How to Forecast Intermittent Products
This article explores one of the most challenging problems in supply chain management: forecasting intermittent demand. We first delve into the intricacies and root causes of erratic sales patterns. Then, we critically examine traditional forecasting methods (Croston) and metrics (MAPE or WMAPE), highlighting their limitations in the face of unpredictability. Ultimately, we propose strategies that will enhance your forecasting accuracy amid demand volatility and provide tactics to optimize your supply chain for increased resilience to intermittent orders.
Note to the reader: this article is twice as long as usual. Dealing with intermittent sales is challenging and requires a holistic approach: I am presenting you with a complete, step-by-step guide on how to cope with this.
Navigating the Complexity of Intermittent Demand
In the realm of supply chain management, products with intermittent demand pose specific challenges to planners. These are products for which demand does not occur regularly but sporadically, with some periods showing strong demand and others showing little to none. Common examples can range from certain pharmaceuticals and spare parts to seasonal clothing. Moreover, if you need to generate more granular forecasts, most products will start displaying intermittency.
Accurate forecasts are crucial to supply chains. They avoid overstocking and piling up dead inventories while simultaneously improving customer satisfaction as better forecasts reduce lost sales. (Read this article for a detailed case study about the business impact of forecasting accuracy). Moreover, for some industries like pharmaceuticals or service parts, getting the right product to the right place at the right time can be a matter of life and death.