Dell saves 60% on shipping by using complex predictive models


If ever anyone tells you that math is all theory and no application, slap them in the face. Then tell them that in just the latest example of applied mathematics having real-world effects, Dell has employed a complex mathematical model to predict shipping rates, inventory problems, and so on, with phenomenal success. Professor of Operations Management Jérémie Gallien says:

“Before our model, the company would use forecasts, but the reality is that forecasts are never right. By using probability theory to quantify the uncertainty of demand forecasts and projected inventory, we have been able to vastly improve and speed up the decision-making process.”

I guess when you have such a huge amount of data, it makes sense to hire some serious crunchers (MIT in this case) to break it down and make sense of it. [Thanks for the tip, Paul!]