A sustainable multiobjective multi-site resource-constrained project scheduling problem

Project scheduling stands as a critical component in managing tasks within projects to optimize resource utilization and ensure efficient project execution. While traditional approaches to the resource-constrained project scheduling problem have primarily focused on single-site contexts, modern project environments often span multiple sites, necessitating the development of models that can accommodate this complexity. On the other hands, the growing importance of social sustainability in project planning underscores the need to address issues such as equitable workload distribution across project sites. This paper introduces a novel sustainable multiobjective multi-site resource-constrained project scheduling problem that integrates multi-site scheduling considerations with the principles of social sustainability. The problem formulation aims to minimize both project makespan and the deviation between maximum and minimum working loads across project sites. Additionally, three types of renewable, non-renewable, and doubly-constrained resources are simultaneously addressed in the problem for the first time to improve its realism. To solve this problem, an interactive fuzzy TH approach is proposed as an exact approach to handle the conflict between the objectives. In the second step, three well-known multiobjective metaheuristic algorithms, including NSGA-II, MOPSO, and SPEA-II, are designed and implemented to address the complexity of the problem in large-sized projects. An efficient repair algorithm is designed to ensure the feasibility of solutions and handle the problems' constraints in the search procedure. Taguchi's design of experiments is proposed to calibrate the input parameters of algorithms and improve their efficiency. Numerical examples from the literature are utilized to illustrate the efficacy of the proposed model and solution methodology. In this direction, the one-way analysis of variance and Kruskal- Wallis parametric and non-parametric statistical tests are also used to provide amore systematic comparison of results. In general, the results reveal the better performance of NSGA-II in terms of the quality of solutions, while SPEA-II outperforms in terms of CPU time. Finally, insights for project managers are discussed and the paper is concluded by suggesting some directions for future research.