Show TOC Entering content frame

Object documentation Croston Method Locate the document in its SAP Library structure

Definition

The Croston method is a forecast strategy for products with intermittent demand. In the univariate forecast profile, choose forecast strategy 80.

The Croston method consists of two steps. First, separate exponential smoothing estimates are made of the average size of a demand. Second, the average interval between demands is calculated. This is then used in a form of the constant model to predict the future demand.

Initialization

The system checks the first time bucket of the historical values. If it finds a value (not zero), it gives Z this as its initial value. X is set to 1. If it does not find a value, Z is set to 1, X to 2.

 This graphic is explained in the accompanying text

 

This graphic is explained in the accompanying text

Calculation of Forecast Parameters

The forecast is made using a modified constant model. The forecast parameters P and X and determined as follows:

This graphic is explained in the accompanying text

The last iteration, that is the iteration for the last time bucket, delivers the parameters Z(f) and X(f) for the forecast.

Forecast

As of Release 4.0 two forecasts are possible:

...

       1.      The forecast quantity is distributed over all time buckets in the forecast period. This method alone was available in previous releases.

This graphic is explained in the accompanying text

       2.      The system distributes the forecast quantity according to the mean interval X that it determined earlier.

This graphic is explained in the accompanying text

This difference is illustrated in the graphic below.

This graphic is explained in the accompanying text

Use

Exponential smoothing is often used to forecast demand in stock control systems. If demand is intermittent, however, this method almost always produces inappropriate stock levels. The Croston method is suitable if demand appears at random, with many ¾or even most ¾time periods having no demand; where demand does occur, the historical data is randomly distributed, independently or almost independently of the demand interval. Such demand patterns are known as "lumpy demand" or intermittent, irregular, random or sporadic demand. One example might be demand for spare parts or equipment that are usually ordered in batches to replenish downstream inventories.

In Releases up to 3.1 no ex-post forecast was calculated with this forecast strategy. This is possible in Release 4.0.

Leaving content frame