Why do zeroes happen? A model-based approach for demand classification
Ivan Svetunkov, Anna Sroginis
TL;DR
This paper tackles the challenge of zeros in demand data by separating naturally intermittent demand from artificial stockouts and by classifying demand into six fundamental types. It introduces the Automated Identification of Demand (AID), a two-stage framework that first detects stockouts using interval-based statistics and then selects among six demand models via information criteria, enabling size and occurrence to be modeled separately when appropriate. Through simulations and a real retail case study, the authors show that incorporating stockout detection, separating demand sizes from occurrence, and refining demand categories significantly improve forecasting accuracy (e.g., RMSSE reductions) and inventory performance, compared with naive approaches that ignore stockouts or treat all zeros uniformly. The findings offer practical guidance for forecasting and inventory management, highlighting the value of stockout-aware features and model-based demand classification in reducing lost sales and improving service levels, while acknowledging limitations around alternative model selection criteria and ground-truth stockout data.
Abstract
Effective demand forecasting is critical for inventory management, production planning, and decision making across industries. Selecting the appropriate model and suitable features to efficiently capture patterns in the data is one of the main challenges in demand forecasting. In reality, this becomes even more complicated when the recorded sales have zeroes, which can happen naturally or due to some anomalies, such as stockouts and recording errors. Mistreating the zeroes can lead to the application of inappropriate forecasting methods, and thus leading to poor decision making. Furthermore, the demand itself can have different fundamental characteristics, and being able to distinguish one type from another might bring substantial benefits in terms of accuracy and thus decision making. We propose a two-stage model-based classification framework that in the first step, identifies artificially occurring zeroes, and in the second, classifies demand to one of the possible types: regular/intermittent, intermittent smooth/lumpy, fractional/count. The framework relies on statistical modelling and information criteria. We argue that different types of demand need different features, and show empirically that they tend to increase the accuracy of the forecasting methods and reduce inventory costs compared to those applied directly to the dataset without the generated features and the two-stage framework.