Enhancing construction site layout planning using hybrid Ant Colony and simulated annealing algorithms

Abstract

Effective construction site layout planning (CSLP) is critical for enhancing logistics efficiency, safety, and space utilization in complex construction environments. In practice, planners must evaluate multiple layout alternatives under spatial, operational, and safety constraints within limited decision time. However, traditional optimization approaches often struggle with the nonlinear and constraint-rich nature of real construction sites, either requiring excessive computational time or compromising solution quality. To address these challenges, this study proposes a hybrid metaheuristic framework that integrates Ant Colony Optimization (ACO) with Simulated Annealing (SA) to support CSLP decision-making. The approach leverages the global search capability of ACO and the local refinement strength of SA to generate high-quality layout solutions with improved computational efficiency. The hybrid mechanism employs adaptive pheromone updating alongside local search enhancement to balance exploration and exploitation effectively. The framework is validated through application cases, including a benchmark problem and a large-scale multi-building construction site. Results demonstrate that the proposed method achieves up to a 95% reduction in computational time compared with standalone approaches, while maintaining or improving solution quality. Beyond algorithmic performance, the study offers managerial value by enabling structured comparison of layout alternatives and supporting data-driven decision-making.