We all know that the first principle of forecasting is that it is always wrong. The second principle is that it is more important to understand what the estimate of error is. It is this estimate that we use to plan for risk, inventory levels and production planning. While it is true that the forecast is always wrong one of the important points about this error that often gets overlooked is that while most of the time it is not your fault, some of the time IT IS YOUR FAULT. Most of the time the error is due to the fact that, no matter how we try to improve our information, math, and processes, ultimately, forecasting is just another term for crystal ball gazing. If we truly could calculate a correct demand number every time it would be called calculating not forecasting. No matter how good we are, something is going to come along and knock all our lovely processes totally out of kilter and we will just have to find a way to deal with the aftermath. My point here though, is that sometimes we cause the errors to happen. Definitely not deliberately but by not consider certain consequences or by not having a process that is flexible enough to deal with non-normal circumstances. This year most of us who live in southern Ontario and many other parts of North Eastern North America are living a perfect example of uncontrollable forecast error and I am wondering how many organizations will deal with this properly in future forecasts. I would even go so far as to suggest that in this particular industry there is going to be a lot of excess inventory in the system come Spring of 2015.
So what am I talking about? SALT. The weather this year has been truly abysmal which has led to a significant and on-going shortage of salt for peoples driveways, Most stores ran out in late December, there was none to be found for weeks and while it is available now, the supply is noticeably sporadic and undependable. Is this the fault of the salt producers or the retail channels they distribute through? Absolutely not. Nobody was calling for the type of weather we have had (and if you do not believe that call any friends you may have in the Atlanta area and ask them how their weather has been). Normally in Southern Ontario the retail outlet as a whole start out the season with total stock levels of most of a year’s usage plus additional levels of inventory available at the producers level. In this case that supply was totally used up in the first couple of storms because they were more ice storms than snow. So was the forecast demand wrong? Absolutely, but it was not their fault because the external factors were too far from normal to anticipate the amount of true demand.
So why am I predicting high excess stock levels next year? A couple of reasons such as the fact that averages are just that, averages, which means that when you have a data point significantly off the mean then it is unlikely the next data point will also be that far off the mean in the same direction. In layman’s terms, it is unlikely next year will be as bad as this year. Most forecast systems, however, are based on history with a higher emphasis on recent history. The recent history in their systems is going to show a huge uptick in actual demand for this year and if they do not smooth out that jump in the historical record the system IS going to forecast higher than normal demand for next year. Also, people being people they are going to lean toward a higher number during the qualitative part of the process as well. Once burnt, twice shy type of thing. All of this is going to lead to higher levels of salt in the system for next year and if the weather is closer to normal (which is very likely) they will be left with a large excess stock situation in the spring. If this actually does happen it will be an example of an incorrect forecast being the fault of the forecast process rather than true demand variance (you creating the error). The key here is that the forecast is the wrong tool for dealing with significant, uncontrollable circumstances. In this case it is actually a job for Risk Management. The industry needs to look at how they reacted to the weather and the shortfall, what other options they have, and how they could do better. The retail channels cannot carry enough inventory to deal with all circumstance or to store large quantities of items like salt during the off season. The challenge is to keep inventory low, without stocking out, and having the flexibility to react as necessary. For those of us in Supply Chain Management, just a normal day at the office.
Having issues with your inventory?
Is your forecast driving your inventory issues? Do you understand what other Drivers are biasing your inventory levels to move up or down? Do you need help understanding and getting control of your inventory? Contact Ed White at Jade Trillium Consulting to discuss whether we can help your organization and how best to proceed.
Hope you enjoyed this posting. Talk to your friends and co-workers about their experience and thoughts on this topic, especially what it means for your organization. And, as always, I would love to hear back on your (and their) thoughts. Just fill in the comment box below along with your contact information to let me know what you think.