Distribute
Use
You use this function when you want to change the distribution of plan data to various characteristic values. The sum of the plan values for the chosen selection is kept constant and the total amount is newly distributed to the characteristics you want. You are able, for example, to determine a specific yearly sales figure as the total and then distribute this to the individual periods in the fiscal year. Typical application areas for the distribution of plan data are:
· Allocation of capital
· Budgeting
· Seasonal distribution of quantities and amounts
Prerequisite
At least one data record must exist for the corresponding selection conditions. The data from this data record is to be distributed using the selected distribution function.
Features
BW-BPS offers the following types of distribution:
· Distribute by key
· Distribute by key from sender to receiver
· Distribute with reference data
· Distribute by reference data from sender to receiver
You use appropriate parameter groups to define distribution keys. You create these parameter groups for a distribution function. You can carry out the distribution for all data records or for a specific combination of characteristics and key figures.
You can use reference data from more than one year, for example, from 010.2003 to 004.2004, and also plan for more than one year. The number of periods in the reference and plan area must, however, be the same. The reference data periods must be selected as individual values (not as an area).
You can decide whether the data to be distributed should be kept at the outgoing data records or deleted after distribution. In addition, you can determine that only data is distributed that was previously contained in data records with the initial characteristic value “non-assigned”.
The types of distribution differ as follows:
Distribute by Key
When distributing by key, you provide characteristics values and the shares by which distribution is to take place. For the fields to be changed you select the characteristics that you want to use and distribute the plan data using their values.
To distribute plan data within periods of a fiscal year in order to seasonally distribute the planned key figure, include, for example, the characteristic period/fiscal year in the list of fields to be changed.
In addition, you determine characteristics for conditions that must be fulfilled so that the planning function can be carried out.
To carry out planning only for those divisions that you specify, include the characteristic division in the list of conditions fields.
To be able to carry out a planning function defined in this way, you must finally define a parameter group. Enter the values necessary for your concrete planning task for the previously determined characteristics.
You could, for example, define as a condition that plan data only is to be generated for the "pharmaceuticals" division. In the parameter group you could determine that the key figure to be planned has a normal distribution over the fiscal year.
Distribution factors
The distribution factors with which you distribute the total amount to be distributed in the parameter groups represent relative values. These relate to the sum of the factors used in the parameter group: The system views the total that is formed from the addition of all factors within a parameter group as 100 % and then determines the relative weight of every factor for distribution. The following table shows the treatment of distribution factors:
Values before distribution
Period |
Amount |
Distribution factor (user entry) |
Relative weight |
001.2003 |
100 |
1 |
12,5 % |
002.2003 |
100 |
2 |
25,0 % |
003.2003 |
100 |
2 |
25,0 % |
004.2003 |
100 |
3 |
37,5 % |
Total: |
400 |
8 |
100 % |
Values after distribution
Period |
Amount |
Distribution factor (user entry) |
Relative weight |
001.2003 |
50 |
1 |
12,5 % |
002.2003 |
100 |
2 |
25,0 % |
003.2003 |
100 |
2 |
25,0 % |
004.2003 |
150 |
3 |
37,5 % |
Total: |
400 |
8 |
100 % |
The system proceeds as follows during distribution:
...
1. The total amount to be distributed is determined (400 in the example).
2. The distribution factors, which are assigned to the individual characteristic values (in the example the 4 periods), are added to a total (8) and then set in proportion to each other relatively. This results in the percentages which are reported in the “relative weight” column and always add up to 100 %.
3. The amounts, which are assigned to the individual characteristic values, are multiplied by the relative weights determined and thus distributed again.
Note the following special features: If the total of the keys is 0, distribution is carried out equally. Negative keys remain negative. The system tries to distribute rounding differences as equally as possible. Rounding differences usually occur because key figures are stored with two or three decimal places. Non-assigned characteristic values are disregarded when the key is created.
Distribute with Reference Data
When distributing with reference data, you make the same settings as when distributing by key. In addition, you select one or more characteristics that are to be used to determine reference data.
For example, you want to generate plan data for the "plastics" division. The data is to have the same seasonal distribution as the (existing) plan data of the "pharmaceuticals" division. Choose the characteristic "fiscal year/period" as the field to be changed and the characteristic division as the reference characteristic. You create a parameter group in which you enter the “pharmaceuticals” division under reference data. When executing the function, the system distributes the values to the periods so as they are already available for the values with the characteristic value “pharmaceuticals”.
Distribution with reference data is suited to two-level sales planning. First the revenue and sales quantities that are to be realized by different customer groups are planned. In a second level you can plan which revenue and sales quantities can be realized by the customers. To obtain proposals for this detailed planning, distribution is carried out with reference to the actual data for the previous year. The field that is supplied with values during this process is the Customer Number field. Plan values for revenue and sales quantity are thereby allocated as in the actual values of the previous year.
Further Settings
Only distribute non-assigned values
For the basic function types distribute with reference data and distribute by key you can also specify that only those data records should be included in distribution that contain the initial value "not assigned" (#) for the characteristics to be changed.
In the example above, this would mean that the values of all data records, to which no seven-digit period (= characteristic to be changed) was assigned, would be distributed to the four periods 001.2003 to 004.2003.
Retain values at sender data record
For the extended function types distribute with reference data from sender to receiver and distribute by key from sender to receiver, you can decide whether the values that you distribute to the receiver data records should remain posted at the sender data records, or whether the appropriate sender data records should be deleted after distribution.
Enhanced distribution function types
Function type |
What you need to know |
Distribution by key from sender to receiver |
This function distributes values from sender to receiver within a characteristic (or several characteristics). When you define the planning function you have to specify within which characteristics you want distribution to take place. Specify the sender, receiver and ratio for the distribution in the parameter groups. |
Distribute by reference data from senders to receivers |
This function distributes values from sender to receiver within a characteristic. When you define the planning function you have to specify within which characteristic you want distribution to take place and the characteristics for reference data. Specify the sender and receiver in the parameter groups. The ratios that distribution is to be carried out by are determined from the reference data. |
See also: