Coupling simulation-based and machine learning methodologies for energy optimization and environmental impact mitigation in buildings
G. Mihalakakou, Athanasios Giannadakis, Sonia Malefaki, Manolis Souliotis, Pantazis Georgiou, Alexandros Romeos, Anastasia Antzoulatou, Pantelis G. Nikolakopoulos, John A. Paravantis
Abstract
Buildings are responsible for a significant share of global energy consumption and carbon emissions, making energy optimization in the built environment a priority for sustainable development. This review investigates how simulation-based optimization techniques and machine learning approaches can be used to improve energy performance, reduce environmental impact, and support climate-resilient building design. The primary aim of this study is to critically evaluate and synthesize existing methods that optimize energy use in buildings while balancing economic, environmental, and occupant comfort objectives. The review systematically analyzes a wide range of optimization strategies, including deterministic mathematical programming, metaheuristic algorithms, and hybrid techniques that combine simulation models with machine learning-based surrogate modeling. Methodologies such as genetic algorithms, particle swarm optimization, support vector machines, and artificial neural networks are discussed in terms of their effectiveness, scalability, and adaptability. The findings indicate that simulation-based optimization, particularly when coupled with data-driven models, provides a powerful framework for improving design decisions and operational strategies across diverse climatic contexts and building typologies. Case studies demonstrate meaningful reductions in energy consumption, life cycle costs, and emissions, confirming the practical value of these methods. The review also highlights current challenges such as data quality, computational complexity, and the gap between academic models and real-world implementation. By bridging analytical modeling and intelligent algorithms, this work offers a comprehensive and practical reference for researchers, designers, and policymakers. The novelty of this research lies in its integrated perspective, which combines optimization theory, artificial intelligence, and sustainability objectives to guide future developments in energy-efficient building practices.