Optimizing construction project management via genetic algorithms – A state-of-the-art review

Abstract

Construction projects involve complex trade-offs between conflicting objectives such as minimizing duration and costs while ensuring feasibility, safety, and quality. This paper reviews research on leveraging genetic algorithms (GAs)–adaptive bioinspired search techniques–to enhance decision-making in construction management spanning planning, scheduling, and control. The versatility of GAs in tackling combinatorial optimization problems is established through various applications optimizing workflows, resource allocation, and crew coordination for enhanced project performance. Empirical analyses validate GAs generating superior trade-off solutions regarding project timeline, budgetary compliance, profitability, and safety over conventional methods such as mathematical programming or regression. Challenges remain regarding further enhancing the solution quality and fine-tuning the GA parameters. Nonetheless, by efficiently exploring vast solution spaces, GAs enable the scientific translation of construction data into actionable insights for augmenting complex project decisions. Exciting innovation opportunities exist in synergistically hybridizing GAs with promising contemporaries, such as machine learning, simulation, and robotics. Collectively, the existing literature underscores GAs' immense and relatively untapped potential of GAs to develop next-generation AI-powered tools for optimizing construction productivity, cost efficiency, sustainability, and lean automation.