Purchase Planning for Minimum Assortment Goods Given the Uncertainty of Demand

Abstract

Some retailers (e.g. pharmacies) are responsible for satisfying the demand for the minimum range of goods, which are generally unprofitable. With respect to such goods, the enterprise seeks to satisfy uncertain demand rather than to make profit. We assume that: (a) the vector of demand for goods of the minimum assortment in the planning period lies “between” the demand vectors of several previous periods (is a convex linear combination of these vectors); (b) the smaller the maximum unsatisfied demand (by product groups and possible vectors of demand), the greater is the reliability of meeting the demand. Under these assumptions, we address the problem of allocating a limited procurement budget among commodity groups to meet uncertain demand most reliably. The article shows that this problem is equivalent to finding an optimal strategy by Wald’s criterion in some game with nature and can be reduced to a linear programming problem. Using the problem features, we propose a fast (having quadratic complexity) algorithm for constructing an optimal procurement plan. The model can be used when planning the minimum assortment goods procurement in order to maximize the meeting demand reliability, achievable within the allocated budget. As far as we know, such a formulation of the problem has not been studied in the previous literature.