A Systematic Literature Review on Integrating Machine Learning Algorithms and Metaheuristic Algorithms in Optimizing Sustainable Digital Architectural Design

Abstract

This systematic literature review investigates how machine learning (ML) algorithms and metaheuristic methods have been integrated into sustainable digital architectural design optimization over the past two decades. The study analyzes 42 peer-reviewed publications, highlighting the critical role of these technologies in enhancing energy efficiency, structural optimization, and environmental sustainability. This review showcases the growing use of ML and metaheuristics to reshape traditional architectural practices by addressing multi-objective optimization models such as generative design and performance improvement. Furthermore, the review outlines the approaches taken in the study, focusing on algorithmic decision-making for material and energy consumption optimization. Despite the promising advancements, several limitations are outlined, including non-standardized procedures and limited applications in practice. The research calls for mixed-method procedures and empirical studies for the validation of algorithmic models in real architectural projects. Practical implications of this study include the potential for ML and metaheuristic algorithms to enhance design procedures, reduce environmental impact, and achieve higher energy conservation, which will result in more sustainable building solutions. Social implications are the broader use of such technologies with the aim to achieve sustainability goals within urban environments, make resource allocation more effective, and enhance occupant comfort. This review contributes to the literature by pointing out the need for standard methods and introducing these technologies into practice to have the maximum impact on sustainable design.

Topics & Concepts