Operations, Information and Decisions Papers

Document Type

Working Paper

Date of this Version

6-2007

Publication Source

Management Science

Volume

53

Issue

6

Start Page

934

Last Page

951

DOI

10.1287/mnsc.1060.0661

Abstract

This paper studies the assortment planning problem with multiple merchandise categories and basket shopping consumers (i.e., consumers who desire to purchase from multiple categories). We present a duopoly model in which retailers choose prices and variety level in each category and consumers make their store choice between retail stores and a no-purchase alternative based on their utilities from each category. The common practice of category management (CM) is an example of a decentralized regime for controlling assortment because each category manager is responsible for maximizing his or her assigned category’s profit. Alternatively, a retailer can make category decisions across the store with a centralized regime. We show that CM never finds the optimal solution and provides both less variety and higher prices than optimal. In a numerical study, we demonstrate that profit loss due to CM can be significant. Finally, we propose a decentralized regime that uses basket profits, a new metric, rather than accounting profits. Basket profits are easily evaluated using point-of-sale data, and the proposed method produces near-optimal solutions.

Keywords

game theory, assortment planning, optimization

Share

COinS
 

Date Posted: 27 November 2017

This document has been peer reviewed.