EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3208. Dynamic pricing strategy for substitute perishable products
Invited abstract in session MA-50: Food Waste, stream Retail Operations.
Monday, 8:30-10:00Room: M2 (building: 101)
Authors (first author is the speaker)
1. | Mariana Sousa
|
Industrial Engineering and Management, Faculty of Engineering of the University of Porto | |
2. | Sara Martins
|
INESC TEC | |
3. | Maria João Martins dos Santos
|
INESC TEC | |
4. | Pedro Amorim
|
Industrial Engineering and Management, Faculty of Engineering of University of Porto |
Abstract
Dynamic pricing is a widely used strategy for adjusting the price of goods and services in response to changing and demanding consumer behavior. In grocery retailing, this approach is often applied to perishable products to encourage consumers to purchase items with limited remaining shelf life. However, existing pricing strategies often overlook the impact of sales cannibalization among similar products and lack differentiation in discounts across the product spectrum.
In this study, we develop a dynamic pricing approach for a variety of perishable items with a focus on optimizing retail profits and reducing food waste. We first test assumptions about substitution effects among similar products using historical data from a European retailer. We then develop a demand forecasting model for a subset of the products and incorporate it into a reinforcement learning algorithm. This algorithm determines the pricing strategy for the selected set of products, establishing continuous discounts throughout their shelf life. Finally, we analyze the impact of considering multiple products rather than a single product at a time to develop the pricing policy.
This research provides insights into a methodology for constructing dynamic pricing strategies and uncovers practical implications.
Keywords
- Revenue Management and Pricing
- Strategic Planning and Management
- Machine Learning
Status: accepted
Back to the list of papers