Two efficient nature inspired meta-heuristics solving blocking hybrid flow shop manufacturing problem

The hybrid flow shop scheduling problem is one of the most relevant optimization problem in manufacturing industry. In this paper, we investigate the blocking hybrid flow shop scheduling problem under the constraint of sequence dependent setup time. The objective is to minimize the total tardiness and earliness with uniform parallel machines under the constraint of sequence dependent setup time. To solve this kind of problems, significant developments of new meta-heuristic algorithms make it possible to implement new metaheuristics inspired by the behavior of living beings or natural phenomena. In this context, we suggest six algorithms based on the migratory bird optimization and the water wave optimization algorithms. We give three new versions for each meta-heuristic in order to solve this optimization problem. The main improvement of the suggested algorithms concerns the exploration phase of the neighborhood system. The enhancement approaches are based on the iterated greedy algorithm, the greedy randomized adaptive search procedure, the path relinking technique and the local search procedures. These modifications in the two nature inspired meta-heuristics make it possible to develop a new neighborhood generation structure constituting hybrid optimization algorithms. A comparative study between the different proposed methods is carried out on a variety of problems ranging from small to relatively large size instances. The simulations show good performances recorded by the water wave optimization algorithm in term of quality and convergence speed towards the best solution.