Assortment Optimization with Position Effects under the Nested Logit Model

Rui Chen,Hai Jiang

DOI: https://doi.org/10.1002/nav.21879

2020-01-01

Abstract:We study the assortment optimization problem with position effects under the nested logit model, whose goal is to find the revenue-maximizing subset of products as well as their corresponding display positions. In this joint assortment-position optimization problem, the choices of products are affected by not only their qualities and prices but also the positions where they are displayed. Despite determining the assortment and their corresponding display positions sequentially, we propose to solve this problem in an integrated way to obtain the optimal solution. We formulate this problem as a nonlinear binary integer programming model and develop a dynamic programming based solution approach to obtain the optimal assortment-position assignments. We carry out extensive numerical experiments to evaluate the benefit of our integrated approach. The most important insight we discover is that it is not necessarily better to put the most attractive products in the best position. Moreover, we show that compared to the sequential approaches, our approach can improve revenue by 10.38% on average, which suggests that firms should take into consideration position effects when making assortment decisions. Finally, we discuss results related to two extensions of this problem, that is, the special case when positions are preassigned to nests, and the joint assortment-position-price optimization problem.