Assortment optimization with position effects under the nested logit model
Authors:
- 27 November 2019
- journal article
- research article
- Published by Wiley in Naval Research Logistics (NRL)
- Vol. 67 (1) , 21-33
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.Keywords
Funding Information
- National Natural Science Foundation of China (71622006, 71761137003)
This publication has 41 references indexed in Scilit:
- Capacitated assortment and price optimization under the multinomial logit modelOperations Research Letters, 2012
- Quantifying Social Influence in an Online Cultural MarketPLOS ONE, 2012
- Location, Location, Location: An Analysis of Profitability of Position in Online Advertising MarketsJournal of Marketing Research, 2011
- Pricing Multiple Products with the Multinomial Logit and Nested Logit Models: Concavity and ImplicationsManufacturing & Service Operations Management, 2011
- Dynamic Assortment Optimization with a Multinomial Logit Choice Model and Capacity ConstraintOperations Research, 2010
- Does In-Store Marketing Work? Effects of the Number and Position of Shelf Facings on Brand Attention and Evaluation at the Point of PurchaseJournal of Marketing, 2009
- A PTAS for capacitated sum-of-ratios optimizationOperations Research Letters, 2009
- Cumulated gain-based evaluation of IR techniquesACM Transactions on Information Systems, 2002
- Shelf management and space elasticityJournal of Retailing, 1994
- A new polynomial-time algorithm for linear programmingCombinatorica, 1984