Discounted Sales of Expiring Perishables: Challenges for Demand Forecasting in Grocery Retail Practice

TL;DR

The paper tackles the problem of forecasting demand for expiring perishables when price discounts are used to reduce spoilage. It implements a two-step regression on a large real-world retailer dataset to separate regular demand from discount-driven uplift: Step 1 estimates baseline demand using forecast, stock, and weekday effects; Step 2 regresses the residuals on the number of discounted sales to quantify uplift. Results show that baseline forecasts systematically underestimate demand during discounted periods, and the uplift grows with the number of discounted sales, with substantial SKU-level variation. These findings suggest that a fixed-share discount rule is inadequate and point to richer modeling approaches and additional data to improve forecast accuracy and minimize spoilage.

Abstract

Grocery retailers frequently apply price discounts to stimulate demand for expiring perishables. However, integrating these discounted sales into future demand forecasts presents a significant challenge. This study investigates the effectiveness of incorporating a fixed share of these sales as \textit{regular} demand into the forecast, as commonly applied in practice. We employ a two-step regression approach on data from a major European grocery retailer, covering over 1,700 products across 676 stores. We reveal that forecasts underestimate actual demand for most SKUs when discounted sales occur. This residual uplift effect is significantly influenced by the number of sales at reduced prices. Our findings underscore the necessity for more precise approaches to integrate discounted sales into demand forecasts, thereby preventing excess inventory and the associated economic and environmental impacts of spoilage in the grocery sector.

Figures (2)

• Figure 1: Illustration of the problem description.
• Figure 2: Illustration of average residuals and estimated coefficients $\hat{\gamma}_{10}$.