Defining line replaceable units

• Published in: European journal of operational research Vol. 247; no. 1; pp. 310 - 320
• Main Authors: Parada Puig, J.E., Basten, R.J.I.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 16.11.2015; Elsevier Sequoia S.A
• Subjects: Capital assets; Cost reduction; Defective products; Integer programming; Line replaceable unit definition; Linear programming; Maintenance; Replacement; Replacement costs; Studies; Integer programming; Line replaceable unit definition; Replacement; Maintenance
• ISSN: 0377-2217; 1872-6860
• DOI: 10.1016/j.ejor.2015.05.065

•We discuss the line replaceable unit (LRU) definition decision problem.•Replacement levels are selected within the indenture structure of capital assets.•We frame the problem in maintenance literature and show how it is treated in practice.•We propose an MILP model to optimize the LRU definition decision problem.•Numerical experiments show large cost reductions compared to heuristics from practice. Defective capital assets may be quickly restored to their operational condition by replacing the item that has failed. The item that is replaced is called the Line Replaceable Unit (LRU), and the so-called LRU definition problem is the problem of deciding on which item to replace upon each type of failure: when a replacement action is required in the field, service engineers can either replace the failed item itself or replace a parent assembly that holds the failed item. One option may be fast but expensive, while the other may take longer but against lower cost. We consider a maintenance organization that services a fleet of assets, so that unavailability due to maintenance downtime may be compensated by acquiring additional standby assets. The objective of the LRU-definition problem is to minimize the total cost of item replacement and the investment in additional assets, given a constraint on the availability of the fleet of assets. We link this problem to the literature. We also present two cases to show how the problem is treated in practice. We next model the problem as a mixed integer linear programming formulation, and we use a numerical experiment to illustrate the model, and the potential cost reductions that using such a model may lead to.