Enhancing safety and psychological capital in construction worker resource scheduling

To tackle the resource scheduling challenges encountered by construction workers, while taking into account safety constraints and psychological capital, we have developed a multi-objective resource scheduling model tailored for multi-skilled construction teams.This model aims to achieve three key objectives: minimizing project costs, reducing completion time, and enhancing safety performance. To address the complexities of this model, we developed an enhanced Grey Wolf Optimization（ GWO） algorithm that overcomes the limitations of the basic GWO, including its vulnerability to local optima, premature convergence, and restricted global search capability. This improved version incorporates a dual-layer encoding scheme to better reflect real-world scenarios, along with opposition-based learning to enhance the quality of the initial population. To further improve the quality and efficiency of the solution set, an adaptive hunting weight mechanism was integrated to optimize position updates. Additionally, crossover and mutation operators from genetic algorithms were applied to enhance overall algorithm performance. The effectiveness of the proposed algorithm was assessed through experiments conducted on both simulated datasets and real-world enterprise cases. The results were compared across three metrics（IGD,C-metric, and iteration speed） and three objective outcomes. For the Inverted Generational Distance（IGD） metric, the Multi-Strategy Hybrid Grey Wolf Optimization（MSHGWO） algorithm demonstrates a value significantly closer to 0, indicating superior convergence, while the IGD values of other algorithms are closer to 1. Regarding theC-metric, the MSHGWO algorithm approaches 1,whereas the C values of the other algorithms are nearer to 0, highlighting its dominance in Pareto optimality. Furthermore, in terms of output objectives, the MSHGWO algorithm achieves the lowest costs, the shortest project duration, and the highest safety performance.The experiments demonstrated that the MSHGWO algorithm outperformed traditional GWO as well as two other improved GWO variants across various evaluation metrics and objectives. This algorithm offers a more efficient and safer resource allocation strategy for construction workers.