Lot-sizing under uncertainty of demand and lead time in a rolling schedule environment

Abstract

This paper examines the influence of uncertainty in demand and lead time on the performance of ten well-known lot-sizing procedures. Demand uncertainty is present in the form of forecast errors. Lead time is assumed to follow a probability mass function with its value bounded. In order to avoid any contamination of the results due to different service levels from confounding cost comparisons, safety stock is introduced which is enough such that no shortages occur during the experiments. The variation of forecast demand and inventory related cost parameters are also considered as experimental factors. The results of this study indicate that uncertainty in demand and lead time not only changes the performance of the lot-sizing rules, but also the character of this performance compared to those under uncertainty in demand only.