Parameterizable algorithm for price optimization on brick and mortar retail products using Multi-Armed Bandit Algorithm (MAB) and Robust Quadratic Programming (RQP)

• Grant number: 19/22822-9
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: July 01, 2020
• End date: June 30, 2021
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Applied Probability and Statistics
• Principal Investigator: Marcus Vinicius Veiga Roggero
• Grantee: Marcus Vinicius Veiga Roggero
• Company: Infopreços S/A
• CNAE: Desenvolvimento e licenciamento de programas de computador customizáveis
• City: São Paulo
• Associated researchers: Daniel Soares Lopes ; Rodrigo Eiji Yamagata Diana
• Associated research grant(s): 22/02635-2 - Parameterizable algorithm for price optimization on brick and mortar retail products using multi-armed bandit algorithm (MAB) and Robust Quadratic Programming (RQP), AP.PIPE
• Associated scholarship(s): 20/12749-0 - Parameterizable algorithm for price optimization on brick and mortar retail products using multi-armed bandit algorithm (MAB) and Robust Quadratic Programming (RQP), BP.TT; 20/13242-6 - Parameterizable algorithm for price optimization on brick and mortar retail products using multi-armed bandit algorithm (MAB) and Robust Quadratic Programming (RQP), BP.TT; 20/12264-6 - Parameterizable algorithm for price optimization on brick and mortar retail products using Multi-Armed Bandit algorithm (MAB) and Robust Quadratic Programming (RQP), BP.TT

Abstract

Pricing brick and mortar retail products is a challenge for retailers. The sheer amount of products and variables involved creates a difficulty in achieving optimal pricing, as stores have between 3,000 to 50,000 different products and the variables involved in calculating demand range from the selling price itself (which is the most representative) to day of the week, seasonality, weather, competition with other retailers, cannibalization and product affinity, among others. The objective of this research is the development of an optimized product pricing algorithm that increases margin or sales of a brick and mortar retailer in a parameterizable way (beta coefficient = weight between margin and sales). AMR Research studies show that the average potential for gross sales increases by 1 to 3% and gross margin by 2 to 5% for regular pricing optimization (excluding promotional and markdown prices). (AU)