6 months after setting up an advanced planning system and S&OP process in China and moving back to the USA, I get an urgent e-mail from the supporting IS manager. The General Manager was angry that the system did not anticipate a rash of new demand from new customers. He therefore does not see the value in the system and wants to stop paying the license fees (despite having 2.5 years left in the contract).
The system was set up like most any planning and forecasting system. The statistical forecasts are based on historical sales for existing customers using a variety of forecasting algorithms. Sales input is solicited and added to the statistical forecasts to create a hybrid-consensus forecast. Supply planning picks up the consensus forecast which then drives make/buy decisions.
The chaos happened when the focus on delivering on new customer forecasts was too high. When sales input was entered into the system and did not materialize the sales force was chastised for not delivering. However, when the sales force did not offer sales input into the consensus forecast and then signed new business, they were not chastised. Over time sales input became conservative and a large percentage of future business was missing.
The conservative consensus forecast was then discussed in the S&OP meeting and one number was hatched. Despite a several-month pattern of new volume showing up that wasn't forecasted, there was no commitment from sales to support a continued over-achievement of the forecast. Downstream the sourcing team failed to purchase sufficient product to satisfy customers, since they were purchasing to the conservative consensus forecast.
Chaos ensued that resulted in the following: missed sales volume, expensive transportation, and up-charges by suppliers smelling blood in the water.
As with most cases of supply chain chaos, it was caused by a failure to communicate. The fix was easy. The blame was rightly placed on the process rather than the system. An additional metric of under-forecasted volumes was added to the S&OP process. The bias for under-forecasting new volume is now discussed and layered onto the consensus forecast during the S&OP meeting and make/buy decisions adjusted accordingly. Time will tell if this avoids future chaos.
I suspect that as the world pulls out of receission there will be a lot of stockout chaos similar to this one. Sometimes it is hard to be hopeful after the slow downs of recent times...