摘要

This paper considers the multi-skill resource-constrained project scheduling problem in which the processing time of each task is modeled as a step-function of its starting time and a specific deteriorating date. For solving the complex problem, four neighborhood structures and two mutation operators based on problem characteristics are proposed to form an improved tabu search algorithm (ITS). Computational experiments are performed on a set of test instances from the recent literature. The proposed algorithm improved the currently best-known results for most instances in a reasonable time. Furthermore, the dataset is revised to integrate the step-deteriorating effect. Numerical results demonstrate the proposed algorithm is a powerful solution methodology for solving the studied problem in terms of solutions quality and show that the mutation phase in ITS has a direct benefit over the solving ability of the ITS.