A few days ago I wrote a piece extolling the folly of relying on forecasting, using the blizzard that pounded Chicago as an example. Today a reader named Sam commented, "While I agree with some of the evils of forecasting in manufacturing, I would say the National Weather Service was spot on this time around. Weather forecasting is getting to be very reliable and they deserve a lot of credit."
From my personal standpoint, I agree with him because the forecast was accurate within my lead time. The storm hit in th early afternoon Tuesday. The National Weather Service first predicted it on Sunday – about 2 days notice. Two days was plenty of time for me to make sure the liquor cabinet was stocked and the fridge was full.
I would guess, however, that many of the local businesses would take issue with Sam's opinion. On the Friday and Saturday before the storm, the forecast was for snow showers. For the most part, retailers had no chance of laying in the snow shovels, snow blowers and salt to meet the demand. Manufacturers and distributors rolled into work on Monday with 24 hours to deal with the impending loss of two days of shipping and receiving and inevitable lost capacity. I imagine most of them thought a little more notice would have been helpful, and that the forecast w week before the storm – closer to their lead time to plan for it – was hopelessly inaccurate.
The point is that forecast accuracy can only be measured relative to the lead time necessary to deal with the implications of whatever is being forecast.
Whether it is the demand for widgets or weather conditions, all forecasting is less accurate the further out it goes.
Forecast inaccuracy in manufacturing translates into excess inventory, missed customer shipments, or both. In any business forecast inaccuracy translates into a misalignment between resources and demand. Forecast inaccuracy directly creates a great deal of waste (usually unperceivable to traditional accounting schemes).
Sourcing from China or India necessitates long lead times and reliance on inevitably inaccurate forecasting and is therefore, by definition, wasteful.
This is why lead time reduction is the cornerstone of lean. It is why demand pull is a cornerstone of lean.
I wrote a piece about Kroger just two weeks ago, and their integrated weather tracking and logistics center. The value of this cross-functional group is its ability to compress Kroger's lead times in relation to forecasted weather events. It was the lead time in our lean blog specifically because it described an organization that cut horizontally to compress lead times. I would guess that Kroger would agree with Sam in saying the National Weather Service was "spot on" because they have the ability to react within the lead time given for this storm. Companies not focused on Kroger-like lean are probably moaning and complaining about poor forecasting.
The same is true with manufacturers who focus on lean. They get the sales, while the non-lean companies waste money and do a lot of internal bickering on forecasting when they should be worrying about their lead times.