Minimizing the sum of completion times on a single machine with health index and flexible maintenance operations

This paper is motivated by the development of Industry 4.0 and the need to better integrate production and maintenance decisions. Our problem considers a single machine on which jobs of different families are scheduled to minimize the sum of completion times. The machine has a health index which decreases when jobs are processed. To restore the machine health, maintenance operations must be scheduled. Moreover, to be scheduled, each job requires the machine to have a minimum health index which depends on the job family. Two cases are studied: (1) The daily case with a single flexible maintenance operation, and (2) The weekly case with two flexible maintenance operations. The second case is shown to be NP-complete. Two Mixed Integer Linear Programming models are presented for each case. The first model uses classicalpositional variables, while the second model improves the first model by using the notion of master sequence. Different valid inequalities are also proposed. Computational experiments show that the second model is much more efficient than the first model when solved with a standard solver, and the impact of the valid inequalities is discussed.

Automated summary

This paper discusses the integration of production and maintenance decisions in the context of Industry 4.0. It proposes two cases of scheduling jobs on a single machine to minimize completion times, taking into account the machine's health index. Two Mixed Integer Linear Programming models are presented, and the second model outperforms the first in terms of efficiency. Computational experiments also consider the impact of valid inequalities.