A hybrid particle swarm optimization with estimation of distribution algorithm for solving permutation flowshop scheduling problem

In this paper we propose PSO–EDA, a hybrid particle swarm optimization (PSO) with estimation of distribution algorithm (EDA) to solve permutation flowshop scheduling problem (PFSP). PFSP is an NP-complete problem, for which PSO was recently applied. The social cognition in the metaphor of canonical PSO is incomplete, since information conveyed in the non-gbest particles is lost. Also, the intelligence of the particles is totally neglected by the canonical PSO and most of other literatures. To tackle such problems, we propose to enable the sharing of information from the collective experience of the swarm by hybridizing an EDA operator with PSO and to add the primitive intelligence to each particle by using a local search mechanism. To enhance the performance of the algorithm proposed, a new local search algorithm, the minimization-of-waiting-time local search (MWL), is applied. The computational experiment on different benchmark suites in PFSP, in which two new best known solutions have been found, shows a superiority of PSO–EDA over other counterpart algorithms in terms of accuracy.