Efficient Frontier and Applications in Product Offering and Pricing
52 Pages Posted: 12 Apr 2022 Last revised: 16 May 2022
Date Written: April 11, 2022
Problem definition: The joint assortment and pricing problem is notoriously challenging, especially with a large problem size and various operational constraints arising from business practices. In this paper, we consider the joint product offering and pricing under the constrained logit choice models with product-differentiated price sensitivities in both static and dynamic settings.
Methodology/results: We develop an efficient methodology that applies efficient frontier and dimensional reduction that transforms the joint optimization into a single-variable one controlling the aggregate resource consumption rate. The efficient frontier is employed to transform the non-concave objective function equivalently into its concave counterpart. We show that the optimal prices of all the products are uniquely determined by a target adjusted markup. The mixed combinatorial optimization can be simplified by searching over efficient sets of polynomial size. For the dynamic joint product selection and pricing, we fully characterize the optimal policy that adopts a simple time-threshold structure that can be pre-computed. We also show the robustness of our methodology by extending the analysis to general settings and the nested logit model.
Managerial implications: The proposed methodology contributes to the increasingly popular topic in the retail management by significantly reducing the computational complexity of joint product selection and pricing. Our results on the impacts of the product set constraints and price bounds provide concrete guidance to practitioners on products offerings and prices in various business scenarios.
Keywords: Efficient Frontier; Efficient Set and Price; Assortment Planning; Pricing; Time-Threshold Policy
Suggested Citation: Suggested Citation