For those of you on LinkedIn, be sure to sign up for the Institute of Business Forecasting and Planning discussion group. This is an active (and addictive) forum for sharing information and perspectives on a wide variety of forecasting & planning topics. The following question was posted by Reno DiGenova, the VP – Replenishment, Inventory & Demand Planning at Geneva Watch Group, and has garnered 27 responses:
The more general question is: Should you ever freeze your forecast?
It is definitely appropriate to freeze a forecast at some point for measuring forecasting performance. (Otherwise, we could wait until the actuals came in and always hit 100% accuracy.) The usual practice is to freeze the forecast at supply lead time – that is, the point after which you can no longer impact supply in the period being forecast. For example, if the supply in February cannot be adjusted after January 31, forecasting performance for February would be based on the forecast as of January 31.
A more contentious question is: Should you ever change your forecast within the supply lead time? Or going back to our example: Does it make sense to change the February forecast in February, even though you can no longer impact February’s supply?
Elif Kotman and others on the blog point out a benefit of changing the forecast within supply lead time: “Even though the Supply Chain may not change the production or purchases for the current month anymore, they can adjust the planned volumes/quantities for periods in the further future.” This is reasonable as long as you have a high degree of certainty in the change, and that the change is significant. Small changes are inconsequential so why waste time on them? Frequent changes terrorize and add nervousness to the planning systems, as Scott Roy and Tommy Jorgensen pointed out in the discussion.
Another situation is when you can make changes to supply within the forecasting bucket.
Many (perhaps most) organizations plan supply in weekly buckets, even though they do sales forecasts in months. Richard Watson makes a strong case for weekly forecasting, “…because it enables 52 data points in the year creating a clearer definition of the demand pattern, allowing the organization to be more dynamic, while at the same time simplifying tracking and monitoring mechanisms.” Aligning forecasting and planning buckets also avoids the messy conversion between months and weeks.
Join IBF’s discussion group on LinkedIn, and make these conversations a regular part of your working week. And if that is not enough, join us in Scottsdale, AZ (February 26-28) for the IBF Supply Chain Forecasting & Planning Conference which includes:
- Monday morning ½ day workshop on “What Management Must Know About Forecasting.”
- Round Robin Roundtable Discussions on various topics including Worst Practices in Forecasting.
- Schedule a 30-minute consultation with me or one of my SAS colleagues from professional services and R&D. We’ll be ready to discuss your biggest challenges related to forecasting process, statistical modeling, or whatever you have to throw at us. Send me an email (email@example.com) to reserve a time, or sign up at the SAS booth at the event.
- Free copy of The Business Forecasting Deal (book signing during breakfast on Tuesday).
Register by January 27 and enjoy early bird pricing and free participation in the IBF Golf Outing. (Based on past scores from the outing, actually being able to golf is not a requirement.)
Hear Mike Speak at: