Authors:
Abdelhamid Boujarif
1
;
David Coit
2
;
Oualid Jouini
1
;
Zhiguo Zeng
1
and
Robert Heidsieck
3
Affiliations:
1
Industrial Engineering Laboratory (LGI) CentraleSupélec, Paris-Saclay University, Gif-sur-Yvette, France
;
2
Department of Industrial and Systems Engineering, Rutgers University, U.S.A.
;
3
General Electric Healthcare, 283 Rue de la Minière, 78530 Buc, France
Keyword(s):
Multi-Component Systems, Opportunistic Maintenance, Reliability, Economic Dependence, Structure Dependence, Genetic Algorithm.
Abstract:
This paper presents an opportunistic maintenance model for a multi-component system. We develop a model that considers the ages and residual values of the non-failed components and component failure time distributions. We also consider the structural and economical dependency between the items by favoring grouped over individual replacement to reduce operation costs. We use a genetic algorithm to derive the optimal opportunistic maintenance plan by minimizing the long-run operational cost considering both maintenance cost and potential penalty costs due to failure in the future. The model recommends additional preventive operations in cases where the reliability does not satisfy the quality condition and to reduce the long-run operational cost. A sensitivity analysis shows that the optimal decision is mainly affected by the logistic cost, the interest rate, and the planning horizon. The model’s performance has been evaluated using several real case problems, demonstrating that the pr
oposed method is very efficient.
(More)