Abstract
In the pursuit of a better world and the United Nations Sustainable Development goals, smarter decisions are now critical in the realm of engineering project management. The integration of Environmental, Social, and Governance (ESG) factors into project cost estimation is now critical and timely, and Artificial Intelligence (AI) presents a transformative tool for this task. This research Aims to develop an AI-assisted methodology to estimate ESG-related project costs for facilities management of fleets of buildings, disrupting traditional cost management approaches and enhancing decision-making for a more sustainable future.
The Research Methodology includes a state-of-art review of 108 journal papers used to synthesise an initial conceptual framework. It includes Estimating Requirements, Data Integration, ESG Costs and Cost Drivers, AI Architecture, and the critical aspect of AI Explainability. The objective of the framework is to guide smarter decisions by providing a clear, accountable, and data-driven approach to ESG cost estimation, thereby contributing to a better world through responsible project management.
The Findings are an analysis of a global facilities management company's estimating requirements, and a dynamic ESG scoring model using Recurrent Neural Networks (RNN) to evaluate the ESG performance of fleets of buildings during the use stage of the full life cycle.
The prediction results are favourable in that a chosen ESG metric has been reasonably predicted for the buildings in the data set. However, Limitations exist in the scope of the data set analysed, the subjectivity in the ESG metric developed, and the lack of synthesis with private cost data.
The Research Methodology includes a state-of-art review of 108 journal papers used to synthesise an initial conceptual framework. It includes Estimating Requirements, Data Integration, ESG Costs and Cost Drivers, AI Architecture, and the critical aspect of AI Explainability. The objective of the framework is to guide smarter decisions by providing a clear, accountable, and data-driven approach to ESG cost estimation, thereby contributing to a better world through responsible project management.
The Findings are an analysis of a global facilities management company's estimating requirements, and a dynamic ESG scoring model using Recurrent Neural Networks (RNN) to evaluate the ESG performance of fleets of buildings during the use stage of the full life cycle.
The prediction results are favourable in that a chosen ESG metric has been reasonably predicted for the buildings in the data set. However, Limitations exist in the scope of the data set analysed, the subjectivity in the ESG metric developed, and the lack of synthesis with private cost data.
| Original language | English |
|---|---|
| Publication status | Published - 22 Oct 2024 |
| Event | 2024 INFORMS Annual Meeting - the Seattle Convention Center and The Hyatt Regency Seattle, Seattle, United States Duration: 20 Oct 2024 → 23 Oct 2024 https://meetings.informs.org/wordpress/seattle2024/ |
Conference
| Conference | 2024 INFORMS Annual Meeting |
|---|---|
| Country/Territory | United States |
| City | Seattle |
| Period | 20/10/24 → 23/10/24 |
| Internet address |