Multi-Objective Scheduling Problems for Re-Entrant Shops

Atefeh Moghaddam (1) , Farouk Yalaoui (1) , Lionel Amodeo (1)

1 LOSI - Laboratoire d'Optimisation des Systèmes Industriels

Production scheduling problems are multi-objective by nature which means that several criteria, in conflict with each other, have to be considered at the same time. In this chapter, we study a bi-objective re-entrant permutation flow shop scheduling problem in which the jobs have strict due dates. If the manufacturer will not be able to respect any job’s due date, that job shall be rejected whether by ignoring it completely or sub-contracting the job. In both cases, the company is charged a fee called “opportunity loss” or “out-sourcing cost”. After a brief introduction, we present the existing research works in this area, then we describe the problem in more details by giving a mixed-integer linear programming model. Since the problem is NP-hard, a genetic algorithm coupled with a new dominance concept, different from Pareto, is developed to find a good estimation of non-dominated solutions set. Finally the computational results are discussed and some directions for future research are presented.

Keywords: Schedule Problem; Completion Time; Pareto Front; Multiobjective Optimization; Total Completion Time

Domains: Operations Research [cs.RO]

Atefeh Moghaddam, Farouk Yalaoui, Lionel Amodeo. Multi-Objective Scheduling Problems for Re-Entrant Shops. Computational Intelligence Systems in Industrial Engineering, 6, pp.271-293, 2012, Atlantis Computational Intelligence Systems. ⟨10.2991/978-94-91216-77-0_13⟩. ⟨hal-02499384⟩