Joint Strategy of Dynamic Ordering and Pricing for Competing Perishables with Q-Learning Algorithm

Jiangbo Zheng,Yanhong Gan,Jiatai Chang,Ying Liang,Qingqing Jiang,Wenqing Wu

doi:10.1155/2021/6643195

Abstract

We use Machine Learning (ML) to study firms’ joint pricing and ordering decisions for perishables in a dynamic loop. The research assumption is as follows: at the beginning of each period, the retailer prices both the new and old products and determines how many new products to order, while at the end of each period, the retailer decides how much remaining inventory should be carried over to the next period. The objective is to determine a joint pricing, ordering, and disposal strategy to maximize the total expected discounted profit. We establish a decision model based on Markov processes and use the Q-learning algorithm to obtain a near-optimal policy. From numerical analysis, we find that (i) the optimal number of old products carried over to the next period depends on the upper quantitative bound for old inventory; (ii) the optimal prices for new products are positively related to potential demand but negatively related to the decay rate, while the optimal prices for old products have a positive relationship with both; and (iii) ordering decisions are unrelated to the quantity of old products. When the decay rate is low or the variable ordering cost is high, the optimal orders exhibit a trapezoidal decline as the quantity of new products increases.

Highlights

Due to the scarcity of resources and the advance of technology, both academia and practice have highlighted the significance of value deterioration, focusing on perishables as a central issue
Our contribution to bridge the existing research gap involves the following: (i) we incorporate the number of old products carried to the period into the joint strategy to better cope with consumer preferences and dynamic demand substitution, with the purpose of maximizing the retailers’ profits when considering fixed order cost and inventory holding cost, which is not included in Sainathan [18]; (ii) we develop the Q-learning algorithm rather than dynamic programming or value iteration to solve the Markov model and gain the multiperiod optimal strategy
Given that few works researched on pricing and inventory optimization for perishables considering multiperiod joint strategies and consumer choice behaviors, this paper conducts a simulation study centered on optimal joint strategy especially when different ages’ products are sold simultaneously

Summary

Introduction

Due to the scarcity of resources and the advance of technology, both academia and practice have highlighted the significance of value deterioration, focusing on perishables as a central issue. When Apple introduces new mobile phones, it will continue to sell old phones at low prices; when the car launches a new model, the old model will continue to sell at a reduced price How should these retailers dynamically develop joint ordering and pricing strategies when considering the purchase. This paper will verify the optimal dynamic data of old products carried into period as well as the optimal dynamic ordering and pricing strategy in a multiperiod when different ages’ products sale simultaneously. We introduce a consumer utility function to develop a demand model that considers customer purchase behaviors and build our Markov decision model to obtain an optimal strategy, including the decision actions selected for each state that can maximize the total expected discounted profit over an infinite horizon.

Literature Review

Solution Approach

Numerical Studies and Observations

Conclusion and Future Research

Findings

Conflicts of Interest