The inventory control literature generally assumes that the demand distribution and all its parameters are known. In practical applications it is often suggested to estimate the demand variance either directly or based on the one-period ahead forecast errors. The variance of the lead time demand, essential for safety stock calculations, is then obtained by multiplying the estimated per-period demand variance by the length of the lead time. However, this is flawed, since forecast errors for different periods of the lead time are positively correlated, even if the demand process itself does not show (process) auto-correlation. In this seminar (and the related journal article), I show that traditional approaches can lead to safety stocks that are up to 30 percent too low and service levels that are up to 10 percent below the target. I also present corrected approaches, and discuss the general need for more research that integrated statistics into decision making.
