Construction waste management model for contractors using lean and BIM tools

Universidade do Minho Escola de Engenharia Mahmoud Abdel Fattah Abdel Athim Karaz Construction Waste Management Model for Contractors Using Lean and BIM Tools October 2024 UMinho | 2024 Mahmoud Abdel Fattah Abdel Athim Karaz Construction Waste Management Model for Contractors Using Lean and BIM Tools Universidade do Minho Escola de Engenharia Mahmoud Abdel Fattah Abdel Athim Karaz Construction Waste Management Model for Contractors Using Lean and BIM Tools October 2024 Philosophy Doctorate Thesis Civil Engineering Work conducted under supervision of: Professor Doctor José Manuel Cardoso Teixeira ii Copyright and Terms of Use of This Work by a Third Party This thesis is academic work that can be used by third parties as long as internationally accepted rules and good practices regarding copyright and related rights are respected. Accordingly, this work may be used under the license provided below. If the user needs permission to use the work under conditions not provided for in the indicated licensing, they should contact the author through the RepositoriUM of Universidade do Minho. License granted to the users of this work Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International CC BY-NC-SA 4.0 https://creativecommons.org/licenses/by-nc-sa/4.0/deed.en iii Acknowledgements This research results from a critical life stage, demonstrating persistence and consistency towards pursuing knowledge. During the past years, many people have been met with favourable and directly and indirectly influenced my life, therefore this investigation. I am grateful to myself and all of them, as they have pushed me and shaped the way I am today, formulated the current knowledge, and shared beautiful experiences. Firstly, I would be specifically profoundly indebted to my advisor, Professor José Manuel Cardoso Teixeira, whose unwavering guidance, invaluable advice, patience, and devoted support have been instrumental throughout this journey. His mentorship was fundamental in shaping this academic endeavour and inspired personal growth and resilience. I extend heartfelt gratitude to the members of this dissertation committee members, for their expertise, constructive feedback, and commitment to excellence. Their collective wisdom has enriched this research and contributed to its scholarly rigour. I am grateful to the University of Minho for providing a conducive environment for intellectual inquiry and scholarly pursuit. The university's resources, facilities, and opportunities have been indispensable in completing this dissertation. I wish to thank my colleagues and peers for their camaraderie, encouragement, and stimulating discussions. Your camaraderie has made this academic journey more enriching and enjoyable. Special thanks are dedicated to my mother, father, sisters, and brothers for their unconditional love, encouragement, and unwavering belief in my abilities. Their steadfast support has been my anchor during challenging times. I would express my gratitude to my wife, who was always beside me to motivate me when I was down. I want to thank my friend Mohammed Al Darabseh for his company while living abroad and during this PhD journey. This research has been supported by the Portuguese Foundation for Science and Technology (FCT) under the doctoral grant SFRH/BD/04751/2021. iv STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism, any form of undue use of information, or falsifying results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. The research is not directed towards any specific demographic warranting distinctive ethical attention. Given its focus on professionals such as contractors, architects, trades, and lean practitioners, the primary ethical concerns centre around safeguarding the privacy, confidentiality, and anonymity of participating companies, subjects, and questionnaire respondents. The literature review upheld meticulous citation and electronic management for all scholarly sources. Additionally, the sources of data were duly acknowledged during the dissemination of research outcomes to enhance the reliability of the data. v Resumo Os resíduos de construção continuam a ser um desafio significativo para o sector, dificultando a produtividade e desencorajando a inovação. Vários relatórios descrevem os seus impactos adversos, tais como o obstáculo ao desenvolvimento da sustentabilidade ambiental, económica e social. Os académicos, os reguladores e as sociedades profissionais apelam continuamente a uma indústria sem resíduos. No entanto, a concretização deste objetivo ambicioso continua a ser contrária à situação atual, uma vez que a regulamentação, as políticas e as práticas existentes ainda têm origem em processos tradicionais de gestão da construção que se centram no tratamento dos resíduos tal como são, no final, em vez de começarem na fonte de produção de resíduos. Esta tese centra-se na exploração do problema generalizado do desperdício de Making-Do (MD) em projectos de construção, com a utilização dos princípios da Lean (LC), em particular o Last Planner System (LPS) reforçado pela Modelação da Informação da Construção (BIM), para desenvolver estratégias de mitigação do MD. Os resíduos de MD gerados pela realização de tarefas de construção sem os pré-requisitos necessários causam ineficiência, aumento de custos e redução da qualidade num projeto. Inicialmente, a revisão sistemática da literatura identificou um conhecimento limitado dos resíduos de MD e a falta de uma abordagem integrada para os mitigar. Esta investigação analisa então os factores subjacentes aos resíduos de MD, os elementos dos resíduos de MD e os seus efeitos, sintetizando os resultados da revisão da literatura e dos questionários estruturados. Com base nesta avaliação, as causas potenciais do desperdício de MD podem estar relacionadas com o planeamento, a comunicação e o conhecimento entre os trabalhadores. Com base nestes resultados, a investigação criou um novo Modelo Dinâmico de Sistema - o Modelo Dinâmico para a Mitigação do Making-Do (D3M) para modelar e avaliar os efeitos do LPS e do BIM na redução dos resíduos de MD. O D3M combina os resultados dos inquéritos por questionário e estudos de caso de projectos de condomínios de vários andares, em que os aspectos técnicos e sociais são quantificados para fornecer tendências gerais na geração e propagação de resíduos de MD. O modelo permite testar vários cenários de implementação do LPSBIM em termos da sua eficácia na redução dos resíduos de MD e na melhoria dos resultados do projeto. Os resultados destacam a importância do planeamento colaborativo e da análise das restrições, da partilha transparente de informações através do BIM e de uma cultura que dê prioridade à prevenção e resolução dos resíduos de MD. Assim, esta investigação oferece um modelo de aplicação e uma ferramenta de simulação baseada em modelos (D3M) para a prática de integração LPS-BIM para a minimização dos resíduos de MD, a fim de beneficiar os profissionais e os investigadores no aumento das possibilidades de sucesso nos projectos de construção. Palavras-chave: Construção Lean (LC); Modelação de Informação de Construção (BIM); Resíduos de Construção; Modelagem Dinâmica de Sistemas (SDM); Resíduos de improvisação. vi Abstract Construction waste remains a significant challenge for the industry, hindering productivity and discouraging innovation. Various reports outline its adverse impacts, such as hurdling environmental, economic, and social sustainability development. Academics, regulators and professional societies continuously call for a zero-waste industry. However, achieving this ambitious objective remains contrary to the actual situation, with the existing regulations, policies and practices still originating from traditional construction management processes that focus on treating wastes as they are, in the end, instead of starting from the source of waste generation. This thesis focuses on the exploration of the pervasive problem of Making-Do (MD) waste in construction projects, with the use of Lean Construction (LC) principles, particularly the Last Planner System (LPS) enhanced by Building Information Modelling (BIM), to develop mitigation strategies for MD. MD waste generated by undertaking construction tasks without the required prerequisites causes inefficiency, increased cost and reduced quality within a project. Initially, the systematic literature review identified limited knowledge of MD waste and the lack of an integrated approach for mitigating it. This research then reviews the factors behind MD waste, elements of MD waste, and its effects by synthesizing the outcomes of literature review and structured questionnaires. Based on this evaluation, potential causes of MD waste can pertain to planning, communication, and knowledge among workers. Based on these result, the research has created a new System Dynamic Model - the Dynamic Model for Making-Do Mitigation (D3M) to model and assess the effects of LPS and BIM on MD waste reduction. D3M combines results from the questionnaire surveys and case studies of multistorey condominium projects where technical and social aspects are quantified to provide general trends in generating and propagating MD waste. The model enables various LPS-BIM implementation scenarios to be tested in terms of their effectiveness in reducing MD waste and enhancing project outcomes. Findings highlight the importance of collaborative planning and constraint analysis, transparent information sharing through BIM, and a culture prioritising MD waste prevention and resolution. Thus, this research offers an application model and a model-based simulation tool (D3M), to the LPS-BIM integration practice for MD waste minimisation to benefit practitioners and researchers in enhancing the chances of success in construction projects. Keywords: Lean Construction (LC); Building Information Modelling (BIM); Construction Waste; System Dynamic Modelling (SDM); Making-Do waste. vii TABLE OF CONTENTS 1. Introduction ________________________________________________________ 1 1.1. Motivation __________________________________________________________ 2 1.2. Objectives __________________________________________________________ 4 1.3. Thesis Outline _______________________________________________________ 5 2. Literature Review ___________________________________________________ 8 2.1. Introduction _________________________________________________________ 9 2.2. Systematic Review Research Questions ____________________________________ 12 2.3. Materials and Methods ________________________________________________ 13 2.4. Data Evaluation _____________________________________________________ 15 2.5. Results ___________________________________________________________ 18 2.5.1. Waste Elimination and Lean Construction _____________________________________ 18 2.5.2. Lean Construction Principles for Waste Elimination ______________________________ 20 2.5.3. Integrated Lean Project Delivery ____________________________________________ 21 Lean Supply Chain Management _______________________________________________ 22 Lean Design Management ____________________________________________________ 24 Last Planner System _________________________________________________________ 25 Location-based Management Systems __________________________________________ 27 2.5.4. Relational and Legal Structures for Waste Elimination ____________________________ 31 2.5.5. Lean Sustainability _____________________________________________________ 31 2.5.6. Value Stream Mapping __________________________________________________ 33 2.5.7. Building Information Modelling as Lean Construction Enabler _______________________ 34 Brief BIM history ____________________________________________________________ 35 BIM definition ______________________________________________________________ 35 BIM and Lean ______________________________________________________________ 36 BIM construction management ________________________________________________ 37 BIM and Lean Production Planning and Control ___________________________________ 37 BIM for lean supply chain management _________________________________________ 38 2.5.8. Lean Construction Research in the Portuguese Context ___________________________ 39 2.6. Discussion _________________________________________________________ 41 2.6.1. How does lean construction conceptualise waste? _______________________________ 41 xiv exponential decay at the right when the balancing loop is dominant. .................................................. 70 Figure 3-9 - Common reference modes (Kirkwood, 1998) .................................................................. 72 Figure 3-10: Basic Stock and Flow Diagram ...................................................................................... 73 Figure 4-1A conceptual diagram of making-do phenomena. ............................................................. 85 Figure 4-2A diagram illustrating the relationships among prerequisites, MD categories and impacts based on the protocols developed by Dos Santos et al. (2020) and Sommer (2010) .......................... 91 Figure 4-3-- Casual Loop Diagram illustrates the structural relationships with prerequisites, MD categories and impacts. .................................................................................................................. 102 Figure 5-1 - A Framework of Policy to Improve the Performance of MD mitigation models. ............... 107 Figure 5-2 -- Educational attainment among the survey participants. ................................................ 115 Figure 5-3 -- Occupational roles within the respondent group. .......................................................... 116 Figure 5-4 -- The percentage of Lean and BIM Education and Training among the Respondents ....... 117 Figure 5-5 -- The Respondents Experience in Lean and BIM (years) .................................................. 117 Figure 5-6 -- The knowledge and use of Making-Do terminology and similar concepts. ...................... 118 Figure 5-7 -- Estimated Percentage of MD in construction workflows according to the respondents. .. 119 Figure 5-8: Examination of respondents' perspectives on the entities accountable for Making-Do (MD) waste generation. ........................................................................................................................... 119 Figure 5-9 -- Unadjusted SEM measurement model ......................................................................... 125 Figure 5-10 -- Adjusted measurement model ................................................................................... 126 Figure 5-11 -- Bootstrap settings ..................................................................................................... 129 Figure 5-12 --- A graphical representation of the applied Mediation Analysis for LPS, BIM, COO, MDK130 Figure 5-13 -- Framework for an LPS and BIM for MD waste mitigation. ........................................... 132 Figure 6-1 -- Generic view of the dynamic systems for making-do ..................................................... 135 Figure 6-2 – Example of raw data collected and classified into Phases, Substages, Prerequisites, MD categories, and MD impacts. This figure is not mentioned in the text ............................................... 138 Figure 6-3 -- General Work Breakdown Structure used for the three projects ..................................... 139 Figure 6-4-- An examination of the distribution pattern within the phases primarily impacted by MD incidents in the analysed cases (according to data collected). .......................................................... 140 Figure 6-5-- An analysis of the distribution among the sub-stages most affected by MD incidents in the examined cases (according to data colleceted). ............................................................................... 141 Figure 6-6 -- Analysing the task prerequisites (not resolved constraints) caused MD incidents: a study of affected tasks distribution in examined cases (according to data colleceted) .................................... 142 xv Figure 6-7 -- Analysing Categories of MD: a study of affected task distribution in examined cases (according to data colleceted). ........................................................................................................ 143 Figure 6-8 -- Analysing the Impact of MD Incidents on Production Systems: A Study of Affected Distribution in Examined Cases ....................................................................................................... 144 Figure 6-9 -- Graphical outputs of a single MLRA too small ............................................................... 148 Figure 6-10-- Stock and Flow Diagram for Traditional Production Planning and Control .................... 152 Figure 6-11 -- Snippet of Planning and Control Subsystem ............................................................... 153 Figure 6-12 -- Waste subsystem ...................................................................................................... 154 Figure 6-13 -- Simulated for averages of planned, backlog and completed tasks based on the traditional planning settings. ........................................................................................................................... 155 Figure 6-14: System dynamics simulation for project A stages infected by MD incidents. Is this the same data of figure 6.4? .......................................................................................................................... 157 Figure 6-15: The accumulated number of simulated tasks included MD incidents. ........................... 157 Figure 6-16: The difference between simulated (expected) and collected data for Stages infected by MD. ...................................................................................................................................................... 158 Figure 6-17: The variance between simulated tasks and observed data across time. ....................... 159 Figure 6-18: Simulation results for the number of constraints Case A. ............................................. 160 Figure 6-19: The variance between simulated constraints and observed data across time. ............... 161 Figure 6-20: Simulation results for the categories of MD encountered in Case A. .............................. 162 Figure 6-21: The variances between observed and simulated MD categories. ................................... 162 Figure 6-22: Simulation of MD impacts on the production system. ................................................... 163 Figure 6-23: The variance between simulated and observed impacts across time. ............................ 164 Figure 7-1 – The research design used for this chapter ................................................................... 173 Figure 7-2 -- Workprogress subsystem with MD and waste subsystems. ........................................... 180 Figure 7-3 -- The dynamic variables affecting the construction productivity. ...................................... 183 Figure 7-4 -- Impact of workspace limitation on productivity. ............................................................ 185 Figure 7-5 -- The dynamic subsystem for Resources. ....................................................................... 189 Figure 7-6: The LPS dynamic subsystem ........................................................................................ 190 Figure 7-7 - The technical LPS dynamic subsystem ......................................................................... 191 Figure 7-8 -- BIM functionalities subsystem ...................................................................................... 192 Figure 7-9 - Time analysis for constraints in Project M ..................................................................... 197 Figure 7-10 -- Time analysis for MD in Project M .............................................................................. 198 xvi Figure 7-11 -- Time series analysis for waste in Project M ................................................................ 199 Figure 7-12 -- Time analysis for constraints in Project N ................................................................... 199 Figure 7-13 -- Time series analysis for waste in Project N ................................................................. 200 Figure 7-14 -- Time analysis for waste in project N. .......................................................................... 200 Figure 7-15 - Comparison between constraints P1 - P6 in project M ................................................. 201 Figure 7-16-- MD categories CAT1 to CAT5 comparison in project M ................................................ 202 Figure 7-17 -- Comparative analysis of the impact of MD I1-I5 in Project M ...................................... 204 Figure 7-18 -- Comparison between constraints P1 - P6 in project N ................................................ 205 Figure 7-19 – MD categories CAT1 to CAT5 comparison in project N ............................................... 206 Figure 7-20 -- Comparative analysis of the impact of MD I1-I5 in Project N ....................................... 207 Figure III-1 -- Cloud application of D3M. ........................................................................................... 249 Figure III-2 -- D3M Welcome Message ............................................................................................. 249 Figure III-3 -- The dynamic framework for MD waste analysis. .......................................................... 250 Figure III-4 -- User input window. ..................................................................................................... 252 Figure III-5 -- preset buttons for extreme value tests. ........................................................................ 252 Figure III-6 -- Slider controllers to set parameter values. ................................................................... 253 Figure III-7 -- D3M dashboard. ......................................................................................................... 254 Figure III-8 -- Control charts of D3M................................................................................................. 255 Figure III-9 -- Work progress and MD subsystems. ........................................................................... 256 Figure III-10 -- Location subsystem. ................................................................................................. 256 Figure III-11 -- Resources subsystem ............................................................................................... 257 Figure III-12 -- Productivity subsystem ............................................................................................. 257 xvii INDEX OF TABLES Table 2-1: Inclusion and exclusion logic used to filter research documents ......................................... 15 Table 2-2 -- Construction production waste categories according to (Formoso et al., 2020)................. 19 Table 2-3 -- Comparative analysis for the integrated methods with VSM to investigate construction wastes. ............................................................................................................................................. 33 Table 2-4-- Current contributions for BIM Construction Management Software and Lean Construction Based BIM ....................................................................................................................................... 38 Table 2-5--Lean construction factors of waste elimination ................................................................... 43 Table 3-1 -- Construction management worldviews adapted from (Creswell & Creswell, 2017; Fellows & Liu, 2015) ........................................................................................................................................ 55 Table 3-2 -- Model fitness measures .................................................................................................. 63 Table 3-3-- Exploratory and Confirmatory Factor Analysis ................................................................... 65 Table 3-4: The components of Stock and Flow Diagram ..................................................................... 72 Table 3-5 -- Comparison between available software packages for System Dynamics Modelling functions ........................................................................................................................................................ 80 Table 4-1 - Summary of Causal Factors Contributing to Making Do in Construction Projects ............... 86 Table 4-2 - Prerequisites for Construction Project Execution and Their Explanations, Sourced from Koskela (2000) and Sommer (2010) ................................................................................................ 91 Table 4-3-- Making-Do Categories and Descriptions in Construction Projects ...................................... 93 Table 4-4 -- Impact of Making-Do Practices on Construction Projects: ................................................ 94 Table 5-1 -- Demographic characteristics of the survey respondents ................................................. 115 Table 5-2 -- Reliability analysis table with means and ranking of LPS and BIM strategies for MD mitigation. ...................................................................................................................................... 120 Table 5-3 -- KMO and Bartlett's Test ................................................................................................ 122 Table 5-4 -- Component labelling and corresponding criteria from factor analysis ............................. 122 Table 5-5 -- Model fitness measures ................................................................................................ 124 Table 5-6 -- Model fitness measures for the adjusted measurement model ....................................... 126 Table 5-7 -- Loadings, Reliability and Convergent Validity ................................................................ 127 Table 5-8. HTMT Analysis ............................................................................................................... 128 Table 5-9--Moderation analysis for the structural model and model fit indices ................................... 129 Table 5-10 -- A summary of Mediation Analysis results .................................................................... 130 Table 6-1-- Constraints, MD, and MD impacts .................................................................................. 135 xviii Table 6-2-- Comparative Overview of Multistorey Condominium Projects: Case A, Case B and Case C ...................................................................................................................................................... 137 Table 6-3 -- Association tabulation test for Task prerequisites (Left Column) and MD categories (Top Row) 6241 ..................................................................................................................................... 145 Table 6-4 - Chi-Square test for prerequisites and MD categories ....................................................... 146 Table 6-5 -- Association tabulation test for MD Categories (Left Column) and construction (impacts) waste (Top Row) ............................................................................................................................. 146 Table 6-6 -- Chi-Square test for MD categories and MD impacts ....................................................... 147 Table 6-7 – Summary of the Multiple Regression Models (P1-P6, CAT1-CAT5, and I1-I5) ................. 148 Table 6-8 -- Regression tabulation for task prerequisites (P1-P6) in relation to sub-stages (SS1-SS11) 149 Table 6-9 – Regression tabulation for MD categories (CAT1-CAT5) concerning constraints (P1-P6). .. 150 Table 6-10 -- Regression tabulation for MD impact (I1-I5) concerning MD categories (CAT1-CAT5) .... 150 Table 6-11 -- LPS -BIM model parameters ....................................................................................... 167 Table 7-1 - Comparative Overview of Selected Construction Projects A and B ................................... 174 Table 7-2Time to discover constraints with master planning completion ......................................... 182 Table 7-3 - Lookup function of the impact of workspace availability on productivity ........................... 184 Table 7-4 - Lookup function of the impact of fatigue on productivity ................................................. 185 Table 7-5 - Lookup function of schedule pressure effect ................................................................... 186 Table 7-6 -- Lookup function for the impact of applying the BIM process . ........................................ 187 Table 7-7 -- Table function of resources allocation needed following the fraction of project completed. ...................................................................................................................................................... 189 Table 7-8 -- Lookup the commitment function to implement BIM processes (Porwal, 2013). ............. 192 Table 7-9 -- Data entered in AnyLogic for critical variables of the baseline scenario (Projects A and B) ...................................................................................................................................................... 194 Table 7-10 -- Input Data of baseline scenario ................................................................................... 194 Table 7-11 -- Comparison of system dynamic model with project data .............................................. 195 Table 7-12. The mix of variables to be tested in Scenarios I to IV. .................................................... 196 Table 7-13 – Constraints P1 to P6 in Project M Simulation .............................................................. 201 Table 7-14-- Making Do Categories CAT1 to CAT 5 in Project M Simulation ...................................... 203 Table 7-15 -- Waste I1 to I5 Project M Simulation ............................................................................ 203 Table 7-16 -- Constraints P1 to P6 in Project N Simulation ............................................................... 204 Table 7-17 -- Making Do Categories CAT1 to CAT 5 in Project N Simulation ..................................... 205 xix Table 7-18 -- Waste I1 to I5 Project N Simulation ............................................................................. 206 Table 8-1 -- D3M vs traditional MD analysis ..................................................................................... 213 Table III-1 -- Parameters of LPS-BIM within D3M .............................................................................. 251 Table IV-1 -- Parameters explanation and scale ................................................................................ 258 Table IV-2 -- Array dimensions used in the Anylogic.......................................................................... 262 Table IV-3 -- Dynamic equations ...................................................................................................... 263 Table IV-4 -- Table functions in D3M ................................................................................................ 270 xx LIST OF ABBREVIATIONS 2D-CAD Two-Dimension Computerised-Aided-Design __________________________________ 33 4D 3D BIM model elements associated with time ____________________________________ 38 AECO Architecture, Engineering, Construction and Operations ___________________________ 33 AHP Analytical Hierarchy Process ______________________________________________ 75 AON Activities On Nodes _____________________________________________________ 27 AVE Average Variance Extracted _______________________________________________ 121 BIM Building Information Modelling ______________________________________________ 33 CBA Choosing-By-Advantage __________________________________________________ 24 CCC Construction Consolidation Centers __________________________________________ 22 CDW Construction and Demolition Waste ___________________________________________ 2 CFI Comparative Fit Index ___________________________________________________ 119 CLD Casual Loop Diagram ___________________________________________________ 98 CPM Critical Path Method ____________________________________________________ 37 CR Composite Reliability ____________________________________________________ 121 CSC The Construction Supply Chain _____________________________________________ 21 FL Flow Line ______________________________________________________________ 32 GIS Geographic Information System _____________________________________________ 39 HTMT Heterotrait-Monotrait __________________________________________________ 122 ICT Information and Communication Technology ____________________________________ 94 IFOA Integrated Forms of Agreements ____________________________________________ 25 IGLC International Group of Lean Construction ______________________________________ 41 IPD Integrated Project Delivery _________________________________________________ 25 ISO International Organization for Standardisation ___________________________________ 33 xxi JIT Just-In-Time ____________________________________________________________ 22 KMO Kaiser-Meyer-Olkin _____________________________________________________ 116 KPIs key performance indicators ________________________________________________ 73 LBMS Location-Based Management Systems _______________________________________ 29 LBS Location-Based-Structure _________________________________________________ 30 LC Lean Construction _______________________________________________________ 37 LDM Lean Design Management ________________________________________________ 16 LPC Lean Planning and Control ________________________________________________ 16 LRA Linear Regression Analysis ________________________________________________ 60 LSCM Lean Supply Chain Management ___________________________________________ 16 NaN Not a Number ________________________________________________________ 182 NFC Near-Field Communication ________________________________________________ 39 NVA Non-Value-Added ______________________________________________________ 23 PERT Program Evaluation and Review Technique ____________________________________ 2 PPC Per cent Plan Complete __________________________________________________ 98 PPC Plan Percent Complete ___________________________________________________ 28 QT Queuing Time __________________________________________________________ 23 RCA Root Cause Analysis _____________________________________________________ 25 RFI Request for Information ___________________________________________________ 40 RFID Radio Frequency Identification _____________________________________________ 38 RMSEA Root Mean Square Residual ____________________________________________ 119 SBD Set-Based-Design ______________________________________________________ 24 SRMR Standardized Root Mean Square Residual ___________________________________ 119 STELLA Experimental Learning Laboratory with Animation _____________________________ 75 TFV Transformation-Flow-Value _________________________________________________ 18 xxii TLI Tucker-Lewis Index ______________________________________________________ 119 TPS Toyota Production System _________________________________________________ 17 TT Takt Time ____________________________________________________________ 23 TTP Takt Time Planning ______________________________________________________ 32 TVD Target-Value-Design _____________________________________________________ 24 WBS Work Breakdown Structure ______________________________________________ 30 WIP Work-In-Progress ______________________________________________________ 23 WMP Material Waste Management Plan ___________________________________________ 25 WWP Weekly Work Planning ___________________________________________________ 27 1. Introduction 1 1. Introduction This chapter presents a general overview of the thesis, opening with the main objectives of the research motivation. A brief description of the research methodology includes the main theoretical and methodological approaches to achieve the thesis objectives. The thesis structure is outlined, with short descriptions for each chapter illustrated. 2. Literature Review 8 2. Literature Review This chapter presents the state of the art of the studied topic and provides a thematic analysis for lean construction methods. Finally, it identifies the research gaps in the literature and presents a theoretical framework. Except for some changes executed for the formatting and organisation purposes of global information in this document, this chapter integrally presents the work: (Karaz & Teixeira, 2023b). Waste Elimination based on Lean Construction and Building Information Modelling: A Systematic Literature Review. U. Porto Journal of Engineering, 9(3), 72–90. https://doi.org/https://doi.org/10.24840/2183-6493_009-003_001808 2. Literature Review 9 2.1. Introduction Construction waste is a high-level concept behind poor productivity and low innovation levels in the industry, which is challenging to measure systematically. The plethora of reports shows the massive consequences of construction waste, which hurdles sustainability development in its three dimensions (environmental, economic, and social). Regarding materials, one-third of the global construction material is landfilled without treatment (Yuan & Shen, 2011). The second type of waste is environmental, the industry's greenhouse gas (GHG) footprint; 33% of the global GHG is released from construction and transportation projects (UN,2017). Construction and Demolition material Waste (CDW) and GHG emissions concern many regulators, such as the European Commission to publish several directives to impose strategies for the negative environmental impact of the construction (i.e., EU waste directive (2008/98/EC), The Energy Performance of Buildings Directive (2010/31/EU) among others). However, none of the provided directives have discussed the root causes of the emerging issues, and most of the proposed regulations and guidelines offer end-of-pipe solutions without proposing countermeasures that tackle waste at source (Osmani et al., 2008). Hence, most developed regulations deal with symptoms of waste rather than targeting how waste is produced in terms of production waste. That takes time units into the formula of waste generation. Production waste is usually called non-value added (NVA), defined as any activity that absorbs resources (e.g., time, location, material, energy, and others) without adding value to internal or external customers. According to the meta-analysis of Horman & Kenley (2005), NVA activities constitute 49.6% of the construction operations. The literature has investigated different types of NVA, including rework (Love & Li, 2000a), product defects (Josephson & Hammarlund, 1999), waiting (Sacks, 2016), transportation (Belayutham et al., 2016), intuitional and intuitional waste (Sarhan et al., 2017),. And the relation between production and environmental waste (Carvajal-Arango et al., 2019; Golzarpoor et al., 2017; Golzarpoor & González, 2013; Nahmens & Ikuma, 2012). This disparity in measuring and defining waste measures increases the difficulty of formulating holistic and hurdles efforts of providing general guidelines for root causes analysis (Formoso et al., 2020). Additionally, many reported types of waste are measured empirically at an operational level or professional experience, which challenges a comprehensive judgment on the nature of generated wastes and their relationships with other types. Toyota Production System (TPS) first attempted a taxonomy for production waste, which was 2. Literature Review 10 supposed to be achieved by attacking overproduction and reducing inventory, operational advantages at the production level, and increasing profits at the organisational level with minimum investment (Ohno, 1988). On this basis, lean production was founded as a waste elimination-focused philosophy to generate customer value with zero waste ideally. Lean production considers waste an actionable language that brings stakeholders' attention to actual causes of inefficiency;Womack and Jones (2003) drew from a set of principles adopted in various industries, including construction. Lean Construction (LC) adopted these principles to reduce cycle time and product and process variability, enforce continuous improvement, and increase transparency (Koskela, 2000; A. Santos, 1999). In addition, other principles, including 'map for value stream,' 'establish for pull planning,' and 'structure the construction process into flows,' are also used (Koskela et al., 2002). From a theoretical point of view, LC is founded on the Transformation-Flow-Value (TFV) theory, which states that traditional construction management sees the process as input-process-output is a counterproductive perspective because it is a black-boxed definition that hides wastes inside the process which resolves production problems after occurring. The TFV suggests adding two more concepts, Flow (F) and Value (V). That forms a new paradigm in construction projects, where F aims to break down the construction process into NVA and VA, as shown in Figure 2-1, to provide technical metrics to understand and control the construction progress and problems proactively. The V perspective adds a social dimension by encouraging the upstream to understand the downstream requirements, which should work collaboratively to reap global project optimisation instead of focusing on local optimisation. Figure 2-1 -- Transformation-Flow-Value theory 2. Literature Review 11 In practice, lean construction cannot be applied only from the early stages until handover but also at the end-of-life/extension-of-life stages of a building (e.g., demolition, renovation, retrofitting, expansion). This application extends across the whole construction supply chain. The ideal Lean Supply Chain Management aims to deliver construction products on time, which relies on pulling information about the product and process from production signals rather than solely forecasting demand (Bortolini et al., 2019; Vrijhoef, 2020). Production signals should be pulled from reliable plans that are retrieved from LC planning and control systems (i.e., Last Planner System® (LPS) and Location-BasedManagement System (LBMS)). The LPS is a context-specific and socio-technical system that aims to improve planning reliability through successive collaborative sessions to shield the downstream from upstream variability (Ballard, 2020). At the same time, LBMS is a spatial, temporal system that technically plans and structures the construction operations according to their locations (i.e., floor, zones, sections, and floors) (Kenley & Seppänen, 2010). LPS and LBMS can be used concurrently to address waste such as work-in-progress, waiting, space congestion, and overproduction(Frandson et al., 2014). Those wastes will be discussed and removed collaboratively after acknowledging work structure, sequencing, resource allocation, and associated constraints. Due to the intensity of the production information, labour and manual inputs for LPS and LBMS can hinder planning reliability. So, Building Information Models (BIMs) are essential to communicate realworld data streamlined from LC planning and control systems and support the decisions both systems take. According to Sacks et al. (2010), BIM functionalities are recognised as a coherent and consistent source of information that can provide up to 56 positive interactions with lean principles. A significant example of LC-BIM is the integration of LPS into the BIM 4D model, which assists in filtering trade activities according to task readiness and enables managers to track the production progress and bottlenecks using the production metrics (Sacks et al., 2010). Those metrics include construction production rate, resource consumption, flow index, Plan Percent Complete (PPC), and task maturity, which can be translated into Andon signals to bring stakeholders' attention to production wastes (Dave, 2013). The empirical research shows that LPS-BIM's effectiveness for waste elimination is partial without real-time tracking, which is improved through digital monitoring, artificial intelligence, and linked data, among other technologies (Dave & Sacks, 2020). Additionally, LBMS can benefit from BIM workflows supporting LBMS forecasts with accurate material take-offs and location interference analysis. Moreover, LC-BIM has a positive impact on accelerating the adoption of other initiatives, including 4.0 construction, circular economy, design-out- 2. Literature Review 12 waste (Karaz et al., 2021), and Design for Manufacturing and Assembly (DfMA) (Gbadamosi et al., 2019). The planning and control-based LC-BIM approach holds vast waste elimination opportunities for sustainability (Saieg et al., 2018). The objectives of this chapter are to explore lean construction approaches to conceptualise construction waste, to identify waste elimination factors presented by lean construction at various stages of the construction supply chain, and to review the relationship between Lean principles and BIM functionalities to address construction waste. The chapter utilises the systematic literature review method to achieve these objectives. The systematic review found four gaps regarding the concept of waste in construction management research as follows: i) persistent construction management theories in the current research and practice, ii) a holistic theory that explains waste propagation and its characteristics is not presented yet, iii) ambiguity in the reported data collection methods for waste, iv) variety in conceptualisation to address similar types of waste. 2.2. Systematic Review Research Questions A plethora of research attempts to investigate the impact of LC and BIM on improving production and planning using the concepts of waste elimination and value generation. However, LPS-based and LBMSbased BIM utilisation remains parallel where models are presented while collaborative planning takes place with low automation and maturity levels. Additionally, the presented research lacks explicit definitions and measures for construction waste, which is rarely captured systematically. A comprehensive and integrative literature review is needed to understand how lean construction and BIM interact to reveal and eliminate waste systemically. Therefore, this chapter investigates these gaps by adopting the SLR methodology. The formulated SLR questions are defined as follows: 1. How does lean construction conceptualise waste? 2. What are waste elimination factors imposed by lean construction (LC)? 3. What are the factors of LC-BIM that contribute to waste elimination? This chapter is structured as follows: Section 2.3 presents a detailed description of the systematic literature review methodology and a descriptive analysis of the publications in this field over time. Section 2.4 presents content and thematic analysis to cluster the existing research into four research themes. Section 2.5 concludes the performed analysis and discusses further development needed to 2. Literature Review 13 improve the current understanding and application of waste elimination in the construction industry. 2.3. Materials and Methods Year by year, enormous research is conducted with conflicting understandings of construction waste and various interventions to tackle it. A Systematic Literature Review (SLR) is a method to understand a context-specific problem and laud the suggested interventions by the literature to address that problem by synthesising the dispersed results from evidence-based literature. An SLR should be a transparent, upgradable, transferable, and quality exclusive review Denyer & Tranfield, (2009). However, the constraints of SLR are limited to a time-consuming review methodology that requires additional resources compared to traditional methods (Mulrow, 1994; Wohlin & Claes, 2014). The adopted SLR methodology is illustrated in Figure 2-2, which shows that SLR comprises four iterative stages: (1) planning for the review, (2) material collection, (3) data evaluation, and (4) results reporting and dissemination (Tranfield et al., 2003). During 'planning for the review,' the review question is developed in the following section as an early-stage process that conceptualises and formulates complex problems into a contextual frame (Flemming et al., 2019). Figure 2-2-- Systematic Literature Review (SLR) Methodology The literature data were collected using the Scopus database, IGLC, and snowballing techniques during the material collection stage. Most retrieved documents are from the Scopus database that was queried by using two entries: the first query is a preliminary search string that combines only the study STAGE 1 PLAN FOR THE REVIEW STAGE 3 DAT A EVALUATION STAGE 4 REPORTING AND DISSEMINATION Ɩ Storing, classifying and sorting records Ɩ Data base creation Reference Management SLR Methodology SLR Tasks Descriptive Analysis References Manager Mendeley Research Database XMind Keyword map Microsoft Excel Thematic Analysis Qualitative Analysis MaxQDA Formulate research question Elaborated in section 1 Conduct SLR? STAGE 2 MATERIAL COLLECTION Ɩ Title-Keyword-Abstract Search Ɩ Search strings in section 1.2 Locating studies in Scopus Meets Criteria? Preliminary Review Title, abstract & keywords Screening Ɩ Full-text screening Ɩ Descriptive evaluation Ɩ Full-text analysis Ɩ Database update Ɩ Inclusion and exclusion In-depth Review Ɩ Perform Thematic analysis Ɩ Results evaluation and extraction Reporting the evidence Ɩ Material synthesis Ɩ Report and disseminate the evidence Develop the review protocol Ɩ Identify the question relevance Ɩ Determine databases, keywords, and exclusion and inclusion criteria Qualitative Analysis Database, Software and Tools 2. Literature Review 14 keywords' lean construction' and 'BIM' and 'waste' using the 'AND' operator; the second query was formulated using the 'AND 'OR' operators to extend the search range and also included the TitleAbstract-Keywords using the streamlined keywords in the tree map (Figure 2 - 3 ). The keywords used were adapted from (Tezel et al., 2020; Viana et al., 2012). At the same time, the snowballing technique followed the procedure suggested by (Wohlin & Claes, 2014). Figure 2-3 -- A mind map of streamlined keywords After querying the selected databases, the number of included documents for a preliminary analysis was (411) records, as illustrated in Figure 2-4. The inclusion/exclusion criteria are summarised in Table 1 to filter the number of documents included. The requirements are applied for publishing year, researched concepts, research domains, and language. After using the specified including/exclusion criteria in Table 2-1, the number of research documents was narrowed to (190) after screening Titles and Abstracts. In contrast, (136) documents were excluded after full-text analysis, so the number of relevant documents (54) is to be analysed in this paper. Keywords BIM LC Lean construction Last Planner System Just In Time Kaizen Kanban CW Visualisation Virtual Design Building Information Model* (BIM) Non-value-add* (NVA) Construction waste 2. Literature Review 15 Figure 2-4 – Prisma diagram applied to researched documents Table 2-1: Inclusion and exclusion logic used to filter research documents Criteria Inclusion Exclusion Publishing Year 1999-2022 Earlier than 1999 and later than 2022. Discussed Topics Combine LC and waste concepts, OR LC, BIM, and waste. Studies are on only one concept: studies on BIM and waste only. Research Domain Construction management domain with a focus on LC concepts. Other domains than LC management. Publication Language The English language only. Other than English. . 2.4. Data Evaluation This section employs descriptive analysis to provide a broad overview of waste elimination development through LC and LC-BIM over time. Firstly, the study aimed to depict how the topic quantitatively developed over time. Figure 2-5 illustrates that the number of waste elimination papers has risen. From Identi�cation Screening Eligibility Included Records identi�ed through scopus (n= 394) Records after duplicates removed (n= 307) Records screened (n= 258) Records removed by author ƎMahmoudƐ (n= 49) Records identi�ed through IGLC* (n= 7) Records excluded title, abstract, or keywords (n= 78) Articles excluded (n= 136) Full articles assessed for eligibility (n= 180) Records included (n= 54) Records included using snowballing (n= 10) 2. Literature Review 16 early 1992 until the first quarter of 2022, most literature recognised waste elimination as a central concept of LC. In 2008, the impact of LC-BIM on waste elimination emerged (Eastman et al., 2008). Figure 2-5-- The number of documents contributed to waste elimination based on LC and LC-BIM from 1992 to 2022 The proportion of relevant documents that used the LC-BIM approach is 53.70%, while 46.30% of selected records utilised LC theory and methods. Figure 2-6 presents an analysis of the objective of waste elimination research, which was classified into (i) conceptual, (ii) literature review, and (iii) empirical. The first class includes papers on theory and predominantly on historical and conceptual analysis of construction waste. The second class represents literature review papers that develop integrative solutions based on secondary data from evidence-based literature. Empirical papers are those aimed to provide applications or adaptations for LC and LC-BIM solutions in a specific context of the construction supply chain, for example, (i) reporting problems and prescribing a solution for that problem, (ii) implementation of LC principles, methods, tools, and techniques, (iii) define the requirements for waste elimination solutions-based LC-BIM (iv) evaluation of LC and LC-BIM solutions, (v) use of IT artefacts; among others. 2. Literature Review 17 Figure 2-6-- A distribution of the research methods from 1992-2021 The analysis demonstrates that the focus of the literature has been mainly empirical rather than theoretical and literature review, as shown in Figure 2-7. Additionally, this highlights that theory is not evolving at the same rate as practical implementation, and there are limited successful examples, hence weak support for waste elimination in LC-BIM implementation. The analysis in Figure 2-7 classified the empirical research into four functional areas of the construction supply chain management, namely (i) Lean Supply Chain Management (LSCM), (ii) Value stream Mapping, (iii) Lean design management (LDM) and (iv) Lean planning and control (LPC). That shows that little empirical research was conducted on post-occupancy stages such as demolition (Elmaraghy et al., 2018), rehabilitation (Pereira & Cachadinha, 2011), and facility management (Bascoul et al., 2018). Figure 2-7 -- Time analysis of empirical research from 2002 to 2021. 2. Literature Review 24 Figure 2-10-- Illustration for the role of Construction Consolidation Centres Lean Design Management Lean Design Management (LDM) provides simultaneous design processes that focus on diminishing waste collaboratively at the earliest stages of BIM projects using social, technical, and socio-technical dimensions (Barkokebas et al., 2021; Uusitalo et al., 2019). The social dimension urges people to adopt BIM and enhance their collaborative production skills based on trust and shared understanding among the involved parties (Arayici et al., 2011). At the same time, the socio-technical dimension eliminates waste in terms of planning and control, customer requirement management, decisionmaking methods, and problem-solving techniques (Herrera et al., 2021; Uusitalo et al., 2017). Again, the production indicators are used in LDM to quantify NVA in design workflow and information flow. The current practice plan for design tasks is Kanban based on software but without complete reflection of LC-BIM integration (Mahalingam et al., 2015). This practice disregards opportunities for BIM functionalities specific to the construction design context. This research recommends more research on applying LDM planning and control methods in a BIM environment to apply waste elimination concepts. At the same time, the research lacks integration of waste elimination in LDM customer management methods, which can potentially steer design processes and products toward customers' value and waste elimination by utilising techniques such as Target-Value-Design (TVD), Choosing-By-Advantage (CBA) and Set-Based-Design (SBD) (Karaz & Teixeira, 2023a). Tire 3 Tire 2Tire 1 Of-site Supply Chain On-site Construction Direct Supply Demand/feedback, design information, production information, Revere Logistics Orders Reverse Logistics Pull Supply SupplyJIT Allo Push Supply Consolidation Center 2. Literature Review 25 Last Planner System The primary objective of the LPS® is to provide reliable production planning that shields downstream from upstream variability (Ballard, 2000). LPS is defined as a socio-technical system that unites the stakeholders, including 'the last planners' or the last responsible people, to optimise production plans in successive detailed levels (Ballard, 2020). Figure 2-11 illustrates the structure of LPS, which divides the planning into four successive levels: Master scheduling (Should), Phase planning (Can), Lookahead planning (Will), weekly, bi-weekly, or daily planning and learning (Did). Master scheduling is an overall project schedule from start to finish, which identifies Activities On Nodes (AON) using the CPM method that delimits the project scope. The knowledge of a master schedule is fed by historical data or practitioners' experience; a master schedule is used as a contractual schedule to sublet works to subcontractors, assess the project feasibility, identify long lead times, and identify milestones to be used in the next stage "phase planning" (Frandson et al., 2013). Lookahead planning spans 4-6 weeks, collaboratively deriving sub-tasks from the milestone schedule, detecting and delegating responsibility to constraints, and thus, trades commit to removing these constraints (Ballard 2000). In the weekly work planning (WWP), the team screens constrained tasks and excludes them from the execution schedule. To streamline continuous and stable flow, LPS practitioners commence working on mature activities only, and a quality check for work backlog is applied for work package definition, soundness, sequence, size, and learning (Ballard & Howell, 1998). In the lookahead planning window, constraints analysis is a critical function of LPS. According to (Koskela, 1999), the constraints of the construction products are seven (1) design, (2) components, (3) materials, (4) materials, (5) space, (6) connecting works and (7) external conditions. Similarly, (Ballard & Howell, 1998) classified the constraints into three clusters: (1) directives: necessary information to commence processing (i.e., specifications, product designs, crew numbers, and speciality, among the others), (2) previous works: prerequisite work that must be finished before starting new work and (3) resources: including labour, equipment, support facilities, materials, and space. 2. Literature Review 26 Figure 2-11-- A generic framework for the Last Planner System (Ballard & Howell, 2003) During meetings to screen tasks and constraints, a language/action can be used to formalise the communication for commitments between trades to describe, request, declare, promise, or assert specific information about work packages (MacOmber et al., 2005). Control indicators such as PPC are used to measure the reliability of promises, which is the percentage of work completed divided by the promised work, which helps investigate and learn potential production bottlenecks (Sacks, Korb, et al., 2018). The LPS practice can diminish wastes such as making do, moving, waiting, transportation, inventory, reworks, and defects. Hence, LPS formalises communication between multidisciplinary trades during work structuring, sequencing, constraint removal, control, and learning. Information synchronisation is necessary to implement LPS, primarily by increasing the planning details to enlarge the immense amount of information that can be retrieved into the system, which can be tackled using BIM functionalities such as 4D planning, clash detection, and site layout planning (Dave & Sacks, 2020). Lean-BIM-based production planning and control systems PCS can streamline flexible production systems that actively respond to bottlenecks. Waste should be revealed and Project Objective Information Resources Production DID Planning the Work CAN WILL SHOULD Last Planner Process 2. Literature Review 27 understood by improving situational awareness, activating root causes of waste, performing what-if scenarios, and activating real-time tracking for production bottlenecks. Figure 2-12 -- Last Planner System stages, durations, details, main activities, addressed wastes, and responsible parties. Additionally, location-based management systems (i.e., takt-planning, flow line, Line-of-Balance) can be applied to structure even and stead workflows across locations of the construction product, supported by 4D functionality to provide additional insights about work sequencing, related time, and spatial conflicts (Björnfot & Jongeling, 2007). The available digital tracking methods are enabled for site conditions by LPS and BIM, including surveying methods, laser scanning, indoor positioning, GPS, BLE Beacon, IoT techs, stational touch screens, PDAs, mobiles, tablets, and RFID (Chen et al., 2020; 2016; Sacks et al., 2010; von Heyl & Teizer, 2017). These technologies deliver additional prospective opportunities in automatic waste elimination decisions through artificial intelligence (AI) algorithms, according to the increasing waste data that can be extracted from Lean planning and control systems, which are requirements for system learning, testing, and validation (McHugh et al., 2022). Location-based Management Systems The goal of Location-based Management Systems (LBMS) is similar to that of LPS, which is to streamline reliable planning and control by planning for continuous workflow and preventing interference between trades. Nevertheless, LBMS is a scheduling and control method rather than LPS and depends on technical measures, unlike LPS, where social measures are apparent. LBMS is preplanning, planning, Phase (Pull) Ɩ Apply reverse planning Ɩ Identify handoffs, durations, and overlaps Ɩ De�ne delivery conditions Ɩ De�ne key milestones Ɩ Identify critical path (CPM) Ɩ Assign start-�nish relations CAN SHOULD WILLDID Master Short-term Learning Main activities Addressed wastes Actors *PPC = Plan Percent Complete; RNC = Reason for Non-Compliance Ɩ Project Managers Ɩ Site/production Manager Ɩ Project Managers Ɩ Site/production Manager Ɩ Project Managers Ɩ Site/production Manager Ɩ Crew managers Ɩ Project Managers Ɩ Site/production Manager Ɩ Crew managers Ɩ Owner Ɩ Portfolio Managers Ɩ Project Managers Planning detail Planning Stage Planning Window Lookahead Ɩ Constraint analysis Ɩ Breakdown Processes Ɩ Design for operations Ɩ Make reliable promises Ɩ Measure PPC* Ɩ Investigate RNC** Ɩ Standardize 4-6 weeksPhases Milestones Ɩ Budget overrun Ɩ Schedule overhead Ɩ ENV measures Ɩ Value measures Ɩ Defects Ɩ Budget overrun Ɩ Schedule overhead Ɩ ENV measures Ɩ Value measures Ɩ Defects Ɩ Makind-do Ɩ Work-in-Progress Ɩ Inventory Ɩ Crews absent, injuries Ɩ Transportation Ɩ Information delay Ɩ Rework Ɩ Idle (waiting) Ɩ Un�nished works Ɩ Rework Ɩ Moving Ɩ PPC Failures Weekly/Bi-Weekly/Daily assignments Pebbles (Assignments) Boulders (Processes) Rocks (Phases)Rocks (Milestones)Pebbles (Operations) 2. Literature Review 28 scheduling, and controlling systems for production units based on their physical location (Kenley & Seppänen, 2010). The LBMS process comprises scheduling, schedule optimisation, forecasting, alarming, and control. The first step of scheduling using LBMS is defining Location-Based-Structure (LBS). LBS is the method of breaking the project into manageable locations, similar to Work Breakdown Structure (WBS), but applied to location logic instead of activity-based logic. Each task is defined at a specific hierarchy level and includes one or more locations (Kenely and Seppanen, 2010; pp 125-126). Defining LBS is a critical decision because it is not only necessary for LBMS purposes (e.g., defining logical relationships, visualising flow line, and controlling the progress), but it could also be applicable for classifying quantitytake-offs and signalling logistics and deliveries (Kenley & Seppänen, 2010: PP 203-204). In the example elaborated in Figure 2-13, LBS first breaks down the project into interdependent structures that can be built separately, then narrows it down to manageable locations where one trade can operate continuously without interruption (waiting). Where actors can filter and communicate activities based on this LBS code. Figure 2-13 -- A generic example of LBS hierarchy for a building project (Kenley & Seppänen, 2010) Accurate estimations of quantities are critical to calculate the schedule durations. Calculating quantities is a subsequent step in LBS definition in LBMS scheduling. The task quantities must define all the work to be completed in a location before a crew moves to the following location. Then, LBMS builds tasks from quantities, defines optimum crew size, and uses layered logic for other tasks. Durations are computed by multiplying quantities with labour consumption and dividing by the crew size. Schedule alignment and optimisation through risk management expand trade-offs between time, risk, and cost sought by alignment, making planning tasks continuous and buffering tasks against interference. LBMS optimises the plan to ensure workers do not wait for work and work does not wait Project I A B C I II 1 III I II 1 III I 2 3 1 2 1 1 2 2 4 3 4 3 4 3 4 2 3 1 2 3 4 4 4 II III 4 3 2 1 4 3 2 1 3 2 1 Project Structure Floor Apartment 2. Literature Review 29 for workers. LBMS needs more data than CPM, and LBS needs to be defined beforehand. At the same time, CPM methods are a free form of planning; they report weekly or daily, while CPM is monthly and LBMS is real-time control. The strategies used for schedule optimisation are 1) changing production rates, 2) adding more resources, 3) splitting tasks, 4) allowing discontinuous work, and 5) adding more for the scope or reducing resources. Besides scheduling information, LBMS also provides controlling information. This information includes baseline schedule, current stage, progress, forecasting, and alarms. The baseline schedule is an owner reporting tool that sets limitations to the current schedule. The current stage enables changing quantities, productivity rates, logic, and planning during production, where each task (detailed task) is linked to one baseline task for comparison reasons. The progress stage monitors the actual performance of the project/location/tasks. In this stage, actual dates do not replace planned ones but are used to compare and detect variations. (Input) in start and finish dates, suspended days, and actual resources, while the (output) is calculated as 1) actual resource consumption and 2) actual production rate (units/day/trade). The forecast combines the current and progress stages information to signal early warning about production problems by assuming that the production will continue with the same productivity and with the planned resources and follow the current logic. LBMS alarms show when the predecessor delays the forecast of a successor. Alarms are two weeks before the planned time (Kenley & Seppänen, 2010: 123-126), that attempt to prevent cascading delays by concentrating production control resources to avert these alarms from happening and correcting the predecessor's production rate or slowing down the successor. In LBMS, stakeholders can reduce the number of interdependencies between activities provided by CPM methods, remove float between tasks and synchronise production rates. LBMS can apply several control indicators to forecast production capabilities and provide information about root causes of cascading delay (i.e., work-in-progress, waiting, rework, and congestion) (Kenley & Seppänen, 2010). Finally, LBMS reduces production complexity by streamlining continuous flow across locations, steering planning targets towards stable and interrupted production, giving clear directions for trade crews, and reducing risk and waiting time (Biotto & Kagioglou, 2020). The comparison conducted by Biotto and Kagioglou (2020) highlights Lean-Based Management System (LBMS) techniques as prominent management methodologies in both the master and phase planning processes. Within master planning, the Line of Balance (LOB) and Flow Line (FL) techniques are commonly applied. LOB primarily focuses on monitoring the delivery rate of completed units, with 2. Literature Review 30 planners allocating production capacity buffers within the duration of work packages per unit between activities. This monitoring function enables LOB to balance production by adjusting the number of crews assigned to specific activities and modifying the composition and quantity of work to be executed. In contrast, the Flow Line (FL) technique is more oriented toward production pace planning, with actors allocating buffers for production capacity within task durations. The principal strategy of FL involves modifying the crew's composition. The slope of FL defines the production rate, calculated by dividing the work quantity by the duration. FL is typically employed in repetitive construction scenarios but can also be applied in complex projects, provided the project can be effectively segmented into equal locations. On the other hand, the Takt Time planning (TTP) technique is commonly used in phase planning because it provides more significant planning details than FL and LOB and requires a higher level of collaboration. TTP is defined as 'the unit of time within which a product must be produced (supply rate) to match the rate at which that product is needed (demand) rate' (Frandson et al., 2013). It is measured using the Takt Time (TT) indicator, which can be obtained by dividing the available production time by the product units demanded by the customer. The primary buffer strategy used in TTP is like the FL method, which uses a production capacity buffer that urges actors to insert different compositions and numbers of workers within the work package. Figure 2-14 – Visualisation comparison between three LBMS techniques: a) Line of Balance; b) Flow Line; c) Takt Time (Biotto & Kagioglou, 2020) 2. Literature Review 31 2.5.4. Relational and Legal Structures for Waste Elimination Legal bonds between parties are essential in guiding construction organisations toward value creation and waste generation (Koskela, 2000). Relational contracts emerged and flourished in the late 20th century to facilitate a road map for construction improvements. Integrated Project Delivery (IPD), Integrated Forms of Agreements (IFOA), Partnering, and alliance contracts are examples of Relational contracts that may tackle construction waste by cultivating collaboration in defining clear assignments and responsibilities toward project goals that align the interests of multiple stakeholders to optimise their projects globally instead of focusing on local optimisation for the production. Partnering contracts reduce the liaisons between actors to resolve issues and involve the downstream in upstream decisions through improved collaboration and transparency. Material waste is highlighted in partnering contracts by imposing a material waste management plan (WMP), but the terms of partnering contracts lack focus on production waste elimination (Matthews et al., 2000). In contrast, alliance contracts support task completion, manage inventory and toolbox sharing, and allocate responsibilities where lean principles apply to decompose complex wastes and discard NVA through tools such as 5whys, waste walks 'Genchi nembutsu,' spaghetti diagrams, quality control histograms and Root Cause Analysis (RCA). Combining lean principles and relational contracts can eliminate about 24% of NVA by facilitating double learning, improving process quality, and implementing safety measures (Vilasini et al., 2014). However, alliance procurement lacks explicit terms for waste elimination, which causes an absence of legal commitment and measures toward waste recognition, analysis, and responsibility for waste elimination. More research is needed to include the concepts of waste elimination in relational contracts, and this gap can also be applied to IPD, alliance contracts, and collaborative Design-Build contracts. 2.5.5. Lean Sustainability From the LC perspective, environmental waste encompasses adverse outcomes within the production system that fail to contribute value to the final customers (Formoso et al., 2015; Koskela et al., 2013). Practices associated with LC have demonstrated the potential to enhance production efficiency and sustainability performance (Kim & Bae, 2010; Rosenbaum et al., 2014; Saieg et al., 2018). Provides evidence elucidating various approaches to studying the impact of lean construction on reducing environmental waste, showcasing reductions achieved in each study. However, it is worth noting that 2. Literature Review 32 specific lean methods, such as Just-in-Time (JIT), may inadvertently lead to increased carbon emissions due to frequent shipments required to maintain zero-inventory conditions (Hussein & Zayed, 2020). Moreover, a comprehensive solution for effective facility management with a lean management philosophy has not yet been investigated (Elmaraghy et al., 2018). 2. Literature Review 33 Table 2-3 -- Comparative analysis for the integrated methods with VSM to investigate construction wastes. Author(s) Method Material waste Carbon Emission Energy consumption Wate r Landfi lls Fuel Electric (Golzarpoor et al., 2017) DES-I/O ● 41% 41% - - ● (Fu et al., 2015) LCA - 42-44% - - - - (Wu et al., 2013) Low carbon lean system ● 29.39% ● - - ● (Kim & Bae, 2010) CEDST ● 10-20% ● ● - ● (Rosenbaum, Toledo and González 2014) VSM 50 to 100% ● ● - - 100% (Belayutham, González and Yiu 2016) VS-PM ● - - - ● - (Vilventhan, Ram, and Sugumaran, 2019) VSM 13.1857.37% - - - - ● (Nahmens and Ikuma 2012) SLIK 64% ● ● ● ● ● DES-I/O = Discrete Event Simulation – Input/output; LCA = Life Cycle Analysis; CEDST = Construction Environmental Decision-Support Tool; VSM= Value Stream Mapping; VS-PM = Value Stream-Process Map; SLIK = Safety and Lean Integrated Kaizen; ● = reported reductions without percentage 2.5.6. Value Stream Mapping Applying the Value Stream Mapping (VSM) method is indispensable for lean initiatives, offering a structured approach to visualize information and material flows while identifying potential improvements through waste elimination principles (Womack & Jones, 2003). Through data collection of production indicators sourced from involved stakeholders or via direct surveys and observations, two distinct VSMs can be developed: current and future maps. Both maps rely on key performance indicators such as NVA (Non-Value-Added), CT (Cycle-time), WIP (Work-In-Progress), Takt Time (TT) and Queuing Time 2. Literature Review 40 integrated into university curricula through respondents' enthusiasm for the potential of lean construction. However, few respondents intended to teach lean construction principles in future courses. Surprisingly, 42% of professional respondents do not know what lean construction is, while 58% confirmed that lean is partially or fully applied in their workflow. Matias & Cachadinha (2010) Evaluated the potentials and barriers to LC adoption in Portugal by interviewing to represent ten contractor companies and identified that the primary cause of delays and waste is Request for Information (RFI). The LC is not well-heard in Portugal, but respondents have shown interest in adopting it and enthusiasm for its potential. Martins & Cachadinha (2013) Interviewed designers and owners to investigate the root causes of waste and evaluate LC potentials to tackle these problems. Surprisingly, most respondents did not know about the LC; they agreed that traditional procurement and RFI are the most frequent causes of construction waste. In interviews with owners and designers in Portugal, only a few companies are testing LC, and most companies are unaware of their potential. Portugal's lack of contractual structure leads to lean construction and the perception that meeting deadlines does not allow them to use LC (Martins, 2011). Despite awareness of LC potentials, another survey showed that the lean construction philosophy is not widely applied in Portugal (Pereira, 2014). 53% of lean applications in Portugal are 5s, and kaizen is mainly used in logistics management; 55% of respondents were not interested in learning lean construction philosophy (Caseiro, 2016). Action research was applied in a Portuguese company, and positive results have been reported for applying lean methods such as Kaizen, visual management, 5s, safety, and team boards (de Sousa, 2019). H. Just-in-time LPS; their work visualised the processes using VSM (Gonçalves, 2009). Comparative research is also performed. For instance, the adoption of LC philosophy experience was compared between Portuguese and Brazilian perceptions of and (Caseiro, 2016). Another example was studying the Danish experience in LC and attempting to propose a model for the Portuguese context based on this experience (Silva, 2008). LC is relatively unknown in Portugal (Gonçalves, 2009), and many critical barriers have been reported. 1) Informal planning and resistance to change were reported as critical barriers. 2) lack of top management support bureaucracy, 3) lack of lean awareness and understanding, 4) fear of new things, and 5) poor constraints analysis and lack of commitment to short-term planning (Gonçalves, 2009), 6) 2. Literature Review 41 the research encountered resistance to change and blockage from companies to share information, which are the most ranked barriers (Pedrosa et al., 2023). 2.6. Discussion By reviewing 54 relevant papers as the systematic literature review framework, efforts were made to ensure a comprehensive understanding of the covered topics, namely, waste elimination, Lean construction, and BIM, by addressing the disadvantages and limitations of a traditional literature review. These included designing a review protocol to avoid ambiguity, increasing search transparency, and identifying the study's suitability for replications. That said, it may be the case that a small number of the databases used (i.e., Scopus and IGLC) may have produced a slightly lower number of found documents. Nevertheless, the findings of this review indicate that waste elimination is a core concept of lean construction philosophy, but when it is integrated with BIM, few results have been reported, especially regarding production waste. Additionally, changing current research on lean construction and BIM does not show explicit guidelines for waste elimination. Understanding the underlying root causes of construction waste was a significant hurdle to building a holistic review of the waste elimination approaches. This assertion agrees with previous research such as (Formoso et al., 2020; Viana et al., 2012), who identified the difficulty of composing a systematic literature review on waste in construction because of four reasons: 1) persistent construction management theories in the current research and practice; 2) a holistic theory that explains waste propagation and its' characteristics is not presented yet.; 3) ambiguity in the reported collection methods for waste data; 4) variety in conceptualisation to address similar types of waste. 2.6.1. How does lean construction conceptualise waste? The concept of waste in lean construction has been integral since its emergence in 1992, emphasising the imperative of waste elimination for ensuring survival and enhancing the competitiveness of companies (Sacks et al., 2018). Within the construction literature, waste is conceptualised as a set of symptoms of an inefficient production system, including material waste, injuries, capital waste, and energy waste. In addressing waste, the lean construction approach diverges from traditional end-of-pipe and reactive strategies, which merely respond to symptoms of internal waste after its occurrence within a production system. Koskela, 2013 criticised the poor foundations in the construction management research in conceptualising waste and production concepts; even the relationship between them is the 2. Literature Review 42 purely economic view that only sees the world as input and output, which refused to include waste concepts with the existing construction management theories. Based on the transformation-flow-value theory, lean construction understands waste concerning value and flow concepts in conceptualizing construction waste. The former tends to refine the process of capturing the customer value from any misconception, and the latter utilises a mixture of social and technical methods to reveal the construction flows and expand the transparency to see wasteful (NVA) and not wasteful activities. These NVAs are introduced as a list of waste and categorized as a “waste list.” The list helps stakeholders provide another layer of analysis for waste root causes in a construction system. Waste is conceptualised in lean literature as a communication tool and unit of measurement. The first consideration of waste is to be conceptualised as an actionable language that facilitates discussion among stakeholders, where a storytelling approach describes the generated waste, the nature of this waste, and the consequences of this generated waste. This approach increases situational awareness among stakeholders who can collaboratively identify and address the root causes of production bottlenecks in their projects or organisations. Secondly, waste is a metric used to measure the root causes of inefficiency within a production system or organisation, and it provides insights into the key processes and operations that require further attention for improvement. This conception of waste also enables the stakeholders to recognise potential gains that can be realised through waste elimination initiatives (Bølviken et al., 2014). Thus, within a lean construction framework, waste elimination catalyses proactive learning, problem-solving, and continuous improvement rather than merely being a byproduct of production processes. 2.6.2. What are waste elimination factors imposed by lean construction? The waste elimination measures provided by lean construction can be categorized into two distinct facets: social and technical, as shown in Table 2-5. The waste elimination concept is centralised in lean construction foundations operationalised in production planning and control. Due to waste elimination concepts and functions embedded within lean production planning and control, planners can provide more reliable plans and trades work within more stable workflows. The literature review revealed eight fundamental principles identified: (1) reducing cycle time, (2) minimising variability, (3) fostering continuous improvement, (4) enhancing transparency, (5) decreasing batch size, (6) minimising rework, (7) reducing inventories and (8) mitigating defects (Koskela, 2000; A. Santos, 1999). Table 2-5 pointed 2. Literature Review 43 to the factors of LC for waste elimination based on lean principles in the Lean Foundations column. The second fold is streamlined from production planning and control functions; this study shortens the lean function to this role because it can be applied to various stages of Integrated Lean Project Delivery (design, supply, assembly, and use). The second aspect of the response stems from insights derived from production planning and control functions. Given its versatility in application across various stages of Integrated Lean Project Delivery (including design, supply, assembly, and utilisation), this study simplifies the lean function to this role. Table 2-5--Lean construction factors of waste elimination Measures Lean Foundations Production Planning & Control Functions Stable and Reliable Production Plans Social • Respect People • Decide by consensus. • Extend the network of partners. • Capture the customer value. • Align internal and external clients. • Feedback from downstream • Collaborative planning and control • Use Language/Action perspective. • Hand-off Management • Provide reliable promises. • Commitment planning • Increase Situational awareness. • Feedback loops • Upstream understand downstream work • Measure commitment Technical • Simplify processes. • Use Reliable Technology • Reduce Cycle Time • Reduce batch size. • Increase Transparency • Standardise • Continuous Improvement • Pull Planning • Work structuring. • Visualise material and information flow. • Pull Planning • Constraint Analysis • Interference Analysis • Alignment Analysis • Root Cause Analysis • Schedule buffers allocation • Control Metrics • Control Charts • Resources Allocation • Product and flow visualisation • Production Stability • Continuous one-piece flow • People commitment • Variability Reduction • Workable backlogs • Quality Improvement • Control Actions • Workflow reliability metrics 2. Literature Review 44 2.6.3. What are waste elimination factors imposed by lean construction (LC) and Building Information Modelling (BIM)? Most collected documents used BIM as a parallel process with lean construction implementation regarding waste elimination based on LC-BIM. This assertion agrees with the results presented by Ratajczak et al. (2017). Regarding waste elimination measures offered by adding BIM to LC, both systems appeared to tackle the root causes of waste caused by fragmentation between process and product information (Dave & Sacks, 2020). However, the literature never quantified waste elimination, which appears somewhat necessary. From a strategic point of view, few lean design studies have actively pursued implementing BIM standards (ISO, 2018) while executing BIM workflows in Lean construction projects. However, no correlation was determined between these standards and waste elimination. The extant literature failed to identify the impact of waste elimination practices on BIM standards. This gap would suggest future research agendas focusing on finding the potential waste elimination-based BIM standards currently not prioritised in the reviewed literature. Again, the literature considered BIM's potential significant in streamlining LC principles, methods, techniques, and tools; therefore, many waste elimination measures were initiated to improve information flow, as shown in Figure 2-16. It is interesting to note consistent overlaps of 'simulation' and 'quantity-take-off 'during production planning and control by streamlining reliable and real-time production Information to stakeholders that use LPS and LBMS. However, only developing LC-BIM integration during lean supply, planning, and control functions. For example, little research has been conducted to study the impact of LC-BIM on the "use" and "lean design" stages, which include renovation, retrofitting, expansion, and demolition (Elmaraghy et al., 2018). This assertion would suggest that the literature considered that most construction waste is caused by planning and control in the lean assembly stage and is rarely generated directly or indirectly during the lean design and supply stages. This perception is a denial acknowledged by (Cheng & Ma, 2013; Uusitalo et al., 2017), whose findings reveal that the BIM process contributes significantly to construction production planning and control reliability. The literature has attempted to develop software that harnesses BIM functionalities to automate LPS implementation (Dave, 2013; Heigermoser et al., 2019; Sacks et al., 2010; Schimanski et al., 2021). These systems share some characteristics: 1) visualising the process and product information; 2) providing digital kanban cards; 3) digitising constraint analysis workflow and managing constraints data; 2. Literature Review 45 4) enabling communication between design management, project management, site management, and crews; 5) monitor project, process and operation progress, 6) enable the production metrics to highlight wastes. Such software advances the planning functionalities offered by commercial BIM systems by increasing the planning details from master schedules only to mediumand short-term planning. Additionally, they enable the feedback from labourers at the coal face when data entry forms are available for each planned task. Moreover, tracking technologies such as IoT (Dave et al., 2016), GPS, beacons, and autonomous drones (Lin & Golparvar-Fard, 2021) enable automatic information capture to monitor production progress and material supply and logistics statuses. Such technologies would be fundamental to realise initiatives such as artificial intelligence, Big Data, Digital twins for construction (DTC), and Industry 4.0 (McHugh et al., 2022). Based on the above discussion, this chapter presents a conceptual framework (Figure 2-16) that assembles production planning and control functions based on lean construction and the existing functionalities of BIM to streamline waste elimination measures, which eventually provide more stable and reliable production plans. The links between concepts in the presented framework, Figure 2-17, will be a basis for hypothesis generation in the following chapter (Chapter 5). 2. Literature Review 46 Figure 2-16-- Critical factors of BIM functionalities supporting lean construction functions. Figure 2-17-Generic conceptual Framework for waste elimination model based on Lean construction and BIM. BIM Processes Produc�on Planning & Control Func�ons Stable and Reliable Produc�on plans Waste Elimina�on Lean Founda�ons 2. Literature Review 47 2.7. Conclusions This chapter presents an in-depth literature review based on the systematic literature review methodology according to (Tranfield et al., 2003). The review objective was to understand: i) how lean construction (LC) conceptualises waste and ii) what waste elimination factors are imposed by LC and Building Information Modelling (BIM). Based on predefined exclusion and inclusion criteria, the number of records has been filtered from 406 to 54 relevant documents. A time analysis was presented to illustrate the accumulation of waste elimination knowledge based on LC and BIM between 1992-2021. This analysis shows that the emergence of the waste elimination concept based on LC has been evolving since 1992, while an integrative approach of LC-BIM was first reported in 2008. The analysis also indicates a small quantity of theoretical and review research compared to empirical research. This assessment shows a lack of theoretical foundations for LC-BIM topics, and even if it exists, waste elimination was implicitly discussed or separately analysed. Waste is a high-level and context-specific concept hindering production performance and negatively impacting social, economic, and environmental dimensions. The theoretical papers discussed the idea of construction waste in terms of nature, conceptualisation, taxonomies, and propagation and reviewed cause-effect relationships between different types of waste. The lean construction community seeks a holistic understanding of the root causes of production waste and the consequences on the overall supply chain. The nature of production waste inherits the characteristics of the construction processes in terms of dependencies between tasks. (Formoso et al., 2015) It shows that the pattern of waste propagation can be mapped into networks of waste cycles, typically initiated by failures in construction management functions (production planning and control, product design, site layout planning, previous stages, and quality management). The taxonomy of construction waste has been adopted from manufacturing research (i.e., Ohno's list of waste (Ohno, 1988). This list has been transferred into practice without reflection on the construction peculiarities. Hence, a context-specific taxonomy is needed for the construction, as Formoso et al. (2020) suggested. Bølviken & Koskela (2016) reported that the concept of waste had been ignored in the construction management research, and the practice lowers the importance of waste elimination in their agenda. This assertion shows that future research should focus on developing new theories based on production theories to understand and capture the characteristics of production waste in the construction context. However, the consensus on the dominance of Making Do and its enormous impact not only generates production waste (NVA) but can also hold responsible behaviour for hindering the overall 2. Literature Review 48 project performance. The current waste taxonomies do not directly address the Making Do waste generation and its effect on the construction industry. For example, Formoso et al. (2020) identify Making Do as one of the types of production waste that occurs at the beginning or continuation of a task with incomplete resources, resulting in reactive decisions such as ‘firefighting’ or ‘low-fruit gripping,’ they do not describe its peculiarities in detail. While these taxonomical frameworks help conceptualize construction waste, such an approach is insufficient in analysing Making Do. A dedicated taxonomy for Making Do waste could consider factors like inaccurate task planning of the project schedule, ineffective communication between trades, unpredictable arrival of materials, and improper control of the resources. In order to gain insight and progress in the comprehension of Making Do, researchers can ground it in a particular taxonomy, measure its duration and the extent of its influence, and create effective strategies for its reduction. The empirical literature themes have been classified following the proposed model of Integrated Lean Project Delivery (ILPD) (Ballard & Howell, 2003). The thematic analysis shows that waste elimination is a primary part of lean construction projects from the early stages, which should be defined in relational contracts as project purpose in the project definition stage. However, the existing contractual structure (e.g., Design-Bid-Build, Integrated Project Delivery (IPD)) does not include implicit clauses for waste elimination in its templates. This gap can hinder stakeholders' commitment to waste elimination due to the absence of regulatory measures (Matthews et al., 2000; Vilasini et al., 2014). Concerning the sustainability dimensions, the literature reported "improving sustainable performance" as an explicit objective within some lean construction projects. The consensus by using Value Stream Mapping (VSM) shows that production waste elimination is not only reducing environmental waste (such as material waste, landfills, gas emission, energy consumption, and wastewater) (Kim & Bae, 2010; Rosenbaum et al., 2014). Nevertheless, social benefits can also reaped by reducing the likelihood of accidents and injuries on-site or in construction factories (Gambatese et al., 2017). The economic impacts of waste elimination are presented in terms of reducing operation costs and shortening the lead time of overall project delivery, which are objectives of any lean initiative (Vrijhoef, 2020). The lean design has an essential role in generating information for fabrication, logistics, assembly, and facility use, but the literature lacks approaches that study the effect of LC-BIM on the whole supply chain. The role of BIM is significant if lean construction is to be continued as the dominant construction management method in the future; the current state of queue shows that the only possible way to automate lean systems is by melting the boundaries with BIM. Thus, the parallel use of both concepts is not enough. A typical integration between LC-BIM exists in production planning and control systems 2. Literature Review 49 (i.e., Last Planner System (LPS) and Location-Based-Management-System (LBMS)). LPS and LBMS have complementary functions in creating reliable and stable plans together. The potential of harmonised use of both systems has been revealed and applied in previous studies (Dave et al., 2016; Seppänen et al., 2010, 2015), which show that both systems are highly correlated; for example, LBMS become deficient without effective social communication for information, on the other hand, LPS requires more metrics than PPC to measure the quality of the construction flow which LBMS offers. Both systems rely on intensive production information in terms of quantity, duration, and task definition, and BIM can provide that, which supports production planning and control with the necessary information when needed. A more critical waste elimination-focused research agenda is currently required to investigate the integration mechanisms of the BIM and LC concepts across the supply chain. Given that the diffusion of BIM and LC have been studied since 2008, it is possibly surprising that so few have investigated their combined implementation for waste elimination purposes. A comprehensive understanding of this combined BIM and LC requires considerations of the processes, technology, and diffusion mechanisms to inform practitioners and policymakers of the industry. To overcome the limitations of this study in terms of covering records collected from databases other than Scopus, databases such as Web of Science (WOS), Google Scholar, and EBSCO literature collection and review can be implemented in the future. 3. Research Methodology 56 Research approach Mixed, often Qualitative Qualitative Quantitative Mixed methods This research will mix constructivism and positivism ontologies, and the former seeks to construct the knowledge based on practitioners' views on construction projects. The latter directly observes the making-do phenomena through collected data from the construction projects. Both worldviews decide the mixed research approach, and the qualitative research approach provides the context of the significant perspective. The role of the researcher can be objective in terms of using closed-ended questions to the research subjects. Alternatively, open-ended questions that the role of the researcher is an actor. These worldviews and their followed research approach are selected because of the available data collected. 3.2.2. Research Design There are several strategies for conducting research, and the literature abounds with contradictory claims regarding the right approach to achieve specific research objectives or problems (Creswell and Creswell, 2017). As such, considerable effort is required to choose the appropriate research approach and data collection methods in response to the research questions. This investigation enables the researcher to plan while properly considering the research paradigm, strategies, and techniques. Three research approaches are commonly used in construction management research: quantitative, qualitative, and mixed methods. Quantitative methods Engaging in quantitative research entails the meticulous measurement of data through suitable scales. Two fundamental questions guide this process: What aspects should be measured, and how should these measurements be executed? (Fellows & Liu, 2015). The choice of measurement scale is paramount, ensuring ease of data collection, accuracy, and the validation of subsequent analyses. In an ideal quantitative approach, the researcher's influence on the collected data is minimised to uphold objectivity and mitigate the potential impact of personal beliefs. Quantitative methods adopt an empirical stance to measure data about key variables using appropriate scales. Building on existing findings to enable replication, researchers formulate inquiries. Predominantly following a deductive research paradigm, quantitative methods emphasise hypothesis and theory testing alongside exploring variable relationships. It commences with a hypothesis or conceptual model construction grounded in existing literature and theories; subsequent stages 3. Research Methodology 57 encompass data collection and analysis to substantiate these initial constructs. This approach resonates with the positivist philosophical standpoint (Fellows & Liu, 2015). Efficiency in data collection and proficiency with recognised quantitative data analysis techniques constitute the primary strengths of quantitative research. Noteworthy examples include experimental designs and survey research. Experimental design assesses the impact of independent variables on one or more dependent variables. In this context, the dependent variable represents the measured response. Between-subjects designs involve randomly assigning subjects to experimental conditions, enabling data collection across various groups. On the other hand, survey research gauges trends, attitudes, or consensus within a population sample to generalise findings to a broader context. It facilitates the extension of conclusions from a sample to the entire population. Survey research predominantly employs questionnaires as data collection tools, encompassing closed-ended and open-ended queries. While self-administered and posted questionnaires are standard, the popularity of web-based questionnaires, utilising platforms like SurveyMonkey, QuestionPro, Google Surveys, and Microsoft Forms, has grown. Leveraging web-based questionnaires enhances best practices, encourages participation, reduces potential errors, and streamlines data processing and analysis. Qualitative methods Qualitative research predominantly adopts an inductive approach centred around theory generation. Within this research framework, textual analysis is a common practice. The research process initiates with data collection and subsequent analysis to discern patterns, culminating in constructing a theoretical framework. Qualitative research offers a notable advantage by delving deeply into the phenomena under investigation, although it often demands a considerable time investment. Qualitative research encompasses diverse methodologies, including ethnography, grounded theory, phenomenology, and case study. Qualitative research methods find their primary utility in comprehending the underlying causes, principles, and behaviours associated with a given problem or issue, as expressed by participants (Fellows and Liu, 2015). The researcher's role includes determining variables and the corresponding measurement methodology. For instance, phenomenological research entails an inquiry design to elucidate individuals' experiences of a phenomenon, frequently accomplished through interviews (Giorgi, 2009). Grounded theory is a research strategy in which a theory of process or behaviour is devised and grounded in the participants' views (Creswell & Creswell, 2017). 3. Research Methodology 58 On the other hand, case studies facilitate an in-depth examination of a case, process, event, or individual within the bounds of specific temporal and activity parameters, utilising an array of data collection techniques (Yin, 2018). The informing case studies will be used in this thesis as a method of research that includes a detailed examination of Making-Do waste as a single instance to conclude broader insights and theories while understanding complex phenomena from different perspectives (F. Chen et al., 2020; Dixon-Woods et al., 2007; Ickis & Omazić, 2013). Mixed methods The mixed methods approach harmonises quantitative and qualitative data, addressing the limitations of individual methods and offering more profound insights into the research question. While terms like multi-methods, integrated methods, and quantitative and qualitative methods are used, "mixed methods" is the most prevalent designation (Creswell & Creswell, 2017). This approach leverages triangulation, wherein diverse data sources are scrutinised to establish research themes. In line with Creswell's categorisation based on the sequence of method application, mixed methods exhibit three primary types: 1) convergent parallel mixed methods, involving simultaneous collection and comparison of qualitative and quantitative data; 2) explanatory sequential mixed methods, commencing with quantitative data collection and analysis before proceeding to qualitative data collection and analysis; and 3) exploratory sequential mixed methods, initiating with qualitative data collection and analysis, followed by corresponding quantitative phases. These typologies provide structure to data integration and offer researchers a framework for conducting comprehensive mixed methods studies. 3.3. Data analysis methods 3.3.1. Pareto (80/20) rule A Pareto chart combines bars and a line graph, with individual values arranged in descending order on the bars and the cumulative total represented by the line. The chart is named after the Pareto principle, which, in turn, is derived from the economist Vilfredo Pareto. The primary objective of using a Pareto chart is to highlight the most significant factors. It is commonly utilised in quality management as a tool for quality control. The chart proves helpful in identifying the most probable causes of defects in products and processes, aiding in focused problem-solving and decision-making within various industries. 3.3.2. Factor Analysis 3. Research Methodology 59 Factor analysis is a statistical method for discovering a rotated subset of principal components that, instead of the original variables, reduce the amount of noise in the data while retaining the information that stands for the actual variables in the "Multivariate Data Analysis" (Hair et al., 2019). It is mainly used in data reduction and dimensionality reduction, where the objective is to decrease the number of variables into fewer factors, subject to the best preservation of the original variability. Factor analysis intends to discover the underlying structure in the data using its ability to recognise common patterns among the observed variables. These unknown factors are the basis that encompasses the aspects of the subject that are not directly measured but are perceived through the gathered variables. The factor analysis procedure differs in how the factors have been extracted from the correlation matrix of the observed variables with the rotated factors that yield a more straightforward and easier-to-interpret solution. The resulting factors shed light on the interconnections between the observed variables and the main factors aiding in understanding the outcomes. In summary, exploratory factor analysis is vital for revealing the distinctive content of multivariate sets and determining the latent factors that produce the complex pattern of associations in the data set. 3.3.3. Chi-Square Association test The chi-square test uses a contingency table to evaluate hypotheses for statistical analyses, especially when the sample size of the collected data is large (Field, 2005; Hastie et al., 2009). This test mainly examines the relation between two categorical variables in a cross-tabulation setting. These crosstabulations classify the data into two discrete or groupings, implying two independent variables in the dataset. The rank of the table refers to the categories of the first variable, while the column indicates the category of the second variable; each variable must have more than one category or factor. Any cell in the chi-square or contingency table sums up the number of occurrences that fall within a particular combination of the categories, making it easier to understand the interrelation between variables. The chi-square test encompasses two primary types of statistics: Pearson's and Likelihood ratio chi-squares. The formula for each is stated below. In Equation 3.1, pi represents the summation of all probabilities pi, across all categories i from 1 to r . In a random sample of n observations from a population, these observations are divided into r mutually exclusive classes, each with corresponding observed numbers x ij (for i = 1, 2, ..., r ). The null hypothesis states the probability pi that an observation belongs to the ith class. X2 represents the Chi-square ( X2 ) statistic, a measure of how observed data deviate from expected data. That tests the interdependence of two categorical variables and the goodness-of-fit between observed and expected frequencies. For each cell in the contingency table, X2 calculates the ratio of the square of observed frequency (Oij) to the 3. Research Methodology 60 expected frequency (E ij ). The result is the X2 statistic, and in order to decide whether the differences observed are statistically significant or not, it is compared with the critical value from the Chi-Suqare distribution with the relevant degrees of freedom. �𝑝𝑝𝑖𝑖= 1 𝑟𝑟 𝑖𝑖=1 Equation 3.1 𝑋𝑋2=�𝐸𝐸𝑖𝑖𝑖𝑖 =�𝑂𝑂𝑖𝑖𝑖𝑖2 𝐸𝐸𝑖𝑖𝑖𝑖 −𝑛𝑛 𝐶𝐶 𝑖𝑖=1 𝑟𝑟 𝑖𝑖=1 Equation 3.2 where 𝐸𝐸𝑖𝑖𝑖𝑖 =(𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑟𝑟𝑡𝑡𝑟𝑟𝑖𝑖) x (𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑐𝑐𝑡𝑡𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐𝑗𝑗) 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑐𝑐𝑐𝑐𝑐𝑐𝑛𝑛𝑛𝑛𝑟𝑟 𝑡𝑡𝑜𝑜 𝑡𝑡𝑛𝑛𝑜𝑜𝑛𝑛𝑟𝑟𝑜𝑜𝑡𝑡𝑡𝑡𝑖𝑖𝑡𝑡𝑐𝑐𝑜𝑜 Equation 3.3 Likelihood-ratio chi-square statistic represents the ratio between observed and expected frequencies in the association table as expressed in Equation 3.5 𝐺𝐺2= 2 ��𝑂𝑂𝑖𝑖𝑖𝑖 ln �𝑂𝑂𝑖𝑖𝑖𝑖 𝐸𝐸𝑖𝑖𝑖𝑖� 𝐶𝐶 𝑖𝑖=1 𝑟𝑟 𝑖𝑖=1 Equation 3.4 Where 𝑂𝑂𝑖𝑖𝑖𝑖is referred to the observed value for frequency in the cell (i,j), 𝐸𝐸𝑖𝑖𝑖𝑖 is the expected frequency value for cell (i,j). A contingency table's degree of freedom (DF) is the number of parameters that can vary independently in a statistical computation. DF is the number of cells in the association table that can be changed freely once the row and column totals are fixed. Equation 3.6 shows the formula of DF, where R is the number of rows and C is the number of columns. DF is essential when using X2 statistics because DF helps determine the critical value of the Chi-Square distribution, allowing the researcher to compare calculated Chi-Square values to decide whether to reject the null hypothesis. 𝐷𝐷𝐷𝐷=(𝑅𝑅−1) × (𝐶𝐶−1) Equation 3.6 3.3.4. Linear Regression Analysis Linear Regression Analysis (LRA) is a statistical modelling technique that examines causal relationships within predetermined datasets that are linear in the parameters. It formulates linear equations, specifically the best-fit line (commonly known as the least square line), to explicate the associations between predictor and dependent variables. LRA proves invaluable for researchers in predicting and 3. Research Methodology 61 forecasting the behaviour of modelled data, exhibiting convergence with methodologies inherent to machine learning. The Linear Regression model serves the purpose of estimating parameters involving unknown variables, denoted as β for parameters; predictors, denoted as 𝑋𝑋𝑖𝑖, dependent variables represented as 𝑌𝑌𝑖𝑖 and errors signified by 𝑒𝑒𝑖𝑖. The resultant linear formula for the regression function 𝑌𝑌𝑖𝑖 is expressed as follows: 𝑌𝑌 𝑖𝑖 =𝑓𝑓(𝑋𝑋 𝑖𝑖 +𝛽𝛽)+𝑒𝑒 𝑖𝑖 Equation 3.7 The resulting regression models should be tested for fitness, and several statistical indicators describe the fitness of regression models, such as standard deviation (S), R2, adjusted R2, and Durbin-Watson statistic. S stands for the standard deviation of the residuals that measures the distance between the data values and the fitted values provided by the regression model. This statistic is measured in the same units as the dependent variable (response). A lower value of S indicates a better fit for the model to describe the data. However, a lower S value alone does not guarantee the validity of the underlying model assumptions. It is essential to examine residual plots to verify potential violations of assumptions, thereby ensuring the model's reliability. R2 (R-squared), or the coefficient of determination, represents the proportion of variance in the dependent variable accounted for by the independent variables in the model. It is calculated as one minus the ratio of the residual sum of squares (representing the unexplained variation) to the total sum of squares (representing the total variation within the data). R-sq is a crucial metric for assessing the goodness of fit of the regression model, with higher values indicating a more significant proportion of explained variance and a better fit to the data. The R-sq value ranges between 0% and 100%, where 0% indicates that the model explains none of the variations, as shown in Figure 3-1 (c), 50% indicates that the model explains 50% of the variation, as in Figure 3-1 (b) and 100% indicates the model explains all the variation Figure 3-1 (a). 𝑅𝑅2= 1 −𝑆𝑆𝑆𝑆𝑅𝑅 𝑆𝑆𝑆𝑆𝑆𝑆= 1 −𝑆𝑆𝑆𝑆𝑆𝑆 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑒𝑒𝑆𝑆 𝑅𝑅𝑒𝑒𝑅𝑅𝑆𝑆𝑒𝑒𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑛𝑛 𝑆𝑆𝑆𝑆𝑆𝑆 𝑅𝑅𝑓𝑓 𝑡𝑡𝑅𝑅𝑡𝑡𝑆𝑆𝑡𝑡 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑒𝑒𝑅𝑅 = 1 −∑(𝑦𝑦𝑖𝑖−𝑦𝑦�𝑖𝑖)2 ∑(𝑦𝑦𝑖𝑖−𝑦𝑦�)2 Figure 3-1-- R squared values representing the goodness of regression line fit (a) (b) (c) 3. Research Methodology 62 The Durbin-Watson statistic tests first-order autocorrelation in time series data without lagged independent variables ranging from zero to four. A value close to 2 shows no autocorrelation, while an appropriate range is about 1.50 to 2.50. Values below 1.50 indicate positive autocorrelation; values above 2.50 indicate negative autocorrelation. However, the Durbin-Watson statistic becomes unreliable in models that contain lagged variables. High autocorrelation, in series with seasonal solid patterns, suggests that ordinary least squares regression may not be the most accurate method to use when forecasting. 𝐷𝐷𝐷𝐷=∑(𝜀𝜀𝑡𝑡−𝜀𝜀𝑡𝑡−1)2 𝑟𝑟 𝑖𝑖=2∑𝜀𝜀𝑡𝑡2𝑟𝑟 𝑖𝑖=1 3.3.5. PLS-SEM Statistical Technique The primary virtue of structural analysis lies in its ability to propound, explore, and implement scientific discoveries, making it an inevitable passage for contemporary researchers of either natural or social sciences. The Partial Least Squares Structural Equation Modelling (PLS-SEM) was a second-generation approach among the multivariate analysis methodologies. On the other hand, Covariance-Based Structural Equation Modelling (CB-SEM) prevailed in social science literature till around 2010, whereas CB-SEM was at the apex. Therefore, researchers should be aware of these historical developments to embrace new possibilities. At the same time, PLS-SEM, which is above the CB-SEM in popularity nowadays, has experienced a significant spike in the number of publications made in many disciplines such as marketing, management, information management systems, accounting, strategic management (Hair et al., 2019), construction management (Wang et al., 2024) and many others. Within data analysis, univariate analysis involves studying a single variable to characterise its distribution with measures of central tendency and dispersion through graphical representation. This kind of approach of bivariate analysis can be taken to the next level by examining the relations between two different variables (or dependent and independent variables). It is done to find the patterns and associations in the data. Here, correlation analysis and regression can be utilised. Next, multivariate analysis deals with relationships among various multi-dimensional variables and uses a diverse set of statistical techniques, e.g., factor analysis, cluster analysis, and logistic regression, to discover unknown structures and intricate patterns within the data. This integrative approach allows researchers to tune into the complexities of interrelationships and the interactions among numerous variables, thus opening an opportunity for a more detailed comprehension of the phenomena (Hair et al., 2019). 3. Research Methodology 63 These multivariate statistical methods are often referred to as first-generation methods (Fornell, 1982) and can be categorised into two main types: conducting exploratory and confirmation studies. The confirmatory approach evaluates the theories developed and accepted a priori via henceforth known methods. Different attachment methods consist of logistic regression, multiple regression, analysis of variance, and confirmatory factor analysis. While prediction techniques aim to forecast existing or future patterns, the primary role of exploratory methods is to gain insights into the data when little or no previous knowledge is available. The exploratory methods include K-means clustering, multidimensional scaling, and exploratory factor analysis. Regarding the choice of software, one has several options for implementing partial least squares structural equation modelling (PLS-SEM) and covariance-based structural equation modelling (CB-SEM). Regarding PLS-SEM, PLS-Graph by (Chin et al., 2003) is the mainly used option, and SmartPLS (Ringle et al., 2005) is also a favourite. Besides that, R and semPLS by Monecke and Leisch (2012) are also considered. While CB-SEM can be carried out using programs such as LISREL, AMOS, CALIS, EQS, and SEPATH, it comes with some constraints like measurement error, sample size, and unobservable errors. The SEM procedure: • Specify the measurement model: this step provides a rigorous analysis of the fitness of data with the modelled variables. Table 3-2 summarises the model indices and their acceptable values. As a composite of CMIN/df, Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Residual (RMSEA), and Standardized Root Mean Square Residual (SRMR), these model-fit indices are commonly used for the overall evaluation of the model. Some calculated statistics significantly differed from the established below-standard value defined in previous research (Bagozzi & Yi, 1988; Bentler, 1990; Hu & Bentler, 1998; Schumacker & Lomax, 2004). Table 3-2 -- Model fitness measures Fit indices Threshold Source Probability level Insignificant (Bagozzi & Yi, 1988) CMIN/DF Between 1 and 3 (Schumacker & Lomax, 2004; Ullman & Bentler, 2012) CFI >0.95 (Bentler, 1990) 3. Research Methodology 64 TLI >0.08 SRMR <0.08 (Hu & Bentler, 1998) • Identify the Structural model: a structural model shows the expected relationships between the constructs (also called unmeasured or latent variables) and, in some cases, between the variables (also called observed variables) that serve as the measurement components of the constructs. The structural model describes the relationship between the variables, including their intermediate and direct relationships, and is often represented through a path diagram. This thesis applies two types of path analysis: mediation analysis and moderation analysis, as illustrated in Figure 3-2. Figure 3-2 -- Mediation analysis and Moderation analysis. • Mediation analysis examines whether the effect of an independent variable (X) on a dependent variable (Y) is transmitted through a third variable known as the mediator. It explains how one variable affects another, including paths from X to the mediator, mediator to Y, and X to Y, and direct effect (a*b) (Baron & Kenny, 1986). The analysis involves the significance of the indirect effect (a*b) and often uses methods such as the Sobel test or bootstrapping for the significance of testing (Hayes, 2013). • Moderation analysis examines whether the effect of an independent variable (X) on a dependent variable (Y) is moderated through a third variable known as the moderator. It defines the conditions of the relation between two variables changes, and this consists of the main effect of X and Moderator and the interaction effect of X*Moderator on Y. Statistical analysis is typically used to analyse the significance of the interaction of the term (X* Moderator) whereby an assessment of its significance is usually carried out; sometimes simple slopes analysis is used for explaining this interaction (Baron & Kenny, 1986). 3. Research Methodology 65 3.4. Exploratory and Confirmatory Factor Analysis This thesis employs Exploratory Factor Analysis (EPA) and Confirmatory Factor Analysis (CFA) to discover the underlying structure of LPS and BIM variables. Table 3-3 summarises the distinction between the two techniques. EPA is often used at earlier stages of the research when no priori hypotheses are available about the relationships between variables and factors; the data drive the procedure and investigate the underlying factor structure by identifying the number of factors and loadings of observed variables without any priory structure (Leandre R. & Duane T., 2011). EPA often utilises factor rotation to achieve a more straightforward and interpretable structure of factors. The CFA is often used to confirm the results of EPA by ensuring that each factor loads exclusively onto one construct; it increases the clarity and precision of the modelled structure for relationships and data. Based on established research findings and theories, CFA tests existing hypotheses about the factors' structure. The researcher assigns the number of factors, several variables associated with each factor, and the relations between factors. Model fit indices such as Chi-square, RMSEA, TLI, and CFI test whether the data fits or rejects the modeled structure(Hair et al., 2019). Table 3-3-- Exploratory and Confirmatory Factor Analysis Exploratory Factor Analysis Confirmatory Factor Analysis Reductionist It is b eneficial for condensing extensive datasets with numerous indicators. It a lso aids in identifying any overlap or ambiguity in indicators, particularly in cases where they may measure multiple constructs simultaneously. Ensure that each indicator loads exclusively onto one construct, preventing ambiguity and enhancing the clarity and precision of the model. Matrices Correlation matrices are commonly utilized , yet they pose challenges when comparing parameters across different samples. It uses a covariance matrix and is more adept at handling comparisons across samples. Using the covariance matrix, SEM is better equipped to manage comparison across samples, offering greater robustness and accuracy in parameter assessments. Rotation Considers data rotation It does not consider rotation 3.5. System Modelling and Simulation Systems thinking provides tools for better understanding complex management problems. A key recommendation from system thinkers for problem-solving is to shift the focus from addressing the 3. Research Methodology 72 Figure 3-9 - Common reference modes (Kirkwood, 1998) 3.5.4. Stock and Flow Diagram (SFD) In the model formulation phases of SDM, CLDs are transformed into Stock and Flow diagrams (SFDs). In typical SFD, there are six types of objects, as shown in Table 3-4. The Stock element refers to a realworld process's levels, accumulation, and aggregations over time. Stocks change through Flow elements, therefore defining the dynamicity of the system. The cloud element of SFD refers to irrelevant information to the model or information that leaves the model without affecting the Stock or the flow. Dynamic variables are auxiliary or intermediate functions that dynamically interact with the stocks and flow in the system. Parameters are similar to dynamic variables but are static values that provide the system boundary variations or conditions. Dynamic variables and parameters are connected to SFD using links identical to those used to formulate CLDs. Table functions represent a series of collected data containing functions as arguments and values. Mathematical formulas are necessary to drive changes in the SFD models, as the next section discusses the foundations of stock and flow calculations. Table 3-4: The components of the Stock and Flow Diagram SFD components Definition Legend Stock System levels, states, accumulation, and integrations are only Stock Exponential Growth Oscillations Goal-Seeking S-Shape Performance Time Performance Time Performance Time Performance Time Goal 3. Research Methodology 73 controlled by change rates (flow); another stock level cannot directly influence the stock level. Flow Changes (or rate of changes) in each situation Cloud It is a state used when the information is irrelevant to the purpose of analysis. Dynamic and static Parameters Intermediate concepts that consist of functions of stocks and affect flows. Constant values form the boundary conditions for the system. Table Functions A function is defined as a table using an argument and returned value. Links A direct line is a flow other than a flow that connects different SFD components. 3.5.5. SDM Mathematical Foundations The Stock and Flow Diagram (SFD) is unchanged without mathematical models working in the simulation software's back ends. From the stock definition as aggregation or accumulation, it is easy to predict the primary mathematical function behind SDM. Figure 3-10 represents the most straightforward Stock and flow diagram that will be used to explain SDM mathematics as expressed in Equation 3-1, which indicates that stock change over time equals the accumulation of difference between inflows and outflows with additions to initial Stock. A numerical approximation is used instead of direct integration, and two standard methods are used for approximation, namely Euler and Runge Kutta methods, which are used for approximation expressed in equations. Figure 3-10: Basic Stock and Flow Diagram fow outflow stock initial_stock inflow 3. Research Methodology 74 𝑅𝑅𝑡𝑡𝑅𝑅𝑠𝑠𝑠𝑠 (𝑡𝑡) = 𝑅𝑅𝑛𝑛𝑅𝑅𝑡𝑡𝑆𝑆𝑡𝑡_𝑅𝑅𝑡𝑡𝑅𝑅𝑠𝑠𝑠𝑠+∫�𝑅𝑅𝑛𝑛𝑓𝑓𝑡𝑡𝑅𝑅𝑖𝑖(𝑡𝑡)−𝑅𝑅𝑆𝑆𝑡𝑡𝑓𝑓𝑡𝑡𝑅𝑅𝑖𝑖(𝑡𝑡)�𝑆𝑆𝑡𝑡 𝑡𝑡 𝑡𝑡0 ……. (Equation 3-1) When t = final time step, t0 = initial time step, initial_stock = the value that simulation begins with Inflow = the flow that feeds the Stock, outflow = is the stock output. 1. The Euler approximation method, a widely utilized numerical technique, is favoured for its simplicity and efficiency. Unlike more complex procedures, it necessitates only a single equation for calculation. This straightforwardness makes it particularly accessible for various applications. The basic form of the Euler approximation formula is as follows: 𝑆𝑆 𝑅𝑅𝑡𝑡𝑅𝑅𝑠𝑠𝑠𝑠 (𝑡𝑡) 𝑆𝑆𝑡𝑡 =�𝑅𝑅𝑛𝑛𝑓𝑓𝑡𝑡𝑅𝑅𝑖𝑖(𝑡𝑡)−𝑅𝑅𝑆𝑆𝑡𝑡𝑓𝑓𝑡𝑡𝑅𝑅𝑖𝑖(𝑡𝑡)� 𝑅𝑅𝑡𝑡𝑅𝑅𝑠𝑠𝑠𝑠(0)=𝑅𝑅(0) 2. The Runge-Kutta approximation formula, often called the RK4 method, is a numerical technique for solving ordinary differential equations. This method is known for its accuracy and reliability but requires a relatively longer computational time compared to more straightforward methods due to the necessity of evaluating four intermediate values, denoted as k's, to determine the value of the stock. The equations governing the RK4 method are typically represented as follows: 𝑠𝑠1=𝑓𝑓(𝑅𝑅∗(𝑡𝑡),𝑡𝑡)=𝑅𝑅𝑛𝑛𝑓𝑓𝑡𝑡𝑅𝑅𝑖𝑖−𝑅𝑅𝑆𝑆𝑡𝑡𝑓𝑓𝑡𝑡𝑅𝑅𝑖𝑖 𝑠𝑠2=𝑓𝑓�𝑅𝑅(𝑡𝑡)+ℎ𝑠𝑠1 2,𝑡𝑡+ℎ 2� 𝑠𝑠3=𝑓𝑓�𝑅𝑅(𝑡𝑡)+ℎ𝑠𝑠2 2,𝑡𝑡+ℎ 2� 𝑠𝑠4=𝑓𝑓(𝑅𝑅(𝑡𝑡)+ℎ𝑠𝑠3,𝑡𝑡+ℎ) 𝑅𝑅𝑡𝑡𝑅𝑅𝑠𝑠𝑠𝑠=𝑠𝑠1+ 2𝑠𝑠2+ 2𝑠𝑠3+𝑠𝑠4 6 3.5.6. Model Validation and Verification Methods Model validation is critical in comparing the model against reality and similar models performed for the same problem under investigation. Validation should end with accepting or refusing the formulated hypothesis at the first step of SDM. According to Sterman, SDM can be validated through ten methods: 1) dimensional consistency test; 2) boundary adequacy test; 3) extreme condition test; 4) structured assessment test; 5) parameter variation assessment test; 6) integration errors test; 7) behaviour reproduction test; 8) failing member test; 9) surprise behaviour test; 10) structural sensitivity test (Sterman, 2000). The dimensional consistency test verifies whether connected variables have a direct relationship. It is used when developing SFD to expose wrong connections between model elements and highlight irregularities in the model behaviour. The structure test verifies model behaviour compared 3. Research Methodology 75 to historical data and produces time-based charts that can be analysed to determine if they are relevant to historical data. Such a test is essential to provide credibility to the formulated SFD. Modellers can compare various custom structures (archetypes) with their model under development. There are two robustness testing methods: (1) Sensitivity analysis, which is used to verify how the developed SFD is sensitive to small changes, and (2) Parameter variation, the opposite testing method to sensitivity analysis because it tests the model against wider variations. No model has ever been, nor will ever be, thoroughly validated. Therefore, terms such as "useful," "illustrative," or "reliable" are more apt descriptions for models than "valid" (Forrester, 1961). Attempting to test the validity of a model without a clear distinction of its purpose is meaningless; a system dynamics model addresses specific problems, not general system issues. The confidence we place in a model to help us analyze a given problem should not depend on whether the model can analyze a different problem (Richardson & Pugh, 1981). The basic structure of this research is validated based on project management models, as previously discussed. The Extreme Condition Test determines whether the newly added components can behave realistically under extreme values or policies (Han et al., 2012). 3.5.7. Construction management dynamics The complexity of construction management is a widely recognized and well-researched area. Overbudget and schedule overhead are common industry problems caused by recurrent changes, workflow interruptions, and information fragmentation. As a result, construction managers tend to use policies to avoid these issues by increasing resources, adding more overtime, push work to execution before refining its constraints, which may have positive outcomes to reduce cost and increase the pace of production, but in the longer run can lower productivity by demotivation, material waste, rework, waiting, quality defects among others. SDM has been demonstrated as an effective analytical tool in construction project management. To manage such complexity, the SDM must be capable of representing the constituting subsystems with their complex interrelationships and properties. Interdependencies among project processes Construction processes exhibit intricate interdependencies, where a change in one process can ripple through others, affecting resource utilisation, spatial allocation, and allocated time. Such changes or interruptions can lead to work blockages, particularly in areas directly impacted by these alterations. 3. Research Methodology 76 Construction dynamics The dynamic nature of construction projects is characterised by various time delays, including schedule execution, error identification, and correction, response to changes in scope or specifications, and waiting for materials or information directives. Additionally, productivity is adversely affected by the diversion of experienced workers' time to train recruits in the short run. To effectively represent, analyse and explain the complexities of socio-technical and managerial systems in construction, Highly developed guidelines that enable the application of system dynamics methodologies Nonlinear interrelationships The complex relationship between causes and effects in construction events often defies simple linear modelling. As an analytical tool, system dynamics underscores the nonlinear nature of these relationships in construction (Ford, 2010). For example, overtime may lead to increased fatigue, affecting motivation and ultimately impacting productivity (Lyneis et al., 2001). Such relationships are complex to be represented using linear mathematical functions because they follow nonlinear relationships. 3.5.8. Application of SDM in Construction Management In the field of construction management research, there is a diverse range of applications, including decision-making and policy analysis, performance assessment, rework and change management, scheduling, risk and contingency planning, resource management, productivity enhancement, cost planning and estimation, project control, bidding and procurement strategies, as well as health and safety considerations (Kedir et al., 2023). Policy analysis Decision-making and policy analysis research utilises SDM to analyse various scenarios and support policy development across system-level, high-level, and industry-level perspectives. That includes employing SDM to manage infrastructure projects, such as examining highway maintenance policies and assessing sustainability. Future research should explore incorporating feedback delay into SD models, refining them to analyze policy effects on subsystems and mitigating issues related to policy optimization and scenario analysis. Studying hybrid models can enhance understanding of social impacts in decision-making. 3. Research Methodology 77 Performance assessment Performance assessment in construction management involves evaluating various metrics against predefined standards or benchmarks. SDM has been utilised to enhance performance at the project and organisational levels, addressing areas such as project management quality improvement and overall organisational performance. Future research should focus on expanding SDM models to include more key performance indicators (KPIs) and capturing dynamic relationships between these parameters to determine overall performance. Additionally, understudied factors affecting performance, such as out-of-sequence work, warrant further investigation. Scope and design changes and rework Rework and construction project changes significantly impact schedule, cost, and quality performance metrics. SDM research is well-suited for managing these issues because it captures the dynamic relationships between planned activities and causes of rework or changes. Previous studies have focused on understanding these relationships to propose effective mitigation strategies. The research has used SDM to simulate change management strategies and their effects on project planning and performance (Love et al., 1999;D Love et al., 2000; Han et al., 2012). However, there is a stagnation in research on rework causation, possibly due to reliance on traditional methods like questionnaire surveys (Love et al., 2016). Further research is needed to explore qualitative and quantitative aspects of rework causation in different construction contexts and to understand the dynamic impact of changes on the recursive nature of rework during both the design and construction stages. Planning and scheduling SDM has been widely employed in project scheduling, particularly in studying the effects of scheduling delays. Early studies focused on mitigating delay and disruption in projects by optimising project duration extensions. Further research explored delays caused by compressing large projects for earlier delivery and investigated dynamic planning and the impact of scheduling in multiple projects. Recent studies have attempted to quantify the effects of various factors on project schedules, highlighting the complexity of achieving planned project milestones solely through schedule-driven management (D. N. Ford, 2002; Lee et al., 2003; Park & Peña-Mora, 2003). Resource management SDM can be considered a tool for managing construction resources, encompassing resource allocation, crew management, workforce learning evolution, and crew planning. Additionally, literature has 3. Research Methodology 78 employed SDM to assess the influence of resources on various aspects of project performance, including cost, schedule adherence, productivity, and overall project success. Productivity SDM offers a comprehensive framework for capturing the dynamics of construction processes and assessing the influence of various factors on productivity. Extensive research has examined numerous factors that affect productivity, including overtime, fatigue, labour motivation, schedule pressure, resource delays, shortages, and workflow interruptions (Lyneis et al., 2001; Porwal & Hewage, 2012). Moreover, recent studies, such as that by Kaya and Dikmen (2024), have specifically focused on evaluating the impact of different working hour arrangements and technologies on labour productivity. This research contributes to understanding workforce management strategies and their implications for overall project performance in construction contexts. 3.5.9. SDM and Lean Construction Research This section provides a comprehensive overview of studies employing System Dynamics Modelling (SDM) to investigate Lean Construction (LC) methodologies and techniques in the construction industry. Zhan et al. (2022) devised a conceptual SDM for quantity surveying practices, integrating Building Information Modeling (BIM) and Lean principles to delineate intricate relationships among people issues, internal impetus for BIM and Lean adoption, external drivers for LC and BIM implementation and impediments to implementation. Nguyen and Sharmak (2022) employed the Value Stream Mapping (VSM) method to evaluate environmental performance, demonstrating how Lean strategies like the Last Planner System (LPS) and Poka-Yoke reduce processing time and CO2 emissions. Meshref et al. (2023) proposed a decision-making framework for managing construction material waste throughout the life cycles of industrial projects, integrating BIM and Lean design into the design phase. Omotayo et al. (2020) diagnosed kaizen costing and budgeting practices for construction projects in Nigeria using SDM alongside the Analytical Hierarchy Process (AHP). Studies by Ko and Chung (2014), Ko and Kuo (2019), and Ko (2020) validated Lean Design Processes in formwork workflows, aiming to enhance formwork design efficiency through Lean principles and BIM. Cano and Rubiano (2020) developed a dynamic model to assess improvements in understanding non-value-adding waste within construction processes to enhance economic performance and behavioural aspects. Francis and Thomas (2019) quantified the effects of Lean construction practices on sustainability using causal loop diagramming techniques. Chinda (2009) evaluated effective Lean policies for fostering safety-oriented cultures within construction projects using SDM to explore diverse scenarios manipulating personnel, 3. Research Methodology 79 leadership dynamics, partnerships, and resource allocation variables. Collectively, these studies underscore the versatility and efficacy of SDM in investigating and enhancing various facets of Lean construction practices within the construction industry. 3.5.10. Available Software Packages There are several software tools available for simulation activity as system dynamic modelling, including Experimental Learning Laboratory with Animation (STELLA), Vensim, AnyLogic©, Powersim, iThink®, TRUE, DynaRisk, and DynamoTM (Kedir et al., 2023). These computer programs are evaluated by comparing the friendly user interface, easy modelling, capabilities of simulation engineering and trying alternative approaches, customizability options, integrability, open source, and programming language, which are displayed in Table 3-5. The assessment was done by AnyLogic© 8.7.11 professional (concerning other systems considered) conditioned by such parameters as error-checking, compatibility, capability, accessibility, and flexibility. The error checker consists of a Java-based set of rules that accept or reject a particular model structure. The rules are designed to identify inappropriate modelling issues and warn the user about the problem. Furthermore, it enables the smooth import of the models from Vensim, empowering the exchange process and, thus, importing them with other compatible software. The tool also provides a pre-set environment for model testing, enabling the researcher to personalise the test with Java code. Furthermore, the cloud service of AnyLogic© allows model sharing with people who do not have a license or an AnyLogic© account, hence increasing the model’s usability among those who lack prior experience in system dynamics. Numerical solvers for differential, algebraic, and mixed equations are designed as an integration to the AnyLogic simulation engine. Simulation (general simulation, interactive simulation (games and pedestrian simulators)), run comparisons, sensitivity analysis, calibration and optimisation, and Monte Carlo prediction are some of the capacities of the experiment. It has been explained that Any Logic runs on (Meshref et al., 2023). Furthermore, the ease of using the AnyLogic© software for simulation predictions can be customised using user input to include people having limited experience or no knowledge of system dynamics. It is worth mentioning that AnyLogic may be used to model in discrete event simulation (DES) and agentbased modelling (ABM), hence increasing the analyser's flexibility concerning different levels of depth that can be required. 3. Research Methodology 80 Table 3-5 -- Comparison between available software packages for System Dynamics Modelling functions Software Package STELLA iThink® Vensim PLE Anylogic® Powersim studio TRUE DynaRisk DynamoTM User-Friendly Interface x x x Intuitive Modelling x Advanced Model Building x x Simulation Capabilities x x x x Analysis Features x x Multi-Method Simulation x Decision Support x x x Customizable x x Interoperability x Open Source x x Programming Language Python Java VBScript and C++ Python License Commercial Commercial Commercial Commercial Freeware Commercial Commercial 3. Research Methodology 81 3.6. Conclusions This chapter serves to delineate the rationale underlying the chosen research methodology. It commences by dissecting established methodologies within construction management, culminating in the discernment of an apt approach harmonising with amassed data and research objectives. The adoption of critical realism and the integration of mixed methods stems from their capacity to transcend singular research paradigms and philosophical orientations. Notably, the strategic inclusion of System Dynamics Modelling (SDM) emerges as a promising avenue for lean construction policy testing, offering a comprehensive framework to address making-do waste and non-value-added inefficiencies. This methodology framework is reinforced by constructing a robust theoretical framework that enhances understanding of the making-do challenge and evolves into a practical research instrument closely intertwined with SDM. Acknowledging limitations, notably the abstract nature of SDM, we envision future possibilities by integrating Agent-Based Modelling (ABM) to enhance realism. In conclusion, this chapter signifies a methodological selection and framework construction, positioning us to methodically unravel the complexities of making-do waste while embracing its boundaries and future potentials. The next chapter assesses 1) understanding from the literature making-do waste and 2) the expectations of stakeholders on Lean construction planning and control methods and BIM for makingdo elimination. Understanding stakeholders' expectations is essential for deploying and accepting the best LC-BIM practices for reducing production waste within the construction industry; the next chapter discusses data collected and analysis processes. 4. Understanding Making-Do 88 "Emergent Behaviour," MD emerges from intricate variable interactions, yielding unexpected production outcomes due to uncertainty and variability. The "Equivalence of Success and Failures" principle highlights the value of learning from positive and negative MD incidents. In "Functional Resonance," MD manifests across levels—individual, team, and organisational—stemming from diverse contextual functions. Again, this non-linear behaviour of MD often traces back to a singular root cause, illuminating complex production system interconnections, which is delineated as interaction within a complex social system (CSS). 4.1.3. The relationship between MD and other construction issues This section interprets the relationship between MD and pivotal construction management phenomena, including buffering, variability, uncertainty, improvisation, and unfinished work. Comprehending these associations makes it feasible to augment comprehension of these phenomena. While exploring MD waste, Koskela (2004) discusses the concept of buffering and its stark contrast with MD. Buffering entails the temporary halt of materials awaiting processing, introducing a pause in the workflow. In contrast, MD involves negative waiting time, depicting scenarios where tasks progress without the complete standard inputs or even commence without at least one essential input. Fireman and Saurin (2020) emphasised that variability in both product and process has a substantial connection to the occurrence of MD. Particularly in complex systems such as construction, unforeseen variability is prevalent. Consequently, implementing a margin becomes imperative, serving as a proactive measure to prevent the immediate conversion of "making-ready" failures into instances of MD waste. In this context, slack is pivotal as an early indicator that operational performance diverges from the parameters defined by standardised operating procedures. Recognising the inevitability of a considerable degree of variability, prudent preparation for such circumstances becomes essential (Bertelsen and Koskela, 2005). Furthermore, the research accentuates that forestalling and mitigating the repercussions of MD can reap benefits from proactively identifying sources of variability and formulating corresponding slack resources. Pikas et al. (2012) establish a correlation between uncertainty and the phenomenon of MD. Pikas cited Winch's (2010) definition of uncertainty, which is characterised as the absence of essential information required by the project team for task execution, which remains unattainable. Notably, 90% of 345 recorded MD cases were attributed to uncertainty (Formoso, 2011). Uncertainty can engender defensive lose-lose behavioural patterns when planning is obscured by uncertainty. Within this context, MD decisions become primarily steered by risk aversion rather than the pursuit of optimal benefits for all parties involved. 4. Understanding Making-Do 89 The phenomenon of MD is closely intertwined with improvisation. Improvisation, a prevalent people practice, finds its presence even within well-structured business organisations (Formoso, 2011). Its significance becomes particularly pronounced when established rules and methods fall short (Formoso, 2011). The frequency of improvisation tends to escalate in the face of unpredictable events or pressing urgencies that necessitate immediate responses (Cunha, 2004; Hamzeh et al., 2012). Hamzeh et al. (2012) establish a connection between people's attitudes towards improvisation and the characteristics of a project, encompassing factors such as complexity, delivery method, time constraints, company rigidity, design challenges, and project type. It is within this interplay that the concepts of MD and improvisation are entwined. It is worth noting that improvisation often intertwines with the notion of MD. While these terms overlap in specific contexts, their distinct differences lack precise delineation within the literature. Fireman (2013) illuminated a strong correlation between MD and the emergence of Unfinished Works waste. The phenomenon of Unfinished Works, as observed by Sukster (2005), can often be attributed to a lack of harmonisation between Production Planning and Control (PPC) and Quality Management (QM) processes. This misalignment can result in tasks within short-term planning packages being prematurely deemed completed, leaving many small tasks unattended for the following week (Fireman, 2013). These instances of unfinished work are frequently associated with rework or incomplete tasks, potentially contributing to an upsurge in Work-in-Progress (WIP). Such scenarios can distort the accuracy of PPC by erroneously marking work packages as finalised when additional work is still required in subsequent weeks. The nature of these unfinished works predominantly involves new work packages and rework, which are often excluded from short-term planning. These situations frequently encompass Non-Value-Added (NVA) activities, such as setup, movement, and residue removal. Similarly, Emmit et al. (2012) noted that MD frequently results in left unfinished tasks, leading to "unfinished works." This circumstance necessitates subsequent rectification of functions. Such an occurrence is often linked to evaluating prior work as acceptable, allowing the following trade to proceed. As introduced by Brodeskaia (2010), the concept of Re-entrant Flow underscores the consequences of planning based on individual tasks rather than adopting a location-based and standardised batch approach. Organising work around specific locations and standardised batch sizes, facilitated by a Location-based Management system (LBMS), minimises unnecessary movement. This strategic organisation empowers trades to complete tasks within a designated unit before transitioning to the subsequent batch, thereby fostering operational efficiency. 4. Understanding Making-Do 90 4.1.4. People's behaviour Towards MD As discussed in the first section of this chapter, the decisions in construction projects can lead people to make the following decisions: (1) Abandonment of the planned work, (2) Improvisation or MD (Pikas et al., 2012). These decisions are influenced by behaviours toward production, which vary among people in the construction industry regarding attitude, modes of thinking, and freedom to engage as necessary proactively (Hamzeh et al., 2012). According to the case study by Hamzeh et al. (2012), construction practitioners differentiated between white-collar and blue-collar people. Blue collars were found to be more pragmatic and innovative but localised to their execution stage. On the other hand, top management's decisions were found to be more sensitive to time and cost restrictions than bluecollar people, as well as there is little innovation in their decisions, where their efforts often focus on strategic planning, establishing timelines, and scheduling tasks. According to Javanmardi et al. (2019), a comparison was made between the behaviour of construction practitioners in China and the USA concerning MD. They found that construction managers in China may not encounter variation in project duration due to their preference for MD, while in the USA, MD may exhibit a significant influence on project duration. Javanmardi et al. (2019) show that construction practitioners in China tend to rely on making-do practices without significantly affecting project durations, whereas, in the USA, making-do behaviours substantially impact project timelines. This result suggests that the approach towards making do among construction managers differs between the two countries, potentially influencing project outcomes and durations accordingly. Javanmardi et al. (2019) study has rejected if there is any variation of MD impact when different management levels are responsible for MD decisions. However, Hamzeh sees that the perception of MD decisions can differ between management levels, where the attitude towards MD differs. Thus, there is a difference in the perception of MD decisions among management levels, suggesting potential disparities in attitudes and approaches towards MD within construction management hierarchies. Therefore, these findings underline the complexity of MD and highlight the need for nuanced analysis when considering its impact on construction projects. 4.1.5. Essential Prerequisites for Construction Process Execution Figure 4-2 depicts a protocol for managing Making-Do practices and corresponding reporting mechanisms. This protocol establishes the relationship between the categories of emerging Making-Do practices and their impacts, highlighting the essential inputs required to execute work packages based 4. Understanding Making-Do 91 on the works of (Dos Santos et al. 2020) and (Sommer 2010). Figure 4-2A diagram illustrating the relationships among prerequisites, MD categories, and impacts based on the protocols developed by Dos Santos et al. (2020) and Sommer (2010) Table 4-2 summarises the essential prerequisites for executing construction processes, sourced from Koskela's (2000) and Sommer's (2010) works. Table 4-2 detailed insights into each requirement per these exploratory studies, defining seven fundamental construction flows comprising information, materials and components, labour, equipment and tools, physical space, interdependent tasks, and external conditions. From the necessity of comprehensive work plans ensuring detailed project information to the availability of skilled labour and functional tools and equipment, each prerequisite holds significant importance. Furthermore, this table details the importance of understanding interdependencies between tasks, accounting for external factors like weather conditions, and providing temporary facilities. Grasping these prerequisites is paramount for project managers, as they ensure a well-prepared and efficiently functioning construction environment, thereby enhancing project outcomes and minimising disruptions. Table 4-2 - Prerequisites for Construction Project Execution and Their Explanations, Sourced from Koskela (2000) and Sommer (2010) Prerequisite Explanation Information The presence of comprehensive work plans with adequate information Materials and Components The availability of materials and components ensures adherence to project specifications and standards regarding quality and quantity. Does it has impacts on the production system? Constraints Analysis? Is there any MD practices? Task Execution Interrupted Workflow Yes Yes No No Yes No 4. Understanding Making-Do 92 Labour Availability of the necessary human resources in terms of both quantity and qualifications. Equipment and Tools The availability and proper functioning of the required tools and equipment are essential for successfully executing tasks. Space Ensuring adequate workspace availability, well-defined circulation routes, and sufficient storage areas for materials is essential for uninterrupted project operations. Interdependent Tasks Activities with strong interdependencies can impact the execution of subsequent tasks. External Conditions External factors, such as weather conditions, including wind and extreme temperatures, play a significant role in project planning and execution. Installations Availability of provisional electrical and hydraulic installations, site security facilities, scaffolding, area closures, and storage zones 4.1.6. Making Do Categories Table 4-3 categorises various aspects of Making-Do (MD) practices in construction projects, shedding light on the complexities involved. The first set of categories delves refers to the physical elements essential for task execution. 'Access/Movement' delineates spatial considerations, emphasising the need for unobstructed access and clear pathways for labourers. 'Component Adjustment' highlights the adjustments made to construction elements, often unplanned, to ease installation. 'Working Area' assesses the appropriateness of the work zone, ensuring conducive surroundings for task execution. 'Storage' underscores the significance of preparedness, providing adequate storage of tools, materials, and components. 'Equipment/Tools' reflects the adaptation or development of tools for efficient usage during tasks, while 'Water and Electricity Supply' focuses on establishing the necessary infrastructure for a seamless energy supply. 'Protection' evaluates the availability of suitable conditions for implementing protective measures. The final category, 'Sequence,' delves into the deviation from the intended construction process, pinpointing instances where tasks stray from the planned sequence. The sources, primarily Formoso et al. (2011), Sommer (2010), Fireman et al. (2013), and Leão et al. (2014), provide a comprehensive foundation for understanding these categories. This detailed categorisation offers invaluable insights into the multifaceted nature of MD practices, providing a basis for construction professionals to address and mitigate these challenges effectively. 4. Understanding Making-Do 93 Table 4-3-- Making-Do Categories and Descriptions in Construction Projects MD Category Explanation Source Access/Movement Relating to the spatial considerations required to execute tasks regarding sufficient access and clear labour pathways (Formoso et al., 2011; Sommer, 2010) Component Adjustment Unnecessary or unplanned adjustments of construction components or elements to facilitate task completion during installation Working Area Lack of consideration given to the appropriateness of work areas or supporting zones during activity execution. Storage Inadequate preparation was given to sort and store tools, materials, and components. Equipment/Tools Inappropriate development or adaptation is improvised for equipment and tools during usage for task execution. Water and Electricity Supply The inappropriate adaptation or establishment of infrastructure for supplying water and energy. Protection Availability of conditions for protection measures (Sommer, 2010) Sequence Deviation from the intended construction process (Fireman et al., 2013a; Leão et al., 2014) 4.1.6. Impact of Making-Do Practices in Construction Projects Table 4-4 provides an in-depth analysis of the diverse impacts of MD practices within construction projects. The social consequences encompass demotivation among workers, particularly when forced into tasks likely to remain incomplete, leading to frustration and extensive communication efforts. This demotivation is exacerbated by the need for return visits to rectify errors, impacting overall productivity. From a technical perspective, MD contributes to delays, extending project durations and processing times. These delays often require additional inspections and corrections, leading to overbudget scenarios due to penalties, rework, and waste. Material wastage occurs when inappropriate materials are employed to compensate for MD-induced delays, sometimes leading to demolitions and corrections 4. Understanding Making-Do 94 upon error discovery. Out-of-sequence activities further disrupt project flow, creating illogical task sequences. Rework becomes a common consequence of MD, requiring additional resources and inspections for rectification. Unfinished works initiated informally through MD are often abandoned due to their inability to meet quality standards. Quality deviation is another notable issue, manifesting in defective products and necessitating rework. Additionally, MD adversely affects productivity, undermining the performance of materials, equipment, and personnel, compromising overall project success. Furthermore, inadequately managed spaces contribute to increased equipment/tool transportation waste and personnel movement across various project locations. This comprehensive analysis underscores the multifaceted negative impact of MD practices, emphasising the urgent need for effective mitigation strategies within construction management practices. Table 4-4 -- Impact of Making-Do Practices on Construction Projects: MD Impact Explanation Source Social Demotivation MD can diminish the motivation of workers who are cognizant of engaging in forced tasks that are unlikely to be completed. (Ronen, 1992; Koskela, 2004; Formoso, et al. 2011; Amaral et al.2020) Frustrations Workers dealing with challenging tasks may become frustrated, often engaging in extensive communication, especially when missing essential inputs. This frustration can be exacerbated when they must return to rectify errors without witnessing any significan t increase in productivity. (Alhava et al., 2019; Neve & Wandahl, 2018) Reduced effort exerted In the early stages of MD-affected activities, efforts rise temporarily but often shift elsewhere in the long run when issues arise. (Koskela, 2004) Violations of Health, Safety and Environment (HSE) regulations. Without Health, Safety, and Environment (HSE) instructions, tools, a clean environment, and adequate training can lead to hazardous activities. (Neve & Wandahl, 2018) 4. Understanding Making-Do 95 Technical Delay (Extended project duration) and extended processing time MD causes a certain amount of delay, resulting in longer lead times, and is often required for error correction and additional inspections, especially when there are stoppages in subsequent tasks. (Ronen, 1992; Koskela, 2004; Formoso et al. 2011; Hamzeh et al. 2012; Alhava et al. 2019; Amaral et al.2020) Material Waste MD can contribute to material waste by necessitating inappropriate materials to compensate for delays or inadequacies in material delivery. MD may also lead to demolishing or correcting mistakes once the problem is discovered. (Koskela, 2004; Formoso, et al. 2011; Amaral et al.2020) Sequence activities Tasks that commence with inadequate preparation, whether due to improper prerequisites or their absence, tend to set off a chain reaction of events that follow an illogical sequence. (Neve and Wnadahl, 2018) Overbudget Delays, material wastage, penalties for delays, rework, energy inefficiencies, accidents, and injuries collectively contribute to exceeding the allocated budget for each task affected by MD. (Amaral et al.2020, 2022) Overload or vacant work locations Workers dealing with MD-infected tasks, often caused by inadequate space management, may face disruptions of empty locations or congestion in tightly packed work areas. (Neve & Wandahl, 2018) Rework Initiating activities without adequate inputs can result in production mistakes and errors, necessitating inspections and allocating additional resources for rectification. (Ronen, 1992; Koskela, 2004; Formoso et al. 2011; Hamzeh et al. 2012; Fireman et al. 2013; Alhava et al. 2019; Amaral et al.2020) 4. Understanding Making-Do 96 Unfinished works Small, informally carried tasks may be abandoned when it becomes apparent that they cannot be executed correctly through the MD approach. (É. M. Dos Santos et al., 2020a; Emmitt et al., 2012; Fireman et al., 2013a) Quality Deviation MD often results in non-conformance with process and product standards, leading to deviations in the required quality. These deviations are typically manifested as defective products that may necessitate rework. (É. M. Dos Santos et al., 2020a; Formoso et al., 2011; Koskela, 2004; Sommer, 2010) Reduced Productivity The negative impacts of MD are closely linked to the poor production performance of materials, equipment, and personnel, ultimately affecting the overall project's success. (É. M. Dos Santos et al., 2020a; Formoso et al., 2011; Koskela, 2004) Moving and Transportation waste Tasks with inadequately managed spaces are closely associated with increased waste in equipment/tool transportation and the movement of personnel across various locations within the construction project. (Perez et al., 2015) 4.2. MD Mitigation Strategies The increased attention that Making Do (MD) waste has received from both the construction industry and academic circles, coupled with recent advancements in Planning and Control (P&PC) methods and Information and Communication Technology (ICT) technologies, has paved the way for the development of several methodologies aimed at aiding professionals in conducting thorough prerequisite analyses for tasks during the planning and control phases to diminish the negative impacts of MD practices. Indeed, the role of computer assistance is vital in construction-related tasks to attain flexibility, reliability, and efficiency (Sacks et al., 2018). Based on these considerations, this section delves into a comprehensive review of existing MD research to evaluate the efficacy and limitations of recommended strategies. 4. Understanding Making-Do 97 This section identifies the best practices from the gathered papers. Subsequently, these identified papers were thoroughly examined by manually scrutinising the references cited and placing an additional 25 best practices. Through a thorough assessment of the primary objectives of these strategies extracted from the literature, four broad classifications of practices emerged, which can be summarised as follows: i) Making Do identification, categorisation, and quantification. ii) Production Planning and Control iii) Quality Management and Control iv) Information Communication Technologies v) Social Empowerment Perspectives Koskela's discourse posited the imperative for a paradigmatic realignment within the domain of production theories to respond to Making Do practices. Within this context, his recommendations adopt a comprehensive approach to reconfiguring the organisational landscape. The discourse expounds upon Koskela's propositions, which ardently advocate for the managing-as-organisation principles, linguisticaction theory, and scientific experimentation as transformative frameworks. Furthermore, his conceptual purview extends to the nuanced conceptualisation of construction as an adaptive complex system, drawing profound resonance from the insights espoused by Bertelsen (2004). 4.2.1. Production Planning and Control Systems Pragmatic recommendations also succeed in the above-mentioned theoretical paradigms advanced by Koskela. Among them, he advocates for implementing the Last Planner System (LPS) as an operational methodology while simultaneously emphasising the pivotal objectives of reduction for lead time and inventory as pivotal catalysts for improvement. LPS, as a technical planning system, fosters the formation of a social system among crews. Within LPS, two primary control types can be identified. Within LPS, two primary control types can be identified. 1Production unit control: this form of control enables improved worker assessment through continuous learning and corrective actions. 2Workflow control: This form of control focuses on optimising the sequence and rate at which workflows are done across production units. (Alhava et al., 2019; Amaral et al., 2022; Dos Santos et al., 2020; Emmitt et al., 2012; Fireman et al., 2013; Formoso et al., 2020; Pikas et al., 2012; Sukster, 2005) Advised to focus on planning and control necessitates rigorous control over construction processes, which identifies the minimum or 4. Understanding Making-Do 104 the construction site conditions that can be connected to the BIM process for further visualisation and analysis steps. Several policies could be applied to regulate the effect of material delivery on project performance. On the site, measures of PP&C associated with planned site layout planning (SLP) were inserted (Cheng et al., 2015). These functions send 'pull' signals for materials to be timely ordered and delivered under the just-in-time (JIT) concept. In the PP&C, the materials should be quantified and allocated before execution. Thus, in any process that commences without the perquisite materials, a making-do waste arises, responsible for other wastes such as Work-In-Progress (WIP) and unfinished works, and hinders labourers' productivity. Another waste not communicated in Figure 4-3 is substitution waste, a phenomenon that urges substituting unavailable materials with other materials to meet scheduled deadlines; this waste is similar to making do but hinders productivity due to reworks and defects. The material delivery also impacts the reliability of production planning and control (e.g., when the required materials are not available when the planned work is released, a making-do effect could be raised, which can snowball to the emergence of rework and defects). The role of SLP (site layout planning) is necessary to coordinate physical locations between the demand of the production system and temporary facilities, delivered materials, crew movements, machine setups, and truck traffic. BIM processes positively reduce design changes, variability, rework, and RFIs (Dave & Sacks, 2020), as shown in Loops R11, R12, and B4. BIM functionalities such as quantity-take-off, 4D planning, visualisations, clash detection, and interoperability are available in commercial software, bringing valuable information control to the PP&C systems. Actual interaction between BIM functionalities and lean-based PP&C is not mature. The literature actively proposed different prototypes, but a real impact on waste elimination has not been presented yet. However, the potential effects of BIM functionalities on waste reduction are evident even on final wastes such as rework, material waste, and capital waste. B3, R9, and R10 are concerned with quality management in controlling variability using quality control charts and continuous improvement methodology based on the PDCA cycle (Deming, 1982). Construction variability is the primary source of product defects and process discontinuity, leading to reworks, poor productivity, capital waste, and cascading delays. Loop R7 illustrates the causality of Making Do with WIP and unfinished works. The impact can be explained by working on activities without its prerequisites (e.g., trades that focus on local optimisation are reluctant to pick the most accessible available work packages at the beginning, "a low-hanging fruit phenomenon", crews move to open spaces until a problem appears, leave unfinished work behind them and prevents WIP to be progressed. 4. Understanding Making-Do 105 Eventually, this phenomenon led to interruptions in workflow and interference with other trades' schedules, leaving labourers to wait until the problem is resolved, where overtime strategy would be a solution to compensate for the delay, which can lead to labour fatigue and labour productivity to be negatively impacted (Loop R8). Again, the impact of making-do waste is complex and cannot be modelled in a unidirectional causality link (Formoso et al., 2020), which contradicts CLD principles (As shown in loops R7 and R8, the link between making-do and unfinished-works is bidirectional). Thus, further investigation is needed to explore intermediate variables between making-do and other wastes. 4.5. Conclusions Despite the extensive body of literature addressing the causes of Making-Do (MD) waste, a noticeable dearth of studies focus on developing planning and control strategies to identify and mitigate MD waste effectively. The existing research primarily revolves around quantifying MD waste and exploring the relationships between its causes and consequences (Amaral et al., 2022; Dos Santos et al., 2020; Formoso et al., 2011; Leão et al., 2014). This observation highlights a significant knowledge gap, indicating the absence of an established framework to aid production planners and trade teams in detecting, estimating, and eliminating MD waste from their construction project processes, cultures, and practices. This chapter analysed the theoretical foundations of Making-Do, delving into its root causes, detection methods, and specific characteristics. The exploration concluded by creating a causal loop diagram connecting prerequisites, MD causes, MD categories, and impacts. Recognising the critical importance of mitigating the adverse effects of MD waste, this chapter is dedicated to bridging this knowledge gap by revealing the causal structure of MD in construction projects. A generic causal loop diagram connects the dots between production planning and control, quality management, harnessing information technology, and social factor empowerment. Based on the built CLD through this chapter, the subsequent chapter aims to simulate the impacts of effective methodologies, including the Last Planner System, Location-Based Management, and Building Information Modelling (BIM) on MD waste, within a comprehensive policy framework. This framework is designed to proactively minimise the impact of MD waste long before the execution phase, offering a promising approach to enhance both lean and non-lean project management practices. 5. Surveying LPS-BIM strategies for MD mitigation 106 5. Surveying LPS-BIM strategies for MD mitigation This chapter presents the results of a survey to investigate the implementation level of Lean Construction and BIM in the Portuguese construction industry. To achieve this, it estimates the levels of awareness of "making-do" practices among participants and identifies the primary group of stakeholders who support such development. In addition, respondents were given the chance to rank a set of mitigative strategies of MD, which were collected through the qualitative data analysis in Chapter 4. 5. Surveying LPS-BIM strategies for MD mitigation 107 5.1. Introduction The significance of both the Last Planner System (LPS) and Building Information Model (BIM) is widely acknowledged in addressing the shortcomings of production planning and control that led to MakingDo (MD); however, the current policies lack validation. Accordingly, this chapter assesses the stakeholders' expectations about LPS and BIM for MD mitigation strategies from the existing literature review that used thematic analysis, a qualitative data analysis technique in Figure 5-1. The following step was embedding the identified factors in a questionnaire survey to explore new terms applied to describe MD practice in the construction industry and to classify the aspects of the developed criteria for evaluating LPS and BIM methods to control MD waste. The survey revealed three principal groups of factors: "BIM-based collaboration for constraint analysis," "Medium-term and Short-term MD analysis," and "Enterprise learning and adaptation," which includes, but is not limited to, components like "improved documentation for MD cases" and "Dynamic reports for MD and constraint analytics." Proper understanding and consideration of these factors are significant in addressing the stakeholders' expectations regarding applying LPS. Along with the expectations of relevant industry practitioners, an LPS-BIM framework for MD mitigation policy was built, including the technological and industrial needs for planning and control of the production and the MD mitigation. Figure 5-1 - A Framework of Policy to Improve the Performance of MD Mitigation Models. 5. Surveying LPS-BIM strategies for MD mitigation 108 5.2. In-depth exploration of the critical factors that mitigate MD through LPS and BIM This section will comprehensively examine the dynamic variables associated with the LPS and BIM to limit MD practices and their adverse consequences. Considering the intersections of these variables, the discussion will explore the social factors inherent in the LPS in terms of collaboration and adaptation. Subsequently, this section will discuss the technical components of the LPS, including aspects such as visual management and medium short-term planning. Finally, the functionality of BIM will be explained within this context. 5.2.1 Collaboration The dynamic variable “Collaboration” constitutes an essential aspect of the LPS. Without collaboration, the efficacy of the LPS will remain limited, and its outcomes will be less promising (Ballard,2020), as it becomes challenging to foster trust and commitment among trades without a collaborative approach to implementing the LPS. This variable underscores the radical innovation brought about by the LPS, particularly in its social dimension. Collaboration facilitates deployment planning and control responsibilities throughout the project organisation and focuses on reliable promises between interdependent players as the key to the successful application of the LPS. Successful implementation of a collaborative LPS includes multiplication of the following parameters: • Handling conflicts among various stakeholders: Addressing conflicts among diverse stakeholders is a critical metric that seeks to gauge the extent of project management's exertion in harmonising interests across multiple sectors. It also encompasses resolving disputes about responsibility for constraints and the mechanisms employed for their removal. • Enabling discussions: Facilitating discussions is a crucial dimension that assesses how project management encourages dialogue among relevant parties during the planning phase and the resolution of their tasks within lookahead and make-ready planning sessions. • Ensuring a high level of coordination: This input specifies the percentage of coordination implemented during Last Planner System (LPS) sessions. • Engagement level in constraints analysis: This parameter enables users to designate the degree of involvement for each trade in the constraints analysis process. A heightened level of engagement can result in more reliable plans and a reduction in MD practices when constraints are removed correctly and collaboratively. 5. Surveying LPS-BIM strategies for MD mitigation 109 • Measure plan performance: This parameter is dedicated to assessing the reliability of measurements for planning performance. It examines the potential for missing data or errors in reporting the PPC metric, including the possibility of overestimating PPC. 5.2.2 Adaptation This dynamic variable assesses the level of adaptation within LPS practices, aimed at fostering a more common or shared understanding among the project parties while applying LPS and addressing root causes of resistance to changes during the LPS implementation (Rooke, 2020). The adaptation variable addresses the root causes of resistance to change by transforming organisational values, encouraging local adjustments, and enhancing individual capabilities through further training and education in lean construction philosophy and LPS principles; additionally, as part of social LPS, the adaptation variable endeavours to mitigate MD practices by generating data banks in computerised forms through active data collection on MD incidents. This process involves the development of meaningful language to store, analyse and organise the ontology of MD. The adaptation variable is the multiplication of the following parameters: • Local adjustments to organisations: Local adjustments to organisational practices encompass the modifications made to accommodate the Last Planner System in alignment with the organisation's culture, resources, projects, priorities, goals, and visions. These adjustments are tailored to integrate seamlessly with the network of stakeholders associated with the organisation. • Coaching and training: Evaluation of coaching and training effectiveness pertains to the level of instruction provided to superintendents and crew leaders in collaborative pull planning. It also encompasses assessing how planners prepare for and sustain the implementation of the LPS. • Generate data bank about LPS metrics and MD incident records.: Timely data collection is imperative for monitoring identified MD incidents and their resolutions. The data repository for MD incidents should comprehensively detail each occurrence, including its unique identifier (ID), timestamps for the incident history, MD category, associated constraints, root cause analysis, involved trades, location, process name, operation details, task specifics, MD impact and any countermeasures employed if corrections were made. • Exchange experiences with other enterprises: Exchanging experiences provides a valuable opportunity to share insights and learnings. Through disseminating successful and unsuccessful stories, enterprises can derive a collection of strategies acquired through experience, aiding in mitigating adverse MD occurrences in their projects. This collaborative 5. Surveying LPS-BIM strategies for MD mitigation 110 approach enables the transfer of knowledge and the cultivation of a more informed and resilient project management framework. • Compare MD records and experience with other enterprises: Analysing MD records and experiences compared with other enterprises offers a competitive advantage, allowing companies to benchmark their waste reduction efforts. This comparative assessment provides valuable insights into industry practices, enabling the identification of best practices and areas for improvement in managing and minimising MDs. 5.2.3 Visual Management One of the components of soft Lean Construction tools involves educating individuals to recognise waste and flow within their projects. This concept, known as “learning to see,” aligns with the principles concept of visual management, which employs different tools that target the human five senses to increase the transparency of the construction process and reveal wastes (Tezel & Aziz, 2017). In this dynamic system, this variable is characterised by the multiplication of three parameters, as outlined below: • Communicate instructional and actionable information.: Communicate the traditional list of wastes around the site clearly and concisely. Also, infographics about the causes of MD and the possible consequences should be distributed. • Keep all plans public: All plans at all levels of the LPS hierarchy should be shared and seen with involved parties. This step increases situational awareness regarding task constraints, which can prevent possible MD. • BIM models available during LPS sessions: In LPS sessions, it is imperative to employ the most up-to-date BIM models to facilitate the essential information required for task planning and execution. It is recommended that functional analysis, quantity take-off schedules, 4D BIM information, and task-specific attributed data be shared with the BIM database. This collaborative approach enhances the effectiveness of planning sessions. 5.2.4 Lookahead Planning This subsection transitions to the technical aspects of the LPS, which focuses on lookahead planning, positioned between long-term and short-term planning. This stage of the LPS hierarchy entails breaking down project phases and milestones into operational tasks, as outlined in Chapter 2. At this stage, the fundamental function of the LPS lies in the constraints analysis, a process hypothesised by the literature 5. Surveying LPS-BIM strategies for MD mitigation 111 that significantly impacts tackling MD when it is executed successfully. A collaborative design for operations emerges as another pivotal function of LPS, enhancing visual management as elucidated in subsection 5.1.4 and leveraging collaboration parameters in subsection 5.1.1. The variable Lookahead planning comprises the following constituents: • Constraints management: The core function of the LPS is constraints analysis, which can be performed at each level of planning. For example, the master schedule is constrained by cost, time, quality, environment, and social factors in a broader view. In pull planning, the constraints are interdependent among processes, and conditions of satisfaction for each process should be screened before moving towards Lookahead planning. • Break down work packages into actionable details: Break down projects into phases and phases into milestones. Within milestones, define processes, processes into operations, operations into tasks, and tasks into elemental motions. The LPS hierarchy must be followed, particularly for highly uncertain critical tasks. MD practices should be classified within the publicly discussed LPS hierarchy during LPS sessions. • Collaborative design for operations: Collaborative operation design is advisable for exploring various scenarios of constraints and anticipating potential issues MD before execution. Methods recommended in the literature include virtual prototyping techniques (such as BIM simulation DES), first-run studies, and physical prototyping. 5.2.5 Weekly, Bi-weekly, and Daily Planning This subsection describes the parameters that comprise short-term planning within the LPS. This operational stage of the LPS involves aligning managerial directives with weekly, bi-weekly, and daily feedback from trades. It entails providing feedback on workable backlogs, engaging individuals in daily constraint discovery and removal, making commitments to work, facilitating constraint removal, and monitoring work progress to calculate the Plan Percent Complete (PPC). The multiplication of the following parameters formulates the short-term planning variable: • Daily huddle meetings: This parameter specifies the percentage of daily meetings between workers and superintendents. These meetings involve discussions on planned and completed tasks, addressing technical task execution details. Workers also directly report the daily work output from the previous day, highlighting any encountered constraints during execution. Furthermore, white-collar personnel are included in these meetings, especially during Gemba walks. 5. Surveying LPS-BIM strategies for MD mitigation 112 • Maintain workable backlog: The workable backlog is a crucial parameter as it ensures the availability of work for execution ahead of time, preventing engagement with constrained tasks. Tasks located in this backlog are more likely to be free from constraints, or their constraints are less uncertain. Additionally, it aids in classifying non-workable tasks, prompting individuals to inquire about their constraints and attempt to resolve them so they can be included in this workable backlog later. • Engagement level in constraints analysis: This parameter enables users to designate the degree of involvement for each trade in the constraints analysis process. A heightened level of engagement can result in more reliable plans and a reduction in MD practices when constraints are removed correctly and collaboratively. • Daily discussion of constraints: The more frequently the constraints are discussed, the more they are understood and removed. A daily conversation is recommended to achieve a higher level of collaboration in removing constraints and preventing MDs. 5.2.6 BIM Functionalities The existing BIM functionalities vary widely based on the purpose of use, according to (ISO, 2018). For MD mitigation, the following functionalities are explained: • Add MD attribute to BIMs: Facilitating the incorporation of MD and constraints information within BIM construction management and authoring software is essential. This parameter measures the percentage of MD information a project team integrates into the parametric BIM model. • Utilise 4D BIM for the tasks at the last planner hierarchy: This parameter quantifies the percentage of planning details incorporated into the 4D BIM. It is advisable to enhance the level of 4D planning in alignment with the LPS hierarchy. • Apply clash detection analysis to discover spatial constraints: This value represents the percentage of applied clash detection across a project. This BIM functionality is recommended to prevent spatial clashes between the works of different trades. Additionally, it can assist decision-makers in determining the sequencing among operations. • Utilising colour-coded visualisation to represent task status: This parameter denotes the percentage of visualisation utilised for tracking task status in response to Andon signals. It encompasses embedded types of constraints associated with tasks and reported MD issues. Additionally, this parameter reports visualised information used by white-collar personnel for data collection and offer guidance on task execution or problem resolution. 5. Surveying LPS-BIM strategies for MD mitigation 113 • Online communication tools using BIM cloud services: This parameter specifies the percentage of communications performed over BIM cloud services to exchange production information on developed BIM models by design teams with site management, crew leaders, and crews. 5.3. Research Methodology A comprehensive literature review identified 30 variables related to using BIM and LPS for MD mitigation. These variables were subsequently structured into a questionnaire survey, and a pilot study was conducted to refine the questionnaire before its distribution to the intended respondents. The pilot study involved five PhD students in Civil Engineering, whose feedback informed the final version of the questionnaire presented in Appendix A. 5.3.1 Questionnaire Design The questionnaire survey commenced with a pilot study employing a preliminary questionnaire containing a compiled list of LPS and BIM strategies for MD mitigation. This initial phase assessed questionnaire relevance, length, complexity, and layout. Participants for the pilot study were selected from two Portuguese universities and comprised PhD students specializing in construction management and BIM research. Feedback from the pilot study participants was instrumental in refining the final questionnaire. The final questionnaire survey comprises the following three sections: SURVEY COVER LETTER – This section explains the purpose of the survey, provides a definition of MD practice, and elaborates on examples from the literature. The respondents were informed that the data collected would be used solely for research purposes to encourage a high response rate. Likewise, the respondents were assured that the confidentiality of all individuals’ responses would be maintained. SECTION A: DEMOGRAPHIC OF THE RESPONDENTS – This section captures demographic information about respondents. The respondents were asked to indicate their experience level, education level, and designation. This section would enable the researcher to identify the respondents' roles within which industry. SECTION B: LEAN AND BIM EDUCATION AND IMPLEMENTATION – This section captures whether the respondents have been in training or a BIM and Lean Construction 5. Surveying LPS-BIM strategies for MD mitigation 120 5.4. Results 5.4.1 Reliability analysis Reliability analysis assessed the internal consistency of the variables related to using BIM and LPS to mitigate MD practices in construction projects. A total of 25 variables were tested if they are, and the Likert scale consistently reflects the construct of the study set out to measure. Accordingly, Cronbach’s alpha coefficient of reliability (α) was calculated for the variables using Equation 5.1. 𝛼𝛼= 𝑁𝑁 𝑁𝑁−1�1− ∑𝜎𝜎2 𝜎𝜎𝑇𝑇 2� Equation 5-1 In this context, N represents the total number of questions. Each question has a score variance denoted as σ where i ranges from 1 to n. The overall test score's total variance, not in percentage form, is represented by the σ T . Cronbach’s alpha αwhich has a value from 0 to 1, and the higher the value of (α), the greater the internal consistency of data (Field, 2005). It is generally believed that a value of α = 0.7 is acceptable, and α > 0.8 depicts good internal consistency. The calculated α for this study is 0.9475, demonstrating an excellent internal consistency. The 25 variables were then ranked using the descriptive statical mean as the ratio of importance. The results of the reliability analysis and ranking of the variables are shown in Table 5-2. Table 5-2 -- Reliability analysis table with means and ranking of LPS and BIM strategies for MD mitigation. No Variable Mean Cronbach’s Alpha Rank VA24 Identify and resolve time and space clashes using BIM Clash Detection tools. 3.736 0.945 1 VA23 Report task information in alignment with product specifications to ensure accuracy. 3.722 0.945 2 VA25 Facilitate the exchange and communication of Making-Do practices through online BIM models. 3.722 0.944 3 VA22 Utilise 4D planning to visualise constraints and their impact on project timelines. 3.681 0.945 4 VA5 Provide coaching, training, and seminars for superintendents and forepersons. 3.653 0.945 5 VA11 Ensure the availability of BIM models, design drawings, and site layout plans for reference during the implementation of the Last Planner System. 3.611 0.944 6 VA21 Facilitate daily discussions between trades to address constraints and coordinate activities. 3.583 0.945 7 VA2 Ensure high-level coordination among project stakeholders. 3.542 0.946 8 5. Surveying LPS-BIM strategies for MD mitigation 121 VA20 Collaboratively design operations using BIM for digital prototyping. 3.486 0.945 9 VA12 Maintain transparency by keeping all plans publicly accessible. 3.472 0.945 10 VA14 Apply constraints analysis proactively to identify and address potential issues as a team. 3.472 0.945 11 VA9 Facilitate knowledge exchange and sharing experiences among different companies. 3.458 0.945 12 VA3 Facilitate discussions to address concerns and foster consensus. 3.444 0.945 13 VA7 Establish a data bank to clarify misconceptions regarding Lean construction, Making-Do, and Last Planner System principles. 3.444 0.946 14 VA1 Handle disagreements and interests effectively to foster collaboration. 3.431 0.947 15 VA6 Process and translate knowledge from experiential learning into actionable insights. 3.403 0.946 16 VA13 Utilis e guiding information across digital and physical environments to enhance understanding. 3.403 0.946 17 VA17 Involve stakeholders in constraints management processes to enhance collaboration in Mitigating MD. 3.347 0.944 18 VA8 Learn from past incidents of Making-Do. 3.306 0.946 19 VA16 Encourage stakeholders to communicate and share any constraints that may impede progress. 3.278 0.945 20 VA10 Compare and analyse multiple cases to understand how Making-Do is managed. 3.264 0.946 21 VA4 Adapt local adjustments to align with organizational requirements. 3.181 0.947 22 VA18 Maintain a workable backlog of tasks to prioritize and manage workload effectively. 3.153 0.944 23 VA15 Delay tasks with uncertain constraints to avoid potential disruptions. 2.931 0.947 24 VA19 Break down tasks from processes to operations and further to individual tasks for clarity of management and control. 2.889 0.947 25 The mean ranking reveals that “Identifying and resolving time and space clashes using BIM Clash Detection tools” is the most significant stakeholder expectation of using BIM for MD mitigation. This rank is because the construction industry is long overdue for BIM-Based tools to identify conflicts. 5.4.2 Exploratory Factor Analysis (EFA) The factor analysis method aims to discover "underlying" structures associated with the variables revealed in the literature. Its goal is to determine the set of dimensions forming variables as the base of their structure, using the reductionist method to substitute them with fewer uncorrelated principal components. The resulting procedures have the added advantage of deleting redundant (highly correlated) variables while at the same time preserving the integrity of the original data. In this research, 5. Surveying LPS-BIM strategies for MD mitigation 122 factor analysis was done by Minitab employing principal component analysis (PCA) with oblique rotation (varimax) of 25 variables. PCA was employed for factor extraction, and varimax rotation was used as a rotation procedure. The Kaiser-Meyer-Olkin (KMO) measure for the sampling adequacy got a value of 0.873, which is higher than the recommended threshold of 0.5, while Bartlett’s Test of Sphericity resulted in a p-value of 2.45 x 10-104 (less than 0. 5) as shown in Table 5-3, suggesting substantial evidence against the null hypothesis of an identity matrix. Table 5-3 -- KMO and Bartlett's Test Kaiser-Meyer-Olkin of Sampling Adequacy. 0.873 Bartlett's Test of Sphericity Approx. Chi-Square 1178.643 df 300 Sig. <0.000 That demonstration previously mentioned confirms that this data set is suitable for factor analysis. The PCA results reorganise the list of variables into five factors, which account for the total variance of 67.871%, as shown in Table 5-4. Accordingly, the groups were deduced and categorised based on the assigned variables. The groups include: • Group A, denoted by Integrated Production and Product Information • Group B is denoted by Adaptation Towards LPS and MD-free culture. • Collaborative Commitment towards MD Mitigation denotes Group C • Group D is denoted by Constraints Analysis with Short and Long-term Planning • Group E denoted by Active Feedback Mechanisms for MD Incident Resolution Table 5-4 -- Component labelling and corresponding criteria from factor analysis No Variable Eigen Values % of Variance Loading Factor AIntegrated Production and Product Information Management 11.338 45.353% VA24 Identify and resolve time and space clashes using BIM Clash Detection tools. 0.767 VA23 Report task information in alignment with product specifications to ensure accuracy. 0.712 VA25 Facilitate the exchange and communication of Making-Do practices through online BIM models. 0.715 VA22 Utilise 4D planning to visualize constraints and their impact on project timelines. 0.803 BAdaptation Towards LPS and MD-free culture 1.761 7.043% VA5 Provide coaching, training, and seminars for superintendents and supervisors. 0.672 5. Surveying LPS-BIM strategies for MD mitigation 123 VA6 Process and translate knowledge from experiential learning into actionable insights. 0.715 VA4 Adapt local adjustments to align with organizational requirements. 0.527 VA11 Ensure the availability of BIM models, design drawings, and site layout plans for reference during the implementation of the Last Planner System. 0.655 VA12 Maintain transparency by keeping all plans publicly accessible. 0.739 VA13 Utilis e guiding information across digital and physical environments to enhance understanding. 0.511 VA21 Facilitate daily discussions between trades to address constraints and coordinate activities. 0.576 VA19 Break down tasks from processes to operations and further to individual tasks for clarity of management and control. 0.698 VA18 Maintain a workable backlog of tasks to prioritize and manage workload effectively. 0.688 VA15 Delay tasks with uncertain constraints to avoid potential disruptions. 0.759 CCollaborative Commitment towards MD Mitigation 1.426 5.702% VA2 Ensure high-level coordination among project stakeholders. 0.743 VA3 Facilitate discussions to address concerns and foster consensus. 0.678 VA1 Handle disagreements and interests effectively to foster collaboration. 0.748 VA20 Collaboratively design operations using BIM for digital prototyping. 0.644 DConstraints Analysis with Short and Long Terms Planning 1.324 5.297% VA14 Apply constraints analysis proactively to identify and address potential issues as a team. 0.644 VA17 Involve stakeholders in constraints management processes to enhance collaboration in Mitigating MD. 0.796 VA16 Encourage stakeholders to communicate and share any constraints that may impede progress. 0.738 EActive Feedback Mechanisms for MD Incident Resolution 1.119 4.476% VA9 Facilitate knowledge exchange and sharing experiences among different companies. 0.528 VA7 Establish a data bank to clarify misconceptions regarding Lean construction, Making-Do, and Last Planner System principles. 0.512 VA8 Learn from past breakdowns and instances of Making-Do. 0.676 VA10 Compare and analyse multiple cases to understand how MakingDo is managed. 0.724 16.968 67.871% 5. Surveying LPS-BIM strategies for MD mitigation 124 5.5. Structural Equation Model (SEM) Based on SEM, a Confirmatory Factor Analysis (CFA) was implemented in AMOS software to verify the measurement model's validity. The SEM comprises defining the measurement model and the structural model. 5.4.1. Measurement Model Fitness The measurement model was established using the graphical representation functionality in AMOS 26.0.0. as shown in Figure 5-9. The model comprised four variables, COO, MDK, LPS, and BIM, representing 25 factors loaded on these variables. Similarly, 25 errors were associated with these factors, and measuring these unobserved errors forms the power of SEM, which other multivariate methods cannot measure. The model assumes covariances among all unobserved variables: COO, MDK, LPS, and BIM. As a composite of CMIN/df, Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Residual (RMSEA), and Standardized Root Mean Square Residual (SRMR), these model-fit indices were used for the overall evaluation of the model. Some calculated statistics significantly differed from the established below-standard value defined in previous research (Bagozzi & Yi, 1988; Bentler, 1990; Hu & Bentler, 1998; Schumacker & Lomax, 2004). The four-factor model, comprising COO, MDK, LPS, and BIM, demonstrated some unsatisfactory fit to the data, as indicated by the following fit indices: CFI = 0.786, TLI = 0.762, SRMR = 0.099, and RMSEA = 0.098. However, the satisfactory fit to the data is as follows: CMIN/df = 2.133 and probability level = 0.000 (Table 5-5 includes model-fit indices; please see below for more details) Table 5-5 -- Model fitness measures Fit indices Threshold Obtained value Evaluation Source Probability level Insignificant 0.000 Passed (Bagozzi & Yi, 1988) CMIN/DF Between 1 and 3 2.133 Excellent (Schumacker & Lomax, 2004; Ullman & Bentler, 2012) CFI >0.95 0.786 Terrible (Bentler, 1990) TLI >0.08 0.762 Terrible SRMR <0.08 0.099 Unacceptable 5. Surveying LPS-BIM strategies for MD mitigation 125 RMSEA <0.06 0.098 Terrible (Hu & Bentler, 1998) Figure 5-9 -- Unadjusted SEM measurement model The validity test shows that the measurement model requires several modifications. First, by looking at factor loadings for every item, It was found that three items, i.e., VA4, VA10, VAR21, had low factor loadings (VA4 = 0.46, VA10 = 0.45, VA21 = 0.48) which are all less than the accepted threshold of 0.5. Therefore, they were taken out of the study. Second, after looking at the medication indices, several covariances are suggested to be drawn in (BIM: e25 ↔ e24, e25↔e21, e25↔e22, e23↔e22) and (LPS: e11↔e12, e12↔e13, e16↔e17, e17↔e18, e18↔e19, e11↔e19), afterwards another analysis was carried out after running the analysis for the model fitness after modifications for the measurement model. These model-fit indices were used to evaluate the model as a composite of CMIN/df, CFI, TLI, RMSEA, and SRMR. Significantly, the means of all calculated statistics were within the established standard values. The four-factor model, comprising COO, MDK, LPS, and BIM, demonstrated a satisfactory fit to the data, as indicated by the following fit indices: CMIN/df = 1.490, CFI = 0.932, TLI = 0.911, SRMR = 0.08 and RMSEA = 0.065 (Table 5-6 includes model-fit indices; please see below for more details). 5. Surveying LPS-BIM strategies for MD mitigation 126 Table 5-6 -- Model fitness measures for the adjusted measurement model Fit indices Threshold Obtained value Evaluation Probability level Insignificant 0.000 Passed CMIN/DF Between 1 and 3 2.133 Excellent CFI >0.95 0.946 Acceptable TLI >0.90 0.932 Excellent SRMR <0.08 0.063 Excellent RMSEA <0.06 0.060 Acceptable Figure 5-10 -- Adjusted measurement model 5.4.2. Model Reliability Analysis Construct Reliability was evaluated using Cronbach's Alpha and Composite Reliability (CR). The Cronbach's Alpha coefficients for each construct in the study exceeded the recommended threshold of 0.70 (Nunnally & Bernstein, 1994). CR values also ranged from 0.813 to 0.839, surpassing the 0.70 benchmark (Hair et al., 2019). Therefore, CR was established for each construct in the study, as documented in Table 5-7. The convergent validity of the scale items was assessed using Average Variance Extracted (AVE) (Hair et al., 2019). The AVE values for BIM functionalities and LPS technical measures exceeded the 5. Surveying LPS-BIM strategies for MD mitigation 127 threshold value of 0.5 (Hair et al., 2019). However, Collaboration, MD knowledge, LPS Functions, and BIM Functionalities exhibited AVE scores below 0.5. Nonetheless, given that the CR values exceeded the required threshold, it can be inferred that these constructs maintain adequate convergent validity for the present study, as summarised in Table 5-7. Table 5-7 -- Loadings, Reliability, and Convergent Validity Items Loadings Alpha CR* AVE** Collaboration 0.815 0.835 0.628 VA1 0.813 VA2 0.806 VA3 0.757 Making-Do Knowledge 0.811 0.813 0.552 VA5 0.696 VA6 0.727 VA7 0.786 VA8 0.676 LPS Functions 0.873 0. 839 0. 397 VA11 0.588 VA12 0.542 VA13 0.623 VA14 0.589 VA17 0.605 VA18 0.720 VA19 0.625 VA20 0.725 BIM Functionalities 0.843 0.838 0.567 VA22 0.730 VA23 0.815 VA24 0.814 VA25 0.638 * CR: Composite Reliability **AVE: Average Variance Extracted The discriminant validity of the research was assessed by utilizing both the Fornell and Larcker Criterion approach and the Heterotrait-Monotrait (HTMT) ratio. The Fornell Larcker Criterion allows discriminant 5. Surveying LPS-BIM strategies for MD mitigation 128 validity when the square root of a construct's Average Variance Extracted (AVE) surpasses the correlation with other study constructs (Fornell & Larcker, 1981). Nevertheless, the Fornell and Larcker Criterion has been the source of recent criticisms, and scholars have accepted new methods, such as the HTMT Ratio, as alternative techniques. Although the Fornell and Larcker Criteria did not provide evidence of discriminant validity in this study, all HTMT ratios were below the 0.85 threshold value, as (Henseler et al., 2015) recommended. Therefore, the HTMT ratio was used to confirm the discriminant validity. The presented content of Table 5-8 is the results of a closely carried out discriminant validity analysis, where the coefficients and significance values of discrimination are highlighted in Table 5-8. Table 5-8. HTMT Analysis COO MDK LPS AVE COO 0.628 MDK 0.616 0.552 LPS 0.603 0.765 0. 397 BIM 0.332 0.504 0.666 0.567 5.4.3. Structural Model Assessment The structural equation model is generated through AMOS and was used to test the relationships among dependent variables. Similar to the measurement model criteria, an excellent fitting model is accepted if the value of CMIN/df is below five, which meets the requirements of (Hair et al., 2019), TLI and CFI below 0.90 (Bentler, 1990). Also, an adequate-fitting model was accepted if the AMOS software computed RMR and SRMR between 0.05 and 0.080 (Hair et al., 2019). The fitness measures for the fell within the acceptable range: CMIN/ df = 1.418, TLI=0.932, CFI= 0.946, SRMR= 0.0597, and RMSEA =0.060. 5.4.4. Mediation Analysis Mediation analysis examines those complex and multi-dimensional draws between two constructs because their influence likely operates not directly but through a third variable called a mediator. In this case, the intermediary variable demonstrates an interfering element in the relationship between the two variables, which explains the underlying mechanisms of the associations. The bootstrapping method investigates mediation within a model with settings of (bootstrap samples = 2000 samples, percentile confidence level (PC) = 90%, Bias-corrected confidence level = 90%). Knowing the basic terms, such as direct and indirect, is vital in mediation analysis. 5. Surveying LPS-BIM strategies for MD mitigation 129 Figure 5-11 -- Bootstrap settings The squared multiple correlation was 0.62 for MDK; this shows that the LPS, COO, and BIM account for 62% of the variance in MDK. The square multiple correlation for COO accounts for 0.44; this signifies that LPS functions and BIM account for 44% variance in collaboration. The study assessed the impact of COO, LPS, and BIM on MDK. The effect of LPS on MDK was positive and significant; hence, H2 was supported. However, the impact of BIM on MDK was negative and insignificant. Therefore, H1 was not supported. The effect of the COO on MDK was positive and insignificant, too. Hence, H1 was not supported. The impact of LPS on the COO was positive and significant; hence, H2 was supported. On the contrary, the BIM impact on the COO was negative and insignificant. Thus, H1 was not supported. Model fit indices and hypothesis results are presented in Table 5-9. Table 5-9--Moderation analysis for the structural model and model fit indices Hypothesised Relationship Standardised Estimates t-value p-value Decision LPS  MDK 0.698 3.15 <0.001 Significant BIM  MDK -0.013 -0.085 0.932 Insignificant COO  MDK 0.118 1.030 0.303 Insignificant LPS  COO 0.803 4.309 <0.001 Significant BIM  COO -0.257 -1.415 0.157 Insignificant R-Square MDK 62% COO 44% References 232 Liu, Z., Osmani, M., Demian, P., & Baldwin, A. (2016). A BIM-aided construction waste minimisation framework . 59 (2015), 1–23. https://doi.org/10.1016/j.autcon.2015.07.020 Love, P. E. D., Edwards, D. J., & Smith, J. (2016). Rework Causation: Emergent Theoretical Insights and Implications for Research. Journal of Construction Engineering and Management , 142 (6). https://doi.org/10.1061/(ASCE)CO.1943-7862.0001114 Love, P. E. D., & Li, H. (2000a). Quantifying the causes and costs of rework in construction. Construction Management and Economics , 18 (4), 479–490. https://doi.org/10.1080/01446190050024897 Love, P. E. D., & Li, H. (2000b). Quantifying the causes and costs of rework in construction. Construction Management and Economics , 18 (4), 479–490. https://doi.org/10.1080/01446190050024897 Love, P. E. D., Manual, P., & Li, H. (1999). Determining the causal structure of rework influences in construction. Construction Management and Economics , 17 (4), 505–517. https://doi.org/10.1080/014461999371420 Lyneis, J. M., Cooper, K. G., & Els, S. A. (2001). Strategic management of complex projects: A case study using system dynamics. System Dynamics Review , 17 (3), 237–260. https://doi.org/10.1002/sdr.213 MacOmber, H., Gregory, G. A., & Reed, D. (2005). Managing promises with the last planner system: Closing in on uninterrupted flow. 13th International Group for Lean Construction Conference: Proceedings , 13–18. Mahalingam, A., Yadav, A. K., & Varaprasad, J. (2015). Investigating the role of lean practices in enabling BIM adoption: Evidence from two Indian cases. Journal of Construction Engineering and Management , 141 (7), 05015006. https://doi.org/10.1061/(ASCE)CO.1943-7862.0000982 Martins, J., & Cachadinha, N. (2013, July). An Exploratory Study of Lean Construction in Portugal: Owners and Designers’ Perspectives. 21th Annual Conference of the International Group for Lean Construction (IGLC 21) . Martins, J. R. M. (2011). Lean Construction na Construção e Engenharia PortuguesasOportunidades e Desafios Para os Donos de Obra . Universidade Nova de Lisboa. References 233 Matias, J., & Cachadinha, N. (2010). Evaluating the Potential Benefits and Challenges of LC Adoption in the Portuguese Construction Industry: A Survey Study. Proceedings IGLC-18 . https://run.unl.pt/handle/10362/10003 Matthews, J., Pellew, L., Phua, F., & Rowlinson, S. (2000). Quality relationships: Partnering in the construction supply chain. International Journal of Quality and Reliability Management , 17 (4), 493–510. https://doi.org/10.1108/02656710010371191 McHugh, K., Dave, B., & Koskela, L. (2022). On The Role of Lean in Digital Construction. In Industry 4.0 for the Built Environment (pp. 207–226). Springer. Meshref, A. N., Elkasaby, El. A. F. A., & Abdel Kader Mohamed Farid, A. (2023). Reducing construction waste in the construction life cycle of industrial projects during design phase by using system dynamics. Journal of Building Engineering , 69 . https://doi.org/10.1016/j.jobe.2023.106302 Michaud, M., Forgues, E.-C., Ing, M., Carignan, V., Forgues, D., & Ouellet-Plamondon, C. (2019). A lean approach to optimize BIM information flow using value stream mapping. Journal of Information Technology in Construction (ITcon) , 24 , 472–488. https://doi.org/10.36680/j.itcon.2019.025 Mulrow, C. D. (1994). Systematic reviews: rationale for systematic reviews. Bmj , 309 (6954), 597–599. Nahmens, I., & Ikuma, L. H. (2012). Effects of Lean Construction on Sustainability of Modular Homebuilding. Journal of Architectural Engineering , 18 (2), 155–163. https://doi.org/10.1061/(ASCE)AE.1943-5568.0000054 Nath, T., Attarzadeh, M., Tiong, R. L. K., Chidambaram, C., & Yu, Z. (2015). Productivity improvement of precast shop drawings generation through BIM-based process re-engineering. Automation in Construction , 54 , 54–68. https://doi.org/10.1016/j.autcon.2015.03.014 NBS. (2018). What is BIM? | NBS. In National Building Specification (NBS) . https://doi.org/10.3917/deba.065.0117 Neve, H. H., & Wandahl, S. (2018). Towards identifying making-do as lead waste in refurbishment projects. IGLC 2018 - Proceedings of the 26th Annual Conference of the International Group for References 234 Lean Construction: Evolving Lean Construction Towards Mature Production Management Across Cultures and Frontiers , 2 , 1354–1364. https://doi.org/10.24928/2018/0236 Nguyen, D. T., & Sharmak, W. (2022). Hybrid Simulation for Value Stream Mapping to Improve the Environmental Performance of the Construction Phase. 30th Annual Conference of the International Group for Lean Construction, IGLC 2022 , 421–432. https://doi.org/10.24928/2022/0145 Nunnally, J., & Bernstein, I. (1994). Psychometric theory. In (No Title) (3rd Edition). McGraw-Hil. Ohno, T. (1988). Toyota production system: beyond large-scale production . CRC Press. Olivieri, H., Seppänen, O., & Granja, A. D. (2018). Improving workflow and resource usage in construction schedules through location-based management system (LBMS). Construction Management and Economics , 36 , 109–124. https://doi.org/10.1080/01446193.2017.1410561 Omotayo, T., Awuzie, B., Egbelakin, T., Obi, L., & Ogunnusi, M. (2020). AHP-systems thinking analyses for kaizen costing implementation in the construction industry. Buildings , 10 (12), 1–24. https://doi.org/10.3390/buildings10120230 Oskouie, P., Gerber, D., Alves, T., & Becerik-Gerber, B. (2012). Extending the Interaction of Building Information Modeling and Lean Construction. 20th Annual Conference of the International Group for Lean Construction , 1 (617), 111–120. https://doi.org/10.1016/j.autcon.2010.09.013 Osmani, M., Glass, J., & Price, A. D. F. (2008). Architects ’ perspectives on construction waste reduction by design. Waste Management , 28 , 1147–1158. https://doi.org/10.1016/j.wasman.2007.05.011 Park, M., & Peña‐Mora, F. (2003). Dynamic change management for construction: introducing the change cycle into model‐based project management. System Dynamics Review: The Journal of the System Dynamics Society , 19 (3), 213–242. Pedrosa, M., Arantes, A., & Cruz, C. O. (2023). Barriers to Adopting Lean Methodology in the Portuguese Construction Industry. Buildings , 13 (8). https://doi.org/10.3390/buildings13082047 References 235 Peneirol, N. L. S. (2007). Lean Construction em Portugal Engenharia Civil: Caso de estudo de implementação de sistema de controlo da produção Last Planner [Dissertação para obtenção do Grau de Mestre]. Instituto Superior Técnico - ULISBOA. Pereira, C. M. N. (2014). Implementação da Lean Construction na Construção Nacional [Dissertação de Mestrado]. Universidade do Minho . Pereira, D., & Cachadinha, N. (2011). Lean construction in rehabilitation works - Suitability analysis and contribution for the definition of an application model. 19th Annual Conference of the International Group for Lean Construction 2011, IGLC 2011 , 157–167. Pereira, F. F. (2019). Implementação de Práticas Lean na Gestão de Subempreitadas: Estudo de Caso [Master Thesis, Universidade Do Porto]. http://www.fe.up.pt Pérez, C. T., & Bastos Costa, D. (2021). Increasing production efficiency through the reduction of transportation activities and time using 4D BIM simulations. Engineering, Construction and Architectural Management , 2954 . https://doi.org/10.1108/ECAM-02-2020-0132 Perez, C. T., Sommer, L., Costa, D. B., & Formoso, C. T. (2015). A case study on causes and consequences of transportation waste. Conf. of the Int’l. Group for Lean Construction , 444– 453. www.iglc.net Pikas, E., Sacks, R., & Priven, V. (2012). Go or No-Go Decisions at the Construction Workface: Uncertainty, Perceptions of Readiness, Making Ready and Making-Do. In I. D. Tommelein & C. & Pasquire (Eds.), 20th Annual Conference of the International Group for Lean Construction . Porwal, A. (2013). Construction waste management at source : a Building Information Modeling based system dynamic approach . University of British Columbia. Porwal, A., & Hewage, K. N. (2012). Building information modeling-based analysis to minimize waste rate of structural reinforcement. Journal of Construction Engineering and Management , 138 (8), 943–954. https://doi.org/10.1061/(ASCE)CO.1943-7862.0000508 Prascevic, N., & Prascevic, Z. (2017). Application of fuzzy AHP for ranking and selection of alternatives in construction project management. Journal of Civil Engineering and Management , 23 , 1123–1135. https://doi.org/10.3846/13923730.2017.1388278 References 236 Proença, V. L. S. (2017). Aplicação do Last Planner System (LPS) em Casos de Estudo [Mestre em Engenharia Civil]. Instituto Superior de Engenharia Do Porto (ISEP). PTPC. (2016). Guia Prático: Lean na Construção . https://www.impic.pt/impic/assets/misc/pdf/Guia_Pratico_Lean_Contrucao_Set_2016.pdf Ratajczak, J., Riedl, M., & Matt, D. T. (2019). BIM-based and AR Application Combined with Location-Based Management System for the Improvement of the Construction Performance . https://doi.org/10.3390/buildings9050118 Ratajczak, J., Schimanski, C. P., Marcher, C., Riedl, M., & Matt, D. T. (2017). Mobile Application for Collaborative Scheduling and Monitoring of Construction Works According to Lean Construction Methods. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) , 10451 LNCS , 207–214. https://doi.org/10.1007/978-3-319-66805-5_26 Richardson, G. P., & Pugh, A. (1981). Introduction to System Dynamics Modeling with DYNAMO. Journal of the Operational Research Society , 48 , 1146. https://api.semanticscholar.org/CorpusID:62752425 Rodrigues, A., & Bowers, J. (1996). System dynamics in project management: A comparative analysis with traditional methods. System Dynamics Review , 12 (2), 121–139. https://doi.org/10.1002/(sici)1099-1727(199622)12:2<121::aid-sdr99>3.0.co;2-x Rooke, J. (2020). People and knowledge: lean organisation. In Lean Construction (pp. 85–101). Routledge. Rosenbaum, S., Toledo, M., & González, V. (2014). Improving Environmental and Production Performance in Construction Projects Using Value-Stream Mapping: Case Study. Journal of Construction Engineering and Management , 140 (2), 04013045. https://doi.org/10.1061/(ASCE)CO.1943-7862.0000793 Rother, M., & Shook, J. (2003). Learning to see: value stream mapping to add value and eliminate muda . Lean Enterprise Institute. Rozenfeld, O., Sacks, R., & Rosenfeld, Y. (2009). “CHASTE”: Construction hazard assessment with spatial and temporal exposure. Construction Management and Economics , 27 (7), 625–638. https://doi.org/10.1080/01446190903002771 References 237 Sacks, R. (2016). What constitutes good production flow in construction? Construction Management and Economics , 34 (9), 641–656. https://doi.org/10.1080/01446193.2016.1200733 Sacks, R., Eastman, C., Lee, G., & Teicholz, P. (2018). BIM Handbook. In BIM Handbook (Third edit). John Wiley & Sons, Inc. https://doi.org/10.1002/9781119287568 Sacks, R., & Goldin, M. (2007). Lean Management Model for Construction of High-Rise Apartment Buildings. Journal of Construction Engineering and Management , 133 (5), 374–384. https://doi.org/10.1061/(asce)0733-9364(2007)133:5(374) Sacks, R., Korb, S., Barak, R., & Geva, G. (2018). Building lean, building BIM : improving construction the Tidhar way (Issue December 2017). Routledge Taylor & Francis Group. Sacks, R., Koskela, L., Dave, B. A., & Owen, R. (2010). Interaction of lean and building information modeling in construction. Journal of Construction Engineering and Management , 136 (9), 968– 980. https://doi.org/10.1061/(ASCE)CO.1943-7862.0000203 Sacks, R., Radosavljevic, M., & Barak, R. (2010). Requirements for building information modeling based lean production management systems for construction. Automation in Construction , 19 (5), 641–655. https://doi.org/10.1016/j.autcon.2010.02.010 Saeed, K. (1998). Defining a problem or constructing a reference mode . Department of Social Science and Policy Studies, Worcester Polytechnic Institute. Saieg, P., Sotelino, E. D., Nascimento, D., & Caiado, R. G. G. (2018). Interactions of Building Information Modeling, Lean and Sustainability on the Architectural, Engineering and Construction industry: A systematic review. Journal of Cleaner Production , 174 , 788–806. https://doi.org/10.1016/j.jclepro.2017.11.030 Santos, A. (1999). Application of flow principles in the production management of construction sites . http://usir.salford.ac.uk/2231/ Sarhan, S., Pasquire, C., Manu, E., & King, A. (2017). Contractual governance as a source of institutionalised waste in construction. International Journal of Managing Projects in Business , 10 (3), 550–577. https://doi.org/10.1108/ijmpb-07-2016-0058 238 Satoshi Hino. (2005). Inside the Mind of Toyota: Management Principles for Enduring Growth (Fujimoto Takahiro, Ed.; 1st ed.). Taylor & Francis. Saurin, T. A., & Sanches, R. C. (2014). Lean Construction and Resilience Engineering - Complementary Perspectives of Variability. In B. T. , Kalsaas, L. Koskela, & T. A. Saurin (Eds.), 22nd Annual Conference of the International Group for Lean Construction. Oslo (pp. 61–71). Schimanski, C. P., Marcher, C., Monizza, G. P., & Matt, D. T. (2020). The last planner® system and building information modeling in construction execution: From an integrative review to a conceptual model for integration. Applied Sciences (Switzerland) , 10 (3). https://doi.org/10.3390/app10030821 Schimanski, C. P., Pasetti Monizza, G., Marcher, C., & Matt, D. T. (2021). Development of a BIMbased production planning and control system for Lean Construction through advancement and integration of existing management techniques. Frontiers of Engineering Management , 8 (3), 429– 441. https://doi.org/10.1007/s42524-020-0105-5 Schumacker, R. E., & Lomax, R. G. (2004). A beginner’s guide to structural equation modeling, 2nd ed. In A beginner’s guide to structural equation modeling, 2nd ed. Lawrence Erlbaum Associates Publishers. Seppänen, O. (2020). Location-based management system now and in the future. In Lean Construction (pp. 276–300). Routledge. Seppänen, O., Ballard, G., & Pesonen, S. (2010). The Combination of Last Planner System and Location-Based Management System. Proceedings IGLC-18 , 467–476. http://p2sl.berkeley.edu Seppänen, O., Modrich, R., & Ballard, G. (2015). Integration of Last Planner System and LocationBased Management System. 23rd Ann. Conf. of the Int’l. Group for Lean Construction , 123–132. Sheramn, B., & Kirwan, B. (2019). To what extent can Lean Construction processes and principles be utilised in the design stage of a BIM project in the Irish design and construction industry . https://arrow.tudublin.ie/schmuldistcap Shewhart, W. A., & Deming, W. E. (1986). Statistical method from the viewpoint of quality control . Courier Corporation. References 239 Silva, P. (2008). Implementing Lean Construction in Portugal-Adaptation after studying a Danish case [Master’s Thesis]. Instituto Superior Técnico. Sommer, L. (2010). Contribuições para um método de identificação de perdas por improvisação em canteiros de obras . Sterman, J. D. (2000). Business dynamics: Systems thinking and modeling for a complex world, . McGraw-Hill. Sukster, R. (2005). A integração entre o sistema de gestão da qualidade e o planejamento e controle da produção em empresas construtoras . Tezel, A., & Aziz, Z. (2017). Benefits of visual management in construction: case from the transportation sector in England. Construction Innovation , 17 (2), 125–157. https://doi.org/10.1108/JBIM-06-2016-0127 Tezel, A., Koskela, L., & Tzortzopoulos, P. (2016). Visual management in production management: A literature synthesis. In Journal of Manufacturing Technology Management (Vol. 27, Issue 6, pp. 766–799). Emerald Group Publishing Ltd. https://doi.org/10.1108/JMTM-08-2015-0071 Tezel, A., Taggart, M., Koskela, L., Tzortzopoulos, P., Hanahoe, J., & Kelly, M. (2020). Lean construction and BIM in small and medium-sized enterprises (SMEs) in construction: A systematic literature review. Canadian Journal of Civil Engineering , 47 (2), 186–201. https://doi.org/10.1139/cjce-2018-0408 Thomas, H. R., Maloney, W. F., Horner, R. M. W., Smith, G. R., Handa, V. K., & Sanders, S. R. (1990). Modeling Construction Labor Productivity. Journal of Construction Engineering and Management , 116 (4), 705–726. https://doi.org/10.1061/(ASCE)0733-9364(1990)116:4(705) Tommelein, I. D., & Gholami, S. (2012). Root Causes of Clashes in Building Information Models. In I. D. Tommelein & C. L. Pasquire (Eds.), 20th Annual Conference of the International Group for Lean Construction . Tranfield, D., Denyer, D., & Smart, P. (2003). Towards a Methodology for Developing EvidenceInformed Management Knowledge by Means of Systematic Review. British Journal of Management , 14 (3), 207–222. https://doi.org/10.1111/1467-8551.00375 References 240 Ullman, J. B., & Bentler, P. M. (2012). Structural Equation Modeling. In Handbook of Psychology, Second Edition . Wiley. https://doi.org/10.1002/9781118133880.hop202023 UN Environmental and International Energy Agency, G. S. R. (2017). Towards a zero-emission, efficient, and resilient buildings and construction sector. In Global Status Report . Uusitalo, P., Olivieri, H., Seppänen, O., Pikas, E., & Peltokorpi, A. (2017). Review of Lean Design Management: Processes, Methods and Technologies. 25th Annual Conference of the International Group for Lean Construction , 571–578. https://doi.org/10.24928/2017/0224 Uusitalo, P., Seppänen, O., Lappalainen, E., Peltokorpi, A., & Olivieri, H. (2019). Applying level of detail in a BIM-based project: An overall process for lean design management. Buildings , 9 (5), 1– 13. https://doi.org/10.3390/buildings9050109 Viana, D. D., Formoso, C. T., & Isatto, E. L. (2017). Understanding the theory behind the Last Planner System using the Language-Action Perspective: two case studies. Production Planning and Control , 28 (3), 177–189. https://doi.org/10.1080/09537287.2016.1233360 Viana, D. D., Formoso, C. T., & Kalsaas, B. T. (2012). Waste in construction: A systematic literature review on empirical studies. IGLC 2012 - 20th Conference of the International Group for Lean Construction . Vilasini, N., Neitzert, T., & Rotimi, J. (2014). Developing and evaluating a framework for process improvement in an alliance project: A New Zealand case study. Construction Management and Economics , 32 (6), 625–640. https://doi.org/10.1080/01446193.2013.874565 Vilventhan, A., Ram, V. G., & Sugumaran, S. (2019). Value stream mapping for identification and assessment of material waste in construction: A case study. Waste Management and Research , 37 (8), 815–825. https://doi.org/10.1177/0734242X19855429 von Heyl, J., & Teizer, J. (2017). Lean Production Controlling and Tracking Using Digital Methods. 25th Annual Conference of the International Group for Lean Construction , II (July), 127–134. https://doi.org/10.24928/2017/0238 Vrijhoef, R. (2020). Extended roles of construction supply chain management for improved logistics and environmental performance. In Lean Construction (pp. 253–275). Routledge. References 241 Vrijhoef, R., & Koskela, L. (2000). The four roles of supply chain management in construction. European Journal of Purchasing and Supply Management , 6 (3–4), 169–178. https://doi.org/10.1016/S0969-7012(00)00013-7 Wang, Y., Zhao, J., Gao, N., & Shen, F. (2024). A Dynamic Evaluation Method for the Development of Intelligent Construction Technology in the Construction Field Based on Structural Equation Model–System Dynamics Model. Buildings , 14 (2). https://doi.org/10.3390/buildings14020417 Wohlin, C., & Claes. (2014). Guidelines for snowballing in systematic literature studies and a replication in software engineering. Proceedings of the 18th International Conference on Evaluation and Assessment in Software Engineering - EASE ’14 , 1–10. https://doi.org/10.1145/2601248.2601268 Womack, J., & Jones, D. (2003). Lean Thinking-Banish Waste and Create Wealth in Your Corporation. In Free Press . Simon and Schuster. Wu, P., Low, S. P., & Jin, X. (2013). Identification of non-value adding (NVA) activities in precast concrete installation sites to achieve low-carbon installation. Resources, Conservation and Recycling , 81 , 60–70. https://doi.org/10.1016/j.resconrec.2013.09.013 Yin, R. K. (2018). Case study research and applications . Sage. Yuan, H., & Shen, L. (2011). Trend of the research on construction and demolition waste management. In Waste Management (Vol. 31, Issue 4, pp. 670–679). Pergamon. https://doi.org/10.1016/j.wasman.2010.10.030 Zhan, Z., Tang, Y., Wang, C., Yap, J. B. H., & Lim, Y. S. (2022). System Dynamics Outlook on BIM and LEAN Interaction in Construction Quantity Surveying. Iranian Journal of Science and Technology - Transactions of Civil Engineering , 46 (5), 3947–3962. https://doi.org/10.1007/s40996-022-00833-w Zhang, X., Azhar, S., Nadeem, A., & Khalfan, M. (2018). Using Building Information Modelling to achieve Lean principles by improving efficiency of work teams. International Journal of Construction Management , 18 (4), 293–300. https://doi.org/10.1080/15623599.2017.1382083 Zlatanovic, D. (2012). System Dynamics Models in Management Problem Solving. Economic Horizons , 14 (1), 25–38. www.ekfak.kg.ac.rs References III.Appendix C: D3M User Guide 249 III. Appendix C: D3M User Guide III.1. Context D3M initially emerged through a dissertation within a research initiative focused on assessing the potential effects of Making-Do (MD) on production practices and construction outcomes. Furthermore, it delves into the possible advantages of merging the LPS and BIM for production planning and control. Figure III-1 -- Cloud application of D3M. Figure III-2 is a snapshot of the welcome message once the model is opened; the message contains a brief introduction with objectives of the model development and simulation and a short description of how D3M can be helpful to the user once the run button is clicked. Figure III-2 -- D3M Welcome Message III.Appendix C: D3M User Guide 250 D3M provides a direct assessment of 25 strategies from LPS-BIM and suggests additional pathways to formulate lean-based policy, as presented in Figure III-3. Figure III-3 -- The dynamic framework for MD waste analysis. III.Appendix C: D3M User Guide 251 Table III-1 delineates each strategy used by D3M, where the parameters and their coefficients were used in the D3M system. All parameters had their weights distributed by considering the estimation results of the SEM mediation analysis. Moreover, the relationship between variables is applied, MDK influenced by LPSF_F, COO_F exerting influences LPS_F and influenced by BIMF_F. Table III-1 -- Parameters of LPS-BIM within D3M Component Parameter Coefficient Formula LPS_F VA10 9.8% ( 0.098* VA10 + 0.072 * VA11 + 0.077 * VA12 + 0.074 *VA13+ 0.073 * VA14 + 0.084 * VA15 + 0.072 * VA16 + 0.064 * VA17 + 0.057 * VA18 + 0.071 * VA19 + 0.077 * VA20) + 0.106 * COO_F VA11 7.2% VA12 7.7% VA13 7.4% VA14 7.3% VA15 8.4% VA16 7.2% VA17 6.4% VA18 5.7% VA19 7.1% VA20 7.7% COO_F 10.6% COO_F VA1 5.7% (0.057 * VA1 + 0.067 * VA2 + 0.055 * VA3 + 0.083* VA4) + 0.160 * BIMF_S VA2 6.7% VA3 5.5% VA4 8.3% BIMF_S 16% MDK VA5 8.5% ( 0.85 * VA5 + 0.047 * VA6 + 0.055 * VA7 + 0.057 * VA8 + 0.081 * VA9) + 0.116 * LPSF_F VA6 4.7% VA7 5.5% VA8 8.1% LPSF_F 11.6% BIMF_F VA21 12.5% (0.125 * VA21 + 0.070 * VA22 + 0.055 * VA23 + 0.061 *VA24 + 0.077 * VA25) VA22 7% VA23 5.5% VA24 6.1% VA25 7.7% III.Appendix C: D3M User Guide 252 D3M was fully developed by System dynamics modelling using Java codes and hosted by AnyLogic 8 Professional 8.7.11. This document serves as a guide for the best practice of D3M. The information in this document does not fully replace a careful understanding of LPS and BIM methods. Figure III-4 -- User input window. Figure III-5 (a) illustrates the options of preset inputs for the scenarios of extreme value simulation, comprising six buttons. Figure III-5 (b) is an example of code used to set all LPS-related parameters in scenarios A to 5 as the maximum value of the Likert scale used throughout the simulation and following SEM results. (a) (b) Figure III-5 -- preset buttons for extreme value tests. III.Appendix C: D3M User Guide 253 Another option to allow users to test variations of the studied parameters is available through sliders, as shown in Figure III-6. Figure III-6 -- Slider controllers to set parameter values. D3M research was funded by the Portuguese Foundation for Science and Technology (FCT), with grant number 2021.04751.BD III.Appendix C: D3M User Guide 254 III.2. D3M folder The D3M folder must not be downloaded to the computer if the user is not using Anylogic software. In case the user holds an Anylogic license, the following files are included: •Executable D3M file (*.alp) •Welcome message (*.png) •D3M Excell spreadsheet (*.xlsx) •D3M database folder (db.properties) and (db. script) •D3M User Guide (*.pdf) III.3. D3M guidelines and outputs To start using D3M, a master plan is required. The dynamic model can be edited or modified in AnyLogic 8 Professional 8.7.11 or later. III.3.1. D3M guidelines The system dynamic model should be run through some user input data that must include the following: •Estimated finish date for each stage or milestone. •Estimated duration in months. •Estimated number of resources used. •Scale of LPS-BIM parameters 1 to 5. III.3.2. D3M output The most vital feature of D3M is not only the simulation for stock and flow diagrams but also real-time information presented in the control dashboard that shows the stored values of constraints, making do incidents and wastes in the (Tasks) unit (As given in Figure III-7). Figure III-7 -- D3M dashboard. III. Appendix C: D3M User Guide 255 Figure III-8 (a) represents how many stages are planned according to the master schedule. (b) showcases the number of tasks in WIP stock and (c) the number of discovered constraints. (d) The progress of finished works is categorized into substages: (e) the number of tasks that emerged wastes and (f) the number of MD incidents categorized. (a) (b) (c) (d) (e) (f) Figure III-8 -- Control charts of D3M. The first page of stock and flow diagrams, presented in Figure III-9, shows the work progress located. The panning cursor / will be activated by pressing on the scrolling roll. (a) represents the master schedule information, (b) the interaction between WIP, finished works and constraints and (c) includes the socks of making do and wastes. Each stock in this figure should be working and it is necessary to III.Appendix C: D3M User Guide 256 identify the area of construction, milestones dates and expected finish dates for each substage to constrain the model into real-world behaviour. Figure III-9 -- Work progress and MD subsystems. Figure III-10 illustrates the space management system; the essential value is the workspaceAvailable variable because this variable will be used to calculate productivity. Figure III-10 -- Location subsystem. The figure below describes the impact of resources on productivity and spaces, TimeToIncreaseWorkforce and resourceAdjustmentTime > 0; it must be more than 0 when needed. LPS, BIM and COO parameters have an impact on resource Efficiency. Part (a) represents the impact of workflow on resources; RR is the conversion between units (tasks) to (person); (b) this section (a) (b) (c) III.Appendix C: D3M User Guide 257 represents the core part of the resources subsystem; (c) is the most vital part where it connects the relationship from resources to the productivity rate. Figure III-11 -- Resources subsystem Figure III-12 showcases how schedule pressure affects productivity (prod), overtime, workflow, space limitation and MD decisions are included in productivity calculation. Figure III-12 -- Productivity subsystem (a) (b) (c) IV.Appendix D: D3M Components 258 IV. Appendix D: D3M Components IV.1. Evaluated LPS-BIM Parameters Table IV-1 explains the parameters of LPS-BIM, which are strategies to mitigate MD and increase the adoption of LPS-BIM. Table IV-1 -- Parameters explanation and scale ID Factor Explanation Rating Scale VA1 Handling conflicts among various stakeholders This parameter measures how the team can address conflicts among diverse stakeholders in resolving disputes about responsibility for constraints and the mechanisms employed for their removal. Likert 1-5 VA2 Coordination level This input specifies the percentage of coordination implemented during Last Planner System (LPS) sessions. Likert 1-5 VA3 Enabling discussions This parameter specifies the level of opening discussions, how project management encourages dialogue among relevant parties during the planning phase and the resolution of their tasks within lookahead and make-ready planning sessions. Likert 1-5 VA4 Local adjustments to organisations This parameter encompasses the modifications made to accommodate the Last Planner System in alignment with the organisation's culture, resources, projects, priorities, goals and visions. Likert 1-5 VA4 Engagement in constraints analysis This parameter enables users to designate the degree of involvement for each trade in the constraints analysis process. A heightened level of engagement can result in more reliable plans and a reduction in MD practices when constraints are removed correctly and collaboratively. Likert 1-5 IV.Appendix D: D3M Components 265 DE15 constraintsRate[CONS TRAINTS] LPSF_S <=60? zidz(constraints[CONSTRAINTS], timetodetectConstraints[CONSTRAINTS]): zidz(zidz(constraints[CONSTRAINTS] , timetodetectConstraints[CONSTRAINTS]), impactOfCOOonConstraints * ImpactofLPSonConstraints(LPSF_S)) Task/time DE16 timetodetectConstraint s effectOFScedulePressureOnConstraintsRemoval (completionFraction.average()) time DE17 sumconsts[CATEGORI ES] constraints[CONSTRAINTS] Task DE18 manHrsNeeded zidz(WIP_value.average(),maxResources.average ()) Task/Person DE19 TaskPerPersonPerMon th zidz(manHrsNeeded,remainingTime.average()) Task / (Person * time) DE20 impacts[CATEGORIES] COO + MDK < 592? MDCAT1(sumconsts[CATEGORIES]) :MDCAT1(sumconsts[CATEGORIES]) /1.5 Task/time DE21 MDtoImpactRate [CATEGORIES] LPSF_S <= 100? MD[CATEGORIES]*percentageOFnegativeMD[CA TEGORIES]: MD[CATEGORIES]*percentageOFnegativeMD[CA TEGORIES] Task/time DE22 percentageOFnegative MD MDK<1? 0.9: 0.5 1/time DE23 innovativeMD [CATEGORIES] MD[CATEGORIES]*(1percentageOFnegativeMD[CATEGORIES]) Task/time DE24 completionFraction [SUBSTAGES] zidz(finishedWork[SUBSTAGES],WIP_value[SUBS TAGES]) unitless DE25 projectIsDone [SUBSTAGES] completionFraction[SUBSTAGES]>=1?1:0 unitless IV.Appendix D: D3M Components 266 DE26 requiredWorkflow [SUBSTAGES] projectIsDone[SUBSTAGES] != 0 && WIP[SUBSTAGES] < 0 ? 0 : max(xidz(WIP[SUBSTAGES], remainingTime[SUBSTAGES], maxWorkflow[SUBSTAGES]) , maxWorkflow[SUBSTAGES]) Task/time DE27 remainingTime [SUBSTAGES] max( 0, completionDate[SUBSTAGES] - time() ) time DE28 completionRate[SUBST AGES] projectIsDone[SUBSTAGES] ==1 ?0 : delay(min(requiredWorkflow[SUBSTAGES] , prod[SUBSTAGES] * resourcesProductivityPerSS[SUBSTAGES]),startM onth[SUBSTAGES]) Task/time DE29 timeToDiscoverMDs[S UBSTAGES] effectOFScedulePressureOnConstraintsRemoval (completionFraction[SUBSTAGES]) time DE30 MD_IncidentsRate[CAT EGORIES] (0.061*MD[CAT1]+0.04*MD[CAT4])/timeToDisc overMDs[SS1] Task/time DE31 MD_IncidentsRate [SUBSTAGES] MD[CATEGORIES]/timeToDiscoverMDs[SUBSTA GES] Task/time DE32 resourcesComing[RES OURCES] resourcesGap[RESOURCES] > TOT_Resources[RESOURCES]? (resourcesGap[RESOURCES]- TOT_Resources[RESOURCES]) /TimeToIncreaseWorkforce:0 Person/time DE33 crowdingEffectOnConst ructionEffeciency [RESOURCES] zidz(requiredResources[RESOURCES], resources[RESOURCES]) time DE34 resources[RESOURCE S] changeInResources[RESOURCES] - dismissals[RESOURCES]* zidz( resources[RESOURCES], TOT_Resources[RESOURCES] ) Person IV.Appendix D: D3M Components 267 DE35 resourcesEfficiency [RESOURCES] crowdingEffectOnConstructionEffeciency[RESOU RCES]*BIMEffectOnCommunication*plannedPro ductivity *LPSEffectOnEfficiency Person DE36 Refficiency_To_SS[SU BSTAGES] Regression Equation Person DE37 plannedProductivity uniform_discr(20,40) Person DE38 TOT_Resources [RESOURCES] resources[RESOURCES]+newWorkForce[RESOU RCES] Person DE39 dismissals [RESOURCES] resourcesGap[RESOURCES] < TOT_[RESOURCES] ? ( TOT_Resources[RESOURCES] - resourcesGap[RESOURCES] )/ dismissalTime[RESOURCES] : 0 Person/time DE40 schedulePressure[SUB STAGES] remainingTime[SUBSTAGES] <= 0 && !(projectIsDone[SUBSTAGES] != 0) ? maxSchedulePressure : zidz( requiredWorkflow[SUBSTAGES], normalWorkFlow[SUBSTAGES] ) unitless DE41 normalWorkFlow [SUBSTAGES] RG_TO_SS[SUBSTAGES]*normalProductivity Task/time DE42 overtime [SUBSTAGES] overtimeTBFTN(schedulePressure[SUBSTAGES]) time DE43 effFatigueProductivity [SUBSTAGES] = fatigueEffProductivityTBFN(averageOvertime[SUB STAGES]) 1/time DE44 prod [SUBSTAGES] effFatigueProductivity[SUBSTAGES]* normalProductivity * overtime[SUBSTAGES]* impactOfWorkSpaceLimitation[SUBSTAGES]+(M DSS[SUBSTAGES]) Task / (time * Person) DE45 workspacePerCrewAvai lable[SUBSTAGES] workspaceAvailable*resourcesProductivityPerSS[ SUBSTAGES] Person * space DE46 impactOfWorkSpaceLi mitation impactOfWorkSpaceLimitations(workspacePerCr ewAvailable[SUBSTAGES]) IV.Appendix D: D3M Components 268 DE47 chanegeInLocations completionFraction.average() <1 && locationsDesign >0 ? 1 : 0 space/time DE48 uncompensableDelays projectDuration*0.85 <= time() && completionFraction.average() < 0.80 ?abs(projectDuration - time()) : 0 time DE49 constructionResources Cost averageOvertime.average()>1? newWorkForce.average()* constructionResourceUnitCostPercentage * averageOvertime.average()*overtimeUnitCost: newWorkForce.average() * constructionResourceUnitCostPercentage R$/time DE50 constructionResourceU nitCostPercentage 7.5 1/ (Person * time) DE51 overtimeUnitCost 1.2 $/time DE52 WIP[SUBSTAGES] MD_IncidentsRate[SUBSTAGES] - completionRate[SUBSTAGES] + constraintsRate[Constraints] Task DE53 wasteRate[SUBSTAGE S] LPSF_S + BIMF_S >=50 ? wasteGenerationRate(completionFraction[SUBST AGES]) *completionRate[SUBSTAGES]*0.7 : wasteGenerationRate(completionFraction[SUBST AGES]) Task/time DE54 replanningRate[CONST RAINTS] zidz(Waste[IMPACTS], timetoDetectConstraints[CONSTRAINTS]) Task/time DE55 d(Waste[IMPACTS])/dt MDtoImpactRate[CATEGORIES]+wasteRate[SUB STAGES]-replanningRate[CONSTRAINTS] Task DE56 spaceChange zidz( durationInHrs *resourcesGap.average() ,finishedWork.average()) 1/time DE57 locationsUtilisationRate occupiedLocations *spaceChange >= initialLocations ? 0 :occupiedLocations *spaceChange space/time IV.Appendix D: D3M Components 269 DE58 workspaceAvailable initialLocations-finishedLocationsoccupiedLocations space DE59 initialLocations 50 space DE60 LPSEffectOnEfficiency LPSF_S >= 50 ? 1.5*impactOfCommitementPlanning : 1*impactOfCommitementPlanning unitless DE61 projectCost indirectCost + zidz(uncompensableDelaysCost,uncompensable Delays)* time() + $ DE62 uncompensableDelays Cost 1.25 $ DE63 indirectCost indirectCostPrecentage * Waste.average() $ DE64 indirectCostPrecentage 0.6 $/ Task DE65 impactOfCommitement Planning COO>=50? 1.5: 1 unitless DE66 MDSS[SUBSTAGE] MDSS_SUBSTAGE(innovativeMD.average()) Task DE67 constraintsRate[CONS TRAINTS] LPSF_S <=60? zidz(constraints[CONSTRAINTS], timetodetectConstraints[CONSTRAINTS]): zidz(zidz(constraints[CONSTRAINTS] , timetodetectConstraints[CONSTRAINTS]), impactOfCOOonConstraints * ImpactofLPSonConstraints(LPSF_S)) Task/time DE68 completionRate projectIsDone[SUBSTAGES] ==1 ?0 : delay(min(requiredWorkflow[SUBSTAGES] , prod[SUBSTAGES] *resourcesProductivityPerSS[SUBSTAGES]),start Month[SUBSTAGES]) Task/time DE69 d(newWorkForce [RESOURCES])/dt resourcesComing[RESOURCES] - changeInResources[RESOURCES] - dismissals[RESOURCES] * zidz( newWorkForce[RESOURCES], TOT_Resources[RESOURCES] ) Person IV.Appendix D: D3M Components 270 DE70 changeInResources [RESOURCES] newWorkForce[RESOURCES]/resourceAdjustem entTime Person/time DE71 resourceAdjustementTi me 2 time DE72 MDtoImpactRate [CATEGORIES] LPSF_S <= 100? MD[CATEGORIES]*percentageOFnegativeMD[CA TEGORIES]: MD[CATEGORIES]*percentageOFnegativeMD[CA TEGORIES] Task/time IV.4. Table Functions Table IV-4 presents the table functions, preset non-linear functions that are not mathematical and represent relationships between different variables; some of these tables are observed from collected data and literature. Table IV-4 -- Table functions in D3M Table Function Data Unit T1 effectOFScedulePressure OnConstraintsRemoval {(0 , 5 ),( 0.1 , 4.5),( 0.2 , 4) , (0.3 , 3 ),( 0.4 , 2) , (0.5,1 ), (0.6,0.9) , (0.7 , 0.8) , (0.8 , 0.7) , ( 0.9 , 0.6) , ( 1 , 0.5) } time T2 resourcesAllocationTBFT N {(0 , 1 ),( 0.5 , 0.8),( 0.8 , 0.2) , (1 , 0 ) } unitless T3 MDCAT1 {(0 , 0 ),( 1 , 0.5),( 2 , 5) , (5 , 12 ),( 10 , 13) , (15,14 ), (30,15) } Task/time T4 MDCAT2 {(0 , 0 ),( 10 , 20),( 20 , 50) , (30 , 100 ),( 40 , 110) , (50,115 ), (60,120) , (70 , 125) , (80 , 130) , ( 90 , 135) , ( 100 , 140) } Task/time T5 MDCAT3 {(0, 0 ),( 5, 3),( 20, 5), (30, 8 ),( 40, 9), (50,10 ), (60,15), (70, 17) } Task/time T6 MDCAT4 {(0, 0 ),( 2, 3),( 3, 30), (4, 35 ),( 5, 40), (6,61.5 ), (9,62), (11, 63), (15, 64) } Task/time IV.Appendix D: D3M Components 271 T7 MDCAT5 {(0 , 0 ),( 3 , 4),( 8 , 6) , (25 , 8 ),( 35 , 12) , (45,20 ), (50,21) , (60 , 21.5) , (70 , 22) , ( 80 , 22.5) , ( 90 , 23) } Task/time T8 overtimeTBFTN {(0 , 0.7 ),( 1 , 1),( 1.2 , 1.2) , (1.5 , 1.4 ),( 2 , 1.45) , (5,1.5 ), (10,1.55) , (20 , 1.6) } time T9 fatigueEffProductivityTB FN {(0 , 1 ),( 0.1 , 0.95),( 0.2 , 0.9) , (0.3 , 0.87 ),( 0.4 , 0.83) , (0.5,0.8 ), (0.6,0.75) , (0.7 , 0.7) , (0.8 , 0.68) , ( 0.9 , 0.65) , ( 1 , 0.6) } 1/time T11 impactOfApplyingBIMT echnology {(0 , 0.1 ),( 0.5 , 0.15),( 1 , 0.35) , (1.5 , 0.55 ),( 2 , 0.725) , (2.5,1.1 ), (3,1.5) , (3.5 , 2.025) , (4 , 2.825) , ( 4.5 , 3.7) , ( 5 , 5) } unitless T12 wasteGenerationRate {(0 , 0.05 ),( 0.1 , 0.09),( 0.2 , 0.25) , (0.3 , 0.4 ),( 0.4 , 0.45) , (0.5,0.55 ), (0.6,0.65) , (0.7 , 0.55) , (0.8 , 0.45) , ( 0.9 , 0.35) , ( 1 , 0.2) } unitless T13 schedulePressureEffectT BFTN {(0 , 0.4 ),( 0.25 , 0.5),( 0.5 , 0.6) , (0.75 , 0.8 ),( 1 , 0.9) , (1.25,1 ), (1.5,1.2) , (1.75 , 0.9) , (2 , 0.6) , ( 0.9 , 0.35) , ( 1 , 0.2) } unitless T14 impactOfWorkSpaceLimi tations {(5 , 0.65 ),( 10 , 0.68),( 15 , 0.7) , (20 , 0.76 ),( 25 , 0.8) , (30,0.93 ), (35,0.95) , (40 , 0.97) , (45 , 0.98) , ( 50 , 0.99) } unitless T15 timeToDiscoverWaste {(0 , 5 ),( 0.1 , 4.5),( 0.2 , 4) , (0.3 , 3 ),( 0.4 , 2) , (0.5,1 ), (0.6,0.9) , (0.7 , 0.8) , (0.8 , 0.7) , ( 0.9 , 0.6) , ( 1 , 0.5) } time T16 ImpactofLPSonConstrain ts {(0 , 0 ),( 20 , 0.1),( 40 , 0.22) , (50 , 0.5 ),( 60 , 0.75) , (100,0.88 ), (140,0.91) , (180 , 0.94) , (200 , 0.96) , ( 250 , 0.98) , ( 260 , 0.99) } unitless T17 ImpactOfLPSonWasteRe duction {(0 , 0.05 ),( 20 , 0.09),( 40 , 0.25) , (50 , 0.4 ),( 60 , 0.45) , (100,0.55 ), (140,0.65) , (150 , 0.75) , (160 , 0.8) , ( 170 , 0.9) } unitless 272 T18 MDSS1 {(1 , 0.04685 ),( 2 , 0.12186118),( 5 , 0.23425) , (10 , 0.4685 ),( 15 , 0.70275)} Task/time T19 MDSS2 {(1 , 0.14344 ),( 2 , 0.37309476),( 5 , 0.7172) , (10 , 1.4344 ),( 15 , 2.1516)} Task/time T20 MDSS3 {(1 , 0.33001 ),( 2 , 0.85805652),( 5 , 1.65005) , (10 , 3.3001 ),( 15 , 4.95015)} Task/time T21 MDSS4 {(1 , 0.06306 ),( 2 , 0.1640118),( 5 , 0.3153) , (10 , 0.6306 ),( 15 , 0.9459) } Task/time T22 MDSS5 {(1 , 0.09973 ),( 2 , 0.2593535),( 5 , 0.49865) , (10 , 0.9973 ),( 15 , 1.49595) } Task/time T23 MDSS6 {(1 , 0.062061 ),( 2 , 0.16138018),( 5 , 0.310305) , (10 , 0.62061 ),( 15 , 0.930915) } Task/time T24 MDSS7 {(1 , 0.06463 ),( 2 , 0.16815084),( 5 , 0.32315) , (10 , 0.6463 ),( 15 , 0.96945) } Task/time T25 MDSS8 {(1 , 0.01873 ),( 2 , 0.0487316),( 5 , 0.09365) , (10 , 0.1873 ),( 15 , 0.28095) } Task/time T26 MDSS9 {(1 , 0.25275 ),( 2 , 0.65722892),( 5 , 1.26375) , (10 , 2.5275 ),( 15 , 3.79125)} Task/time T27 MDSS10 {(1 , 0.22201 ),( 2 , 0.5774409),( 5 , 1.11005) , (10 , 2.2201 ),( 15 , 3.33015)} Task/time T28 MDSS11 {(1 , 0.07174 ),( 2 , 0.18663684),( 5 , 0.3587) , (10 , 0.7174 ),( 15 , 1.0761)} Task/time