Inventory Control for Periodic Intermittent Demand



Intermittent demand is difficult to forecast, as many periods have no demand at all. Forecasting methods for such demand usually create separate estimates for the time between demand occurrences and the size of a demand occurrence. These methods implicitly assume that the time between demand occurrences is memoryless. Data from practice, however, indicates that the times between demand events is often not memoryless but –contrary to implicit model assumptions— displays periodicity. Consequently, the time since the last demand occurrence is an important predictor for future demand. We propose a demand model that accommodates such periodic intermittent demand. We show that the optimal inventory policy is a state-dependent base-stock policy, where the order-up-to-levels depend on the time since the last demand. We benchmark the performance of our approach against heuristic policies both in a theoretical experiment and on five real data sets in terms of average inventory costs.

Zoom link:

Meeting ID: 974 6506 0334