Automated Decision Intelligence Framework for Infrastructure Project Scheduling via Seven-Dimensional Many-Objective Evolutionary Optimization and OPA-TOPSIS Ranking

Abstract The management of complex infrastructure projects necessitates the simultaneous optimization of multiple competing objectives beyond the traditional time, cost, and quality triangle. This study proposes a seven-dimensional multiobjective scheduling framework integrating duration, cost, quality, resource leveling, environmental impact, social cost, and safety. A primary contribution to the body of knowledge is a dynamic Safety Risk Index employing an exponential penalty function to quantify hazards from workspace congestion. Ten metaheuristic algorithms were evaluated across two case studies. Pareto-optimal fronts were ranked using the Ordinal Priority Approach (OPA) and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). Results demonstrate holistic benefits, reducing durations by 39% compared to baselines, while simultaneously improving resource leveling by 29.70%, environmental impact by 19.02%, social costs by 14.95%, and safety risk by 20.02%. The Nondominated Sorting Genetic Algorithm III and Adaptive Geometry Estimation Multiobjective Evolutionary Algorithm proved most robust. This research provides a data-driven decision support tool balancing execution efficiency, environmental sustainability, and workforce safety.