A New Metric for Lumpy and Intermittent Demand Forecasts: Stock-keeping-oriented Prediction Error Costs

TL;DR

This paper introduces a new forecasting error metric tailored for lumpy and intermittent demand, integrating statistical accuracy with business cost considerations to improve predictive model evaluation.

Contribution

The paper proposes a novel metric that better captures demand prediction errors by considering both statistical and business factors, addressing limitations of traditional metrics.

Findings

The new metric outperforms MAPE and RMSE in evaluating intermittent demand forecasts.

Validation on simulated data shows improved error assessment.

Application to real automotive aftermarket data demonstrates practical relevance.

Abstract

Forecasts of product demand are essential for short- and long-term optimization of logistics and production. Thus, the most accurate prediction possible is desirable. In order to optimally train predictive models, the deviation of the forecast compared to the actual demand needs to be assessed by a proper metric. However, if a metric does not represent the actual prediction error, predictive models are insufficiently optimized and, consequently, will yield inaccurate predictions. The most common metrics such as MAPE or RMSE, however, are not suitable for the evaluation of forecasting errors, especially for lumpy and intermittent demand patterns, as they do not sufficiently account for, e.g., temporal shifts (prediction before or after actual demand) or cost-related aspects. Therefore, we propose a novel metric that, in addition to statistical considerations, also addresses business aspects. Additionally, we evaluate the metric based on simulated and real demand time series from the automotive aftermarket.