A model and algorithm for reactive multi-objective multi-skilled project scheduling under resource disruptions

The reactive multi-objective multi-skilled project scheduling under resource disruptions is more complex than the general scheduling problem. This complexity stems from the incorporation of overtime and proactive preemption as response strategies, as well as the consideration of the impact of a resource's skill level on setup time. We propose a reactive multi-objective scheduling model to address this problem. The objective is to generate a new schedule that minimizes deviations in activity start times and resource allocations from the baseline schedule, while also minimizing resource usage costs. To enhance the efficiency of commercial solvers in solving this model, some of the formulas are linearized. Furthermore, we propose a hybrid multi-objective evolutionary algorithm (HMOEA). To boost the performance of the HMOEA, several improvement strategies are introduced, including an adaptive procedure for fitness assignment and density estimation, a hybrid evolutionary strategy, and a local search strategy. Numerical experiments demonstrate the effectiveness of both the linearization measures and the improvement strategies. Five performance metrics and convergence comparisons are employed to assess the solution quality of the proposed algorithm in terms of convergence, diversity, and distribution. Computational results demonstrate that HMOEA leads to significant improvements in all metrics compared to five state-of-the-art algorithms.