Operations Research · 2022 · 24 citations · 35 references
BusinessBusiness AnalyticsRetail ManagementDemand ForecastingInventory ManagementPredictive AnalyticsAggregate Demand ForecastingDemand AnalysisDemand PlanningInventory ForecastingRetail ForecastingRetail Inventory ManagementData-driven Demand ForecastingDemand Forecasting Under Data ScarcityDemand PredictionFlexibilityForecasting With Sparse DataMulti-item ForecastingPredictive Demand AnalyticsRetail Demand Analytics
High‑accuracy demand prediction helps retailers manage inventory and avoid stock‑outs or excess supply, yet many items have limited historical data. The authors introduce data aggregation with clustering (DAC), a practical method that balances data aggregation and model flexibility for demand forecasting. DAC predicts demand by optimally selecting which features to estimate at the item, cluster, and aggregate levels. Theoretically, DAC provides consistent estimates and lower prediction errors than a per‑item model, and empirically it achieves higher accuracy than many common benchmarks on a large online retailer’s data.
High accuracy in demand prediction allows retailers to effectively manage their inventory and mitigate stock-outs and excess supply. A typical retail setting involves predicting the demand for hundreds of items simultaneously, some with abundant historical data and others with scarce data. In “Data Aggregation and Demand Prediction,” Cohen, Zhang, and Jiao propose a novel practical method, called data aggregation with clustering (DAC), which balances the tradeoff between data aggregation and model flexibility. DAC empowers retailers to predict demand while optimally identifying the features that should be estimated at the item, cluster, and aggregate levels. Theoretically, DAC yields a consistent estimate, along with improved prediction errors relative to the benchmark that estimates a different model for each item. Practically, DAC yields a higher demand prediction accuracy relative to many common benchmarks using a real data set from a large online retailer.
35