The job‐shop scheduling problem (JSSP) is considered one of the most difficult NP‐hard problems. Numerous studies in the past have shown that as exact methods for the problem solution are intractable, even for small problem sizes, efficient heuristic algorithms must achieve a good balance between the well‐known themes of exploitation and exploration of the vast search space. In this paper, we propose a new hybrid parallel genetic algorithm with specialized crossover and mutation operators utilizing path‐relinking concepts from combinatorial optimization approaches and tabu search in particular. The new scheme relies also on the recently introduced concepts of solution backbones for the JSSP in order to intensify the search in promising regions. We compare the resulting algorithm with a number of state‐of‐the‐art approaches for the JSSP on a number of well‐known test‐beds; the results indicate that our proposed genetic algorithm compares fairly well with some of the best‐performing genetic algorithms for the problem.

No abstract

No abstract

No abstract

In today's rapidly evolving management world, the scheduling of multiple projects where each one's execution depends on another's successful completion, is of great importance. This paper presents a hybrid metaheuristic algorithm composed of an external Genetic Algorithm (GA) and an encapsulated Ant Colony Optimization (ACO) algorithm for the flexible resource constrained multi-project scheduling problem (MPFRCPSP). The proposed idea is grounded on the concept of prioritizing the sub-projects' scheduling based on: a) the number of external (to other sub-projects) relations and b) the resource requirements as compared to the resource shortage for each resource type and each subproject. The implementationuses the Genetic Algorithm to deal with the classification and prioritization of the projects to be scheduled and the inner ACO algorithm, to perform the activity list optimization for each project. The proposed method was validated using a consistent number of PSP Lib[1] data sets.