In this paper, a new hybrid metaheuristic algorithm based on the scatter search approach is developed to solve the well-known resource-constrained project scheduling problem. This algorithm combines two solutions from scatter search to build a set of precedence feasible activity lists and select some of them as children for the new population. We use the idea presented in the iN forward/backward improvement technique to dene two types of schedule, direct and reverse, and the members of the sequential populations change alternately between these two types of schedule. Extensive omputational tests were performed on standard benchmark datasets and the results are compared with the best available results. Comparative computational tests indicate that our procedure is a very eective metaheuristic algorithm.

Keywords