Why Forecast the Past?


Picture source: futureindicator.com



Matching supply to demand is one of the most common goals of industrial engineering and an important one in any business endeavor, healthcare not excluded. Whether matching bed supply to patient demand, nursing skill resources to clinical needs or operating rooms to surgeries, this effort is critical for both the smooth flow of resources. But people attempt to match supply to production, instead of supply to demand. In effect, they forecast the past.

Instead, you need to look at demand-true demand. What should you or could you have done if you were able to set the supply correctly. This is not going to be found by looking at historical productivity data. But many people still do.

I have seen a lot of improvement and so called “forecasting” tools lately concentrating on matching supply and demand, but they tend to match supply to historical production. We need to inject the use of more sophisticated demand capturing tools and methods. After we know what our true demand, we as industrial engineers can provide more useful models and techniques. A great source for some of these other models like newsvendor, linear optimization, game theoretic and simulation.

However, these engineering techniques require not only the data, but the understanding that we must match supply with the true, dynamic demand for healthcare services. Otherwise we will only provide what we have in the past, and that is not good enough.