CHAPTER 2Driven by Real Demand

Until very recently a manufacturer had to try and predict what it would have to produce based on the past. It would have to develop a sales forecast based on what had been sold last year or last month. That sales forecast, in turn, was the basis for a production plan for the factory. And the factory's output was the basis for what got shipped to the customer.

But historical sales are not the most reliable predictor of the future. Too often the unexpected occurs in the marketplace. A product becomes fashionable for some unexplained reason or an unforeseen event drives up sales. And consumers are just plain fickle.

Until the decade of the 1990s, forecasting was a solo exercise done by companies all on their own and often without consulting supply chain partners. Then it occurred to some supply chain executives that two heads might be better than one in trying to predict the future. In 1995, retail giant Wal-Mart Stores Inc. and the pharmaceutical company Warner-Lambert (now part of Pfizer) did a test of collaborative forecasting for inventory replenishment along with the assistance of the consulting firm Benchmarking Partners and two software companies, SAP and Manugistics (now part of the JDA Software Group). The first pilot between Wal-Mart and Warner-Lambert involved a joint forecast for keeping the amount of Listerine mouthwash in stores. Initially, the two companies exchanged pieces of paper to compare their sales and production forecasts. Later, ...

Get Protean Supply Chains: Ten Dynamics of Supply and Demand Alignment now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.