PROCEEDINGS OF THE WABER 
SuDBE CONFERENCE 2024
30 – 31 July 2024
University of the Witwatersrand
Johannesburg, South Africa

EDITORS

Prof. Samuel Laryea, University of the Witwatersrand, South Africa
Prof. Baizhan Li, Chongqing University, China
A/Prof. Emmanuel Adu Essah, University of Reading, United Kingdom
A/Prof. Sarfo Mensah, Kumasi Technical University, Ghana
Prof. Hong Liu, Chongqing University, China
Prof. Runming Yao, University of Reading, UK/Chongqing University, China

ISBN: 978-0-7961-6032-4

In collaboration with:



i 

Proceedings of the WABER SuDBE 2024 Conference 

30th – 31st July 2024 

University of the Witwatersrand, Johannesburg, South Africa 

© Copyright 

The copyright for papers in this publication belongs to the authors of the papers. 

ISBN: 978-0-7961-6032-4 (e-book)

The ISBN for this publication was provided by the National Library of South Africa. Legal deposits of the 

publication have been supplied to the National Library of South Africa, Library of Parliament, and other places of 

Legal Deposit. 

First published in July 2024 

Published by: 
WABER SuDBE Conference 2024 

C/o Professor Samuel Laryea, Conference chair 
School of Construction Economics and Management 

University of the Witwatersrand, Johannesburg, South Africa 
Email: info@wabersudbe.com / samuel.laryea@wits.ac.za

Website: www.wabersudbe.com 

Editors 

Prof Samuel Laryea, University of the Witwatersrand, South Africa 

Prof Baizhan Li, Chongqing University, China 

A/Prof Emmanuel Adu Essah, University of Reading, United Kingdom 

A/Prof Sarfo Mensah, Kumasi Technical University, Ghana 

Prof Hong Liu, Chongqing University, China 

Prof Runming Yao, University of Reading, UK / Chongqing University, China 

Declaration 

All papers in this publication have been through a review process involving a review of abstracts, peer review of 

full papers by at least two referees, reporting of comments to authors, revision of papers by authors and re-

evaluation of the revised papers to ensure quality of content. 



ii 

TABLE OF CONTENTS 

Table of Contents ii 

Foreword iii 

Conference organising and scientific committee v 

Confirmation of 60/40 rule ix 

Programme and profile of speakers xi 

Conference themes xv 

Conference papers xvii
Index of authors 169 

Index of keywords 173 



iii 

FOREWORD 

It is my pleasure to welcome you to this special WABER SuDBE Conference 2024 which is a 

joint international conference co-organised by the West Africa Built Environment Research 

(WABER) Conference and the International Conference on Sustainable Development in 

Building and Environment (SuDBE) in collaboration with various partners.  

 The International Conference on Sustainable Development in Building and Environment 

(SuDBE) was initiated in 2003 by the National Centre for International Research of Low-

carbon Green Buildings at Chongqing University, China. The West Africa Built Environment 

Research (WABER) Conference was initiated in 2008 as an initiative of the School of 

Construction Management and Engineering, University of Reading, UK, and provides a 

platform for sharing the latest ideas in built environment research on the African continent.  

I am pleased to welcome everyone to Wits University, Johannesburg. I hope you enjoy this 

Conference in the beautiful environment of our university. There are two days of technical 

presentations and an industry panel discussion, with a welcome cocktail party on the first night, 

and a conference dinner on the second night. This is followed by two days of technical tours, 

which provide an opportunity to see some landmark buildings in Pretoria, and two green rated 

developments by Growthpoint Properties in Rosebank and Sandton, Johannesburg. 

The technical presentations consist of seven keynote speeches and 130 paper presentations. 

The keynote speeches focus on an array of interesting topics that relate to the general 

conference theme of sustainable built environments. We have four keynote speeches relating 

to the theme of adaptability of the built environment to climate change and the sustainable 

development goals. The three other keynote speeches address matters of resilient and 

sustainable futures, the use of digital technologies to improve the sustainability of buildings, 

and artificial intelligence and carbon neutrality. The academic and industry leaders speaking 

on these topics are very experienced and their keynote presentations are expected to stimulate 

new ideas and discussion in the conference.  

The accepted papers to be presented in the parallel sessions relate to eight themes namely: 

• Climate Responsive Built Environments

• Air Quality and Healthy Building

• Thermal Comfort and Intelligent Operation

• Low Carbon Technology and Energy System

• Sustainable Urban Renewal

• Building Technology and Performance

• Construction and Project Management

• Real Estate and Property Management

Thank you and congratulations to all the authors of papers in this publication. I also want to 

sincerely recognise and appreciate the efforts of the 154 reviewers from 21 different countries 

who assisted enormously with the scientific work of reviewing the abstracts and full papers 

submitted for this Conference. We initially received 191 abstracts that were thoroughly 

reviewed to provide comments and decisions to authors. 161 full papers were subsequently 

received, and we would like to express our deepest appreciation to the reviewers below who 

did the hardwork of reviewing the 161 full papers and providing review reports which assisted 

the editors to make decisions on the papers and control the quality of the papers accepted for 

publication in the proceedings. Ultimately, 123 papers were accepted and published in this 



iv 

conference proceedings. I extend special thanks to A/Prof Emmanuel Essah and A/Prof Sarfo 

Mensah for leading the review processes and performing extensive editorial duties. 

There are also two workshops taking place during the conference. The first one is a Workshop 

on "Low-carbon urban and city development towards carbon neutrality" which is led by Prof 

Yong Ding from Chongqing University. The second one is a Workshop on Sustainable 

construction industry growth which is facilitated by the cidb Centre of Excellence at Wits 

University.  

Lastly, we also have an industry panel discussion on the risks associated with South African 

renewable energy projects. This will be facilitated by A&O Shearman whom we are proud to 

have as our partners for this conference. 

The conference programme presents a valuable package that will facilitate intellectual and 

practical discussions on sustainable built environments and the construction sector’s role in 

addressing the global challenge of climate change and the sustainable development goals 

(SDGs).  

Special thanks to all our conference partners particularly A&O Shearman, the Construction 

Industry Development Board (cidb), Gauteng Tourism, and Growthpoint Properties for your 

valuable support. 

I would conclude by wishing all participants a stimulating, rewarding and enjoyable 

conference. I look forward to enjoying your presentations, debates and company over the 

conference period of 29th July to 2nd August. Thank you. 

Professor Samuel Laryea 

WABER SuDBE 2024 Conference Co-Chair 

Wits University, Johannesburg, South Africa 

29th July 2024 



v 

CONFERENCE ORGANISING COMMITTEE 

• Professor Samuel Laryea, Wits University, South Africa, Conference Co-Chair

• Professor Baizhan Li, Chongqing University, China, Conference Co-Chair

• Professor Runming Yao, University of Reading, UK / Chongqing University, China,

Conference Secretary

• Professor Hong Liu, Chongqing University, China, Conference Secretary

• A/Professor Emmanuel Essah, University of Reading, UK, Conference Secretary

CONFERENCE SCIENTIFIC COMMITTEE AND REVIEW 

PANEL 

1. Prof. Samuel Laryea, Wits University, Johannesburg, South Africa

2. A/Prof Emmanuel Essah, University of Reading, United Kingdom

3. A/Prof Sarfo Mensah, Kumasi Technical University, Kumasi, Ghana

4. Prof Runming Yao, University of Reading, UK / Chongqing University, China

5. Prof Zhiwen Luo, Cardiff University, United Kingdom

6. Prof Bankole Awuzie, University of Johannesburg, South Africa

7. A/Prof Collins Ameyaw, Kumasi Technical University, Ghana

8. A/Prof Ebenezer Adaku, Ghana Institute of Management and Public Administration, Ghana

9. A/Prof Ehsan Saghatforoush, Wits University, Johannesburg, South Africa

10. A/Prof Elvis Attakora-Amaniampong, SDD-UBIDS, Ghana

11. A/Prof Haruna Musa Moda, University of Doha for Science and Technology, Doha, Qatar

12. A/Prof Ian Ewart, University of Reading, Reading, United Kingdom

13. A/Prof Justus Agumba, Tshwane University of Technology, South Africa

14. A/Prof Kathy Michell, University of Cape Town, South Africa

15. A/Prof Kola Akinsomi, Wits University, South Africa

16. A/Prof Lekan Amusan, Covenant University, Nigeria

17. A/Prof Mehdi Shahrestani, University of Reading, Reading, United Kingdom

18. A/Prof Moshood Fadeyi, Singapore Institute of Technology, Singapore

19. A/Prof Norhayati Mahyuddin, University of Malaya, Kuala Lumpur, Malaysia

20. Dr. Ogbeifun Edoghogho, University of Jos, Nigeria / Univ. of Johannesburg, South Africa

21. A/Prof Shen Wei, University College London, United Kingdom

22. Prof Divine Ahadzie, Kwame Nkrumah University of Science and Technology, Ghana

23. A/Prof Yakubu Aminu Dodo, Najran University, Saudi Arabia

24. A/Prof Yewande Adewunmi, Wits University, South Africa

25. A/Prof. Alex Opoku, University of Sharjah, United Arab Emirates

26. A/Prof. Callistus Tengan, Bolgatanga Technical University, Ghana

27. A/Prof. Ebenezer Boakye, Takoradi Technical University, Takoradi, Ghana

28. A/Prof. Emmanuel Manu, Nottingham Trent University, United Kingdom

29. A/Prof. Evans Zoya Kpamma, Sunyani Technical University, Ghana

30. A/Prof. Francis Kwesi Bondinuba, Kumasi Technical University, Ghana

31. A/Prof. Mavis Osei, Kwame Nkrumah University of Science and Technology, Ghana

32. A/Prof. Richard Ohene Asiedu, Koforidua Technical University, Ghana



vi 

33. Dr Adedeji Afolabi, Covenant University, Nigeria

34. Dr Adwoa Ofori, Trinity College Dublin, Ireland

35. Dr Afolabi Dania, University of Westminster, London, United Kingdom

36. Dr Akosua B.K. Amaka-Otchere, Kwame Nkrumah University of Science and Technology,

Kumasi, Ghana

37. Dr Amna Shibeika, University of Reading, Reading, United Kingdom

38. Dr Baba Adama Kolo, Ahmadu Bello University, Nigeria

39. Dr Bashar Mohammed Al-Falah, Imam Abdulrahman Bin Faisal University, Dammam,

Saudi Arabia

40. Dr Bismark Duodu, Kwame Nkrumah University of Science and Technology, Ghana

41. Dr Bruno Lot Tanko, University of Reading-Malaysia, Johor, Malaysia

42. Dr Christopher Amoah, University of the Free State, South Africa

43. Dr Christos Halios, University of Reading, United Kingdom

44. Dr Cynthia Adeokun, O.N.A Architects Ltd., United Kingdom

45. Dr Emmanuel Selorm Adukpo, University College London, United Kingdom

46. Dr Eng L Ofetotse, Greenwich University, London, United Kingdom

47. Dr Faizah Mohammed Bashir, University of Hail, Hail, Saudi Arabia

48. Dr Folake Ekundayo, Design Studio, United Kingdom

49. Dr Frank Ametefe, University of Cape Town, South Africa

50. Dr Gabriel Nani, Kwame Nkrumah University of Science and Technology, Ghana

51. Dr Geoff Cook, University of Reading, United Kingdom

52. Dr Hafizah Mohd Latif, Universiti Teknologi MARA, Malaysia

53. Dr Hua Zhong, Nottingham Trent University, United Kingdom

54. Dr Immanuel Darkwa, Trinity College Dublin, Ireland

55. Dr Jeremy Gibberd Wits University, Johannesburg, South Africa

56. Dr Jesse Nor, Abuja Metropolitan Management Council, Abuja, Nigeria

57. Dr Koech Cheruiyot, Wits University, Johannesburg, South Africa

58. Dr Kwadwo Oti-Sarpong, University of Cambridge, United Kingdom

59. Dr Lewis Abedi Asante, Kumasi Technical University, Ghana

60. Dr Lovelin Obi, Northumbria University , United Kingdom

61. Dr Lungie Maseko, Wits University, Johannesburg, South Africa

62. Dr Luqman Oyewobi, FUT, Minna, Nigeria

63. Dr Mate Janos Lorincz , University of Reading, United Kingdom

64. Dr Maxwell Antwi-Afari Fordjour, Aston University, United Kingdom

65. Dr Michael Peters, University of Reading, United Kingdom

66. Dr Mojtaba Amiri, Wits University, Johannesburg, South Africa

67. Dr Neil Govender, Wits University, Johannesburg, South Africa

68. Dr Oluseyi Odeyale, University of Ibadan, Nigeria

69. Dr Oluwaseun Dosumu, University of Rwanda, Rwanda

70. Dr Prisca Simbanegavi, Wits University, South Africa

71. Dr Qingqin Wang, China Academy of Building Research, China

72. Dr Ravi Rangarajan, University of Doha for Science and Technology, Qatar

73. Dr Razaq Sherif , Abuja Metropolitan Management Council, Abuja, Nigeria

74. Dr Rolien Terblanche, University of Cape Town, South Africa

75. Dr Ronan Champion, University of Reading, Reading, United Kingdom

76. Dr Sena Agbodjah, Academic City University College, Ghana



vii 

77. Dr Shiyu Han, Chongqing University, China

78. Dr Stanley Okangba, Wits University, Johannesburg, South Africa

79. Dr Thabelo Ramantswana, Wits University, South Africa

80. Dr Tunji-Olayeni Patience Fikiemo, Covenant University, Nigeria

81. Dr. Elijah Boadu Frimpong, Kumasi Technical University, Ghana

82. Dr. Elizabeth Ojo-Fafore, Wits University, South Africa

83. Dr. Emefa Amponsah, Takoradi Technical University, Ghana

84. Dr. Godwin Kumi Acquah, Kwame Nkrumah University of Science and Technology, Ghana

85. Dr. Isaac Mensah, Department of Feeder Roads, Accra, Ghana

86. Dr. Kofi Agyekum, Kwame Nkrumah University of Science and Technology, Ghana

87. Dr. Kofi Owusu Adjei, Kumasi Technical University, Ghana

88. Dr. Kwabena Fosuhene M-Ansong, Kumasi Technical University, Ghana

89. Dr. Kwadwo Twumasi-Ampofo, Center for Scientific and Industrial Research, Ghana

90. Dr. Michael Adabre, Hong Kong Polytechnic University, Hong Kong

91. Dr. Michael Nii Addy, Kwame Nkrumah University of Science and Technology, Ghana

92. Dr. Olumuyiwa Bayode Adegun, Federal University of Technology, Akure, Nigeria

93. Dr. Stephen Agyefi-Mensah, Cape Coast Technical University, Ghana

94. Dr. Kwabena Abrokwa Gyimah, Kwame Nkrumah Univ. of Science & Technology, Ghana

95. Dr. Timothy Crentsil, Kumasi Technical University, Ghana

96. Dr Bernard Acheampong, University of Reading, United Kingdom

97. Mr Faranani Gethe, Wits University, Johannesburg, South Africa

98. Mr Hillary Chanda, University of Reading, United Kingdom

99. Mr. Kingford Mkandawire, University of Reading, United Kingdom

100. Ms Emma Ayesu-Koranteng, Nelson Mandela University, South Africa

101. Ms Lerato Mompati, Wits University, Johannesburg, South Africa

102. Ms Zamageda Zungu, Wits University, Johannesburg, South Africa

103. Oscar Kwasafo, Wits University, Johannesburg, South Africa

104. Prof  Cheng Sun, Harbin Institute of Technology, China

105. Prof  Da Yan, Tsinghua University, China

106. Prof  Dengjia Wang, Xi’an University of Architecture and Technology, China

107. Prof  Xinhua Xu, Huazhong University of Science and Technology, China

108. Prof  Yong Ding, Chongqing University, China

109. Prof  Zhijian Liu, North China Electric Power University, China

110. Prof Abimbola Windapo, University of Cape Town, South Africa

111. Prof Ahmed Doko Ibrahim, Ahmadu Bello University, Nigeria

112. Prof Armando Oliveira, Porto University , Portugal

113. Prof Borong Lin, Tsinghua University, China

114. Prof Carmel Lindkvist, Norwegian University of Science and Technology, Norway

115. Prof Chi Feng, Chongqing University,

116. Prof Chimay Anumba, University of Florida, USA

117. Prof Christopher Pain , Imperial College London, United Kingdom

118. Prof Deji Ogunsemi, Federal University of Technology, Akure, Nigeria

119. Prof Eziyi Ibem, University of Nigeria, Nigeria

120. Prof Fidelis Emuze, Central University of Technology, Bloemfontein, South Africa

121. Prof George Ofori, London South Bank University, United Kingdom

122. Prof Guangyu Cao, Norwegian University of Science and Technology, Norway



viii 

123. Prof GWK Intsiful, University of Liberia, Liberia

124. Prof Immaculate Nwokoro, University of Lagos, Nigeria

125. Prof Jianlei Niu, Hong Kong Polytechnic University , China

126. Prof Joy Maina, Ahmadu Bello University, Nigeria

127. Prof Kemiki Olurotimi, Federal University of Technology, Minna, Nigeria

128. Prof Kulomri Adogbo, Ahmadu Bello University, Nigeria

129. Prof Larry Bellamy, University of Canterbury, New Zealand

130. Prof Linhua Zhang, Shandong Jianzhu University, China

131. Prof Nathaniel Aniekwu, University of Benin, Nigeria

132. Prof Neng Zhu, Tianjin University, China

133. Prof Nishani Harinarain, University of Kwazulu Natal, South Africa

134. Prof Prashant Kumar, University of Surrey, United Kingdom

135. Prof Ron Watermeyer, Infrastructure Options / Wits University, South Africa

136. Prof Runping Niu, Beijing University of Civil Engineering and Architecture, China

137. Prof Salman Azhar, Auburn University, USA

138. Prof Santiago Gassó-Domingo, Polytechnic University of Catalonia, Spain

139. Prof Sarah Hayes, Bath Spa University, United Kingdom

140. Prof Sasan Sadrizadeh, KTH Royal Institute of Technology, Sweden

141. Prof Shijie Cao, Southeast University, China

142. Prof Shilei Lyu, Tianjin University, China

143. Prof Stephen Oluigbo, Ahmadu Bello University, Nigeria

144. Prof Winston Shakantu, Nelson Mandela University, South Africa

145. Prof Xianting Li, Tsinghua University, China

146. Prof Xudong Yang, Tsinghua University , China

147. Prof. Humphrey Danso, University of Education, Winneba, Ghana

148. Dr Matthew Ikuabe, Wits University, South Africa

149. Dr Timothy Ayodele, Obafemi Awolowo University, Nigeria

150. Dr Chidinma Emma-Ochu, Federal Polytechnic Nekede Owerri, Nigeria

151. Dr Miriam Chukwuma-Uchegbu, Federal University of Technology Owerri, Nigeria

152. Dr Partson Paradza, BA ISAGO University, Botswana

153. Dr Lynda Mbadugha, Wits University, South Africa

154. Patricia Kio, Fitchburg State University, USA

THANK YOU VERY MUCH TO ALL 154 REVIEWERS BASED IN 21 DIFFERENT 

COUNTRIES!!! 



ix 

CONFIRMATION OF DHET 60-40% CONFERENCE RULE 

29th July 2024 

TO WHOM IT MAY CONCERN 

I confirm that the papers accepted for publication in the WABER SuDBE 2024 Conference 

proceedings in July 2024 were peer-reviewed by at least two referees. The peer review process 

entailed initial screening of abstracts, review of full papers by at least two referees, reporting 

of the review reports to authors, revision of papers by authors, and re-evaluation of re-

submitted papers to ensure quality of content. A paper is only accepted for publication in the 

conference proceedings based on the reviewers' recommendation. I also confirm that the 

accepted papers were from multiple institutions as detailed below. 

Institution No of papers % 

Ahmadu Bello University, Nigeria 3 2.4% 

Arusha Technical College, Tanzania 1 0.8% 

Ashesi University, Ghana 1 0.8% 

Beijing Institute of Technology, China 1 0.8% 

Central University of Technology, South Africa 3 2.4% 

Chongqing Jiaotong University, China 1 0.8% 

Chongqing University, China 51 41.5% 

Dalian University of Technology, China 2 1.6% 

Durban University of Technology, South Africa 3 2.4% 

Federal University of Technology, Akure, Nigeria 1 0.8% 

Federal University of Technology, Minna, Nigeria 1 0.8% 

Fitchburg State University, United States of America 1 0.8% 

Guangdong Midea Air-Conditioning Equipment Co. Ltd, China 3 2.4% 

Hebei University of Technology, China 2 1.6% 

Henan Polytechnic University, China 2 1.6% 

Huazhong University of Science and Technology, China 5 4.1% 

Kano State Polytechnic, Nigeria 1 0.8% 

Kumasi Technical University, Ghana 2 1.6% 

Kwame Nkrumah University of Science & Technology, Ghana 1 0.8% 

Norwegian University of Science & Technology, Norway 1 0.8% 

Obafemi Awolowo University, Nigeria 1 0.8% 

The University of Sheffield, United Kingdom 1 0.8% 

Tsinghua University, China 2 1.6% 

Universiti Teknologi MARA, Malaysia 1 0.8% 

University of Johannesburg, South Africa 1 0.8% 

University of KwaZulu-Natal, South Africa 4 3.3% 

University of Reading, United Kingdom 4 3.3% 

University of the Free State, South Africa 2 1.6% 



x 

University of the Witwatersrand, South Africa 12 9.8% 

Walter Sisulu University, South Africa 1 0.8% 

Wuhan University of Science and Technology, China 1 0.8% 

Wuhan University of Technology, China 2 1.6% 

Xi'an University of Architecture and Technology, China 1 0.8% 

Zhongkai University of Agriculture and Engineering, China 4 3.3% 

As shown in the table above, only 9.8% of the papers emanated from the University of the 

Witwatersrand, with the remaining 90.2% coming from diverse institutions. Based on the 

above, the WABER SuDBE 2024 Conference, met the 60-40 percentage policy. 

Papers accepted for publication were published via the conference proceedings. The conference 

proceedings’ ISBN is: 978-0-7961-6032-4 (e-book) 

Yours Sincerely, 

Prof Sam Laryea 

University of the Witwatersrand  

Co-Chair of WABER SuDBE 2024 Conference 



xi 

PROFILES OF KEYNOTE SPEAKERS 

Chimay Anumba is a Professor and Dean of the College of Design, Construction and 

Planning at the University of Florida. A Fellow of the Royal Academy of Engineering, 

FREng, he holds a Ph.D. in Civil Engineering from the University of Leeds, UK; a higher 

doctorate – D.Sc. (Doctor of Science) - from Loughborough University, UK; and an 

Honorary Doctorate (Dr.h.c.) from Delft University of Technology in The Netherlands. 

He has over 500 scientific publications and his work has received support worth over 

$150m from a variety of sources. He has also supervised 57 doctoral candidates to 

completion and mentored over 25 postdoctoral researchers. He is the recipient of the 2018 

ASCE Computing in Civil Engineering Award and is a member of the US National 

Academy of Construction (NAC). 

Tim Broyd is Director of the Institute for Digital Innovation in the Built Environment and 

Professor of Built Environment Foresight at the University College London, The Bartlett 

School of Sustainable Construction. He moved to UCL following a career in industry, and 

has substantial experience as corporate director of technology, innovation and 

sustainability for globally operating engineering design consultancies, including both 

Atkins and Halcrow. In addition, he was CEO of construction industry research body 

CIRIA from 2002 to 2007. Within his new role he works with leading individuals in 

industry and government to understand and prepare for the challenges and opportunities 

that lie ahead. This recently included the responsibility for a 'foresight' input to the UK 

Government's developing strategy for construction as well as leading the strategic 

planning of the UK's development and implementation of BIM. Tim is a Fellow of the 

Royal Academy of Engineering, the Institution of Civil Engineers and the Royal Society 

for Arts, manufactures and Commerce. He is also a Visiting Professor in Construction 

Management at the University of Reading and in Civil Engineering at the University of 

Dundee. He has maintained an active engagement in the development and deployment of 

BIM techniques for over a decade, is Vice Chairman of BuildingSmart (UK) Ltd and sits 

on the UK Government's BIM Task Force. He is also a Director of CEEQUAL Limited, 

which is responsible for developing and marketing CEEQUAL, the world's leading 

Keynote Title:  

Resilient and Sustainable Futures – Opportunities to Leverage 

Emerging Technologies 

Professor Chimay Anumba 

FREng, Ph.D., D.Sc., Dr.h.c., CEng/PE, FICE, FIStructE, 

FASCE, NAC 

University of Florida, USA  

Keynote Title:  

The use of digital techniques to improve the sustainability of 

buildings  

Professor Tim Broyd 

University College of London, UK 



xii 

technique for assessing the sustainability of infrastructure projects. He became Vice 

President of the Institution of Civil Engineers in 2011, with particular responsibility for 

Public Voice and Policy. He was subsequently elected the 152nd President of the 

Institution of Civil Engineers, taking office in November 2016.  

Borong LIN is the Professor and Deputy Dean in the Faculty of Building Science and 

technology at Tsinghua University. He also serves as the Dean of Key Laboratory of Eco 

Planning & Green Building commissioned by the Ministry of Education, and the Fellow 

of IBPSA (International Building Performance Simulation Association). Prof. Lin was the 

winner of The National Science Fund for Distinguished Young Scholars in 2018, awarded 

the National Changjiang Scholar at 2019 and 2020 Xplorer Prize. He was selected as 

World's Top 2% Scientists by Stanford University and Elsevier China Highly Cited 

Scholars. Prof. Lin’s research focuses on the whole life cycle technology innovation for 

enhancing built environment quality, energy efficiency and carbon neutrality from the 

architectural design phase to new product development and real operation. As PI, Prof. 

Lin won the second prize of national S&T award and 5 first prize of provincial or 

ministerial S&T awards. He has published over 150 SCI journal papers and is an editorial 

member of five international peer reviewed journals. 

Jeremy Gibberd is a Co-coordinator of the Construction Industry Board (CIB) Working 

Group (W116) on Smart and Sustainable Built Environments (SASBE) and a member of 

the Multi-stakeholder Advisory Committee (MAC) of UNEP’s Sustainable Building and 

Construction programme. He has developed specialist expertise in sustainable built 

environments, education and community architecture, building performance analysis and 

facilities management.  Jeremy completed his PhD at the University of Pretoria in 2003 

on methodologies for integrating sustainability into built environments in developing 

country contexts. Jeremy has lectured on environmental design, advanced technology and 

design at SCAD in the USA and has developed and taught courses on urban design, 

architectural design, materials, sustainability, energy and facilities management with 

various organisations. He regularly works as a research scientist and consultant to 

government, business and community organisations. Jeremy has published widely and 

Keynote Title:  

Artificial intelligence and carbon neutrality: several case studies 

Professor Borong Lin 
Tsinghua University, China 

Keynote Title:  

Sustainability through smart infrastructures 

Professor Jeremy Gibberd 

University of the Witwatersrand, South Africa 



xiii 

presented as invited or keynote speaker at numerous international meetings and 

conferences. 

Humphrey Danso is a Professor and Dean of the School of Graduate Studies at Akenten 

Appiah Minkah University of Skills Training and Development in Ghana. Previously, he 

was also Dean of the Faculty of Technical Education. He holds a PhD in Civil Engineering 

from the University of Portsmouth, United Kingdom; MPhil in Civil Engineering from 

Voronezh State University of Architecture and Civil Engineering, Russia; MSc in Strategic 

Management and Leadership, Kwame Nkrumah University of Science and Technology, 

Ghana, 2019-2021; and MTech in Competency-Based Training from the University of 

Education Winneba in Ghana. Humphrey has published over Ninety (90) publications in 

international outlets in the areas of construction materials, construction management, and 

sustainable construction. 

Kathy is on the full-time academic staff at the University of Cape Town and was Head of 

the Department of Construction Economics and Management from 2017 to 2020 and the 

Deputy Dean for Undergraduate Studies (Teaching & Learning) in the Faculty of 

Engineering and the Built Environment from 2021 – 2023. Kathy holds a Doctorate in 

property and facilities management from the University of Salford (United Kingdom), an 

MPhil in cost and systems engineering and BSc (Hons) in Quantity Surveying from the 

University of Cape Town. She is the Director of the Sustainability oriented and Cyber 

Research Unit for the Built Environment at UCT. She is a registered Professional Quantity 

Surveyor with the South African Council for the Quantity Surveying Profession and a 

member of the Royal Institution of Chartered Surveyors, the Association of South African 

Quantity Surveyors, and the South African Facilities Management Association. Kathy is a 

Past-President of the South African Council for the Quantity Surveying Profession and  

served a four-year term as a Council Member on the Council for the Built Environment in 

South Africa. She was the Africa Market Seat on the Governing Council of the Royal 

Institution of Chartered Surveyors (2020 – 2023), is a current member of the Board of the 

Keynote Title:  

Sustainability through construction materials 

Professor Humphrey Danso 

AAMUSTED University, Ghana 

Keynote Title:  

Sustainable urban development and management 

Professor Kathy Michell 

University of Cape Town, South Africa 



xiv 

CIB International Council for Building and Construction Research, a member of the 

International Facilities Management Association Research Advisory Committee and is a 

founding Director of the Africa Facilities Management Association. 

. 

Ron Watermeyer served as the South African Institution of Civil Engineering’s 101st 

President in 2004. In 2009 he obtained a senior doctorate (Doctor of Engineering) from 

the University of the Witwatersrand for his engineering development work which has 

significantly contributed to the delivery of infrastructure for the advancement of a 

changing South African society. He has published more than 100 papers, articles and book 

chapters on various aspects on the delivery of infrastructure. He is currently a Trustee of 

Engineers Against Poverty (London based international charity), a visiting adjunct 

professor, School of Construction Economics and Management, University of the 

Witwatersrand, the Chair of ISO TC 59 / SC18 (Construction procurement), a Member of 

the Certification Board of FIDIC Credentialing Limited and a Director of Infrastructure 

Options (Pty)Ltd. Ron has been at the forefront of many development initiatives in South 

Africa since the early 1990s. He has reinterpreted building regulations, developed systems 

for the classification of sites in terms of geotechnical characteristics and building practice 

and established technical requirements for a structural warranty scheme for houses. He has 

also changed construction methods, technologies and practices to facilitate socio-

economic development imperatives and pioneered the development of construction 

procurement procedures, practices, tactics and strategy and client delivery management 

practices aimed at improving infrastructure project outcomes.  

Keynote Title:  

Role of the client in achieving sustainable built environments 

Professor Ron Watermeyer 

DEng, CEng, PrEng, PrCM, PrCPM, Hon.FSAICE, 

FIStructE, FICE, FSAAE 

University of the Witwatersrand, South Africa 



xv 

CONFERENCE THEMES 

• T1 Climate Responsive Built Environments

• T2 Air Quality and Healthy Building

• T3 Thermal Comfort and Intelligent Operation

• T4 Low Carbon Technology and Energy System

• T5 Sustainable Urban Renewal

• T6 Building Technology and Performance

• T7 Construction and Project Management

• T8 Real Estate and Property Management



xvii 

CONFERENCE PAPERS 

T1 CLIMATE-RESPONSIVE BUILT ENVIRONMENTS 1 

A unified data mining framework for air source heat pump performance prediction and key influencing factor 

analysis – Yang, Y., Lin, B., Geng, Y., Pei, X. and Ji, W 2

Addressing compliance checking matters of buildings to green standards using natural language processing: a 

review – Yamusa, M. A., Abdullahi, M., Ibrahim, Y. M., Ahmadu, H. A., and Abubakar, M. 11

Analyzing urban spatial agglomeration based on POI data: a case study of Shihezi city, Xinjiang – Han, Y., Liu, 

Q., Wu, X. and Su, Y. 21

Climate adaptation mechanism of traditional Yi dwellings from an e[m]ergy perspective – Ahou, Y., Yang, Z. 

and Xia, Q. 30

Exploration of the sustainable design strategies for the social houses of rural area – Wang, M. and Duan, D. 45

Female students’ perceptions of environmental sustainability: a case study of a university building in the UAE – 

Shibeika, A. 46

Hydrophobicity optimization and exploration of a novel building envelope material – Zhao, H., Wu, S., Wu, Y., 

Sun, H. and Lin, B. 57

Infrastructure in Johannesburg from a sustainable development perspective – Jia, S. and Yang, Y 65

Numerical simulation study on the effect of water diffuser on the performance of heat storage tank – Ren, Y., 

Ren, Z., Xiao, Y., Zhang, Z., Yang, Z., Pang, Y.. and E, Reaihan. 76

Research and application of ecological environment functional materials in China – Liang, J., Lei, Y., Han, X., 

Dong, B. Zhang, H., Zhang, N. and Wang, L. 88

Study of the moisture buffering characteristics of building envelopes with double-layer hygroscopic materials – 

Liu, S., Yan, T., Xu, X., Wan, H. and Huang, G. 99

Towards a decision support for green public procurement implementation: a review of the primary decision-

making factors – Yamusa, M. A., Abubakar, M., Nasir, R. M. and Abdulzaziz, M. 110

T2 AIR QUALITY AND HEALTHY BUILDING 118 

A study of the effect of air purifiers on the concentration of particulate matter in primary school classrooms – 

Luo, H., Chen, Y., Yuan, F. and Yao, R. 119

A study of the effect of indoor glare on personnel's emotions based on the PAD (Pleasure-Arousal-Dominance) 

emotion model – Li, H. Zhu, Y., Song, B. and Li, B. 127

An experimental study of human activity patterns on particle resuspension in a test chamber – Yuan, F., Luo, H. 

and Yao, R. 138

Analysis on occupant behavior and energy consumption characteristics of air conditioning usage in residential 

buildings – Ao, J., Chen, Z., and Du, C. 149

Characteristics and prediction of air conditioners use in residential rooms based on fractal theory – Fu, C., Liu, 

M. and Li, Z. 158

Distribution and characteristic analysis of indoor thermal environment monitoring points during air conditioning 

heating – Ding, Y., Yu, Z. and Liu, Y. 168

Evaluation of the aerosol transport behaviour and infection risk in an isolation ward by a CFD modelling – 

Wang, F., Wang, Y., Zhang, Y. and Xu, X. 179

Experimental study of the disinfection efficacy of microwave radiation on A. Variegatus attached to the filter – 

Zhang, Y., Xu, X., Wang, F. and Yan, T. 191

Implications of indoor heating terminals on allergic and respiratory diseases in childhood: repeated cross-

sectional surveys in China – Wang, C., Yu, W., Wei, S., Zhou, H. and Zhang, Y. 201

Indoor particles exposure and air filtration intervention association with children health-review – Zhong, T. and 

Du, C. 212

Investigating the impact of the indoor environmental quality of vehicles with different fuel transmission: 

emphasis on particulate matter –Mohamed, A. and Essah, E., A. 229

Relationship between indoor environment factors of residential settings and rheumatic diseases in older adults – 

Gao, N. and Yu, W. 240

Research on the correction method of carbon dioxide monitoring sensor in human respiratory zone – Wang H. 

and Yu W. 249



xviii 

Study on efficient removal technology of toluene from indoor ambient air – Wang, J., Chen, J., Chen, D. and 

Yang, C. 257

The effect of air purifiers on the concentration of particulate matter 2.5 -a review – Cheng, L. and Li, M. 267

The impact of different indoor mould concentrations on lung tissue inflammation in mice – Wu, M., Du, C., Ma, 

P. and Yang, X. 277

The impact of short-term exposure to different air-conditioning environments on human thermal adaptation – 

Pan, Y., Shi F., Sun Z., Guo S. and Yan H. 286

The optimal parameters of airflow comfort and validation in air conditioners – Liu, Y., Wang, B., Wang, Y., 

Fan, J., Zhang, L. and Zhu, Y. 296

T3 THERMAL COMFORT AND INTELLIGENT OPERATION 306 

A study of corrected skin wettedness levels in a downward outdoor-indoor temperature transient environment – 

Guo, J., Liu, H. and Chen, G. 307

A study of human perception and response under a high temperature local radiation – Wang, W., He, B. and Liu, 

H. 319

A study of the effect of long-term thermal history on thermal comfort of indoor occupants in summer climate 

chamber – Fu, Y. and Liu, H. 329

A study of thermal comfort experienced by postpartum women during sleeping hours – Wang, Y., Yu, W. and 

Shi, W. 339

Analysis of thermal comfort in indoor air conditioning environment for summer rest in hot summer and warm 

winter zone – Ma, X., Yu, W., Guo, L. and Zhang, Y. 349

Assessment of the effect of differences in human constitutional characteristics on thermal comfort – Ming, R., 

Miao, T., Wang, B., Wu, X. and Li, J. 358

Dynamic thermal sensation prediction model of the elderly using a machine learning method – Zhou, S. and Li, 

B. 367

Effects of different vegetation types on outdoor thermal comfort – Lei, H and Yuan, M. 377

Evaluation of the indoor environment and perceived IEQ: a case study in Norwegian primary schools – 

Chaulagain, A., Mathisen, H., M., Alam, B., G., Bartonova, A., Fredriksen, M., Høiskar, B., A., K., Gustavsen, 

K., Canet, A., M., Fredriksen, T. and Cao, G. 387

Experimental study on the thermal environment demand of underground stations in Chongqing – Ding, Y., Liu, 

Y. and Jiang, X. 399

Heat transfer analysis of separated gravity heat pipe used in a self-activated PCM wall – Xu, D., Yan, T., Xu, 

X., Wu, W., Long, W, Ming, T. and Wu, Y. 411

Human thermal comfort based dynamic regulation of air conditioning during cooling in residential buildings – 

Jing, M., Du, C., Zhang N. and Yu W. 422

Investigating the impact of infant BMI values on heat comfort perception in hot summer and cold winter regions 

– Shi, W., Yu W.，Zhou, H. and Wang Y. 439

Predictive control model for regional cooling system combined with ice storage technology – Tang, C. Bao, L. 

and Li, N. 449

Research on human thermal response in naturally slightly hotter indoor environment – Ding, Y. Zhou, Z. and 

Zhang, X. 459

Research on thermal comfort influence and improvement strategy of air conditioning dynamic environment – 

Guo, L., Yu, W., Zhang, Y. and Guo, R. 471

Study on the effect of light environment on human comfort in a warmer office environment – Bai, S., and Li, Z.

482

Study on thermal comfort of postpartum mothers in air-conditioned environment in summer – Huang, X., Yu, 

W., Du, C. and Zhou, H 494

Study on thermal sensation prediction and temperature satisfaction of air conditioning dynamic environment in 

winter – Zhang, Y., Yu, W. and Guo, L. 505

The spatiotemporal variation pattern of indoor thermal environment under different set temperatures in summer 

intermittent convective cooling environment – Guo, S., Sun, Z., Shi, F., Pan, Y. and Yan, H. 515

Thermal comfort in hot summer and cold winter area with retrofitted traditional electric heating devices (Huo 

Xiang) – Huang, D. and Liu, H. 527



 

 

xix 

 

Thermal comfort in urban parks: a review – Zheng, P. and Yao, R 536 
Thermal comfort prediction model based on optimized random forest algorithm – Jiang, Y. 545 

T4 LOW CARBON TECHNOLOGY AND ENERGY SYSTEM 555 

A general energy saving potential evaluation method of a pipe-embedded wall integrated with natural energy 

based on revised degree hour – Fan, S., Yan, T., Tang, X., Yu, Z., Li, X., Lyu, W. and Xu, X. 556 
A low-carbon distributed energy system suitable for residents in mountainous areas of southwestern China: a 

case study of Weining County in Guizhou Province – Zhang, Z. and Xiao, Y. 567 
Achieving sustainability in student housing: nexus of student housing design and energy use behaviour in 

northern Ghana – Appau, W. M., Anugwo, I. C., Attakora-Amaniampong, E. and Simpeh, F. 579 
Comprehensive life cycle assessment of carbon emissions in the construction industry: a review of methods, 

tools, and applications – Zhu, T., Hua, J., Huang, L. and Zhang, X. 591 
Developing an integrated real-time urban construction carbon emission monitoring framework: towards 

sustainable urban development – He, Y., Ding, Y., Jiang, X. and Zhao, W. 602 
Efficient matching method of cold and heat sources under dynamic load demand characteristics – Ding, Y., Yu, 

X., Jiang, X. and Zhao, W. 615 
Energy-saving optimal control strategy of an ASHP integrated central air-conditioning system – Gao, J., Yang, 

Y., Yan, J., Xu, X. and Liu, Y. 627 
Global energy-saving potential estimation of Radiative Sky Cooling (RSC) used in the pipe-embedded wall 

cooling system – Yan, T., Fan, S., Xu, X., Lyu, W., Ming, T. and Wu, Y. 639 
Low carbon concrete formulation and construction technology in construction phase – Zhao, W., Ding Y., Lai, 

W., Jiang, X. and He Y. 650 
Main accounting indexes of building carbon emissions – Ding, Y. and Chen, W. 661 

T5 SUSTAINABLE URBAN RENEWAL 672 

Application of Chinese traditional mural materials in modern architectural wall decoration – Liu, C. and Syed, 

A. S. A. B. 673 
Correlation between summer outdoor thermal environment and comfort in urban block of Northern Xinjiang – 

Su, Y., Huang, Z. and Wu, X. 674 
Effective integration of traditional mural elements to enhance the attractiveness and artistry of installation 

artworks through Ryan’s narrative theory – Li T., Feng Y., Ye Q. and Liu C. 685 
Enlightenment of American buildings to China of sustainable integrated design –Zhu, X. 696 
Environmental safety assessment of street road lighting combining visual characteristics and physical quantities 

– Liang, B., Huang, Z., Qin, Y., Li, Z.. and Luo, H. 707 
Exploration of interactive dream analysis installation in art design in community public facilities – Ye, Q., Zeng, 

C., Li, T. and Liu, C. 717 
Population, texture, green volume - the suitable density of historic districts based on intrinsic ecology – Hu, C. 

and Gong, C. 727 
Public art and cultural education function of digital exhibition in museums – Li, L. 728 
Research on design of interactive installation based on cultural sustainable intangible cultural heritage – Zeng, 

C., Ye, Q. , Li L. and Liu, C. 729 
Temporal and spatial distribution characteristics of thermal environment of subway stations - an experimental 

study – Ding, Y., Jiang, X., He, Y., Zhao, W., Liu, Y. and Hou, Y. 739 
The impact of road expansion on nearby infrastructure – the case of N11 in Mokopane, South Africa – Mogale, 

W., Musonda, E. and Harinarain, N. 752 
Utilising African epistemologies to augment Collaborative Online International Learning (COIL) initiatives in 

the built environment field – Qumbisa, N. and Makhwabe, S. 763 

T6 BUILDING TECHNOLOGY AND PERFORMANCE 775 

A review of the impact of office lighting environment on employees’ emotional state – Li, M. and Cheng, L.

 776 
A review on indoor lighting evaluation regarding its' effects and indicators – Lin, S. and Du, C. 785 



 

 

xx 

 

A study on the impact of enhancing window airtightness on residential building energy consumption in hot 

summer and cold winter region – Yu, Z., Zhang, C., Xu, X., Tang, X. and Yu, Z. 796 
An advanced design method of intelligent buildings in sustainable development – Xia, Q., Yang, Z. and Ahou, 

Y. 804 
Assessing the performance of PCM embedded non-linear thermal wall – Li, S., Ji, W., Liu, S., Kwanda, L, T. 

and Kusakana, K. 815 
Development and prospection of occupant behavior in residential building – Liao, X, and Li, B. 826 
Effect of elevated temperature on the mechanical properties of high volume recycled coarse aggregate concrete 

containing volcanic ash – Gambo, S., Yahaya, M. W. and Ibrahim, A. G. 837 
Effects of masonry materials characteristics on painted external wall surfaces – Afful, M. O., Mensah, S., 

Orgen, N. K. and Ameyaw, C. 845 
Experimental investigation on post-fire mechanical properties of Q960 ultra-high-strength steel after cold-

forming process – Wang, J. and Shi, Y. 855 
Housing characteristics and heat perception: comparison across formal and informal neighbourhoods in Lagos, 

Nigeria – Adegun, O. B., Morakinyo, T. E., Akinbobola, A., Obe, B. and Olusoga, O. O. 870 
Impact of PCMC roof on indoor thermal-humidity environment and air conditioning energy consumption – 

Jiang, L., Gao, Y., Liu, S. Rashidov, J., Zhang, X. and Fan, Z. 881 
Numerical study on lateral behavior of cold-formed steel composite shear wall – Xiaowei, R., and Yu, S. 893 
Research on the structural regulation of sepiolite fibre and application as self-humidity-control functional 

building materials – Han, X., Tang, R., Hao, L., Dong, B., Wang, L. and Liang, J. 903 
Sociotechnical system failure in construction projects: a distributed situation awareness of sky central roof 

damage – Mkandawire, K., Kabiri, S. and Connaughton, J. 915 
The effect of Melanopic equivalent daylighting illuminance (m-EDI) on satisfaction and productivity in the 

workplace – Li, Z., Yao, R., Bai, S. and Zhu, Y. 927 
Thermal properties of surrounding rock in deep-buried metro station fresh air shafts enhanced by phase change 

materials: a case study from Chongqing, China – Ren, Z., Ren, Y., Yang, Z., and Xiao, Y. 933 

T7 CONSTRUCTION & PROJECT MANAGEMENT 945 

A digital skills gap analysis of building inspectors: the case of the City of Johannesburg Metropolitan Authority 

– Gethe, F., Awuzie, B., Simbanegavi, P. and Chiloane, M. B. 946 
Academia-industry linkages: a missing link in TVET institutions in Tanzania - Mhando, Y., Mamboya, F. and 

Chacha, M. 957 
An evaluation of the quantitative risk assessment simulation undertaken during the planning stage of mega-

projects – Zwane, S., Schutte, D., Maila, S. and Jones, R. 967 
Ascertaining the knowledge of Ghanaian construction professionals on the use of clay bricks as a sustainable 

construction material – Darko, P., D., Nani, G., Mensah, N., A., A., O., Yusif, M. and Badii, P. 979 
Barriers to digitalization of procurement – a review – Ojo-Fafore, E. and Laryea, S. 989 
Bibliometric analysis of virtual reality in construction education – Kio, P., Ohochuku, C., Aduloju, T., and 

Agidani, J. 999 
Bibliometric review of social value in construction literature – Laryea, S., Kwasafo, O., K. and Mensah, S.

 1009 
Buried alive: the challenges facing the emerging contractors in the Limpopo province, South Africa – Moeti, 

M., Amoah, C. and Le Roux, L. 1021 
Challenges associated with differential measurements in stairs construction in low rise residential buildings – 

Boadi, E. O., Mensah, S., Ameyaw, C., Orgen, N. K. and Bondinuba, F. K. 1034 
Construction materials management techniques used in building projects in Kano Metropolis, Nigeria – Wudil, 

B. I., Bashir, K., K., Sani U., and Aikawa, I. U. 1044 
Detecting and preventing unbalanced bidding in South African public sector construction – Tilese, N., Makhaga, 

T., Mphahlele, M. and Zungu, Z. 1052 
Establishing success and failure factors of circular economy transitions in property development firms: a 

servitized business model approach – Nemakhavhani, R., Awuzie, B. and Aigbavboa, C. 1062 
Evaluating the new universities project outcomes using the PMBOK project performance domains – Mosalaesi, 

T. and Laryea, S. 1072 



 

 

xxi 

 

Exploring the challenges in the performance of small-medium contractors in South Africa: a consultants' 

perspective  – Simpeh, F., Baba, V. and Anugwo, I., C. 1083 
Fostering construction firm resilience through persuasive narratives of strategy: a conceptual framework – 

Zungu, Z., Laryea, S. and Nkado, R. 1093 
Investigate the potential impact of individual tracking technology in the Construction Industry – Lai, H., Y. and 

Essah, E., A. 1104 
Investigating management practices in the construction and delivery of electricity projects in Nigeria – Oladiran, 

O., J. and Oguntona, O., A. 1114 
Key barriers to green building implementation in South Africa – Mompati, L., Mandlate, M., Kabini, K. and 

Nomvalo, U. 1125 
Modelling leadership development determinants in Ghana’s construction industry: the moderating role of 

professional capability – Sam, A.. Aigbavboa, C. O. and Thwala, W. D. 1137 
Perceptions of tender document quality and its impact on construction estimates – Nezambe, B. Laryea, S. and 

Govender, N. 1151 
Risk factors that contribute to the collapse of major construction companies: the case of fallen South African 

construction giants – Scholtz, R., Deacon, H., A., Le Roux, L. and Amoah, C. 1159 
Team communication in the built environment: the South African land surveyor’s perspective – Harinarain, N. 

and Mbanjwa, S. 1175 
The job satisfaction of black female quantity surveyors – Punungwe, F. and Terblanche, R. 1185 
Understanding mason training in South Africa – Khuzwayo, B., Walker, M. and Graham, B. 1195 
Using dynamic BIM to improve construction safety culture  – Amiri, M., Saghatforoush, E. and Laryea, S.

 1205 
Wearable technology to reduce fatigue risks for construction workers: a scoping review – Mtetwa, S. I., Mollo, 

L., G. and Emuze, F., A. 1219 

T8 REAL ESTATE AND PROPERTY MANAGEMENT 1232 

Assessment of void periods in residential buildings in Minna, Nigeria – Ogunbajo, R. A. and Kuma, S. S.

 1233 
Prop-tech trend in Nigerian real estate practice: adoption and challenges – Araloyin, F. M., Fateye, T. B. and 

Adebowale, O. O. 1245 
Remote sensing to map and estimate the extent of flood damage – a South African case study – Malusi, B., 

Musonda, E. and Harinarain, N. 1257 
Residential choice and preferences in Ashesi University: comparative study of stated and revealed preferences – 

Doamekpor, N., A., A., Nyarko, G., K. and Adeku, V. 1269 
The impact of inflation on house prices in South Africa: effects of COVID-19 – Mpofu, B., Simbanegavi, P., 

Moobela, C. and Weaich, M. 1281 
Utilisation of digital elevation modelling to determine areas affected by floods in KwaZulu-Natal – Zwane, S., 

Musonda, E. and Harinarain, N. 1297 

INDEX OF AUTHORS 1309 

INDEX OF KEYWORDS 1313 

 



 

1 

 

T1 CLIMATE-RESPONSIVE BUILT ENVIRONMENTS 

 

A unified data mining framework for air source heat pump performance prediction and key influencing factor 

analysis – Yang, Y., Lin, B., Geng, Y., Pei, X. and Ji, W 2 
Addressing compliance checking matters of buildings to green standards using natural language processing: a 

review – Yamusa, M. A., Abdullahi, M., Ibrahim, Y. M., Ahmadu, H. A., and Abubakar, M. 11 
Analyzing urban spatial agglomeration based on POI data: a case study of Shihezi city, Xinjiang – Han, Y., Liu, 

Q., Wu, X. and Su, Y. 21 
Climate adaptation mechanism of traditional Yi dwellings from an e[m]ergy perspective – Ahou, Y., Yang, Z. 

and Xia, Q. 30 
Exploration of the sustainable design strategies for the social houses of rural area – Wang, M. and Duan, D. 45 
Female students’ perceptions of environmental sustainability: a case study of a university building in the UAE – 

Shibeika, A. 46 
Hydrophobicity optimization and exploration of a novel building envelope material – Zhao, H., Wu, S., Wu, Y., 

Sun, H. and Lin, B. 57 
Infrastructure in Johannesburg from a sustainable development perspective – Jia, S. and Yang, Y 65 
Numerical simulation study on the effect of water diffuser on the performance of heat storage tank – Ren, Y., 

Ren, Z., Xiao, Y., Zhang, Z., Yang, Z., Pang, Y.. and E, Reaihan. 76 
Research and application of ecological environment functional materials in China – Liang, J., Lei, Y., Han, X., 

Dong, B. Zhang, H., Zhang, N. and Wang, L. 88 
Study of the moisture buffering characteristics of building envelopes with double-layer hygroscopic materials – 

Liu, S., Yan, T., Xu, X., Wan, H. and Huang, G. 99 
Towards a decision support for green public procurement implementation: a review of the primary decision-

making factors – Yamusa, M. A., Abubakar, M., Nasir, R. M. and Abdulzaziz, M. 110 

 

 



Yang, et al. (2024) A unified data mining framework for air source heat pump performance prediction and key 

influencing factor analysis  In: Laryea, S. et al. (Eds) Proceedings of the WABER SuDBE Conference, 30 to 31 

July 2024, Johannesburg, South Africa 2-8 

2 

WABER SuDBE Conference 2024 

30 - 31 July 2024 

Johannesburg, South Africa 

ISBN: 978-0-7961-6032-4 

A UNIFIED DATA MINING FRAMEWORK FOR AIR 

SOURCE HEAT PUMP PERFORMANCE PREDICTION AND 

KEY INFLUENCING FACTOR ANALYSIS 

Yuren Yang1, Borong Lin2, Yang Geng3, Xingyu Pei4 and Wenjie Ji5 
1, 2, 3Key Laboratory of Eco Planning & Green Building, Tsinghua University, China; School of Architecture, 

Tsinghua University, China 
4, 5School of Mechanical Engineering, Beijing Institute of Technology, China 

Air source heat pumps (ASHPs) have significant emission reduction potential, low operating 

costs, low maintenance requirements, and suitability for various geographic conditions. 

Accurate prediction and analysis for the operational performance of ASHPs is crucial for 

developing optimal control strategies and ensuring the efficient and stable operation of heat 

pump systems. This study proposes a framework utilizing machine learning algorithms for 

performance prediction and interpretability analysis of ASHP systems. The proposed framework 

comprises three modules: data preprocessing, model construction, and interpretability analysis. 

It focuses on predicting the performance of ASHP systems based on real-time operational 

parameters, meteorological parameters, time indices, and other features, as well as extracting 

key influencing features and analysing the effects of each feature on system performance across 

different intervals. The proposed framework was applied to an ASHP system in a secondary 

school in Beijing, China, achieving an accuracy of 78.24% in predicting its heating capacity for 

the next hour. The interpretability analysis was performed using the SHAP method, which is 

based on cooperative game theory. The analysis revealed that instantaneous flow rate had the 

greatest impact on the heating capacity, but its influence varied significantly across different 

flow rate intervals. 

Keywords: air source heat pump (ASHP), heating capacity, interpretability analysis, machine 

learning 

INTRODUCTION 

Electrification of thermal energy can be achieved through traditional direct electric heating or 

heat pumps. Compared to direct electric heating, heat pumps offer higher efficiency and lower 

carbon emissions. The efficiency of heat pumps is three to four times that of fossil fuel systems  

(Wang, Wang & He 2022), thereby improving global heating energy efficiency. In 2020, the 

heat pump market exceeded $53 billion, with air source heat pumps (ASHPs) comprising over 

90% of the market (Huang et al. 2023). This dominance is attributed to their substantial 

emission reduction potential, low operating costs, minimal maintenance requirements, and 

adaptability to various geographic conditions. ASHPs are systems that utilize low-temperature 

heat energy in the air for heating or cooling purposes. They transfer heat energy from the air to 

1 yangyuren@mail.tsinghua.edu.cn 
2 linbr@tsinghua.edu.cn
3 gengy@tsinghua.edu.cn 
4 pxy0937@gmail.com
5 jiwenjie@bit.edu.cn  



Yang, et al. 

3 

 

the heating or cooling system, thereby regulating indoor temperatures (Carroll, Chesser & 

Lyons 2020). Compared to traditional heating and cooling systems, ASHPs offer advantages 

such as high efficiency, energy savings, environmental friendliness with no pollution, and 

stable operation safety. They are suitable for various building types, including residential, 

commercial, and industrial premises, making them a sustainable energy utilization method. 

This technology is of significant importance for reducing energy consumption and lowering 

carbon emissions. 

Accurate prediction for the operational performance of heat pumps is crucial for developing 

optimal control strategies and ensuring the efficient and stable operation of heat pump systems, 

especially considering the increasing prevalence of heat pump applications (Chen et al. 2023). 

For accurate prediction of heating capacity in ASHPs, it's essential to consider factors like 

outdoor temperature and humidity, heat carrier flow rate, compressor operating status, and 

supply and return water temperatures, among others. The operational state of ASHPs used in 

buildings or district heating stations is a complex nonlinear system. Modelling this system 

using conventional mathematical or physical models poses significant challenges. 

Additionally, ASHPs exhibit strong adaptability to their installation locations, which means 

that compared to other types of heat pump units like ground source heat pumps or water source 

heat pumps, the environmental conditions surrounding ASHPs are highly unstable, adding 

complexity to mathematical and physical model construction (Congedo et al. 2023). 

In recent years, the rapid advancement of artificial intelligence (AI) technologies has led to 

data-driven algorithms, represented by machine learning (ML), gradually replacing traditional 

modelling methods for various types of complex nonlinear systems. ML algorithms leverage 

large volumes of system operational data recorded and stored based on IoT technology. They 

use pattern recognition and data analysis to capture the complexity and nonlinear 

characteristics of systems, enabling accurate prediction of system behaviour and optimization 

control. Chen et al. trained an artificial neural network to predict the Coefficient of Performance 

(COP) of ground source heat pump units using environmental parameters and hourly power 

consumption data as input features, and the findings indicated that the ML algorithm achieved 

greater prediction accuracy compared to the empirical regression model (Chen et al. 2023). 

Eom et al. used a deep learning-based single model to quantitatively predict the variations in 

heating capacity, power consumption, and COP  of ASHPs due to frosting, which facilitated 

the optimization of defrost startup control strategies (Eom et al. 2021). Therefore, for complex 

nonlinear systems like ASHPs, ML algorithms hold the promise of better handling the 

complexity and variability of their operational states, providing new solutions for achieving 

efficient and stable system operation. 

Despite many researchers applying machine learning algorithms to predict the performance of 

ASHPs and other types of heat pump units, it is important to note that the black-box nature of 

these algorithms poses a significant barrier to their development and application (Yang et al. 

2022). The black-box nature refers to a model's capability to produce accurate predictions 

without offering clear explanations or insights into the underlying decision-making process. 

This lack of transparency makes it challenging to fully trust models, even if they perform 

exceptionally well when trained and deployed (Ribeiro, Singh & Guestrin 2016). Additionally, 

a thorough comprehension of how selected features impact final prediction outcomes enables 

improved model optimization through techniques like feature engineering (Ribeiro et al. 2016). 

Therefore, the application of ML algorithms in predicting the operational performance of 

ASHPs should balance model accuracy and interpretability. It should help building operators 

understand the impact of dynamic operational parameters such as environmental conditions 

and supply/return water temperatures on ASHPs' performance, facilitating the implementation 

of optimal control. 



Yang, et al. 

4 

 

The study proposes a comprehensive framework encompassing data pre-processing, model 

construction, and interpretability analysis, offering a dependable framework for predicting the 

operational performance and analysing key influencing factors of ASHPs. The performance of 

the methodology was validated using actual ASHP system from a campus in the cold region of 

China as the case study, and the key influencing factors affecting the heating performance of 

the case ASHP units were analysed. 

METHOD 

Data preprocessing 

Data preprocessing plays a crucial role in machine learning, often involving tasks such as 

feature engineering, and outlier cleaning to improve data quality and model performance (Yang 

et al. 2022). The proposed framework selects seven features across three dimensions: real-time 

operational parameters, meteorological parameters, and time indices, as shown in Table 1, as 

input parameters for the ML algorithm, with the heating capacity of ASHP system as the 

prediction target. Additionally, the proposed framework identifies and removes outlier data 

from the original data using the quartile method due to potential occurrences of abnormal data 

records during the operation of air source heat pump units, such as startup, shutdown, 

maintenance, or sudden changes in external environmental conditions.  

Table 1: Input parameters of ML algorithm 

Dimension Feature 

Real-time operational parameters Transient Flow Rate 

Real-time Power 

Supply Water Temperature 

Return Water Temperature 

Meteorological parameters Outdoor Temperature 

Time indices Weekday 

Hour 

Model construction 

Model selection 

Tree-based ML models are popular in nonlinear prediction models, particularly for tabular data. 

These models are often more accurate than deep learning models based on neural networks. 

Therefore, three typical tree-based ML algorithms, including Random Forest (RF), LightGBM, 

and XGBoost, are recommended in the proposed framework. RF operates by constructing 

multiple decision trees and aggregating their predictions, making it effective for both 

classification and regression tasks. LightGBM and XGBoost are efficient gradient boosting 

frameworks based on gradient boosting decision tree improvements. LightGBM adopts a 

histogram-based learning strategy and a Leaf-wise growth method, offering faster training 

speed and lower memory consumption (Ke et al. 2017). XGBoost, on the other hand, 

incorporates more regularization control and split point selection strategies, showcasing 

excellent predictive performance and robustness (Chen & Guestrin 2016). The final ML 

algorithm to be adopted can be determined based on actual data and hyperparameters 

optimization results. 

Hyperparameters tuning 

In the proposed framework, it is necessary to tune these hyperparameters within a certain range 

(as shown in Table 2) to find the optimal-performing model. 

 

 



Yang, et al. 

5 

 

Table 2: Hyperparameters tuning range 

Hyperparameters Tuning range Tuning step 

Number of trees 150-300 10 

Maximum depth of trees 1-30 1 

Minimum sample size of nodes 2-5 1 

 

Interpretability analysis 

In the proposed framework, SHAP (SHapley Additive exPlanations) is used for interpretability 

analysis of ASHP performance prediction models. The emphasis is on identifying critical 

features that have a significant impact on ASHP performance and evaluating the effects of each 

feature on ASHP performance across various intervals. SHAP is based on the Shapley value 

concept from game theory, where weighted averages are computed over permutations of 

feature values to assess each feature's contribution to the model's predictions (Lundberg et al. 

2020). This method treats different combinations of feature values as players in a game, with 

the model predictions as the game's outcome, and quantifies each feature's contribution to the 

outcome through Shapley value calculations. The key advantage of the Shapley value lies in 

its ability to meet the criteria for local accuracy, missingness, and consistency, making it a 

superior attribution method (Lundberg et al. 2020)(Molnar 2019). SHAP utilizes SHAP values 

as a standardized metric for feature importance in ML models (Lundberg & Lee 2017). It 

attributes output values to the Shapley value of each feature, operating under the additive 

feature attribution assumption. This approach provides a straightforward representation of 

complex functional relationships within black-box models by attributing features to the model 

output. 

CASE STUDY 

Introduction of the case dataset 

The ASHP system used in a middle school campus in Beijing, China, which was put into 

operation in 2022, is selected as the case in this study. The campus has an area of 30,500 m2 

and uses the ASHP system for both cooling and heating purposes. The ASHP system consists 

of 26 modular variable-frequency air-cooled heat pump units, as shown in Fig. 1(a), with the 

number of units starting and stopping controlled based on the return water temperature. 

Additionally, the system includes water pumps, a softening constant-pressure supplementary 

water system, a water treatment system, pressure differential control, electric valves, 

temperature and pressure monitoring, and other electric components, as shown in Fig. 1(b).  

In this study, the hour-by-hour heating data for the case system in December 2023 was used to 

produce the dataset. It should be noted that only data from the case system during stable 

operation were used for model training. Data during startup, shutdown, and maintenance 

periods were excluded to avoid interference with model performance from data during special 

operational stages. Finally, a total of 354 detailed hourly stable system operation data were 

retained for training of the ML algorithm after outliers were removed. 



Yang, et al. 

6 

 

 
Fig. 1 (a) modular variable-frequency air-cooled heat pump units; (b) schematic diagram of the case ASHP 

system 

Model training and selection 

During the training phase, 60% of the data were allocated for model training, while an 

additional 10% were set aside for validation to prevent overfitting. The remaining 30% of the 

data were kept separate and used solely to assess the performance of the trained model. After 

training within the tuning range, LightGBM exhibited the best predictive performance among 

the three tree-based ML algorithms on the testing data. Using input information such as ASHP 

system operational parameters, outdoor meteorological parameters, and time indices, it 

achieved an accuracy of 78.24% in predicting the ASHP system's heating capacity for the next 

hour. 

Interpretability analysis 

Global feature importance ranking and local explanation summary 

The feature importance ranking, combined with the summary plot, effectively illustrates the 

overall impact, prevalence, and trends of features, along with their local distributions across 

the dataset. To determine global feature importance, the absolute average of each feature's 

SHAP values was calculated and sorted in descending order. This approach helps mitigate 

heterogeneity effects in partial dependency analysis. A scatter plot was utilized to visually 

present all data points, arranged vertically based on feature importance. Each scatter point's 

position on the x-axis corresponded to the SHAP value of that feature for each instance, while 

the color gradient from red to blue indicated the feature's values, providing a clear 

representation of each instance's value. The presence of long tails on many features indicated 

significant impacts on specific samples, which may be overlooked in global analysis due to 

limited sample sizes or instances (Yang et al. 2022). 

 
Fig. 2 Feature importance ranking and summary plot 

For the case ASHP system, as shown in Fig. 2, the transient flow rate, real-time power, and 

return water temperature had the most significant impact on the heating capacity for the next 



Yang, et al. 

7 

 

hour. Analysing the local explanation summary plot revealed a notable long tail in the transient 

flow rate, indicating that higher flow rates had a significant positive effect on increasing the 

heating capacity for the next hour, while lower flow rates had a significant negative effect. 

Similarly, the long tail in outdoor temperature indicated a strong correlation between lower 

outdoor temperatures and higher heating capacity for the next hour. The combination of global 

feature importance ranking and local explanation summary effectively revealed the key 

features influencing ASHP heating performance. Based on the theoretical foundation of 

cooperative game theory, these features could be quantitatively assessed for their impact using 

a unified metric, providing clear insights into their respective influence levels.  

Feature effects across various intervals 

As observed in the local explanation summary plot, certain features had varying or even 

opposing impacts on ASHP heating performance across various intervals. The outdoor 

temperature and transient flow rate, two features with evident long tails in the local explanation 

summary plot, exhibited varying impacts on the heating capacity across different intervals, as 

shown in Fig. 3. The horizontal axis showed the range of the selected feature values, while the 

vertical axis displayed the corresponding changes in SHAP values. The colour gradient from 

red to blue in the colour bar represented the change in feature values from large to small. From 

Fig. 3(a), it could be observed that when the outdoor temperature is below -7.5°C, there was a 

significant positive effect on increasing the heating capacity of the ASHP system for the next 

hour. However, when the temperature was above -7.5°C, the impact of outdoor temperature on 

ASHP heating capacity didn’t show significant differences. This indicated that -7.5°C was a 

critical threshold for the outdoor temperature's influence on the heating load of the case campus 

buildings. The effect of outdoor air temperature on the amount of heating capacity of ASHPs 

reflects the actual combined effect on the heating demand of the case building with the fixed 

envelope and the heat storage capacity of the case building envelope (Ling et al. 2020). Existing 

studies have also shown that for buildings with fixed envelope parameters, projecting heating 

demand based on time-by-time outdoor air temperatures and implementing further optimized 

control can achieve significant energy savings (Ling et al. 2020)(Zhou et al. 2020), which also 

proves the significance of the proposed framework for building operation control. From Fig. 

3(b), it could be observed that although transient flow rate was the feature with the largest 

overall impact on the heating capacity for the next hour, its effect was relatively small when 

the flow rate was between 440 and 490. However, when the flow rate was above 490, there 

was a significant positive effect on increasing the heating capacity, while below 440, there was 

a notable negative effect on the heating capacity. 

 
Fig. 3 Feature effects across various intervals (a) outdoor temperature; (b) transient flow rate 

CONCLUSIONS 

In this study, a framework based on ML algorithms for performance prediction and 

interpretability analysis of ASHP systems is proposed. The proposed framework comprises 



Yang, et al. 

8 

 

three modules: data preprocessing, model construction, and interpretability analysis. It focuses 

on predicting the performance of ASHP systems based on real-time operational parameters, 

meteorological parameters, time indices, and other features, as well as extracting key 

influencing features and analysing the effects of each feature on system performance across 

different intervals. The proposed framework was applied to an ASHP system in a secondary 

school in Beijing, China, achieving an accuracy of 78.24% in predicting its heating capacity 

for the next hour. The performance interpretability analysis of the ML algorithm was performed 

using the SHAP method, which is grounded in cooperative game theory. The analysis indicated 

that the instantaneous flow rate had the most significant effect on the ASHP system's heating 

capacity, with its impact differing greatly across various flow rate intervals. The proposed 

framework demonstrates significant potential in predicting the heating capacity of ASHP 

systems and providing precise analysis, which can greatly assist building maintenance 

personnel in fully understanding ASHP system performance and achieving optimal control.  

ACKNOWLEDGEMENTS 

This study is supported by the China National Key Research and Development Program (Grant 

No. 2023YFC3306400), the Key Program of National Natural Science Foundation of China 

(Grant No. 52130803), the International (NSFC-NWO) Joint Research Project of National 

Natural Science Foundation of China (Grant No. 52161135201), and the Young Scientists Fund 

of the National Natural Science Foundation of China (Grant No. 52208113). 

REFERENCES 

Carroll, P., Chesser, M. & Lyons, P., 2020, Air Source Heat Pumps field studies: A systematic literature 

review, Renewable and Sustainable Energy Reviews. 

Chen, T. & Guestrin, C., 2016, XGBoost: A scalable tree boosting system, Proceedings of the ACM 

SIGKDD International Conference on Knowledge Discovery and Data Mining. 

Chen, Y., Kong, G., Xu, X., Hu, S. & Yang, Q., 2023, ‘Machine-learning-based performance prediction 

of the energy pile heat pump system’, Journal of Building Engineering. 

Congedo, P.M., Baglivo, C., D’Agostino, D. & Mazzeo, D., 2023, ‘The impact of climate change on 

air source heat pumps’, Energy Conversion and Management. 

Eom, Y.H., Chung, Y., Park, M., Hong, S. Bin & Kim, M.S., 2021, ‘Deep learning-based prediction 

method on performance change of air source heat pump system under frosting conditions’, 

Energy. 

Huang, S., Yu, H., Zhang, M., Qu, H., Wang, L., Zhang, C., Yuan, Y. & Zhang, X., 2023, ‘Advances, 

challenges and outlooks in frost-free air-source heat pumps: A comprehensive review from 

materials, components to systems’, Applied Thermal Engineering. 

Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q. & Liu, T.Y., 2017, LightGBM: A 

highly efficient gradient boosting decision tree, Advances in Neural Information Processing 

Systems. 

Ling, J., Tong, H., Xing, J. & Zhao, Y., 2020, ‘Simulation and optimization of the operation strategy of 

ASHP heating system: A case study in Tianjin’, Energy and Buildings. 

Lundberg, S. & Lee, S.-I., 2017, ‘A Unified Approach to Interpreting Model Predictions’, Advances in 

Neural Information Processing Systems, 2017-Decem(Section 2), 4766–4775. 

Lundberg, S.M., Erion, G., Chen, H., Degrave, A., Prutkin, J.M., Nair, B., Katz, R., Himmelfarb, J., 

Bansal, N. & Lee, S.-I., 2020, ‘From local explanations to global understanding with 

explainable AI for trees’, Nature Machine Intelligence, 2(January), 56–67. 



Yang, et al. 

9 

 

Molnar, C., 2019, ‘Interpretable Machine Learning. A Guide for Making Black Box Models 

Explainable.’, Book. 

Ribeiro, M.T., Singh, S. & Guestrin, C., 2016, ‘Model-Agnostic Interpretability of Machine Learning’, 

(Whi). 

Wang, Y., Wang, J. & He, W., 2022, Development of efficient, flexible and affordable heat pumps for 

supporting heat and power decarbonisation in the UK and beyond: Review and perspectives, 

Renewable and Sustainable Energy Reviews. 

Yang, Y., Yuan, Y., Han, Z. & Liu, G., 2022, ‘Interpretability analysis for thermal sensation machine 

learning models: An exploration based on the SHAP approach’, Indoor Air, 32(2), 1–24. 

Zhou, C., Ni, L., Wang, J. & Yao, Y., 2020, ‘Investigation on the performance of ASHP heating system 

using frequency-conversion technique based on a temperature and hydraulic-balance control 

strategy’, Renewable Energy. 

 

 



 
 

 

Yamusa, et al. (2024) Addressing compliance checking matters of buildings to green standards using natural 

language processing: a review  In: Laryea, S. et al. (Eds) Proceedings of the WABER SuDBE Conference, 30 to 

31 July 2024, Johannesburg, South Africa 11-19 

11 

 

WABER SuDBE Conference 2024 

30 – 31 July 2024 

Johannesburg, South Africa 

ISBN: 978-0-7961-6032-4 

ADDRESSING COMPLIANCE CHECKING MATTERS OF 

BUILDINGS TO GREEN STANDARDS USING NATURAL 

LANGUAGE PROCESSING: A REVIEW 

Muhammad Aliyu Yamusa1, Muhammad Abdullahi2, Yahaya Makarfi Ibrahim3, 

Hassan Adaviriku Ahmadu4 and Mu’awiya Abubakar5 
1,2Department of Quantity Surveying & Public Procurement Research Centre, Ahmadu Bello University, Zaria, 

Nigeria 
3,4Department of Quantity Surveying, Ahmadu Bello University, Zaria, Nigeria 
5Department of Building & Public Procurement Research Centre, Ahmadu Bello University, Zaria, Nigeria 

The incorporation of sustainability objectives in green building (GB) projects adds complexity 

to their design, construction, and management. Current developments in the area of artificial 

intelligence, precisely natural language processing (NLP) techniques have provided great 

potential in analysing voluminous regulatory documents to draw insightful information relating 

to the standards, requirements, and codes to enhance the efficiency and accuracy of compliance 

checking. However, there is a dearth of attempts to tap the potential of NLP to facilitate 

automated compliance checking, especially within the context of green buildings. This paper, 

therefore, aims to assess the benefits and limitations of the current advancements in NLP-based 

methodologies for automated compliance checking of regulatory documents in green buildings. 

This paper conducts a systematic review of literature to achieve its aim. The challenges and 

benefits, as well as the areas of the application of NLP in automated compliance checking of 

regulatory documents in green buildings, are highlighted. The research offers a guide for future 

investigations aimed at broadening the utilisation of NLP in automating the compliance 

verification process for regulatory documents in green buildings and the construction sector as 

a whole. 

Keywords: compliance checking, green building, NLP, regulatory documents, standards 

INTRODUCTION 

It is necessary to evaluate design alternatives in light of the projects’ objectives and anticipated 

benefits. Before the design is finalised and approved, an assessment is necessary to identify 

and expose discrepancies between requirements and design, as well as regulatory constraints 

(). These cycles of analysis and assessment enhance the quality of design in the context of 

building design by guaranteeing that the goals of clients and stakeholders are met (Kamara et 

al., 2000) and that design solutions adhere to regulations as well as legal requirements 

 
1 yamusajf@yahoo.com 
2 bnabdallah02@gmail.com 
3 makarfi@gmail.com 
4 ahmaduhassan@rocketmail.com 
5 muawiyaabubakar1@gmail.com 



Yamusa, et al. 

12 

 

(Dimyadi & Amor, 2013). Thus, design assessment offers a chance to enhance value generation 

and facilitates the removal of flaws (Formoso et al., 2011). 

The suggestion to utilise automation for design compliance checking has been put forward as 

a significant approach to navigate through this particular process. Existing literature 

emphasizes numerous potential advantages that can be derived from this approach (Eastman et 

al., 2009). Automation enables the connection and coordination of various forms of information 

within building models (Laakso & Kiviniemi, 2012), thereby facilitating the verification of 

design compliance in a quicker, more effective, and dependable means (Eastman et al., 2009). 

Several studies have proposed various approaches to tackle the challenges associated with 

automated compliance checking (ACC). These challenges include converting rule 

interpretation and representation into a format that can be understood by computers (Zhang et 

al., 2023), preparing building design model data for verification (Solihin et al., 2020), and 

developing automated compliance checking systems using various techniques that include 

Artificial Intelligence (AI) and Natural Language Processing (NLP) (Kim et al., 2020). Despite 

these efforts, the practical implementation of automated compliance checking has faced limited 

success. This is primarily due to the complexities associated with complying with numerous 

regulatory requirements (Eastman et al., 2009) and the technological limitations that hinder the 

support of this process (Solihin et al., 2020). 

Ensuring compliance becomes even more challenging within the context of Green Buildings 

(GB) design. GBs, together with their sustainable technologies, are regarded as innovative 

solutions that require various changes at different levels, including process, organization, 

industry, and policy. These changes bring about fundamental transformations in decision-

making stages, design approach, procurement process, actors, tasks, roles, competencies, and 

team cultures necessary for the successful delivery of GB projects (Qazi et al., 2021). Such 

transformations demand the implementation of a novel range of technologies for designing, 

constructing projects, as well as procurement processes, thereby introducing extra tiers of 

significant unpredictable risks and uncertainties (Hwang et al., 2017). GB projects are required 

to implement intricate environmental considerations during the standard design phases to 

ensure the optimal functioning of the building over its entire lifespan. This objective is 

accomplished by integrating numerous requirements that address the environmental impact of 

construction projects. These requirements that enable GB projects meet the desired green 

standards are outlined in the green building regulations. To address these challenges and 

facilitate the incorporation of these requirements, the implementation of automated checking 

systems in design were proposed as a crucial approach to manage and enhance quality, as well 

as addressing the intricacies of designs (Nicholas, 2012). This can effectively mitigate the 

considerable uncertainties and unpredictable risks associated with GB projects by ensuring 

conformance to green standards. Makisha & Rybakova (2021) and Rybakova & Makisha 

(2021) proposed a model for evaluating the conformity of contemporary building design and 

construction with "green" certification standards. Their study identified the BRE 

Environmental Assessment Method (BREEAM), Leadership in Energy and Environmental 

Design (LEED), and Deutsche Gesellschaft für Nachhaltiges Bauen (DGNB) to demonstrate 

strong compliance checking susceptibility. 

Recent advances in AI and NLP techniques have provided great potential in analysing 

voluminous regulatory documents to draw insightful information relating to the standards, 

requirements, and codes to enhance the efficiency and accuracy of compliance checking. 

However, there is a dearth of attempts to tap the potential of NLP to facilitate automated 

compliance checking, especially within the context of green buildings. Therefore, this paper 

aims to assess the benefits and limitations of the current advancements in NLP-based 



Yamusa, et al. 

13 

 

methodologies for automatically checking the compliance of regulatory documents in green 

buildings. This will provide response to the research question: what are the benefits and 

limitations of NLP-based methodologies to automate the compliance checking of regulatory 

documents in green buildings? 

The structure of this paper is organised as follows. The second section establishes the 

methodology employed for the research. Subsequently, the review and findings of this research 

are presented. Lastly, the paper concludes by offering some final remarks. 

METHOD 

This research adopted an exploratory research approach by identifying publications related to 

Natural Language Processing (NLP) techniques for Compliance Checking of building 

construction towards green standards. The search queries utilized for this purpose encompass 

NLP techniques such as 'Artificial Intelligence', 'Machine learning', 'Natural Language 

Processing', 'NLP', and 'Semantic-NLP'. Additionally, the queries include terms related to 

building construction such as 'Building', 'Construction', 'Design', 'Architecture Construction 

and Engineering', and 'AEC industry'. Furthermore, terms related to compliance checking such 

as 'Compliance Checking', 'Automated compliance checking', 'Regulatory requirements', 

'Regulatory taxonomy', and 'Regulatory requirements' were also incorporated. The Scopus 

database was chosen for the search as it contains relevant publications and has been utilized in 

similar studies (Yamusa et al., 2023). The inclusion criteria for the publications were limited 

to those in the English language and falling under the categories of 'Construction Building 

Technology' and 'Engineering Civil', without any restrictions on the publication year or 

document type. The final publications were reviewed to identify and consolidate the key 

findings and lessons learned from these studies, which serve as the foundation for the outcomes 

of this research. 

LITERATURE REVIEW 

Introduction to natural language processing and its application 

The objective of Natural Language Processing (NLP), a division of Artificial Intelligence (AI), 

lies in enabling computers to comprehend and analyse speech as well as text in natural language 

in a way similar to humans (Cherpas 1992). NLP has facilitated the development of various 

applications, such as automated text summarization and automated natural language translation 

(Marquez 2000). Tokenization, semantic role labelling, part-of-speech (POS) tagging, (Gildea 

and Jurafsky 2002), named entity recognition (Roth and Yih 2004), and other NLP subtasks 

exemplify the techniques employed in NLP. Two primary methods for handling NLP problems 

exist: the machine learning (ML) and rule based. While the ML-based method utilizes ML 

algorithms for processing of text (Pradhan et al., 2004), the rule-based technique includes 

establishing code rules manually (Soysal et al., 2010). The rule-based approach makes use of 

additional human input in developing the rules. However, if often exhibits superior 

performance in processing the text (Crowston et al., 2010). In the field of construction, several 

noteworthy research endeavours have employed NLP techniques. Several studies have 

developed machine learning-based models for tender classification and spend categorisation 

(Abdullahi et al., 2024a, 2024b; Yamusa et al., 2024). 

Application of NLP in Compliance Checking 

Text analysis using NLP technology 

NLP is widely utilized for various purposes including information retrieval, incident 

investigation, and document correlation in the architecture, engineering, and construction 



Yamusa, et al. 

14 

 

(AEC) sector (Fernando Sanchez-Rada et al., 2020). NLP has been applied for the 

identification and extraction of semi-structured and structured information in documents. This 

is achieved through the utilization of techniques such as ML approach and rule-based approach 

(van Dinter et al., 2021). In order to ensure the accurate retrieval of semantic information from 

text, Altuncan and Tanyer (2018) employed shallow parsing (SP) rules in extracting semantic 

knowledge in construction contract text. Furthermore, Zhou & El-Gohary (2017) developed an 

NLP technique based on rule processing that incorporated semantic rules to automatically 

extract information from construction contract documents. Scholars have also explored the 

extraction of information using semantic relations and semantic similarity, thereby enhancing 

the application of semantic rules (Xu and Cai, 2020). NLP integrated with ML is well-suited 

for documents with high complexities and irregularities and holds great capability for analysing 

extensive documents. Sinoara et al. (2019) proposed an ML tool for training the semantic 

connection among corresponding abstracts and patents, enabling the identification of key terms 

(such as sentences, phrases and words) in patent texts. The field of topic and sentiment analysis 

from texts is rapidly advancing in the realm of NLP. Its purpose is to capture opinions and 

perspectives expressed by writers or a wider populace (Balahur et al., 2012). The sentiments 

found in documents are highly subjective, encompassing emotions, opinions, specifications, 

sentiments, and beliefs (Montoyo et al., 2012). Implicit emotions, in contrast to explicit ones, 

are more prone of being hidden within the text (Peng et al., 2020). 

Within the AEC industry, the trend of digitization has resulted in project documents that reflect 

the perspectives of managers regarding project attributes and work sections. For example, 

Marzouk and Enaba (2019) identified corpus related to conditions in construction contracts and 

mined significant keywords from these contracts. NLP techniques are utilized to develop a 

comprehensive framework for the analysis of contract conditions, which represent the desires 

and expectations of project owners (Hassan & Le, 2020). 

Information extraction 

Information extraction (IE) is a process that aims to automatically extract structured 

information from unstructured textual data. This information includes entities and their 

associated attributes. The main challenge in IE is that computers often struggle to comprehend 

and process textual data. Fader et al. (2011) considered the lexical and syntactic text 

characteristics for developing rules for extracting information. Their objective was aimed at 

extracting web statements to support common-sense knowledge and question answering. 

Gutierrez et al. (2016) applied error detection IE rules on ontologies for biology for extracting 

information from documents within the biology domain. 

In addition to these efforts, researchers have explored various procedures in reducing IE rules 

development costs. One approach involves using statistical learning algorithms to learn IE rules 

from natural text (Chambers & Jurafsky, 2011). Another approach involves the design of 

straightforward interactive environments and coding for developing rules (Valenzuela-

Esc´arcega et al., 2015). Furthermore, there have been attempts to integrate current rule NLP 

applications and programming languages into a unified platform for developing rules (Kluegl 

et al., 2016). These endeavours aim to streamline the process of creating IE rules and improve 

its efficiency. 

Several IE techniques leverage on simple supervised learning algorithms, in conjunction with 

manually engineered semantic and syntactic characteristics, including the named entity 

recognition (NER) technique proposed by Zhou and Su (2002) utilizing the hidden Markov 

algorithm, an NER method introduced by Li et al. (2004) based on support vector machines, 

and an IE approach developed by Finkel et al. (2005) using conditional random fields. In the 

field of architectural, engineering, and construction (AEC), Zhang & El-Gohary (2021) have 



Yamusa, et al. 

15 

 

created an IE approach based on deep neural network to extract syntactic and semantic 

information from regulatory texts. Additionally, Wu & Ma (2024) have designed an NLP-based 

framework for retrieving requirement-related documents. 

Rule-based NLP using pattern-matching-based rules 

Pattern-matching-based rules play a crucial role in a variety of NLP activities, such as text 

understanding (Goh et al., 2006), POS tagging (Yin and Fan, 2013), and IE (Califf and Mooney, 

2003). The core idea behind these rules is to define specific outcomes when a particular 

sequence pattern of elements is identified. These rules are implemented in diverse ways to suit 

different purposes and fields. Nonetheless, they all follow the mapping of "from pattern to 

result" or a common rule format of "if pattern then result". Zhang & El-Gohary (2015) 

introduced a semantic NLP approach based on the rule approach to automate the processes of 

information extraction and transformation. 

Semantic modelling and semantic NLP 

Semantic models are structured frameworks designed to represent the meanings within a 

specific domain or topic. Among the various types of semantic models, ontologies are widely 

utilized. An ontology is a formal and explicit specification of a conceptualization, typically 

comprising axioms, concept hierarchies, and relationships between concepts. These axioms, in 

conjunction with concepts and their relationships, serve to describe the semantic interpretation 

of the conceptualization. Ontologies are valued for their reusability and extensibility, allowing 

for easy adaptation and expansion. The utilization of semantic models, such as ontologies, can 

prove advantageous in natural language processing tasks. Research indicates that semantic-

based information extraction yields superior performance compared to approaches reliant 

solely on syntax (Zhang and El-Gohary, 2013). 

CONCLUSIONS 

Automated compliance checking continues to be a significant area of interest for researchers 

in the construction industry due to its crucial role in facilitating the verification of design 

compliance in a more efficient and dependable manner. The application of artificial intelligence 

(AI), specifically natural language processing (NLP), in automating compliance checking 

offers numerous advantages for ACC practices. One domain that can significantly benefit from 

the potential of NLP to facilitate its automated compliance checking is the green buildings 

domain. This study therefore explores the potential of NLP for compliance checking and how 

the green buildings domain can benefit from its application for its compliance checking. 

The findings from this study show that the utilization of NLP in compliance checking can assist 

in automatically identifying and extracting organized information from regulatory documents, 

typically in unstructured textual data format. Furthermore, the findings reveal that NLP is 

employed in conjunction with other techniques, such as rule-based approaches that rely on 

pattern matching, semantic modelling based on ontologies, and machine learning (ML) that 

relies on data trends and patterns from training text data. However, it is observed that the ML-

based approach to utilizing NLP, which requires no human input, in ACC has demonstrated 

lower accuracy in comparison to the rule-based approach. 

To enhance the accuracy of the ML-based approach, it is crucial to have high-quality and 

structured data. Additionally, the size of the dataset should be sufficient to prevent overfitting 

or underfitting of the models. Nevertheless, it is essential to observe that the use of AI in 

compliance checking cannot replace the expertise of green building professionals. This is 

because domain knowledge plays a vital role in fitting the model, selecting appropriate 



Yamusa, et al. 

16 

 

predictors, understanding the data, and effectively deploying the models in practice, especially 

within a complex domain such as green buildings. 

Moreover, green building professionals should adopt and exploit these techniques rather than 

perceive them as rivals or dangers. They ought to acknowledge the significance that NLP 

techniques bring to their field and utilize them as instruments to augment their own proficiency. 

Additionally, it is crucial to recognize that this research constitutes a crucial analysis centred 

on the influence of NLP on automated compliance checking practices. Nevertheless, the 

analysis is restricted by the scope of the utilized publications. Future investigations should 

employ a systematic review methodology to comprehensively assess the impact of NLP on 

automated compliance checking practices. These investigations should also address the 

obstacles and prospects associated with NLP implementation and offer case validations to 

further enhance the comprehension of its potential advantages. 

ACKNOWLEDGEMENT 

The study was conducted under the Public Procurement Research Centre of Ahmadu Bello 

University, Zaria. 

REFERENCES 

Abdullahi, B., Ibrahim, Y.M., Ibrahim, A.D., Bala, K., Ibrahim, Y., & Yamusa, M. A. (2024a). 

Development of Machine Learning Models for Classification of Tenders based on UNSPSC 

Standard Procurement Taxonomy. International Journal of Procurement Management, 19(4), 

pp. 445-472. http://dx.doi.org/10.1504/IJPM.2023.10056755  

Abdullahi, B., Ibrahim, Y.M., Ibrahim, A.D., Bala, K., Ibrahim, Y., & Yamusa, M. A. (2024b). 

Development of Machine Learning Models for Categorisation of Nigerian Government’s 

Procurement Spending to UNSPSC Procurement Taxonomy. International Journal of 

Procurement Management, 19(1), pp. 106-121. https://doi.org/10.1504/IJPM.2024.135175 

Altuncan, I. U., & Tanyer, A. M. (2018). Context-Dependent Construction Conflict Management 

Performance Analysis Based on Competency Theory. Journal of Construction Engineering and 

Management, 144. 

Balahur, A., Hermida, J. M., & Montoyo, A. (2012). Detecting implicit expressions of emotion in text: 

A comparative analysis. Decision Support Systems, 53, 742–753. 

Califf, M. E., and Mooney, R. J. (2003). Bottom-up relational learning of pattern matching rules for 

information extraction. J. Mach. Learn. Res., 4, 177–210. 

Chambers, N. & Jurafsky, D. (2011). Template-based information extraction without the templates, in: 

Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: 

Human Language Technologies, June, pp. 976–986. https://www.aclweb.org/anthology/P11-

1098 (Aug. 01, 2023). 

Cherpas, C. (1992). Natural language processing, pragmatics, and verbal behaviour. Anal. Verbal 

Behav., 10, 135–147. 

Crowston, K., Liu, X., Allen, E., and Heckman, R. (2010). Machine learning and rule-based automated 

coding of qualitative data. Proc., 73rd ASIS&T Annual Meeting: Navigating Streams in an 

Information Ecosystem, Association for Information Science and Technology, Silver Spring, 

MD, 1–2. 

Dimyadi, J., & Amor, R. (2013). Automated Building code compliance checking – Where is it at?, in: 

Proceedings of CIB WBC 2013, pp. 172–185, https://doi.org/10.13140/2.1.4920.4161. 



Yamusa, et al. 

17 

 

Eastman, C., Lee, J., Jeong, Y., & Lee, J. (2009). Automatic rule-based checking of building designs, 

Autom. Constr. 18 (8), 1011–1033, https://doi.org/10.1016/j.autcon.2009.07.002.  

Fader, A., Soderland, S. & Etzioni, O. (2011). Identifying relations for open information extraction, in: 

Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, 

July, pp. 1535–1545. https://www.aclweb.org/anthology/D11-1142 (Aug 01, 2023). 

Fernando Sanchez-Rada, J., Araque, O., Iglesias, C.A., 2020. Senpy: A framework for semantic 

sentiment and emotion analysis services. Knowledge-Based Systems 190. 

Finkel, J. R., Grenager, T. & Manning, C. (2005). Incorporating non-local information into information 

extraction systems by Gibbs sampling, in: Proceedings of the 43rd Annual Meeting on 

Association for Computationa