Using enhanced standard particle swarm optimization for solving multi-mode project scheduling problem

Abstract

Multi-mode project scheduling problem is a complex and confirmed to be NP-hard problem. Many researchers have devoted themselves for solving a variety of scheduling problems. Meta-heuristic is a promoting scheme. Among them, particle swarm optimization (PSO) has been well applied for solving different problems. However, PSO usually leads to premature convergence and trapped on local optimal. Hence, a modified global best experience communication with random links to make stable convergence is proposed in this study. Moreover, a correction mechanism for infeasible solution is also provided. The efficiency of proposed scheme is verified via testing the largest scale problem in benchmark problems, named multi-mode resource-constrained project scheduling problem that is a generalized project scheduling problem collected in PSPLIB. Experimental results demonstrate that the proposed approach is effective and can make stable convergence. Moreover, this approach is able to efficiently solve MRCPSP class problems.