MSPT2 Heuristic and Dynamic Programming Method for the Parallel Machine Scheduling Problem with scheduled Preventive Maintenance

Racem Mellouli (1) , Cherif Sadfi (1) , Chengbin Chu (1) , Imed Kacem (1)

1 LOSI - Laboratoire d'Optimisation des Systèmes Industriels

In this paper, we consider the parallel-machine scheduling problem with scheduled maintenance periods to minimize the total (non-weighted and weighted) completion time. For the case of single maintenance period on each machine, we provide an adapted definition of the "SPT" algorithm and we propose an MSPT2 heuristic. For the general case, we present a dynamic programming model to solve optimally the problem. In addition, improved dominance properties are proposed. Experimental simulations are done to evaluate the method performances.

Keywords: Dynamic scheduling; Dynamic programming; Parallel machines; Preventive maintenance; Job shop scheduling; Scheduling algorithm; Availability; Performance evaluation; Machinery production industries; Electric breakdown

Domains: Operations Research [cs.RO]

Racem Mellouli, Cherif Sadfi, Chengbin Chu, Imed Kacem. MSPT2 Heuristic and Dynamic Programming Method for the Parallel Machine Scheduling Problem with scheduled Preventive Maintenance. 2006 International Conference on Service Systems and Service Management, Oct 2006, Troyes, France. pp.1264-1270, ⟨10.1109/ICSSSM.2006.320690⟩. ⟨hal-02498253⟩