| Author: | Zhang, Yaning |
| Title: | Work package-based project planning for optimization of carbon emissions in modular integrated construction |
| Advisors: | Teng, Yue (BRE) Li, Xiao (LSGI) Shen, Qiping Geoffrey (BRE) |
| Degree: | Ph.D. |
| Year: | 2026 |
| Department: | Department of Building and Real Estate |
| Pages: | xxi, 326 pages : color illustrations |
| Language: | English |
| Abstract: | The construction industry faces mounting pressure to reduce its carbon footprint, as it accounts for approximately 38% of global energy-related CO2 emissions. Modular Integrated Construction (MiC) has emerged as a promising approach to enhance sustainability in building projects, offering potential reductions in material waste, on-site construction impacts, and delay risks. However, existing efforts to improve MiC's carbon performance primarily focus on technical construction solutions, such as adopting low-carbon materials and implementing lean production practices to reduce waste. Research on using project management methods to improve MiC's carbon performance is notably limited. Work package-based project planning, an important project management method, has been shown to enhance cost-effectiveness and schedule performance by rationally sizing and sequencing work. It therefore also has significant potential to improve a project's carbon performance. This reveals a critical, underexplored opportunity for carbon mitigation through optimized work package configurations in MiC projects. The primary aim of this study is to develop work package-based approaches to optimize carbon emissions in MiC projects while accounting for inherent trade-offs and economic considerations. The specific objectives are: (1) to establish a mathematical foundation linking work package characteristics to carbon emissions through a productivity-based framework; (2) to develop a single-objective genetic algorithm for minimizing carbon emissions in MiC work packages; (3) to formulate a multi-objective optimization approach that balances carbon reduction with cost efficiency; (4) to develop an LLM-driven heuristic optimization framework to lower adoption barriers; and (5) to operationalize these approaches in an integrated platform for practical industry application. This study first establishes the theoretical relationship between work package configurations and carbon emissions using the Cobb-Douglas production function. Building on this foundation, three complementary optimization approaches are developed: a Two-List Genetic Algorithm (TLGA) for single-objective carbon minimization; evolutionary algorithms (NSGA-II and SPEA2) for multi-objective optimization balancing carbon and cost; and a novel LLM-based heuristic framework that provides accessible optimization capabilities. These approaches are validated through comprehensive computational experiments using both simulated project instances and a real-world case study involving a large-scale MiC development. The key findings demonstrate substantial carbon reduction potential through work package optimization. The TLGA achieved carbon emission reductions ranging from 49.93% to 81.15% compared with traditional work package schemes, while the multi-objective approaches identified balanced solutions achieving approximately 68% carbon reductions with only an 11% cost increase. The analysis revealed important patterns in carbon-optimized work packaging: larger projects benefit from smaller, more uniform work packages; different approaches are optimal for parallel factory production versus sequential on-site assembly; and the execution mode significantly influences optimal work package configurations. The LLM-based framework effectively bridged implementation barriers, with performance gaps narrowing markedly for larger projects and under more flexible scheduling conditions. This study makes original contributions to sustainable construction management from both theoretical and practical perspectives. Theoretically, it establishes a productivity-based carbon emissions framework for MiC projects; advances genetic algorithm methodology through an innovative chromosome representation; develops a comprehensive multi-objective optimization approach that balances environmental and economic objectives; and introduces a paradigm for human-AI collaborative optimization through a structured prompt-chain methodology. Practically, it provides accessible decision-support tools for identifying carbon-efficient work package configurations, quantifies trade-offs between carbon reduction and cost, and offers implementation guidelines tailored to specific project characteristics. The integrated optimization platform operationalizes these approaches in a cohesive computational environment, addressing the persistent gap between theoretical optimization capabilities and practical implementation in MiC project management. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14602