When you work with demand forecasting in Business Central, the period option you choose in the forecast view changes how the planning engine interprets your forecast entries. The three options are Forecast, Period (such as month), and Week.
The view setting only changes how figures are displayed. It does not change how the planning engine calculates demand. The actual periods are defined by the dates on your forecast entries.
If you place forecast entries on inconsistent dates, the planning engine splits the demand into uneven periods that may not match what you see on screen. To avoid this, follow a fixed strategy: enter all forecast entries on the first of each month, or on every Monday if you forecast weekly.
If you create an entry by mistake, you can open the forecast entries and delete the specific entry to return the forecast to normal.
The three ways to view forecast periods
When you work in the forecast view, where you see the forecast per item in periods, you can choose between different ways to display those periods.
The first option is Forecast. Every time the system finds a forecast entry, it defines a period based on that entry. In a typical scenario, the periods follow the month because the forecast entries are all placed at the beginning of each month.
The second option is per month. Here you also see the months that contain no forecast at all. For example, in May, from May 1st to May 31st, you might have 60 pieces.
The third option is per week. When you switch to this view and scroll down, you see the forecast broken into weekly periods. The 60 pieces appear in the period that runs up to May 1st.
How forecast entry dates affect the planning engine
The date you put on a forecast entry is what really matters, and this is something you have to be very aware of.
Say you have 60 pieces forecast for May 1st, and then you enter another 15 pieces on May 9th. This is possible to do, but it changes how the planning engine reads your demand. The engine now assumes the 60 pieces cover the period from May 1st until May 9th, because the next forecast entry starts on May 9th.
If you switch back to the Forecast view and update it, you will see that May now has two forecast entries: 60 and 15. This is how the planning worksheet actually distributes the demand. The first forecast period covers only seven days, and the next forecast period covers about three weeks.
If you then view this per month, you see 75 pieces in May as a single figure. But in real life, the planning engine runs this as two separate periods. The monthly view hides the split that the engine actually uses.
Set a consistent forecast strategy
Because the entry dates drive the calculation, you need a clear and consistent strategy for how you create forecast entries.
The most common approach is one of these two:
- Forecast per month, with all entries on the first of the month
- Forecast per week, with all entries on Mondays
Pick one and stick to it. If you mix dates, you end up with uneven periods like the seven-day plus three-week split shown above.
If you do make a mistake, you can always go into the forecast entries and delete the specific entry. Once you update the forecast, you are back to normal.
Q&A
What do the Forecast, Month, and Week view options do in the forecast view?
They change how forecast periods are displayed. Forecast defines a period for each forecast entry it finds. Month shows every month including those with no forecast. Week breaks the figures into weekly periods. The view setting only affects display, not the planning engine’s calculation.
Does changing the period view change how the planning engine calculates demand?
No. The view only changes how figures are displayed. The planning engine calculates periods based on the dates of your forecast entries, regardless of which view you select.
What happens if I place forecast entries on inconsistent dates?
The planning engine splits the demand into uneven periods. For example, an entry on May 1st followed by an entry on May 9th creates a seven-day first period and a roughly three-week second period, even though a monthly view shows them as one combined figure.
What is the recommended way to enter forecast entries?
Use a consistent strategy. Either enter all forecast entries on the first of each month, or enter them on every Monday if you forecast weekly. Choose one approach and apply it consistently.
How do I fix a forecast entry I created on the wrong date?
Go into the forecast entries, delete the specific entry, and update the forecast. The forecast returns to normal.