A multistage algorithm for the job shop scheduling problem

This paper presents a multistage algorithm for the job shop scheduling problem (JSP). A theorem is proved that the calculability of a solution is the sufficient and necessary condition for its feasibility. Therefore, an infeasible solution will be discarded immediately and the improved critical path algorithm is applied to perform the local search for a feasible solution. The computational results of the standard benchmark problems achieved by the new multistage algorithm demonstrate that the algorithm obtains high-quality solutions within reasonable time.