Question: Serial Correlation (MLR/Fourier) and Inventory Problems

Someone asked me a question about the impact of serial correlation (a forecast performance/error measurement) on Inventory problems.

In short, if someone consistently forecasts above, will they always have too much stock? Conversely if one consistently forecast too low, will they always have out of stocks?

On its face this seems right. However there always are buffers in the system. We’ve all heard of the bullwhip effect in which supply chain participants hedge up and down the supply chain to a point that someone always has too much or too little based upon the accuracy of the forecast. I would add, to Dr. Lee’s approach, that this also depends upon the education of the supply chain planning community within an organization.

Serial correlation (above) will result in too much stock unless you’re managing safety stock statistically and hedging against forecast error. The smoothing in most safety stock calculations manage against that. More to the point (above or below with regards to serial correlation), your planning group has to understand how the planning system is related. They must understand how others are building in a reaction to their performance.

Its interesting that my friend posed this question to me in this fashion because I was thinking about it in quite the opposite fashion. I was thinking about it from the perspective of looking at projected stock outs and correlating that with serial correlation.

Further, I was in the process of conducting a coverage duration analysis in the hopes of determining whether or not coverage duration relative to lead times and positive serial correlation were leading indicators of areas needing an inventory reduction strategy.

Net-net, this is one more screaming need for companies to not just view key performance indicators one-by one, but understand the nuances of how they are related (or not related).

Our work has shown even more indicators, that sitting untouched in tables in your JDA SCM suite, that could help save or make a quarter — especially when viewed through the lens of a profitability analysis.

Leave a Reply

You must be logged in to post a comment.