We just received the nice news that our paper “Demand Seasonality in Retail Inventory Management” is accepted for publication in European Journal of Operational Research. This joint work with Joachim Ehrenthal and Dorothee Honhon investigates the value of accounting for demand seasonality in inventory control (as shown in Figure 1 below).
Our problem is motivated by discussions with retailers who admitted to not taking perceived sea- sonality patterns into account in their replenishment systems. We consider a single-location, single-item periodic review lost sales inventory problem with seasonal demand in a retail environment. Customer demand has seasonality with a known season length, the lead time is shorter than the review period and orders are placed as multiples of a fixed batch size. The cost structure comprises of a fixed cost per order, a cost per batch, and a unit variable cost to model retail handling costs.
Our paper yields some important insights for retail management. By accounting for non- stationary demand in inventory management, retailers can reduce inventory holding, handling and stock-out cost substantially. Cost savings are higher for fast-moving items with high demand variability across review periods and low case pack size. Yet, in some cases, accounting for more demand seasonality can lead to an increase in costs, which suggests that the upgrade in the ordering system should be done carefully. Our numerical analysis with real life data indicates potential cost savings that are considerable for the retailing industry with its tight margins and rigorous focus on cost-efficiency (an average reduction of 4.91% in optimality gaps). We show that it is most important to incorporate across-days variations and that large cost savings can be already achieved with a simple distinction between weekday and weekend sales, provided that the weekend is defined appropriately for each item (the optimality gap goes down to an average of 1.60%). Doing so improves the return on investment in automated store ordering systems that are equipped with the capabilities of accounting for non-stationary demand.
We apply the problem to a real life setting, using Point-of-Sales data from a European retailer. We show that a simple distinction between weekday and weekend sales can lead to major cost reductions without significantly increasing the complexity of the retailer’s automatic store ordering system. Our analysis provides valuable insights on the tradeoff between the complexity of the automatic store ordering system and the benefits of incorporating demand seasonality.
Contact me if you are interested in receiving the full paper.