Scott Armstrong is pretty knowledgeable about forecasting processes. In fact, he might be the world’s leading authority. Armstrong is a professor at the Wharton School of Business at the University of Pennsylvania. He’s dedicated his whole life (and he’s 74) to studying forecasting and his book, Principles of Forecasting, should be considered a bible to anyone interested in the field.
We supply chain planners could learn a lot from old Scotty.
For example, one of his gems of wisdom is about complexity of the forecasting process/model. He states, “there’s been no case in history where we’ve had a complex thing with lots of variables and lots of uncertainty, where people have been able to make forecasting models or any complex model work. The more complex you make the forecasting process the worse the forecast gets”.
Hmmm…makes you wonder, eh?
We have entire companies and self-proclaimed experts who are advising companies about how to factor in Big Data, the weather, and a whole host of other variables, all with complex models aimed at sensing and translating and responding.
Maybe most of these folks are fools? After all, they are forecasting what should be calculated. And the farther away from the consumer you’re forecasting, the more variables and constraints you’ll need to try to factor in. Exactly Armstrong’s point.
Don’t be a fool.
Never forecast what you can calculate!