ONLINE LEARNING FOR REVENUE MANAGEMENT PROBLEMS WITH ADD-ON DISCOUNT

Abstract: In E-commerce industry, many sellers use product recommendation to enhance their profit. One of the recommendation methods is that they offer discounts on the supportive product (e.g. games) if the customer purchase the core products (e.g. game consoles). So called add-on discount. This paper mainly discusses the the revenue management problem related to the product recommendation policy with add-on discount. We formulate this problem as an optimization problem to determine the prices of different products and the selection of products with add-on discounts. We propose an upper confidence bound (UCB) learning algorithm to solve the optimization problem efficiently. In order to overcome the curse of dimensionality problem we come across when we use the learning algorithm, we propose the fully polynomial time approximation scheme (FPTAS) algorithm as a subroutine. We can show that the learning algorithm can converge to the optimal algorithm that has access to the true demand functions. And we prove that the convergence rate is tight up to a certain logarithmic term.

Advisor information: Huanan Zhang Assistant Professor of Industrial and Manufacturing Engineering

Files

Metadata

Work Title ONLINE LEARNING FOR REVENUE MANAGEMENT PROBLEMS WITH ADD-ON DISCOUNT
Access
Penn State
Creators
  1. Yajie Wu
Keyword
  1. revenue management
  2. add-on discount
  3. online learning algorithm
License All rights reserved
Work Type Research Paper
Publication Date December 4, 2020
Deposited December 04, 2020

Versions

Analytics

Collections

This resource is currently not in any collection.

Work History

Version 1
published

  • Created
  • Added Creator Yajie Wu
  • Updated Keyword, Description, Publication Date Show Changes
    Keyword
    • revenue management, add-on discount, online learning algorithm
    Description
    • Abstract:
    • In E-commerce industry, many sellers use product recommendation to enhance their profit. One of the recommendation methods is that they offer discounts on the supportive product (e.g. games) if the customer purchase the core products (e.g. game consoles). So called add-on discount. This paper mainly discusses the the revenue management problem related to the product recommendation policy with add-on discount.
    • We formulate this problem as an optimization problem to determine the prices of different products and the selection of products with add-on discounts. We propose an upper confidence bound (UCB) learning algorithm to solve the optimization problem efficiently. In order to overcome the curse of dimensionality problem we come across when we use the learning algorithm, we propose the fully polynomial time approximation scheme (FPTAS) algorithm as a subroutine. We can show that the learning algorithm can converge to the optimal algorithm that has access to the true demand functions. And we prove that the convergence rate is tight up to a certain logarithmic term.
    • Advisor information:
    • Huanan Zhang
    • Assistant Professor of Industrial and Manufacturing Engineering
    Publication Date
    • 2020-12-04
  • Added Yajie Wu_Master Paper.pdf
  • Updated License Show Changes
    License
    • http://www.europeana.eu/portal/rights/rr-r.html
  • Published