Even world-class companies have average forecast accuracy percentages in only the high 70′s, especially at the lowest SKU or component level, which affects service levels and financial performance. Forecast accuracy can be improved through best practices, but will always be affected by the unpredictable:
- Chronic bias, which can be minimized with effective Sales/Inventory/Operations Planning (SIOP). Of course, world-class businesses do this already, yet they still achieve accuracy of less than 80%.
- Unforeseen and unforeseeable special causes, such as natural disasters.
Forecast accuracy is limited in part because it predicts the timing and magnitude of only three types of variation:
- Trend
- Seasonality
- Certain foreseeable special causes, such as promotions
There is always room for improvement, but these can be predicted only so well. Realistically, then, even world-class companies experience a significant forecast-accuracy gap that – if not bridged – compromises target fill rates, inventory performance and financial goals.
Fortunately, a financially beneficial service-level bridge from forecast to reality is right there in your data – but where?
One more form of variation still exists, and it can be quantified using your data: Common-cause random variation in demand (or usage, for components) and replenishment lead time.
Yes, the magnitude of random variation can be quantified by analyzing historical demand and/or lead-time data. Of course, the timing of random variation is unpredictable (that’s what makes it random), and this means that common-cause random variation must be addressed with properly-determined safety stock.
At this point, you may be saying, “I’m already doing that. I use a statistical technique to determine my safety stock levels.”
Forgive the blunt reply, but – How’s that working for you?
Honestly, is your safety-stock technique consistently achieving your fill-rate targets on an item-by-item basis? Does it instead put too much inventory in place – few stock-outs, but suboptimal inventory performance? Does it sometimes recommend too little inventory, so you must subjectively override or increase the calculated safety-stock level?
Unreliable safety-stock calculations do not mean that there is no statistical solution. Instead, you need a safety-stock approach that:
- Properly identifies and isolates the common-cause random variation from all the other variations, and
- Uses a correct, comprehensive statistical safety-stock approach addressing not just common-cause random variations in demand and lead- time, but all six factors that affect safety stock, service levels, inventory performance and expediting.
The next blog entry will discuss how to go about designing the correct and comprehensive statistical safety-stock approach.
(But if you can’t wait, then please visit the TopDown Lean Systems website to learn more now!)
