All you need is a better forecast

In today's International Herald Tribune, I learned that Tesco, British Grocer, Uses Weather to Predict Sales

[The classic variability of British weather] randomness has prompted Tesco, the country’s largest grocery chain, to create its own weather team in hopes of better forecasting temperatures and how consumer demand changes with them.

My first reading of this article was that it sounds like a good idea. But then one of my Theory of Constraints colleagues reminded me of the danger of forecasts: they are always wrong.

Instead, it might be a better idea to design the supply chain to be flexible and fast to respond to actual changes in consumer demand.

2 Comment(s)

Mark Gould

said:

Reading the (rather brief) story, I think Tesco have actually done what you are suggesting. It is described as weather forecasting, but this section:

After three years of research, the six-person team has created its own software that calculates how shopping patterns change “for every degree of temperature and every hour of sunshine,” Tesco said last month.

suggests to me that they are using existing weather data (presumably provided by the Met Office, one of the world's leading meteorological services) to influence the supply chain.

September 3, 2009 2:46 AM

Jack Vinson

said:

Thanks, Mark. The danger of forecasts for demand is that they are wrong, no matter how sophisticated the methods. A much more flexible supply chain that can respond to ACTUAL demand will help them more.

That said, it can't hurt to have advance knowledge of demand spikes (or wells). The concern is the time and effort here. A supply chain project should provide benefits (increased inventory turns) in less than six months.

September 3, 2009 4:08 PM