Forecast measures

Forecast measures define numeric value that are stored for specified item + location + customer + time bucket combinations.

FrePPLe provides a number of standard measures. With this table you can add project-specific values to manage more complex demand forecasting processes and workflows.

The measures are visible and editable in the Forecast editor, Forecast report and Inventory planning screen.

Important

After any change to this table, you should regenerate the plan to assure the planning engine works with the latest definitions of the measures.

In addition to the measures you define in this table, the forecasting engine uses a number of standard measures.

  • Calculation of the statistical forecast:
    This process uses as input historical demand, which is read from the sum of “orderstotal” and “ordersadjustment” measures.
    The statistical forecast populated in the measure “forecastbaseline”.
  • Calculation of the forecast consumption:
    This process reads as input the measure “forecasttotal”, and all open sales orders will consume from this gross forecast. By default “forecasttotal” is computed as “if(forecastoverride == -1, forecastbaseline, forecastoverride)”, but you can easily use a different formula.
    The output is populated in the “forecastnet” and “forecastconsumed” measures.
  • Calculation of the supply planning:
    This process considers the measure “forecastnet” as the remaining sales forecast to plan.
    The outputs of the process goes in many tables, and also populates the measures “forecastplanned” and “ordersplanned”.

Field

Type

Description

name

non-empty string
A unique name.
The measure data are stored using this key in the database. Only alfanumeric characters are allowed.

label

string

The string used in the user interface. This will be translated automatically to the language configured in your browser.

description

string

Free text description.

type

string

Define the storage and calculation behavior this measure.

  • Aggregate
    At higher levels of the item, location, customer and time hierarchies the value of this measure is computed as the sum of all children.
    Edits at higher levels are disaggregated to all children, proportional to the existing value of each child.
    Examples: total sales, total sales value…
  • Local
    The value of such measures is only valid at the item, location, customer and time period where it is defined.
    Examples: cost, price, margin…
  • Computed (beta-state)
    The value of these measures is computed with the expression specified in the compute-expression field.
    Examples: forecasttotalvalue = forecasttotal * cost

mode_past

string

Visibility and editability policy for the current and past time periods.

  • Edit: Read and write access

  • View: Read-only access.

  • Hide: Don’t allow user to display the value.

mode_future

string

Visibility and editability policy for the current and future time periods.

  • Edit: Read and write access

  • View: Read-only access.

  • Hide: Don’t allow user to display the value.

formatter

string

Controls how the values should be displayed.

Accepted values are “number” (default) and “currency”.

discrete

bool

Specifies whether fractional numbers are acceptable for this measure or not.

defaultvalue

number

Specifies the value for the measure when it is left unspecified for a certain item + location + customer + time bucket combinations.

0 is the default value.

If the default value is -1, unspecified cells will display as an empty cell.

compute expression

string

For computed measures, this expression defines how the value is computed.

The expression can use the following:

  • All standard and custom measures

  • All item attribute with a function call: item(“myattribute”)

  • All location attributes with a function call: location(“myattribute”)

  • All customer attributes with a function call: customer(“myattribute”)

  • Mathematical operators: +, -, *, /, %, ^

  • Equality & Inequality operators: =, ==, <>, !=, <, <=, >, >=

  • Logic operators: and, mand, mor, nand, nor, not, or, shl, shr, xnor, xor, true, false

  • Mathematical functions: abs, avg, ceil, clamp, equal, exp, floor, frac, log, log10, log2, logn, max, min, mul, ncdf, nequal, root, round, roundn, sgn, sqrt, sum, swap, trunc

  • Control structures: if-then-else, ternary conditional, switch-case

Here are some concrete examples:

  • round(forecastoveride * item(“cost”))

  • if(forecastoveride != -1, forecastoverride, forecastbaseline)

  • if (x, y, z)

  • if (x > y) z; else w;

  • if (x > y) z; else if (w != u) v; else a;

  • x ? y : z

  • switch { case x: y; case a: b; default: i }

update expression

string

This expression is evaluated when the computed measure is updated by the user.

It will use assignments to update the the measures used in the compute-expression. The variable newvalue is populated with the new value provided by the user.

An example shows how the update expression basically inverses the compute expression:

  • compute expression: forecastoverride * cost

  • update expression: forecastoveride := newvalue / cost