To compute the demand forecast

Use the Generate Demand and Inventory Plan (cpdsp1210m000) session to generate the demand forecast based on historical demand data. The parameters of the Plan Items - Forecast Settings (cpdsp1110m000) session control the calculation method.

Historical demand data

LN forecasts the future demand on the basis of historical demand data from the item master plan or the channel master plan. This historical demand data consists of the customer deliveries between the start date of the scenario and the current plan period.

If the Dependent Demand Forecast check box in the Items - Planning (cprpd1100m000) session is selected, LN bases the demand forecast on the sum of:

  • Customer deliveries
  • Internal deliveries
  • distribution deliveries

Structure of the demand forecast

The demand forecast has three parts:

  • Trend influence

    Describes how fast the demand increases or decreases.
  • Seasonal influence

    Describes any recurring pattern in the demand. For example, the demand for umbrellas is higher in rainy seasons.
  • General demand level or average demand

    The demand without trend or seasonal influence.

Demand forecasting procedure

First, the trend and seasonal influences are determined. Then, the future demand is forecast by means of one of the following methods:

  • Moving average
  • Exponential smoothing
  • Forecast method: polynomial regression
  • Forecast method: time-series analysis

The details of the calculation are described in the Algorithm for demand forecasting topic.