Abstract
Sales promotion based on cash discounts is one of the important marketing tools for retail stores. The progress of IT has enabled easy accumulation of POS data. However, it is difficult to estimate the effects. Because the number of items handled is huge, and their relationships are so complex, even if it is a small store. In addition, it is difficult to optimize a combination of discount items and their price.
In our study, we propose a method which provides an optimal pricing strategy. From computational experiments using certain practical POS data, we show that our method has good performance.
