An inventory management model for an inventory consisting of lumpy items, slow movers and fast movers

This paper described an algorithm developed for the implementation of a new computorized inventory management system for spare parts in a large chemical plant, located in Belgium. The spare parts inventory contains about 34,000 different types of items; 90% of the items are classified as lumpy items, as their demand frequency is less than 4 times a year. Empirical research by means of computer simulations has resulted in the proposal of the ‘Package Poisson’ as a good fitting general distribution function for the demand pattern of the spare parts. This ‘Package Poisson’ forecasts the demand occurrences by means of a Poisson distribution, taking into account that on each occurrence a package of several pieces is asked for. The quantity f of this package can be calculated from the variance and the average of the demand history data of each item. The performance of the algorithm has been measured by comparing the resulting reorder points with the reorder points calculated on the basis of other theoretical distribution functions. In all test cases the ‘Package Poisson’ algorithm gives satisfying results.