Supply chain planning has always been difficult. COVID-19 pandemic creates yet another of these challenges. Supply chains have been disrupted and demand forecasts are changing drastically. We have experienced challenging situations like this in the past and the current crisis is certainly not the last one. Good supply chain processes should be able to deal with normal day-to-day activities and with crisis situations. And here’s one of them.
Three key takeaways:
Please read on for a more detailed description
When and why would you change history?
COVID-19 induces anomalies that will pass and soon become history. This one-time aberration may not be indicative of a regular pattern and should be considered when planning orders or shipments for the future. Even during normal times, there are several instances where we know there are unusual transactions and errors in our history:
Cautionary note when you do make the changes.
You should not be changing the official record of the actual transactional history. That information has to tie to the corporate financial reporting. But in today’s IT world of more flexible apps and improved computing power, conditioning history can come out of the Excel closet of individual planners and be made part of good demand planning process.
There must be discipline and accountability in making the changes that end up in the conditioned history line. Judgments of what are unusual orders that will most likely never be repeated, and therefore omitted or significantly reduced in the conditioned history line and missed demand as a result of supply issues may require guidelines and even some level of approvals. When changes are made, they should be annotated as to why the change was made and, in enough detail, to make it meaningful.
How would you go about changing history?
You should use every tool, trick and approach you have to generate a plan for the future – including using math (statistics, analytics, etc.) that may automatically correct for outliers (by zero weighting, dampening to a demand filter, etc.) and “unusual” actuals. Large orders won that may not be repeated may require a trip down memory lane with sales input. Same will be true with orders that were lost that should have been won under any normal circumstances.
Conditioning history does not guarantee great results. But it does guarantee the best possible starting point in the demand planning process. There will still be unexplained outliers, etc. that we will still need math to help us handle. Patterns in the demand, such as trend and seasonality, will still need to part of a projection into the future. And, of course, we still need to get the input from the various functions that can supply direct future activity insights (such as sales & marketing).
Conditioned history also allows to better discuss and interact with those other participants. When we work with them with conditioned history (that they may very well have been a big part of identifying), we can create more meaningful conversations.
Conditioning history is all about capturing the knowledge of what should have happened and annotating why changes were made so ties back to the transactional actuals can be readily made. It is not a standalone process. It should be an ongoing part of what a good demand planner does in the course of their job – from the initial review of last month’s actuals to the forecast through the day-to-day happenings as the month unfolds. When well done, the conditioned history line becomes one of the most valuable assets in the demand planning world.