How Standard and Median Forecasting Works

The system offers two different methods for forecasting product demand. The default is Standard. Use:

Use the Forecast Parameters For Demand Calculation control maintenance record to define how you want the standard forecasting to calculate. For more information about these settings, see Entering Control Forecast Parameters.

Note: For options in advanced forecasting, see How Advanced Demand Forecasting Works which works independently as a complement to this process.

Standard Forecasting Method

For example, Product A does not have a BTQ set. The exceptional sales percent equals 50%. Product A sells ten times in a forecast period of 365 days in the following quantities: 200, 100, 9, 8, 7, 6, 5, 4, 3, and 2. The sale of 200, being 50% more than the second largest sale of 100, is considered exceptional and eliminated from demand forecasting. The system adds the remaining quantities together (100+9+8+7+6+5+4+3+2=144) and divides the sum by the days in the demand period (144/365=0.394). The system then multiplies the daily demand by one month to produce a monthly demand of 12 units (0.394*30=11.82).

Consider that if Product A had a BTQ set to 100, the sale of 200 would be eliminated and the sale of 100, being 50% more than the second largest sale of 9, would be considered an exception and eliminated from demand forecasting. Adding the remaining quantities together (9+8+7+6+5+4+3+2=44), dividing by the days in the demand period (44/365=0.12), and multiplying by 30 days would produce a monthly demand of only 4 units (0.12*30=3.61).

Median Forecasting Method

For example, Product A does not have a BTQ set. The exceptional sales percent equals 50%. Product A sells ten times in a forecast period of 365 days in the following quantities: 200, 100, 9, 8, 7, 6, 5, 4, 3, and 2. The sale of 200 is proven to be exceptional and eliminated from demand forecasting. The median of the remaining quantities (100, 9, 8, 7, 6, 5, 4, 3, 2) is 6. The system multiplies the median quantity by the number of hits counted (6*9=54) and divided by the days in the forecast period (54/365=.148). The system then multiplies the daily demand by 30 days to produce a monthly demand of 5 units (.148 x 30=4.4).

See Also:

Entering Forecast Parameters for Products