Optimal Merchandise Testing with Limited Inventory

The practice of \merchandise testing" refers to the deployment of fashion products to stores in limited quantities so that a retailer may learn about demand prior to the main selling season. We consider the optimal allocation of inventory to stores in a merchandise test, focusing on the tradeo between the quantity of stores tested and the quality of observations, which can be impacted by demand censoring due to inventory stockouts. We nd that the visibility into the timing of each sales transaction has a pivotal impact on the optimal allocation decisions. When such timing information is unobservable, the retailer may need to consolidate inventory in few stores to increase service levels during the test and thereby to minimize the negative impacts of demand censoring. When sales timing information is observable, the retailer is better o maximizing the number of sales during the test period without regard to stockouts in individual stores. Motivated by our analysis, we propose two heuristic allocation policies for the cases with and without sales timing information, respectively. A numerical study shows the heuristics to be near optimal for instances for which we can compare them to the optimal solution. On the other hand, inecient inventory allocations