Dynamic Pricing Model for E-Commerce Products Based on DDQN
DOI:
https://doi.org/10.47852/bonviewJCBAR42022770Keywords:
e-commerce products, dynamic pricing, deep reinforcement learning, DQN algorithm, DDQN algorithmAbstract
This paper addresses the challenges of intense price competition, price elasticity, and significant demand fluctuations in e-commerce product markets by adopting a dynamic pricing approach. Focusing on a product from the JD.com e-commerce platform, historical data spanning the past three years are analyzed, considering factors such as shipping costs, product inventory, product costs, and the impact of holidays. The study employs the Double Deep Q Network (DDQN) for dynamic pricing optimization of the product and compares its performance with the DQN model. The results indicate that both the DQN algorithm and DDQN algorithm lead to varying degrees of profit improvement for dynamic pricing of products. Specifically, the pricing profit with the DQN algorithm increased by an average of 1.925% compared to the original pricing profit, while the pricing profit with the DDQN algorithm increased by an average of 11.975% compared to the original pricing profit. These findings demonstrate practical significance.
Received: 6 March 2024 | Revised: 1 April 2024 | Accepted: 28 April 2024
Conflicts of Interest
The authors declare that they have no conflicts of interest to this work.
Data Availability Statement
The data that support this work are available upon reasonable request to the corresponding author.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
Funding data
-
Social Science Foundation of Shaanxi Province
Grant numbers No. 2020R043