Optimal and Heuristic Solutions for the Spare Parts Inventory Control Problem

Abstract

One inventory problem that has not changed much with the advent of supply chains is the control of spare parts which still involves managing a very large number of parts that face erratic demand which occurs far in between. For most items it has not been feasible to establish a control system based on the individual item's demand history. Hence either the parts have beenbundled into one group as Wagner did and only one demand distribution has been used for all or they have been divided into different groups and group distributions have been used as in Razi. We picked two popular rules by Wagner and extensively tested their performance on data obtained from a Fortune 500 company. Comparisons were made with the optimal solution and optimal cost obtained from our procedure based on the Archibald and Silver's optimizing algorithm. For some problems with very small mean demand and small average number of demand occurrences, finding the optimal solution required an inordinate amount of CPU time, thus justifying the need to use heuristics.