In the previous two articles we have looked at Top Down Planning & Locking and floating. What other planning specific functionality might you reasonably expect to see in a system?
One of the most common planning specific functions is the mapping or balancing of variables. When we talk about variables being mapped, we mean that we provide the user with the ability to enter data into either on of two related variables.
For example, suppose we are planning a profit margin based on planned sales. We might choose to do this by applying a percentage to the sales plan as in the following example.
Margin Value = Sales Plan * Margin Percentage / 100
Equally, we might wish to enter the Margin Value and calculate the resulting Margin Percentage.
Margin Percentage = (Margin Value / Sales Plan) * 100
In a spreadsheet based system we have to decide which approach to adopt. In the example above we would take Sales Plan as a given, and then we would decide which variable to enter. We would then write a formula to calculate the result for the third, dependent variable.
In sophisticated planning systems we often find the ability to map the two related formulae onto each other. This would allow us to enter either the Margin Value or the Margin Percentage and have the other related variable calculated. However, as you might expect, such sophistication can raise further issues.
As planning is an iterative process, we often find ourselves recasting sales forecasts. Ask yourself what should happen in our example above when we change the value for Sale Plan.
Should it be the Margin Percentage or the Margin Value that remains constant? There is no doubt in my mind that it should be the Margin Percentage that is fixed, but I know of at least one major planning system where it is not possible to map these two variables and preserve the percentage.
Other features that we might look for include the provision of planning specific calculation functions. A good example of this is the “Cover” function. This function calculates stock requirement based on sales forecasts and value for number of weeks forward cover. Whilst it is possible to specify and write this functionality “longhand” it is not a trivial task. The “Uncover” function would, of course, reverse the calculation to calculate the number of weeks forward cover represented by a given amount of stock. These two calculations could then be mapped to allow entry of either Weeks Cover or Closing Stock in a Stock, Sales and Intake plan.
Another feature that can be extremely powerful is the ability to provide for automatic grading based on one or more attributes. Automatic grading allows us to create automatically an alternative store hierarchy that would be set according to given parameters, which we could then use in distribution planning. For example we might wish to group stores together for distribution purposes based on last season’s sales. Alternatively we might wish to group stores demographically by customer type. Equally we may wish to combine these attributes and grade by both turnover and customer type. Whatever the case, automatic grading means that store groupings can be dynamically created with little effort on the part of the planner.
These, then, are some of the things that differentiate planning systems from standard, “vanilla”, spreadsheet or database products. In all cases they are designed to reduce the amount of number crunching required of the planner and to allow them to concentrate on the quality of the input.