Improved Genetic Local Search Heuristic for Solving Non-permutation Flow Shop Scheduling Problem

In this paper we address the two-machine flow shop scheduling problem for minimizing the total completion times under release dates. We used two phases versions of genetic local search algorithms. The first phase is used to find a good permutation schedule. The second phase explores non-permutation schedules to improve the initial solution. In our experiment we compare the quality of permutation and non-permutation schedules. As result, we prove that the improved genetic local search method is able to improve the sum of the completion times solution comparing with the state-of-the-art permutation solution.