Abstract
The Architecture, Engineering, Construction, and Operation (AECO) industry consumes one-fourth of the total raw materials and generates two-thirds of solid waste in the UK. With increasing technologies and regulations, the recovery rates of construction waste (CW) are increasing, but the waste generated at first instance is still high. Previous studies emphasise that accurate quantification and classification of CW in the design stages is a prerequisite to optimising waste generation and enhancing proactive waste management (WM). Although several models exist to quantify, classify, optimise, and report CW generation, significant limitations hinder their applications on a broader scale. The critical limitations identified from a systematic literature review are that these models have an ununified data structure, lack of dynamics with Building Information Modelling (BIM) platforms, absence of established semantic relationships, and lack of standardised units of measurement. These limitations challenge automation, interoperability, consistency, accuracy, reusability, quality decision-making, and integration with actual project workflows. Thus, to address these gaps, the research aims to explore the functionalities of BIM and Semantic Web Technologies (SWT) that promise automation and interoperability within the CW quantification and optimisation process and provide standardised and consistent outputs to make informed decisions from early design stages.The research also explores the industry requirements for CW quantification, classification, and management to make robust and timely decisions. This was guided by a thematic analysis of semi-structured interview data gathered from stakeholders in the UK's AECO industry. The four main themes identified from the analysis include the following: waste data-driven decisions in the design phase, digital innovation in CW quantification, carbon-influenced decision-making, and collaboration within the whole supply chain. The main conclusion drawn from the analysis is that a lack of granular and scalable data impeded comprehensive CW planning, management, and reporting across the life-cycle stages. Further, an exploratory analysis was performed by quantising the interview data to identify what information eases decision-making. An extensive list of 22 information types vital to making informed decisions are identified and characterised as technical, environmental, economic and generic CW indicators.
Accounting for the existing limitations and industry requirements, the research proposed a BIM-integrated semantic CW quantification and optimisation system, which guided the development of a semantic CW data model and a plugin in the Autodesk Revit environment named ‘Product Circularity Ontology (PRODCIRO)’ and ‘Building Waste Tool’, respectively. The developed system facilitates CW optimisation utilising the quantification and classification results from the early design stages, ensuring dynamic BIM-CW information flows and user interaction. The outputs of the developed system proved that in addition to the generic Waste Generation Rates (WGRs), technical and environmental impact data are vital for making comprehensive CW-related decisions. Also, the semantic relationships mitigate the inconsistencies within the material and CW characterisation codes and achieve a unified data structure, thus allowing the data model to deal with diverse data sources, enhance interoperability and automation, and support collaborative decision-making. Despite the fact that recycling of CW is sustainable and diverts waste from landfilling, the GWP analysis of different WM scenarios proved that the top layer (prevention/reduction) of the Waste Hierarchy is less impactful than the middle layers (reuse and recycling). So, the research emphasises that recycling should be preferred only in circumstances where waste prevention/reduction is challenging. Finally, the findings demonstrated that the proposed semantic model could act as an integrated CW data hub supporting BIM Level 3 adoption.
The research concludes that standardised and unified CW quantification and classification data are critical pillars enhancing the optimisation process. This integration further facilitates the CW domain to own a single-collaborative data environment over multiple-fragmented data silos. Further to technological enhancements, the semantic CW data becomes a key player for efficient communication and collaboration among the supply-chain members to make informed WM decisions.
| Date of Award | 8 Dec 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Sergio Rodriguez (Supervisor), Farzad Rahimian (Supervisor) & Nashwan Dawood (Supervisor) |