Are Multi-Echelon Inventory Methods Worth Implementing in Systems with Low-Demand-Rate Items?

Abstract

In practice, most multi-echelon inventory systems are managed using adaptations of single location methods. This paper shows that such methods can be dramatically inferior to methods designed to take advantage of a system's structure. This is especially true in repair parts inventory systems where most items have low demand rates. In this paper we describe a multi-echelon method adapted to a particular situation. Data taken from a large (over $100 million investment) industrial inventory system are used to provide comparative results on overall inventory required to meet a given service level. The multi-echelon method used involves an (s − 1, s) ordering policy and Poisson demand, both of which are appropriate for low demand items; this method requires much lower total investment than the single-location method requires to achieve the same average level of performance. Thus it appears that multi-echelon methods are worthwhile in many situations.