VOLUME 70 : 2022

The present paper is concerned with the problem of controlling repairable inventories of critical items, so that these items are available when they are needed. The context used is that of the rail transport, or train system. We present a nonlinear integer programming formulation for a single-level repairable inventory system. The formulation strives to minimize the sum of the expected annual costs of carrying inventories and of operational delay due to item unavailability on demand. We apply a constraint limiting the allowable probability of item availability to a pre-set minimum value. Using hypothetical data, we solve the model using a spreadsheet-based solver developed for this purpose. The formulation, taken for each item, is convex. Our model promises to be a useful tool for repairable-item inventory control in the railway industry, or in the transportation industry in general.