There are a few things you can do to let the system "cleanse" the historical data
1) specificy an outlier correction value correctly. Try different values for different kinds of data..(observe visually)
2) specificy an approproate outlier correction method
3) Use historical value markings
4) Nominate the promotion key figure in master forecast profile
5) specify work days (time stream)
6) Use an approprate forecast model by VISUALLY look at the history in graphic mode
7) Have key figures like lost sales, returns etc in planning area.
😎 Have key figures for corrected history and corrected forecast in your planning area.
Doing first 6 results in both a corrected history and corrected forecast in addition to system forecast. You can chose to update all these key figures in the planning book by making the settings in the planning area > extras
In addition to this you can have key figure for lost sales, returns, unplanned promotional sales if any. Lost sales exraction can also be automated using a data source like VBFA (if you are using ECC) or simply some arbitray difference between customer orders and delivery quantities.. both are possible with some effort.. esle let planners enter lost sales manually based on their knowledge.
Hope this helps.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.