Impact of Explainable Artificial Intelligence for Sustainable Built Environment

Innovative and complex applications driven by Artificial Intelligence (AI) embedded systems have enhanced the efficiency, accuracy and sustainability of the built environment. Such solutions often face barriers to adoption due to the lack of trust, transparency, understanding, and inherent complexity. Explainable Artificial Intelligence (XAI) encompasses techniques and methodologies to make the decision-making processes of AI models transparent, interpretable, and comprehensible. The objectives of this study are to evaluate the potential of incorporating XAI techniques to provide solutions for the challenges and to enhance interpretability. The impact of XAI on promoting sustainable practices in the built environment is systematically analysed. Further, it aims to identify key application areas, analyse XAI methods, evaluate impacts on decision-making, and explore challenges in implementing XAI for built environment sustainability. The methodology of the study has incorporated a systematic literature search, thematic analysis, and synthesis of findings across multiple domains. The analysis reveals that XAI is increasingly incorporated into energy efficiency, urban planning, construction management, and sustainable design, enhancing model transparency and stakeholder trust. Key XAI techniques, including SHapley Additive exPlanations, Gradient-Weighted Class Activation Mapping, Local Interpretable Model-agnostic Explanations and causal discovery methods, are applied to the AI-driven solutions considered through the analysis of the study. The findings of the study indicate that XAI can significantly contribute to achieving sustainability goals in the built environment by fostering trust, facilitating knowledge transfer, and enabling more informed decision-making. The paper outlines future research directions and practical recommendations for implementing XAI in sustainable built environment practices.