An AI-Enabled Framework for Enhancing Sustainable Decision-Making in Engineering Project Management

Abstract

Decision-making in sustainable engineering project management presents a complex, multi-objective challenge. Traditional methods are often reactive and struggle to balance conflicting economic, environmental, and social goals. Although Artificial Intelligence (AI) offers strong predictive capabilities, many existing decision-support approaches remain fragmented or operate as opaque “black-box” systems, limiting trust in high-stakes decisions. To address this gap, this paper develops a novel conceptual framework through a clarified, PRISMAinformed synthesis of 46 retained studies spanning five domains: AI in project management, sustainability and lifecycle assessment, multi-criteria decision-making, explainable AI, and enabling digital technologies. The framework’s key contribution is a multi-layered architecture that integrates predictive machine learning for forecasting lifecycle impacts, Multi-Criteria Decision-Making (MCDM) for trade-off analysis, and Explainable AI (XAI) to provide human-interpretable justifications for recommendations. As a result, the proposed architecture shifts decision-making from reactive assessment to proactive, transparent, and data-driven optimisation. As a conceptual contribution, the framework provides both a defensible reference model for practice and a platform for future prototype development and empirical validation.