and implementation of BIM technology by Region in Malaysia [33] Region % of BIM Implementation and Adoption Northern (Perlis, Kedah, Penang & Perak) 2% East Coast (Johor, Kelantan, Terengganu & Pahang) 6% Central (Selangor, Putrajaya, Melaka, N. Sembilan & KL) | 78% Sabah & Sarawak 6%

2 2.1.

Literature Review Occupational Safety and Health (OSH) in the Malaysian Construction Industry

Occupational safety and health (OSH) is an area concemed with improvement, advancement, and upkeep of the working environment condition, policies and programs that guarantee the psychological, physical, and emotional well-being of employees, in order to keep the work environment condition moderately free from genuine or potential dangers that could harm workers [36]. The 1994 Occupational Safety and Health Act (OSHA 1994) establishes the regulatory basis for the safety, health and wellbeing of all Malaysian workers, driven by the concept of avoidance and protection of employees from dangers and risks connected with their job activities [37]. The legislation was enacted based on a principle that “the burden for maintaining protection and health in the construction industry lies with those who generate the danger and those who operate with the danger.” 2.2

The Use of BIM in Safety Management

Building Information Modeling (BIM) is a 3D model-oriented platform that provides data and resources for AEC experts to plan, design, create and handle construction projects effectively [13]. BIM can also be defined as an advanced technology that can combine different types of construction information into a 3D digital model and be used in all stages of a project lifecycle [38]. This technology will be of great benefit to the AEC industry as a resource that can contribute to safety by scheduling, accident avoidance, progress tracking during construction, consistency in concept and simulation, data review, costing and better collaboration

among

team

members

[39]. It can

612

1. Othman et al.

also determine the virtual construction of a building or structure before its real physical development to reduce ambiguity, improve safety, identify problems and test and evaluate potential hazards

[12].

The idea to integrate safety within BIM technology was pioneered by Hinze and Wiegand [40], by surveying 35 design firms in the US. At that time, essentially designers used 2D at the design stage, the results indicated that only one-third of the respondents consider the contractor’s safety during design. In the continuation of the effort made by Hinze, Gambatese, et al. [41] built a computer program titled, “Design for Construction Safety Toolbox.” The intended purpose of the tool was to assist designers in defining project-specific hazards and incorporating design recommendations into the design ofa project. These researchers gave birth to the idea of ‘Prevention through Design (PtD)’ similarly known as ‘Design for Safety (DfS)’. Regardless of the fact that PtD has been around for almost two decades now, the usage of BIM-based safety approaches is comparatively a recent phenomenon and the research in this field is still at an early stage [21, 42]. This method may dramatically minimize construction injuries, as per numerous researches; but up to now, a vast majority of PtD tools are predominantly text-based, stand-alone checklists that sometimes do not involve BIM [43-45]. Changing the present safety planning approaches seems to be the main research topics [45]. Some of the previous researches regarding BIM-based tools implementation in safety management are briefly explained below with their weaknesses. Hongling, et al. [44] developed a BIM-safety rule-integrated approach that can automatically identify potential unsafe factors (e.g. slab edge, hole etc.) by combining Autodesk Revit and Unity 3D with the design safety rules. By combining the safety guidelines and the building information contained in the BIM database, hazardous construction variables can be defined and visualized in the model. However, this approach does not consider dynamic construction process and only very few safety rules for the selected scenarios were considered for the framework development. Zhang, et al. [46] also developed a mechanism that can detect exposed edges of slabs and holes, and automatically mount a guardrail system using BIM-based tools. The drawback of this study is that additional work is required using manual modeling whenever there are design changes. Although manual modeling offers the benefit of involving a person in every move, it is time wasting and highly vulnerable to error. Zhang, et al. [11] built a framework using Tekla Structures for the rule-based checking system for safety planning and simulation by intergrating BIM and safety. The developed framework could help in identifying possible fall related hazards like falling from slab or roof edges and openings. However, the proposed approach is not applicable to more complex construction conditions. Hossain, et al. [47] proposed an intelligent BIM-integrated risk management system that would aid designers in recognizing possible hazards coupled with design recommendations for eliminating or at least reducing the degree of danger during construction, operation and maintenance. However, very few hazards were incorporated into the knowledge library for the research. Li, et al. [48] aslo built an intellingent interlocking BIM-based Automated Safety Risk Recognition (SRIS) model in the pre-construction process and Safety Risk Early Warming (SREWS) coupling analysis in the construction phase of the Wuhan Metro construction in China. However, these approaches are focused on the deterministic interpretation of existing system and it is impossible to address all the variables and complexities of real engineering

Development of Framework for BIM-Based Tools

613

uncertainty. Kim, et al. [49] developed a computational algorithm that can automatically recognize possible safety risks associated with scaffolding by evaluating project details stored in BIM-based model and work schedules. The algorithms were introduced as a plug-in in commercially available BIM software and tested with a real-world construction project. However, some of the limitations of the study are as follows: not all potential hazards have been considered since only work spaces and scaffolding have been used as inputs; only supported scaffolds have been used in the research; and when considering more complex geometric building conditions, there needs to be an additional automation method to help users generate work paths based only on crew directions. New and modern approaches to safety training should be regarded, as conventional techniques might not be adequate for complex constrution situation [50]. Previous work has shown that BIM-based tools can be used to mitigate the impact of injuries on construction sites 2.3

Safety Training

Visualization technology offers a visual solution to safety learning, where construction and related events can be clearly identified and viewed in a 3D environment

[51, 52].

The digitization of the construction site permits virtual reviews and data-based investigation of development stages in order to identify the potential hazards that may arise at the early project stage and during construction [45]. Workers can easily view and identify potential hazards in a visual environment and thus quickly improve and understand safety training [8]. Integrating visualization technology with safety-related management rules and standards helps workers to better understand potential hazards throughout the life cycle of a project [7, 53]. Another challenge is the increasing number of foreign workers in multinational construction projects that require a lot of visual-based simple safety training approaches and strategies to better understand and identify safety risks [54].

3

Research Methodology

In this section, a brief description of the methodology to be adopted for this research is presented. The compilation of data in this study is highly necessary in order to accomplish the required objectives within the framework of the research and the allocated period. The research will be carried out by using the triangulation method. It involves the use of a survey questionnaire, semi-structured interview with the BIM experts, safety management experts and other relevant practitioners, use of case study accessible and available in Malaysia and virtual prototyping (VP).

614

4

I. Othman et al.

The Proposed Framework

The proposed framework for the comprehensive safety training in BIM-based tools is illustrated in Fig. 1. The steps for the framework development are briefly explained as follows:

eee

Problem Identification —p:

Inefficient safety

i

Types of accidents identification and ranking according severity andoccurrence

Data from the critical literature review

building using API or IFC

readable language

Process simulationusing Virtual Prototyping (VP)

————* _

Vividand comprekensive safety training for workers

Incomplete safety

|

Data engine ‘Output (Application)

Fig. 1. Basic framework design Step

1: Problem

identification

(see Table 3). It was discovered

that; most of the

incidents happened as a result of managerial failure or a lack of appropriate supervision or safety training or a combination of all according to the preliminary results from DOSH. Step 2: Determination of the type of accidents and their ranking according to their severity and the factors causing them through intensive literature review. Step 3: Collection and validation of relevant safety data through survey questionnaire, semi-structured interview, and national safety standard and best practices around the globe. Step 4: Then data conversion into machine-readable language (algorithm development). Compared to traditional methods, the proposed system will use current safety tules, standards, and best practices (e.g., DOSH Building Safety Guidelines or Best Practices in Construction Industry) and add them to the BIM model. Safety checking guidelines can be implemented either within a BIM-based framework (e.g. Revit, ArchiCAD, Tekla) or instead within a model-readable application (e.g., IFC apps, Navisworks).

Development of Framework for BIM-Based Tools Table 3.

Recorded number of investigated construction a

Rank | Types of accidents

Frequency Percentage (%) | Causes

1

108

38.16

2

| Struck-by

86

30.39

3

| Caught-inbetween

50

17.67

4

| Drowning/ Asphyxiation

40

9.89

v

Fall related hazards

O71

5

Electric shock

6

| Fire & Explosion

7

| Exposure to hazardous chemical substance

0.35

615

dents in Malaysia (2010-2018) [15]

Natural cause; no standard safe procedures for the job, lack of appropriate measures in place, inability to use and improper use of PPE, lack or inadequate PPE required for the job, dangerous working platforms when operating at high altitude, lack of monitoring while work at height, structural collapse, scaffolding installations not according to standard, inability to carry out risk analysis for the scaffolding dismantling procedure, fencing not mounted on the scaffold to avoid falling, communication gap between workers, and lack of safety training Lack of PPE for the workers, lack of engineering controls, machines not well inspected before usage, no safe working procedure for lifting works, failure of the crane’s operator to read load chart, failure to comply with mobile crane standard procedure for lifting works, failure of structure, no fence or barrier to prevent the general public from entering the premises of the construction sites, drunken personnel, lack of safety training and other environmental factors Lack safe work procedure, machine not well checked and inspected before usage, no safe working procedure for lifting works, lifting of excessive load, failure of the crane’s operator to read load chart, failure to comply with mobile crane standard procedure for lifting works, failure to comply with safe operating procedure, environmental factors, and lack of safety training Lack of safe working procedure in water and confined space, lack safe work procedure for piling works, lack or insufficient PPE for work in water, failure to conduct hazard identification, risk assessment, and risk control (HIRARAC) for water treatment plant tank cleaning works, and lack of safety training Carelessness, natural cause, lack of qualified personnel for the job and lack of safety training Lack of standard safe procedure for the job, lack or inadequate PPE required for the job, inadequate supervision and lack of safety training Lack of standard safe procedure for the job, lack or inadequate PPE required for the job, inadequate monitoring and lack of safety training

616

1. Othman et al.

Step 5: Virtual Prototyping (VP) is a very important tool in developing the VPbased safety management platform [7]. VP can model construction activities in a realistic way and collect design and construction information that can be replicated in future projects [55]. Virtual prototype also known as digital mock-up is a computer simulation of a physical product that can be presented, analyzed and tested on the basis of aspects of the product life cycle, such as civil engineering design, manufacturing, service and recycling, as if on a real physical model [56]. Virtual Prototyping (VP) is considered the construction and testing of a virtual prototype. One of the purposes of using VP is to detect or mitigate or eliminate the potential construction hazards via digital mockup of processes at the design stage and project planning [55]. According to Wang [56], a complete virtual prototype must basically comprise three types of models namely: 3D models, a human-product interface and models relating to perspective testing. Data obtained from the survey questionnaire, semi-structured interview and case study will be used for 3D models building and then the processing simulation using the available and accessible BIM-based tools to us. Step 6: The output is expected to be clear and comprehensive safety training module for workers.

5

The Preliminary Result

About 295 are number of fatal accidents that were actually recorded in the DOSH document but around 796 fatal accidents were actually reported from 2010 to 2018. Based on the number of recorded cases obtained from DOSH website the frequency of the types of accidents plaguing Malaysian construction industry were found according to severity (see Table 3). The types of identified accidents will be used as the basis for the development of framework for workers’ safety training in Malaysian construction sector with the aid of available BIM-based tools. The most common types of accidents in Malaysian construction industry were found to be: fall-related accidents, struck-by accidents, caught-in-between accidents, drowning & asphyxiation, electrocution, exposure to hazardous substance and fire & explosion in order of occurrence. The drawn table below gives the rank, accident type, frequency, percentage, and the causes of the identified types of accidents. Of the all the analyzed fatal accidents, the most common causes were found to be: lack of standard safe procedures, lack or inadequate PPE required for the job, lack of safety training and inadequate supervision or monitoring (see Table 3). However, emphasis will only be given to safety training for this research.

6

Conclusion

The preliminary data will be used for relevant safety data collection for the research and validation through survey questionnaire, case study and semi-structured interview. Data obtained from the survey questionnaire, semi-structured interview and case study will be used for impact analysis, algorithm development, 3D models building using either API or IFC and then the processing simulation using the available and accessible

Development of Framework for BIM-Based Tools

617

BIM-based tools to us. This research is still ongoing. The end results are expected to provide a standard framework for the BIM-based tools to be widely used in Malaysia for workers’ comprehensive safety training. It is expected to be completed by 2021. Acknowledgment. Universiti Teknologi Tel: +6053658000, Fax: +6053656716.

PETRONAS,

32610 Seri Iskandar, Perak, Malaysia,

Supporting Fund Organization. FRGS grant Ministry of Higher Education Malaysia References

v

. Bontis,

N., Chua

Chong

Keow,

W.,

Richardson,

S.:

Intellectual

capital

and

business

performance in Malaysian industries. J. Intell. Capital, 1(1), 85-100, (2000) . Kefela, G.T.: Knowledge-based economy and society has become a vital commodity to countries. Int. NGO J. 5(7), 160-166 (2010) . Zhou, Z., Goh, Y.M., Li, Q.: Overview and analysis of safety management studies in the construction industry. Safety Sci. 72, 337-350 (2015) . Davis, V., Tomasin, K.: Construction site safety. ed: Thomas Telford, London,

publication (1990)

. Fatal and Non-fatal a

Internal

cidents in construction industry (2007-2018)

. Carter, G., Smith, $.D.: Safety hazard identification on construction projects. J. Constr. Eng.

Manage. 132(2), 197-205 (2006) . Guo, H., Yu, Y., Skitmore, M.: Visualization technology-based construction safety management: A review. Autom. Constr. 73, 135-144 (2017) . Clevenger, C., Lopez Del Puerto, C., Glick, $.: Developing a BIM-enabled bilingual safety training module for the construction industry. In: Construction Research Congress 2014: Construction in a Global Network. pp. 1792-1800 (2014) . Lin, K.-Y., Tsai, M.-H., Gatti, U.C., Lin, J.J.-C., Lee, C.-H., Kang, S.-C.: A user-centered information and communication technology (ICT) tool to improve safety inspections.

Autom. Constr. 48, 53-63 (2014)

. Guo, H., Li, H., Li, V.: VP-based safety management in large-scale construction projects: A conceptual framework. Autom. Constr. 34, 16-24 (2013) . Zhang, S., Teizer, J, Lee, J-K., Eastman, C.M., Venugopal, M.: Building information modeling (BIM) and safety: automatic safety schedules. Autom. Constr. 29, 183-195 (2013)

checking

of construction

models

and

. Zhang, J., Hu, Z.: BIM-and 4D-based integrated solution of analysis and management for conflicts and structural safety problems during construction:

1. Principles and methodolo-

gies. Autom. Constr. 20(2), 155-166 (2011) . Eastman, C.M., Eastman, C., Teicholz, P., Sacks, R., Liston, K.: BIM handbook: A guide to

building information modeling for owners, managers, designers, engineers and contractors. John Wiley & Sons (2011) . PKirs, P.J., Sanders, G.L., Cerveny, R.P., Robey, D.: An experimental validation of the Gorry and Scott Morton framework. MiS Quarterly, pp. 183-197 (1989)

. M.o. H. R. Department of Occupational Safety and Health (DOSH), Malaysia. Department of

Occupational Safety and Health (DOSH), Malaysia Ministry of Human Resources, Malaysia. http://www.dosh. gov.my/index.php/component/contenVarticle/352-osh-info/accident-case/ 955-accident-case Accessed 2019 . Bryde, D., Broquetas, M., Volm, J.M.: The project benefits of building information modelling (BIM). Int. J. Project Manage. 31(7), 971-980 (2013)

618

1. Othman et al.

17. Wong, A.K., Wong, F.K., Nadeem, A.: Government roles in implementing building information

modelling

systems: Comparison

between Hong

Kong

and the United States.

Constr. Innov. 11(1), 61-76 (2011) 18.

ACCAsoftware. BIM adoption in USA: the first country to implement BIM is now falling behind in infrastructure technology Biblus. http://biblus software.com/en/bim-adoption-

in-usa-the-first-country-to-implement-bim-is-now-lagging-behind/

19.

Azhar, S., Behringer, A., Sattineni, A., Mqsood, T.: BIM for facilitating construction safety

planning and management at jobsites. In: Proceedings of the CIB-W099 International Conference: Modelling and Building Safety, Singapore. pp. 10-11 (2012) 20. Ward, R.: REBIM. How will Technology Impact the Construction Industry? https://rebim.io/ how-will-technology-impact-the-construction-industry/

21. Wetzel, E.M., Thabet, W.Y.: The use of a BIM-based framework to support safe facility management processes. Autom. Constr. 60, 12-24 (2015) 22.

N. Lo. B.S. (NIBS), United States, National Building Information Model Standard, Version

1 - Part 1: Overview, Principles, and Methodologies. https://buildinginformationmanagement. files. wordpress.com/201 1/06/nbimsv1_p1.pdf 23.

Latiffi, A.A., Mohd, S., Kasim, N., Fathi, M.S.: Building information modeling (BIM) ap-

plication in Malaysian construction industry. Int. J. Constr. Eng. Manage. 2(4A), 1-6 (2013) 24. Zahrizan, Z., Ali, N.M., Haron, A.T., Marshall-Ponting, A., Abd, Z.: Exploring the adoption of Building Information Modelling (BIM) in the Malaysian construction industry: A qualitative approach. Int. J. Constr. Eng. Manage. 2(8), 384-395, (2013) 25.

Ahmad Latiffi, A., Mohd, S., Brahim, J.: Building information modeling (BIM) roles in the Malaysian construction industry. Sustain. Solut. Struct. Eng. Constr. 749-754 (2014)

26. Wenegbuma, W.I., Ali, K.N.: Hypothesis analysis of building information modelling penetration in Malaysian construction industry. In: Proceedings of CIB World Building Congress. pp. 5-9 (2013) 27. Zahrizan, Z., Ali, N.M., Haron, A-T., Marshall-Ponting, A., Hamid, Z.A.: Exploring the barriers and driving factors in implementing building information modelling (BIM) in the Malaysian construction industry: A preliminary study. J. Instit.

Eng. Malaysia 75(1), 1-10

(2014) 28.

Latiffi, A.A., Brahim, J. and Fathi, M.S.: Transformation of Malaysian construction industry with building information modelling (BIM). In: MATEC Web of Conferences, 2016, vol.

66: EDP Sciences, p. 00022 (2016) 29. CIDB, Construction Industry Development Board Malaysia (CIDB). Kuala Lumpur (2014)

30. Mohammed, W.: Public Works Department (PWD), Malaysia, Public Works Department (PWD). Kuala Lumpur (2013)

31.

Memon, industry:

A.H., Rahman, L.A., Memon, I., Azman, N.I.A.: BIM in Malaysian construction status, advantages, barriers and strategies to enhance the implementation level. Res.

J. Appl. Sci. 32.

Eng. Technol. 8(5), 606-614 (2014)

Ibrahim, F.S., Shariff, N-D., Esa, M., Rahman, R.A.: The barriers factors and driving forces

for bim implementation in malaysian aec companies. J. Adv. Res. Dynam. Control Syst. 11, 275-284 (2019) 33. CIDB, C.LM.P.: Construction Industry Development Board Malaysia (CIDB). Kuala Lumpur (2016) 34.

Haron, A.T., Marshall-Ponting, A.J., Zakaria, Z., Nawi, M., Hamid, Z., Kamar, K.A.M.: An

industrial report on the Malaysian building information modelling (BIM) taskforce: issues and recommendations. Malaysian Constr. Res. J. (MCRJ) 17, 21-36 (2015) 35.

Mohd-Nor, M., Grant, M.P.: Building information modelling (BIM) in the malaysian architecture industry. WSEAS Trans. Envir. Develop. 10, 264-273 (2014)

Development of Framework for BIM-Based Tools

619

. Nyirendaavwil, V., Chinniah, Y., Agard, B.: Identifying key factors for an occupational health 37.

and

safety

risk

38. Cerovsek, T.: A review 39.

estimation

tool

in

small

and

medium-size

enterprises.

IFAC-

PapersOnLine 48(3), 541-546 (2015) Rampal, K.G., Nizam, J.M.: Developing regulations for occupational exposures to health hazards in Malaysia. Reg. Toxicol. Pharmacol. 46(2), 131-135 (2006) and outlook for a ‘Building Information Model’(BIM):

A multi-

standpoint framework for technological development. Adv. Eng. Inf. 25(2), 224-244 (2011) Volk, R., Stengel, J., Schultmann, F.: Building Information Modeling (BIM) for existing buildings—Literature review and future needs. Autom. Constr. 38, 109-127 (2014)

40. 41.

Hinze, J., Wiegand, F.: Role of designers in construction worker safety. J. Constr. Eng. Manage. 118(4), 677-684 (1992) Gambatese, J.A., Hinze, J.W., Haas, C.T.: Tool to design for construction worker safety.

J. Arch. Eng. 3(1), 32-41 (1997) 42. Toole, T.M., Gambatese, J.: The trajectories of prevention through design in construction.

J. Safety Res. 39(2), 225-230 (2008) 43.

Ka, K., Mills, T.: Research needs for building information modeling for construction safety. In: International Proceedings of Associated Schools of Construction 45nd Annual

Conference, Boston, M, (2010)

44. Hongling, G., Yantao, Y., Weisheng, Z., Yan, L.: BIM and safety rules based automated identification of unsafe design factors in construction. Procedia Eng. 164, 467-472 (2016) Getuli, V., Ventura, S.M., Capone, P., Ciribini, A.L.: BIM-based code checking for construction health and safety. Procedia Eng. 196, 454-461 (2017) 46. Zhang, S., Sulankivi, K., Kiviniemi, M., Romo, I, Eastman, C.M., Teizer, J.: BIM-based fall 45,

47.

hazard identification and prevention in construction safety planning. Safety Sci. 72, 31-45 (2015) Hossain, M.A., Abbott, E.L., Chua, D.K., Nguyen, T.Q., Goh, Y.M.: Design-for-safety knowledge library for BIM-integrated safety risk reviews. Autom. Constr. 94, 290-302 (2018)

Li, M., Yu, H., Jin, H., Liu, P.: Methodologies of safety risk control for China’s metro construction based on BIM. Safety Sci. 110, 418-426 (2018) 49, Kim, K., Cho, Y., Zhang, S.: Integrating work sequences and temporary structures into 48.

50.

safety planning: automated scaffolding-related safety hazard identification and prevention in BIM. Autom. Constr. 70, 128-142 (2016) Haron, A.T.: Organisational readiness to implement building information modelling: a

framework for design consultants in Malysia. University of Salford (2013) Sl. Teizer, J., Cheng, T., Fang, Y.: Location tracking and data visualization technology to advance construction ironworkers’ education and training in safety and productivity. Autom. 52.

Constr. 35, 53-68 (2013) Inyang, N., Han, S., Al-Hussein, M., El-Rich, M.: A VR model of ergonomics and productivity assessment in panelized construction production line. In: Construction Research

Congress. pp. 1084-1093 (2012)

53. Zhou, W., Whyte, J., Sacks, R.: Construction safety and digital design: a review. Autom.

Constr. 22, 102-111 (2012) 54. 55. 56.

Kim, Y.-J., Park, K.-H., Kim, J.-J.: A study on effect analysis that employment permit system affects construction field. Korean J. Constr. Eng. Manage. 6(3), 61-71 (2005) Li, H., et al.: Integrating design and construction through virtual prototyping. Autom. Constr.

17(8), 915-922 (2008) Wang, G.G.: Definition and review of virtual prototyping. J. Comput. Sci. Eng. 2(3), 232236 (2002)

® ‘upaates

Factors of Safety Misconduct Affecting Safety Performance of Tall Building Construction Site Varunesh Thinakaran! and Idris Othman?®

' Aspen House Aspen Vision Development Sdn Bhd, 300, Jalan Macalister, George Town, 10450 George Town, Pulau Pinang, Malaysia varunesh. thinakaran@aspen. com. my

> Department of Civil and Environmental Engineering, University Technology, PETRONAS, 32610 Seri Iskandar, Perak, Malaysia idris_othman@utp. edu. my

Abstract. This is a review paper to analyze factor of safety affecting the construction industry which is mainly involved in building projects all over the world. Every country has their own unique tall building or known as skyscrapers is constructed to elevate the country pride. On the other hand, tall building save

space and accommodate more residents as compared to low rise buildings. However, vast problems encountered by construction team to ensure the safety performance of the construction site is well maintained. In all over the world, we can observe that there are many accidents involve during building construction especially due to safety misconduct. Those accidents affect directly on the

project performance of works delay and financial constraint due non-compliance of safety procedure. A definitive objective for the construction industry is to lessen workplace accidents, wounds, and fatality to zero. The less accident there are, the more popular the construction company name will be, this popularity can contribute to the company revenue whereby client favours to award the project to the construction company which has a good reputation and wellrounded in terms of safety management as well. In this study, the factor of safety misconduct that affecting safety performance of tall building construction site

will be identified through literature review and questionnaire survey. The factors will be validated as well using the case studies project. The statistical strategies

including Relative Importance Index (RID) and Average Index (AVI) will be used to break down the information accumulated, while the statistical package

for the social science (SPSS) will be used to gauge the Spearman’s rank correlation between different gatherings of respondents, the Cronbach’s alpha (reliability test) and legitimacy of the study. The sub factors, which were grouped into five main groups according to past studies namely:

1) worker involvement

2) Management 3) Material and Equipment 4) workplace & 5) Software and Technology.

Keywords: Safety management - Tall building - Construction projects « Safety misconduct

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 620-630, 2021. https://doi.org/10.1007/978-98 1-33-631 1-3_71

Factors of Safety Misconduct Affecting Safety Performance

1

621

Introduction

Safety is very important in all field of industry. Safety is very crucial in the construction industry. It has dependably been a noteworthy issue as it is considered as among the most exposed areas with regards to occupational accidents. Albeit colossal upgrades have been made in safety execution in a few countries, the construction industry keeps on lingering behind most different enterprises. Roughly 80% of all accidents at the building construction site are brought about by dangerous human practices, with most of fatalities being because of struck by objects, falling from height or being struck explicitly by moving vehicles. Amount of accident happen at tall building construction is much higher compared to low rise building. Therefore, as the building gets higher and higher like a skyscraper, the wind force will create an unfavorable working condition. Tower crane at such height exposed to wind force that can hit up to 140 km/h depends on the region. Proper safety measures have to be taken into consideration especially for tall buildings. Labors should be equipped with extra safety care to prevent any fatality. According to Malaysian 25th productivity report, at the period of 2011-2017, the construction sector productivity growth were mainly driven by MFP (Table 1).The multifactor productivity(MFP) growth in the construction sector was a great extend due to intensive demand resulting from mega construction projects nationwide.

2

Malaysian GDP from Construction

Past studies show that in an investigation directed to dissect the connection between, quality and safety was found that a task with a low-quality execution had a higher probability of injuries and fatality. (Navi, et al. 2017). The outcome from another study on construction- safety best practices furthermore, connections to wellbeing execution demonstrated that safety execution is improved as the quantity of safety implementation expanded (Choe, et al. 2017). Clearly, it is imperative that so as to improve the physical condition, safety risk evaluation and representative’s information, and the association must receive an all-encompassing technique that shoulder on the above components. What increasingly imperative is to cultivate a solid safety society in construction through representatives’ convictions and frames of mind prompting safe conduct (Zou 2011).

In Malaysia, Department of Occupational Safety and Health (DOSH) is a department under the Ministry of Human Resources. This division is in charge of guaranteeing the safety, health and welfare of individuals at work just as shielding other individuals from the safety and health risks emerging from the exercises parts which includes manufacturing, construction, mining & quarrying, public services and etc. Figure 2 shows the Occupational Accident Statistics by Sector updated every 2 months. The latest updated was 2/11/2018. Non-permanent disability (NPD) is the highest for

622

V. Thinakaran

and

I. Othman

manufacturing sector as compared to other sectors. Construction sector has high number of deaths which is 81 victims whereby manufacturing sector has 25 death victims, from here we can conclude that the number of death victims in construction is three times higher than manufacturing sector. Therefore, the statistical result proves that safety management is highly crucial for construction site.

3

3.1

Factors of Safety Misconduct Which Affect the Safety Performance Attitude and Behavior

Past investigations demonstrated that a standout amongst the most significant variables impacting development wellbeing is frame of mind and conduct. The audited papers presented diverse factors. Fifteen sub-factors were extricated from substance investigation, including director’s conduct, boss’ disposition, director viability, laborer’s conduct, specialist’s disposition, saw conduct control, conduct criticism, investment for security improvement (specialists’ association, psychological and enthusiastic commitment), security exertion, moral duty regarding wellbeing, risk taking outlook/conduct, passionate state, chance discernment, saw wellbeing state, and security consistence. In the first place, the word related mishap event is profoundly related to the laborer’s dangerous demonstration (Awwad

et al. 2016).

However, such acts are not only an impression of laborers’ individual unfortunate behavior since Fang also, Wu (2015) found that administration dangerous practices play a significant job in specialists’ risky practices. In a refinement between working operators and working conditions, one should include that risky demonstrations of laborers become the real underlying drivers of development mishaps when they are joined with hazardous working conditions on a development site (Chen et al. 2017). On an evidential premise, be that as it may, the laborers’ practices associated with wellbeing preparing, procedures, and projects have been acknowledged as the overwhelming factors causing work site mishaps as of late (Shin et al. 2015). As an answer for the issues brought about by such factors, X. Wu

et al. (2015) demonstrated that the

great passionate state is imperative to keep representatives to work more successfully. 3.2

Safety Programs and Management Systems

Fifteen sub-factors were separated from substance examination, including security programs, wellbeing the board frameworks, correspondence and data, restricted administration time, the board responsibility, chance appraisal execution/exhaustiveness, wellbeing approaches and systems, safety gatherings/association/groups/supervisors, the board work weight, pre-procure screening of representatives, the executives concentrate on wellbeing, the board concerm/association, security directions, wellbeing control instruments, safety the board rehearses what’s more, aptitudes. The presence of a strong

Factors of Safety Misconduct Affecting Safety Performance

623

workplace, in any case, compares to the workers’ anxiety for the security of themselves and their collaborators so that it develops a decent connection between them (Ab Rahman. 2015). It is evident that, at the administrative dimension, view of the chief’s wellbeing

mentality and work practices straightforwardly influence specialists’ practices (Raheem and

Issa 2006;

Grill

and

Nielsen

2019).

Zhou

et al. 2015

pronounced

that

security

execution is particularly solid when top administration is noticeably associated with security. Top-level directors ought to think about security as significant as different parts of the association, for example, creation in request to touch base at an entrenched security culture. The executives’ responsibility to safety not exclusively is a genuine case for specialists, yet in addition one of the most significant variables of safe work conduct among representatives in the development business (Jiang et al. 2014; Stoilkovska et al. 2015). 3.3.

Work

Pressure

The work weight factor incorporates creation weight, work over-burden, exhaustion and burnout, working pace, working time, additional time work, what’s more, plan delay. Different weights that exist nearby will in general undermine site safety (Guo et al. 2015). Usually development ventures are under impressive generation weight on account

of calendar delay

(Guo et al. 2015; Chen et al., 2017). The job that the last

plays in the expansion of work weight can be clarified all the more unmistakably with respect to the investigation of Han et al. (2014) underscoring that plan delay is one of the huge factors influencing injury event on the undertakings. Under generation weight, the administrative need may not be given to safety (Chen et al. 2017), and generation is bound to win the “fight” against security (Guo et al. 2015). This generation weight may thusly cause chiefs and bosses to put less time and vitality in security furthermore, even urge specialists to take easy routes to meet generation calendars to stay aware of the effectively postponed timetable objectives. 3.4

Work

Condition

It is all around acknowledged that word related injury are associated with either the character of the work frameworks as well as conditions or the person laborers’ mental and conduct qualities (Shin et al. 2015). Most development exercises happen in quickly evolving situations and under developing site conditions. Hence, this class covers an allinclusive scope of sub-factors including workplace, introduction to risk/perilous work circumstance, venture danger level, furthermore, working environment safety and security states of specialists. Security administrators can control explicit hazardous demonstrations of specialists by disposing of the related dangerous working conditions and the other way around (Zhou et al. 2015). In this light, notwithstanding, Taufek et al. 2016 finished up that the word related injury event is exceedingly identified with the

624

V. Thinakaran

and

I. Othman

hazardous condition. Higher venture peril level, as it were, will in general be related with higher hazard level nearby (Feng et al. 2014). To take a gander at this issue all the more methodically, one can make reference to Grill and Nielsen 2019 who brought up that, when all is said in done, there are three difficulties to the support of safe situations on building destinations. In the first place, safety is hard to quantify, as getting a protected site relies upon emotional judgment subject to one’s close to home meaning of safety. Second, human blunder isn’t controllable, and people must be accused for carelessness and controllable conditions inside their obligations. 3.5

Organization

The association factor contains organization’s income, organization notoriety, organization’s costs, organization measure, customer’s control, the association of subcontractors, number of subcontractors, and number of employee’s/team measure. There is a substantial volume of past related investigations portraying the general however conclusive job of the association on the security execution of the development ventures. For instance, Choe

and Leite

(2017) announced

that higher tumover rates are related

with the higher damage rates. Likewise, development in organization measure observed to be related with improved wellbeing execution (Choe and Leite 2017). By methods for outline, Muhammadi et al. (2018) presumed that the normal number of mishaps and

cost of mishaps is directly related with the normal number of subcontractors and the allout number of laborers, separately. On the other hand, look into shows that little development organizations altogether influence the advancement of security on development locales, since they have higher mishap rates contrasted and extensive development organizations (Yiu et al. 2018; Ozmec et al. 2015). At the point when thought about, the littler development organizations recommend the likelihood that this inquisitive higher rate of worksite mishaps may predominantly trait to less sufficient administration aptitudes, less capacity to actualize wellbeing and wellbeing works, deficient specialist security mindfulness, disappointment in labor wellbeing laws and guidelines consistence, and deficient security

4

insurance measures

and offices (Yiu et al. 2018).

Gap Analysis

Gap analysis as tabulated in Table 1 below.

Factors of Safety Misconduct Affecting Safety Performance

625

Table 1. Gap analysis on factors of safety misconduct affecting safety performance FACTOR

AUTHORS

Sounders et al. 2017) Motivation

Chen etal

(2015) Mohammadi

WORKER INVOLVEMENT

et al, (2018) Choe (2017)

Rules and pean eautatio

Rahim & Esha (2016) Chen et al ‘Aba“ Rahman et al, (2015) Babalola etal. 2018) Guo et al. 2015)

KEY STATEMENT In the event that a labor gains lower compensation, it makes the profession unattractive to him/her. This is on the grounds that utilizing harsh penalties to consider specialists in charge of security infringement may cause dread inthe work

environment. Peer

weight is the impact of a

companion group : that can change group individuals’ safety attitude or beha Higher dimensions of inspiration towards safety can be strengthened by the level of their cooperation in safety elated exercises Clear desires on occupation site security standards ought to be given by successfl security programs. Furthermore, compelling correspondence between site the executives and specialists should be set up to * support laborers comprehend safety rules for better consistence A. prohibitive strategy including increasingly inspection reviews, higher fines for resistance, and higher accident expenses can prompt superior usage of the safety norms.

GAP How to garantee that the site worker ccan have positive gathering standards?

How to develop

the enthusiasm of laborers to take an interest insafety program?

How to improve effective communication between site management and ‘ site workers? What are the alternatives safety standards that ean be implemented at site?

626

V. Thinakaran

and

FACTOR

I. Othman

AUTHORS

KEY STATEMENT Leaming through preparing builds the skill of administrators and has @ positive impact on their safety

+ Taek al. 2016)et

Tite su ervisor is skilled in pre work aanging, ower dose have the chanceee to give contribution because of time constraint, the learning s won't be compellin, The laborers utilizedpenn.by small development ventures normally have poor capabilities, which may cause the laborers deficient! y mindful of safety perils bringing about a higher rate of word related injury event. The word related accident event is

© HSE

Competency 55 a 2Sg 2

Fang et al © (2015) Choe& Leite . (2017) © Yivet al, 018) © Misiurek & Misiurek 017)

& =

GAP

How to ensure

thecompany construction improves the HSE ‘competency of the: site workers in order to mitigate accidents theat the construction site?:

profoundly identified with the labor's age and type of work carry

out by the labors.

Yearly allotments ought to be made

. Safety Investment and Cost

+ |» ©

Raheem and Issa (2016) Zhou etal, 015) Saurin

2016),

in the financial plan for the buy and support of defensive equipment, compensations of safety staff and different parts ofthe safety programs, Safety frameworks are advantageous

speculations with vital ramifications,

as the experience on safety ‘management into a basie es that maycantumturn apparatus altogetherinto bas improve safety execution.

Wi hat are the effective vecntation to implementation safety management system which cten can an ant ae ee. How tones sat ieee

aie construction ee vor

Factors of Safety Misconduct Affecting Safety Performance FACTOR

AUTHORS

KEY STATEMENT Rework is a basic factor that influences tnderakings injury event on the The task cost is diminished regularly

RESOURCES & FINANCIAL ASPECTS

Financial| aspeets and Productivity

Muhammad et al, (2018) Chen et al. 017) Littlewood et‘ al (2017) Guo et al. ars)

by cutting safety spending plan, By and lage, safty endenors ate decreasedes as salety safety spending cuts, thus the higher danger of exposure to work related harms and injury in long haul which will unavoidably in the end more costly than the merdemonstrate cost adaments Researchers underlines the way that

numerous development injury are brought about by insufficiencies in the development and design period

project. ofthe the projec

‘Absence of arrangement of protection equipment gear can assume a noteworthy job in the expansion of work wounds, Employer did not give Personal

Resources " and Equipment

. Lietal, 01s) , Ghodrati et al. (2018) Chen, et 017) Vin (2018)et al.

Protection Equipment (PPE) to the

laborers, PPEs were definitely not accurately utilized, and when laborers neglected to receive protection or disregarded danger cautioning signs inn the work environment. , 1e arrangement of such protective equipment alone can't ensure the ‘quip

decrease of site injury because of the way that a few development injury result since, first, the safety equipment important to play out the activity securely is absent at the area of the work.

627

GAP How to reduce rework in project which effect on safety formance? low Teredace the deficiencies in the development and design period which can cause 1 construction accidents? How project th hin ‘ewes e

influence the safety?

How wo Tee th culavate awareness thein employers regarding the importance of providing protection equipment for the employees?

628

V. Thinakaran

and

FACTOR

AUTHORS

© Client interference

© ©

+ Change Order

5

I. Othman

+ ©

Taufek et al. (2016) Alarcon et al. (2016) Yiu et al. (2018)

Rahim & Esha (2016) Chen etal, 2017) Torner & Pousette (2015)

KEY STATEM! Absence of experience of client/developer in development venture A portion ofthe client/developer command contractual — worker to complete extra work without ask consultant A client who did not ask consultant advice before direct instruct the contractors may represent a high hazard because ofa lack of consultant advice in constructing something without considering the knowledge of engineering. Elimination or decrease of dangers can be connected through design or elective techniques for development. Construction design and development — process are interlinked, with the procedure being directed by the plan and choices from the design group. Many architects are as yet neglecting to recognize their impact on the safety of the development procedure because of profound situated custom and practice and absent of safety instruction and preparing

GAP

How to limit interference from the client/developer to prevent delay in schedule of a project?

What initiatives, required in limiting the change order in the projects?

Conclusion

The high-risk places in construction is worksites. There are many safety cases reported at construction sites. This study will investigate how management ensure that their sites are safer by identifying factors. Sites accident can be minimized by monitoring all these factors closely. The findings from this study can be a guide for all construction practitioners to reduce site safety cases and accidents. All factors from various aspects such as workers, employers, government, contractors and consultants will be identified.

References

‘Ab Rahman, R.: Managing safety at work issues in construction works in malaysia: a proposal for legislative reform. Modern Appl. Sci. 9(13), 108 (2015) Akcay, C., Aslan, S., Sayin, B., Manisa, E.: Estimating OHS costs of building construction projects based on mathematical methods. Safety Sci. 109, 361-367

(2018)

Factors of Safety Misconduct Affecting Safety Performance

629

Alarcon, L-F., Acufia, D., Diethelm, S., Pellicer, E.: Strategies for improving safety performance in construction firms. Acc. Anal. Prevent. 94, 107-118 (2016) Alhajeri, M.: Health and Safety in the Construction Industry: Challenges and Solutions in the UAE (Doctoral dissertation, Coventry University) (2011)

Awwad, R., El Souki, O., Jabbour, M.: Construction safety prac eastern developing country. Safety Sci. 83, 1-11 (2016) Babalola,

O., Ibem,

E.O.,

industry: A systematic

Ezema,

L.C.: Implementation

ces and challenges in a middle

of lean practices in the construction

review. Building and Environment (2018)

Chen, Y.: Factors Affecting Safety Performance of Construction Workers: Safety Interpersonal Conflicts at Work, and Resilience (Doctoral dissertation) (2017)

Climate,

Chen, Y., McCabe, B., Hyatt, D.: Impact of individual resilience and safety climate on safety performance nd psychological

stress of construction workers:

a case study of the Ontario

construction industry. J. Safety Res. 61, 167-176 (2017) Chen, Y., McCabe, B., Hyatt, D.: A resilience safety climate model predicting construction safety performance. Safety Sci. 109, 434-445 (2018) Choe, S., Leite, F.: Construction safety planning: Site-specific temporal and spatial information

integration. Autom. Constr. 84, 335-344 (2017) Ghodrati, N., Yiu, T.W., Wilkinson, S.: Unintended consequences of management strategies for improving labor productivity in construction industry. J. Safety Res. 67, 107-116 (2018) Grill, M., Nielsen, K.: Promoting and impeding safety—A qualitative study into direct and indirect safety leadership practices of constructions site managers. Safety Sci. 114, 148-159 (2019) Guo, H., Yu, Y., Skitmore, M.: Visualization technology-based construction safety management:

A review. Autom. Constr. 73, 135-144 (2017) HSE (Health and Safety Executive). (2009). “Phase 1 report: Underlying causes of construction fatal

accidents—A

comprehensive

review

of recent

work

to consolidate

and

summarise

existing knowledge. http://www.hse.gov.uk/construction/resources/phasel.pdf (Jul. 2, 2015) Li, H., Lu, M., Hsu, S.C., Gray, M., Huang, T.: Proactive behavior-based safety management for construction safety improvement.

Safety Sci. 75, 107-117 (2015)

Littlewood, J.R., Alam, M., Goodhew, S., Davies, G.: The ‘Safety Gap’ in buildings: Perceptions of Welsh Fire Safety Professionals. Energy Procedia 134, 787-796 (2017) Loosemore, M., Malouf, N.: Safety training and positive safety attitude formation Australian construction industry. Safety Sci. 113, 233-243 (2019)

in the

Misiurek, K., Misiurek, B.: Methodology of improving occupational safety in the construction industry on the basis of the TWI program. Safety Sci. 92, 225-231 (2017) Mohammadi, A., Tavakolan, M., Khosravi, Y.: Developing safety archetypes of construction industry at project level using system dynamics. J. Safety Res. 67, 17-26 (2018) Nawi,

M.N.M.,

Ibrahim,

S.H., Affandi, R., Rosli, N.A., Basri,

F.M.: Factor affecting safety

performance construction industry. Int. Rev. Manage. Market. 6(8S), 280-285 (2017) Ostman, B., Mikkola, E., Stein, R., Frangi, A., Konig, J., Dhima, D., Bregulla, J.: Fire safety in timber buildings, p. 19. Technical guideline for Europe, SP (2010) Oswald, D., Sherratt, F., Smith, S.: Problems with safety observation reporting: a construction,

industry case study. Safety Sci. 107, 35-45 (2018) Othman, [., Majid, R., Mohamad, H., Shafiq, N., Napiah, M.: Variety of accident causes in construction industry. MATEC

Web of Conf. 203, 02006 (2018a)

Othman, L, Ibrahim, M.F.H., Shafig, N., Mohamad, H., Kamil, M.S.: HSE management system for hot work operation at high elevation in shipbuilding project. MATEC Web of Conf. 203, 02005 (2018b) Othman, L, Harahap, M.LP., Mohamad, H., Shafiq, N., Napiah, M.: Development of BIM-based safety management model focusing on safety rule violations. MATEC Web of Conf. 203, 02007 (2018c)

630

V. Thinakaran

and

I. Othman

Othman, I., Shafiq, N., Nuruddin, M.F.: Effective safety management in construction project. IOP

Conf. Series: Mater. Sci. Eng. 291(1), 012018 (2017) Othman, I, Napiah, M., Nuruddin, M.F., Klufallah, M.M.A.: Effectiveness of preventive safety management in construction. engineering challenges for sustainable future. 155 (2016) Othman, I., Napiah, M., Nuruddin, M.F., Klufallah, M.M.A.: Effectiveness of safety management in oil and gas project. Appl. Mech. Mater. 815, 429-433 (2015) Othman,

I.: Safety management practices at construction site. In: Proceeding 3rd International

Conference on Environment (2010) Othman, I., Mohamad, H., Napiah, M., Hashim, Z., Cai, Z.: The framework for effective safety control and implementation at construction project. Int. J. Eng. Technol. Manage. Res. 5(12),

28-42 (2018) Othman, I., Mohamad, H., Sapari, N., Shafiq, N., Ibrahim, F., Kamil, $.: HSE management system at high elevation in shipbuilding project. Int. J. Eng. Technol. Manage. Res. 5(11), 117-127 (2018)

Othman, I., Harahap, M.LP., Mohamad, H., Shafiq, N., Napiah, M.: Development of BIM-Based Safety Management Model Focusing on Safety Rule Violations. In: MATEC Web of Conferences. vol. 203, p. 02007. EDP Sciences (2018) Phoya, Health and safety risk management on building constuction sites in Tanzania:

the

practice of risk assessment, Communication and Control (2012) Poh, C.Q., Ubeynarayana, C.U., Goh, Y.M.: Safety leading indicators for construction sites: A machine learning approach. Autom. Constr. 93, 375-386 (2018) Raheem, A.A., Issa, R.R.: Safety implementation framework for Pakistani construction industry. Safety Sci. 82, 301-314 (2016) Saunders, L.W., Kleiner, B.M., McCoy, A.P., Ellis, K.P., Smith-Jackson, T., Wernz, C.: Developing an inter-organizational safety climate instrument for the construction industry.

Safety Sci. 98, 17-24 (2017) Saurin, T.A.: Safety inspections in construction sites: A systems thinking perspective. Acc. Anal.

Prevent. 93, 240-250 (2016) Taufek, F.H.B.M., Zulkifle, Z.B., Kadir, $.Z.B.A.: Safety and health practices and injury management in manufacturing industry. Procedia Econ. Finance 35, 705-712 (2016) Yiu, N.S., Sze, N.N., Chan, D.W.: Implementation of safety management systems in Hong Kong construction industry—A safety practitioner’s perspective. J. Safety Res. 64, 1-9 (2018) Zaira, M.M., Hadikusumo, B.H.: Structural equation model of integrated safety intervention

practices affecting the safety behaviour of workers in the construction industry. Safety Sci. 98, 124-135 (2017) Zhou, Z., Goh, Y.M., Li, Q: Overview and analysis of safety management studies in the construction industry. Safety Sci. 72, 337-350 (2015)

Othman, I., Harahap, M.LP., Mohamad, H., Shafiq, N., Napiah, M.: Development of BIM-based safety management model focusing on safety rule violations. In: MATEC Web of Conferences. vol. 203, p. 02007. EDP Sciences (2018)

®

Check updatesfor

Characteristics of Interlocking Concrete Bricks Incorporated Crumb Rubber and Fly Ash Bashar S. Mohammed, Amin Al-Fakih, and M. S. Liew Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS (UTP), 32610 Bandar Seri Iskandar, Perak, Malaysia bashar. mohammed@utp. edu. my Abstract. The application of 10% crumb rubber and 56% fly ash to partial replacement

of fine aggregations

and

cement,

respectively,

by

volume

has

formed rubberized interlocking bricks (RIBs). RIBs have been produced by semi-automatic

interlocking

pressing

machine.

This

study

aimed

to extend

previous research on the characteristics of newly developed RIBs. The initial rate of suction, dry density, efflorescence, thermal conductivity, and the elevated temperature was conducted. It was found that RIBs exhibited a lower density and higher initial rate of suction, as well as better thermal conductivity. The RIB’s compressive strength decreased while the porosity increased with the increase in the elevated temperature. Keywords: Crumb rubber - Interlocking brick - Rubberized concrete Elevated temperature - And physical characteristics

1

Introduction

The scrap tires disposal has resulted in many environmental problems worldwide. In Malaysia, every year, where much of it ends up in a landfill or dumpsite, 8 million tires was recorded to meet their end of lifespan [1]. By the end of 2020, In Malaysia, it is expected to disposal 20 million units of tyres [2]. Globally, scrap tires production is exceeding 1 billion units per year [3]. As for Malaysia, government agencies must encourage and implement the recovery and proper management system of scrap tires by providing specifically designed disposal facilities for scrap tires to reduce the stockpiling problem. Prior to use in other applications, discarded tires may be transformed into crumb rubber. The crumb rubber production involves shredding, chipping, and grinding of scrap tires into small sizes which ranges between 25 mm to 50 mm [4-6]. The end product of recycling scrap tire makes it possible for concreting applications as a replacement for the concrete aggregates, crumb rubber may be used. Depending on the volume of aggregates substituted and the size of crumb rubber, crumb rubber application has many impacts on concrete properties. Crumb rubber has a variety of characteristics that hypothetically improve concrete performance. Crumb rubber has hydrophobic nature, lower stiffness, and low specific gravity [7, 8] which contribute to a lower density of rubbercrete. Moreover, crumb rubber concrete has better toughness, fatigue resistance, high permeability, thermal and

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): [CCOEE2020, LNCE 132, pp. 631-639, 2021. https://doi.org/10.1007/978-98 1-33-63 11-3_72

632

B. S. Mohammed

et al.

acoustic as well as impact resistance [9, 10]. Therefore, the concrete incorporated crumb

rubber has high thermal insulation, better resistance to abrasion, and ductility behaviour. Utilizing crumb rubber in concrete reveals several advantages. There have been various forms of work with the goal of replacing fine aggregates instead of coarse aggregates [11, 12]. The usage of concrete crumb rubber gives the rubbercrete a higher electrical insulation

feature

than conventional

concrete

[13]. Besides,

crumb rubbers

have the ability to improve the concrete’s surface resistivity. In tum, pozzolanic compounds are applied to the concrete mix process, which also increases rubbercrete surface strength. The strong resistivity of the surface shows that the rubbercrete is well electrical resistant. In fact, the rubbercrete’s thermal resistance is higher than the standard concrete. The increase in the amount of crumb rubber replacement increases the thermal resistivity [14,

15]. 50%

of crumb rubber replacement reduces the thermal

conductivity by up to 50%. In addition to that,

adding fillers such as gypsum and fly

ash are also able to enhance thermal and fire resistance [16]. The air voids formed due

to the non-polarity of the crumb rubber also enhance thermal resistivity [17]. This is due to the lower thermal conductivity of air as compared to concrete and also the ability of the crumb rubber to opposes the thermal transmittance, within the rubbercrete due to its high insulation property [14]. Recently the brick industry has undergone a major improvement in the manufacture of non-fired bricks. modern strategies. The unfired bricks differ dramatically from the traditional firing bricks, from an economic and environmental point of view. Brick construction would add to the conservation of the atmosphere through a more sustainable, sophisticated, constitutional and eco-friendly solution [18]. Interlocking bricks is a quick and easy technique because it eliminates the use of morter in the construction process. Rubberised interlocking brick is therefore a masonry unit formed by the partial substitution of concrete and cement, by using a semi-automatic locking system, with fly-ash and crumb-rubber residues. This study focuses to extend previous research on the characteristics of newly developed RIBs to investigate its physical characteristics including the initial rate of suction, dry density, efflorescence, thermal conductivity, and elevated temperature.

2

Materials and Methods

In producing rubberized interlocking brick, fine aggregates, cement, crumb rubber, fly ash, and water had been used. the details of materials and chemical composition of cement and fly ash used in the production of rubberized interlocking brick mixtures has been

mentioned

elsewhere

[19-22].

Rubberized

interlocking brick

is a masonry

unit

produced by utilizing 56% fly ash and 10% crumb rubber as partial replacement of cement and sand, respectively by volume using a semi-automatic interlocking machine. After removing the RIBs from the mold, the specimens were left for 28 days at room temperature for the moist curing process as shown in Fig. 1. The specimens then were ready to undergo the rate of suction, dry density, efflorescence, thermal conductivity, and elevated temperature tests. The number of specimens required for each test is shown in Table 1.

Characteristics of Interlocking Concrete Bricks Incorporated Crumb.

633

Fig. 1. Curing process of rubberized interlocking bricks Table 1. Specimens requirement for testing Test Dimensions (mm) | No. of specimen | Standard Initial rate of suction | 250 x 125 x 105] 5 ‘ASTM C67 [23] Density 250 x 125 x 105) 5 BS EN 12390-7 [24] Thermal conductivity 10 x 10x 10 | 5 ASTM D7984 [25] Efflorescence 250 x 125 x 105) 10 ASTM C67 [23] Elevated temperature 50 x 50 x 50 | 18 -

2.1

Physical Characteristics Test Procedure

2.1.1 Density The dry density test was conducted by measuring the mass of as-received, watersaturated, and oven-dried hardened specimens. The specimens were cled to room temperature in dry air tight vessel before weight was recorded. The water displacement process was used to calculate the volume of specimens due to the irregular shape of rubberized interlocking bricks as shown in Fig. 2. The volume and density of specimens are then calculated. IRS test was carried out in accordance to ASTM C67 by using five specimens [23]. Two metal plates were placed in a big and shallow container at least 75 mm to 100 mm apart before water was poured in the container to approximately 3 mm, covering the metal plate as in Fig. 2. Then, the bricks were measured to obtain the dry mass of the bricks. After that, the bricks were placed on top of the steel plates for 60 s. Lastly, the brick is removed and the wet mass min) is calculated as Eq. 1 below.

is measured.

1,kg/n? kg/m? «minmin = where A = immersed

The

initial rate of suction,

I kg/m’.

1000(svetmass — drymass)

surface area (m7) and T

i

is the time of the submersion (min)

(1)

634,

B. S. Mohammed

©

et al.

@

©

Fig. 2. Testing set-up of RIBs (a) Initial rate of suction test; (b) Thermal conductivity test; (©) Efflorescence test; (d) Elevated temperature test (at 400 °C); (e) Density test 2.1.2 Thermal Conductivity The thermal conductivity test was conducted according to ASTM D 7984 by using thermal conductivity analyzer C-Therm Tci [25]. This equipment employed the Modified Transient Plane Source (MTPS) technique in measuring thermal conductivity and effusivity of specimens in 1 to 3 s. Figure 2 shows the RIB undergo thermal conductivity test by using C-Therm thermal conductivity analyzer. 2.1.3 Efflorescence It was conducted in accordance to ASTM C6. The most common salt that existed is magnesium sulphate, calcium, sulphate, and carbonate. Magnesium present in the bricks must be less than 0.05% and the soluble salt content shall not exceed 0.1% in order to allow them to be used in construction. In this test, ten bricks that were free from any dust were required to get an accurate result. The bricks were divided into two sets, which consisted of 5 bricks per set. The 5 bricks were partially immersed in the pan that filled with distilled water at a depth of 2.54 mm for 7 days and kept in the drying room. Another 5 bricks, were placed in the drying room without any contact with water as in Fig. 2 Each brick was separated at least 50.8 mm from one another. After 7 days, the bricks were dried in the drying oven for 24 h. The bricks then were observed for any difference. 2.1.4 Elevated Temperature Test This test was conducted by placing the 50 mm x 50 mm cube at room temperature and in the furnace as in Fig. 2 at 100 °C, 200 °C, 400 °C, 600 °C and 1000 °C for two hours. Three identical cubes were tested for each temperature. The 50 mm cube specimens were first dried in the oven at 67 °C for 24 h to ensure the water in the pore

Characteristics of Interlocking Concrete Bricks Incorporated Crumb.

635

is fully removed, then the weight is recorded. After heated to a particular temperature for two hours, the specimens were left in the furnace to cool down. The cooled specimens are then weighted again. Any changes in the appearance and color of the cubes were observed. Later, the cubes undergo a compressive strength test.

3 3.1

Results and Discussion Initial Rate of Suction

(IRS)

The initial water suction of rubberized interlocking brick ranges from 1.37 kg/m x min to 4.1 kg/m? x min which is equivalent to 40 g to 120 g of water absorption within one minute. According to ASTM C67, the acceptable value ranges from 10 g to 30 g [23]. The nature of crumb rubber to trapped air during the mixing led to the high void formation in the hardened rubberized interlocking brick. The porous microstructure of the rubberized interlocking brick then contributed to the high initial water suction. According to Boral, the optimum value of initial rate of suction is considered to be between 0.5 and 1.5 kg/m*/min. However, brick with initial rate of suction between 10 to 35 g is preferable for its superior bonding property. In conclusion, a high initial rate of suction of rubberized interlocking bricks tend to absorb large quantities of water from the grout, thus resulted in the poor bond connection between grout and bricks. Therefore, the total water-cement ratio maybe increases to reduce IRS. 3.2

Density

As a results

of density test (Table 2), the density of rubberized interlocking bricks

ranges from 1799 kg/m* to 1979 kg/m? with an average of 1894 kg/m? which is classified as medium weight brick. Ortega et al. [26] have recorded a dry density of 1930 kg/m? at 10% crumb rubber replacement in rubbercrete brick. This reduction in the density of RIBs is contributed by the rubber low specific gravity in comparison with the traditional aggregates [12, 26].

Table 2. Density of rubberized interlocking brick Specimen

Mass (kg)

1

5.613

Volume (m*) 836

Density (kg/m*) 1979.44]

2 3

5.454 5.410

2.835 2.815

1923.533 1922.019

4

5.293

2.870

1844.537

5

5.140

2.856

1799.580

Average

1893,822

636

3.3

B. S. Mohammed

et al.

Efflorescence

It can be seen in Fig. 3 the presence of the salt when it was partially immersed in the distilled water for seven days. After drying, a normal eye observation has been conducted on all faces of the rubberized interlocking brick. One out of five bricks was affected by efflorescence, which can be concluded that the brick is not effloresced in accordance to ASTM C67. However, the presence of salt deposited might be due to the existing salt from the container, salt deposited on the sand, or crumb rubber itself.

Fig. 3.

3.4 The

Efflorescence test result of RIBs

Thermal Conductivity thermal

conductivity

0.863 W.m/m*(K)

with

of rubberized an

average

of

interlocking

brick

0.729 W.m/m?(K).

ranges Thermal

from

0.613

to

conductivity

obtained from the experimental test is lower than illustrated in the previous study. Low thermal conductivity of rubberized interlocking brick is due to the microstructure of the crumb rubber which trapped air on its surface, thus increasing the air content. According to Mohammed et al. [14], the thermal conductivity of the air is 0.025 W. m/m?(K), less than that of concrete which is 1.7 W.m/m7(K). Therefore, the air voids

inside the rubberized interlocking brick can deter the thermal transfer throughout the rubberized interlocking brick. Besides, crumb rubber particles also restrain thermal flow due to low thermal conductivity of the crumb rubber particles which is 0.16 W. m/m?(K), less than the fine aggregate with 1.5 W.m/m* Other than that, the thermal conductivity of fly ash is also lower compared to cement, leads to a further reduction in the thermal conductivity. 3.5

Elevated Temperature

It is presented in Fig. 4 that the average compressive strength of rubberized interlocking bricks started to decrease when the temperature increase. The compressive strength at room temperature (27 °C) shows the highest compressive strength compared to the compressive strength of temperature 100 °C to 1000 °C. The average compressive strength of specimens at 27 °C, 100 °C, 200 °C, 400 °C, 600 °C and 1000 °C was recorded at 12.45 MPa, 9.31 MPa, 8.5 MPa, 6.93 MPa, 6.45 MPa, and 2.98 MPa, respectively. The most significant loss in temperature was recorded at the highest temperature tested, 1000 °C.

Characteristics of Interlocking Concrete Bricks Incorporated Crumb.

é

14

0.025

=” = 10 2 28 36

0.02

5 £2

0.005

2 0.018 BZ 0.01

go4 UC

637

0

3zs

0 27

100

200

400

600

1000

Temperature (°C)

mms Compressive strength (Mpa)

—®= Weight loss (kg)

Fig. 4. Compressive strength and weight loss based on the elevated temperature The decreasing of the compressive strength at higher temperatures is due to the melted crumb rubber as according to Guo et al. crumb rubber melt at 170 °C [27]. As

the crumb rubber melt, more voids were formed inside the rubberized interlocking brick, hence weaken the bond between the crumb rubbers and the cement matrix. The weak bond then led to a further decrease in compressive strength. Furthermore, the presence of voids due to the melted crumb rubber also provided the passage for the water vapors to escape from the brick and release pore pressure under high temperature and hence avoid any occurrence of explosive brick

[27].

Other than that, the average weight loss of the specimens is increasing as the temperature increases. The specimens heated at 100 °C, 200 °C, 400 °C, 600 °C and 1000 °C show 0.001 kg, 0.006 kg, 0.012 kg, 0.017 kg and 0.022 kg loss, respectively. The highest weight loss was recorded in the specimen at the temperature of 1000 °C. However, the weight loss started to show a significant increment at 200 °C onwards. According to Lim [28], C-S-H shrinkage was observed at a temperature higher than 300 °C. Therefore, it can be concluded that at 400 °C, 600 °C and 1000 °C, the C-S-H gel and crumb rubber degraded, thus forming a higher volume of voids and reduce the weight of the specimens. The visible changes on specimens color occur after the cubes were placed in the hot temperature furnace in 2 h is shown in Fig. 5.

(b) Fig. 5. RIBs after 2 h in the furnace (a) Heating up to 600 °C; (b) Heating at 1000 °C

638

4

B. S. Mohammed

et al.

Conclusion

The study shows encouraging results and confirmed that the newly developed rubberized interlocking brick exhibits acceptable physical properties. Rubberized interlocking brick is called medium-sized concrete based on the density check. The hydrophobic behavior of a crumb rubber particle that traps air along the surface and repels water decreases density, enhances initial rate of suction levels and improves thermal conductivity. The influence on the compression strength and porosity of the rubberized interlocking brick is demonstrated by elevated temperature.

References 1. Linnenkoper, K.: Malaysian tyre recycling facility to supply footwear majors. https:// recyclinginternational.com/rubber/malaysian-tyre-recycling-facility-to-supply-footwearmajors/4971/. Accessed 1 Aug 2019 “Room for Malaysian tyre exports to grow.” https://www.thestar.com.my/business/business-

news/2013/05/09/room-for-malaysian-tyre-exports-to-grow/. Accessed 1 Aug 2019 3. Nehdi, M.L., Najjar, M.F., Soliman, A.M., Azabi, T.M.: Novel eco-efficient two-stage concrete incorporating high volume recycled content for sustainable pavement construction. Constr. Build. Mater. 146, 9-14 (2017). https://doi.org/10.1016/j.conbuildmat.2017.04.065

4. Turatsinze, A., Garros, M.: On the modulus of elasticity and strain capacity of selfcompacting concrete incorporating rubber aggregates. Resour. Conserv. 1209-1215 (2008). https://doi.org/10.1016/j.resconrec.2008.06.012

Recycl.

52(10),

5. Najim, K.B., Hall, M.R.: A review of the fresh/hardened properties and applications for plain- (PRC) and self-compacting rubberised concrete (SCRC). Constr. Build. Mater. 24(11), 2043-2051 (2010). https://doi.org/10.1016/j.conbuildmat.2010.04.056 6. Cecich, V., Gonzales, L., Hoisaeter, A., Williams, J., Reddy, K.: Use of shredded tires as

lightweight backfill material for retaining structures. Waste Manage. Res. 14(5), 433-451 (1996). https://doi.org/10.1006/wmre.1996.0043 7. Mohammed, B.S., Loong, R.C.H.: structural behavior of reinforced rubbercrete beams in shear. Appl. Mech. Mater. 752, 513-517 (2015). Trans Tech Publ 8. Gerges, N.N., Issa, C.A., Fawaz, S.A.: Rubber concrete: mechanical and dynamical

properties. Case Stud. Constr. Mater. 9, e00184 (2018). https://doi.org/10.1016/j.cscm.2018. 00184

9. Thakur, A., Senthil, K., Sharma, R., Singh, A.P.: Employment of crumb rubber tyre in concrete masonry bricks. Materials Today: Proceedings (2020). https://doi.org/10.1016/j. 10.

matpr.2020.02.106 Mohammed, B.S.,

Liew,

M.S.,

Alaloul,

W.S.,

Al-Fakih,

A., Ibrahim,

W.,

Adamu,

M.:

Development of rubberized geopolymer interlocking bricks. Case Stud. Constr. Mater. 8, 11.

401-408 (2018). https://doi.org/10.1016/j.csem.2018.03.007 Aiello, M.A., Leuzzi, F.: Waste tyre rubberized concrete: Properties at fresh and hardened

state. Waste Manage. 30(8), 1696-1704 (2010). https://doi.org/10.1016/j.wasman.2010.02. 005 12. Sadek, D.M., El-Attar, M.M.: Structural behavior of rubberized masonry walls. J. Cleaner

Prod. 89, 174-186 (2015)

Characteristics of Interlocking Concrete Bricks Incorporated Crumb.

639

. Onuaguluchi, O.: Effects of surface pre-coating and silica fume on crumb rubber-cement matrix interface and cement mortar properties. J. Cleaner Prod. 104, 339-345 (2015). hitps:// doi.org/10.1016).jclepro.2015.04,116 . Mohammed, B.S., Anwar Hossain, K.M., Eng Swee, J.T., Wong, G., Abdullahi, M.: Properties of crumb rubber hollow concrete block. J. Cleaner Prod. 23(1), 57-67 (2012). https://doi.org/10.1016/j,jclepro.201 1.10.035 . Aliabdo, A.A., Abd

Elmoaty, A.E.M.,

AbdElbaset,

M.M.:

Utilization of waste rubber in

non-structural applications. Constr, Build. Mater. 91, 195-207 (2015). https://doi.org/10. 1016/j.conbuildmat.2015.05.080

. Leiva, C., Arenas, C., Vilches, L.F.,

Alonso-Farifias, B., Rodriguez-Galan, M.: Development

of fly ash boards with thermal, acoustic and fire insulation properties. Waste Manage. 46, 298-303 (2015). https://doi.org/10.1016/j. wasman.2015.08.027

. Hall, M.R., Najim, K.B., Hopfe, C.J.: Transient thermal behaviour of crumb rubber-modified

concrete and implications for thermal response and energy efficiency in buildings. Appl. Therm. Eng. 33, 77-85 (2012)

. Al-Fakih, A., Mohammed,

B.S., Nuruddin, F., Nikbakht, E.: Development of interlocking

masonry bricks and its’ structural behaviour: a review paper. In: IOP Conference Series: Earth and Environmental Science, vol. 140, no. 1, p. 012127. IOP Publishing (2018)

). Al-Fakih, A., Wahab, M.M.A., Mohammed, B.S., Liew, M.S., Zawawi, N.A.W.A., As’ ad,

S.: Experimental study on axial compressive behavior of rubberized interlocking masonry 20.

walls. J. Build. Eng. 29, 101107 (2020). https://doi.org/10.1016/j.jobe.2019.101107 Al-Fakih, A., et al.: Mechanical behavior of rubberized interlocking bricks for masonry

21.

Al-Fakih,

structural applications. Int. J. Civ. Eng. Technol. 9(9), 185-193 (2018) A., Mohammed,

B.S.,

Liew,

M.S.,

Alaloul,

W.S.:

Physical

properties of the

rubberized interlocking masonry brick. Int. J. Civ. Eng. Technol. 9(6), 656-664 (2018) 22.

Al-Fakih, A., et al.: Characteristic

compressive strength correlation of rubberized concrete

interlocking masonry wall. Structures 26, 169-184 (2020). https://doi.org/10.1016/.istruc. 2020.04.010 23.

ASTM, C.: 67-03, Standard test methods for sampling and testing brick and structural clay

24.

tile. American Society for Testing and Materials, Philadelphia, PA (2003) EN, B.: 12390-7 Testing hardened concrete-density of hardened concrete. British Standards,

BSI Group Headquarters, vol. 389 (2009) 25.

Standard test method

for measurement of thermal effusivity of fabrics using a modified

transient plane Source (MTPS) Instrument, A. D7984-16, West Conshohocken, PA (2016). www.astm.org 26.

Sodupe-Ortega,

E., Fraile-Garcia, E., Ferreiro-Cabello, J., Sanz-Garcia,

A.: Evaluation of

crumb rubber as aggregate for automated manufacturing of rubberized long hollow blocks and bricks. Constr. Build. Mater. 106, 305-316 (2016). https://doi.org/10.1016/j. conbuildmat.2015.12.131 J.-H., Chen, G.-M., Xie, Z.-H.: Compressive behaviour of concrete 27. Guo, Y.-C., Zhang, structures incorporating recycled concrete aggregates, rubber crumb and reinforced with steel fibre, subjected to elevated temperatures. J. Cleaner Prod. 72, 193-203 (2014). https://

doi.org/10.1016/j,jclepro.2014.02.036 28.

Lim,

S.: Effects of elevated

temperature exposure on cement-based

University of Illinois at Urbana-Champaign (2015)

composite

materials.

® ‘upaates

Image - Based Change Detection in Concrete Beam Krisada Chaiyasarn®®, Apichat Buatik, Kuntapit Jirakasemsuk, Pakkapong Khuangsimma, and Suraparb Keawsawasvong

Department of Civil Engineering, Thammasat University, Khlong Luang, Pathum Thani, Thailand ckrisada@engr. tu.ac. th, Pond_529364016@hotmail. com, [email protected], uraparb@hotmail. com, boss tun0089@gmail. com Abstract. Detection of crack change using images requires accurate preprocessing steps including geometrical and photometrical registration. In this paper, a change detection system is proposed to detect changes of cracks in images taken from different viewpoints in concrete structures. The geometrical registration step was done by the image-based 3D photogrammetry technique. With this technique, images taken at different times and viewpoints can be rectified to remove noise that is not real changes. The queried and reference images were then used to detect changes between the images by a simple change detection algorithm. The experiment was conducted on images of a concrete

beam sample with various stages of crack development. The results were compared against the real measurements of crack width changes. It was shown that the proposed method provides accurate results in detecting changes in a crack in the sample datasets.

Keywords: Computer vision - Crack detetcion - Deep learning - Inspection « Structural analysis

1

Introduction

There are several methods and instruments which can be used in visual inspection to find cracks on concrete structures. However, in many areas, the site accessibility can be quite limited, and there is a risk of an accident when inspectors enter to sites. In addition, visual inspection is a laborious and time-consuming procedure, as it requires inspectors with particular experience and skills. Human-errors remain a big problem for visual inspection, therefore; an alternative method for automatically inspecting cracks on concrete structures more effectively and efficiently is required to facilitate and to improve the inspection procedure. The image processing technique has been currently adopted to detect and examine the state of health of the concrete structural system. This technique uses different views of images taken in several periods and synthesizes those images to create a 3D model. This 3D model can be used in conjunction with the change detection system to evaluate the changes of cracks on a concrete structure in various periods. In order to avoid an error due to viewpoint and lighting difference, the

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 B. S. Mohammed et al. (Eds.): JCCOEE2020, LNCE 132, pp. 640-647, 2021. https://doi.org/10.1007/978-98 1-33-63 1 1-3_73

Image — Based Change Detection in Concrete Beam

641

identical views of recorded images are required for the inspection. However, it is difficult to obtain the identical views of recorded images taken in different times as images are normally taken at different conditions, including viewpoint and lighting difference. As a result, the image synthesis technique has been developed in order to overcome these problems. By using an image-based technique, images can be rectified to correct viewpoint different so that the identical viewpoints and positions of the images taken in different times can be achieved and the changes of cracks at various positions on a concrete structure can be efficiently detected. In addition, unrelated environment or interferences in the images will be removed in order to obtain images of cracks.

2 2.1

Related Works Change Detection

To reduce time and cost of damage assessment, image-based change detection can be employed to evaluate the damage of the structure. Stent [1, 2] applied Structure from Motion (SFM) technique to create a 3D model for monitoring changes of concrete tunnel linings. Apichat [3] used 3D model-based image registration for detecting changes of the historical structures via an Unmanned Aerial Vehicle. The images obtained from the aerial vehicle were used in conjunction with the SfM technique to create a 3D model of historical structures through Agisoft photoscan software. After obtaining the 3D model, the camera registration technique was applied to the 3D model to obtain synthesized images. These images were processed by the pixel-difference technique were to detect changes on any location of historical structures. 2.2

Crack Detection

To detect cracks on concrete or brick structures, visual inspection by human is an extremely time-consuming process. At present, several techniques to inspect cracks such as image-based crack detection have been developed. Mayank et al. [4] used Convolutional Neural Network (CNN) and Support Vector Machine (SVM) to examine

and classify images with or without cracks by. The results showed that more than 90.76% of accuracy was obtained, which is better than only using CNN alone. Chaiyasam et al. [5] also employed the CNN and SVM integrated architecture to detect cracks on historical brick structures, and the accuracy of 74.90% was achieved. Yang et al. [6] applied a Fully Convolutional Network (FCN) technique to automatically detect and measure the length of cracks from 800 images from a trained CNN. FCN utilizes the VGG19 network to identify regions that contain cracks. Then crack skeleton can be developed using the morphological technique and the length and width of cracks can be estimated. In this work, the ground truth was created to validate the accuracy of the results from FCN.

642K, Chaiyasarn et all. 3

Methodology

Figure | shows the outline of the proposed method. The system starts with collecting images of cracks at time T1 to create a 3D model. Then images of the later times (ie. T2, T3 and T4) when cracks were developed were also collected. The images from

TI were used to create a 3D model via the image-based photogrammetry. To remove geometrical errors, the 3D models at the times T2 to T4 were synthetized based on the 3D model at the time T1 in order to acquire the identical views as that of T1. The crack detection was carried out to extract cracks from the images. These extracted cracks are later used to detect the changes in the step of change detection.

Reference

Image

datasets _| Image-based 3D modeling

synthesis

Crack detection

|-+|

Change detection

Query images Fig. 1. The outline of the proposed method 3.1

Image Acquisition

In this study, the concrete beam was statically tested until failure. Every steps of loading, images around the beam were recorded by a DSLR camera. Note that the images recorded at the time T1 were used as the reference model and must contain an overlap of more than 50%. Other images at the times T2, T3 and T4 were also collected and later synthetized using the 3D model from T1. 3.2

Image — Based 3D Modelling

The SFM software, Agisoft Photoscan, was used to synthetize images from a 3D model created from images. With the aid of this software, a sparse point cloud model, a dense point cloud model, a mesh model of a concrete beam and a textured model of a concrete beam were then created as shown in Fig. 2(a), 2(b), 2(c) and 2(d) respectively.

3.3

Image Synthesis

The images relation.

of the surface

of the 3D

model

can

I, = Ky [Rat] P-

be synthesized

by the following

()

where I, is synthesized image; Ky is intrinsic matrix; Rg is rotation matrix; ty is translation vector; P, is set of reference 3D surface; I, is query image (real image).

Image — Based Change Detection in Concrete Beam

pemep

643

gs

@

()

Me @

Fig. 2. (a) Sparse point cloud: (b) Dense point cloud: (c) Mesh model: (d) Textured model 3.4

Crack Detection

Once query and reference images are synthetized, cracks in the images can be detected and localized using the integrated CNN architecture. This section describes the processes of crack localization (with the aid of CNN

and SVM)

and the crack extraction

technique.

3.4.1

Crack Localization by CNN and SVM

According

to Chaiyasarn

et al. [5], crack

can

be

localized

by

the CNN

and

SVM

techniques. In this system, 4,000 images with cracks and 6,000 images without cracks were used in training the CNN and SVM. All images are 64 x 64 pixels. After training, the crack detection system can classify images with or without cracks. The small

boxes

in Fig. 3(a)

indicate

the

locations

of cracks,

and

Fig. 3(b)

shows

the

original image of cracks.

@

()

Fig. 3. (a) Crack localization image: (b) Original image 3.4.2

Crack Extraction

After cracks were localized by the CNN and SVM technique, simple thresholding technique was used to obtain some line features that may resemble cracks. Then the cracks regions from the CNN and SVM are merged with crack regions from the threshold technique to obtain the final crack maps in images. The following equation is the relation between both masks.

644K, Chaiyasarn et all. MLO Tow = Merack where

Mr

is a crack

localization

mask

from

(2)

CNN-SVM,

Ibw

is a mask

from

the

thresholding technique, and Morac is a final crack mask. 3.5

Change Detection

3.5.1. Change Mask The change mask of cracks can now be created by subtracting synthesized images with identical viewpoints. The following equation for creating change mask is used.

B(x) = [ly where

B(x) is change

mask; I,

is query

image

|

3)

(real image

from T2, T3 and T4); I, is

synthesized images from 3D model. 3.5.2, Confusion Matrix The accuracy of the change masks from the system can be compared with the manual change masks from the following equations.

TP

+1N

Accuracy = 5 TNL FP TEN Recall

TP

= TP AEN

(5)

precision