Subcontractor selection for a construction project: Optimal plan

Since general contractors outsource a significant number of work packages to several subcontractors, subcontractor performance is a primary indicator of project outcomes. This study proposes a multiobjective optimization model for setting up optimal subcontracting based on trade-offs between critical project objectives (i.e., cost, duration, quality, safety, and environmental impact). The proposed model aligns the subcontractor selection process with the critical objectives of a project and improves the general contractor's opportunities to achieve these objectives. Unlike existing approaches that involve selecting subcontractors' per-work-packages often after construction has started, the proposed model optimizes the entire subcontracting plan at the project level, allowing for more strategic and more holistic decision making before construction begins. The proposed model combines (1) Grey Relational Analysis (GRA) to evaluate subcontractor candidates for each work package, (2) Nondominated Sorting Genetic Algorithm-II (NSGA-II) to find the best subcontracting plans for a given project, and (3) Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to find the best compromise optimal subcontracting plan that ensures the best project outcomes. The model is validated using a real-world case study of an international airport project, demonstrating its effectiveness in improving project outcomes across multiple dimensions. The contribution of the proposed model to the state-of-the-art in subcontracting practice is its focus on project objectives rather than only cost-based considerations in selecting subcontractors, its novel methodology that involves many objectives, its use of only limited and easily obtainable information to run, its compatibility with any project delivery system, its use before construction starts, and its use of advanced analytical tools to run the model.