What This Article Covers
- A Zero Value Forecast Accuracy Measurement
- Reviewing Intermittent/ Service Parts Error Measurements and Safety Stock Literature
In this article, we will describe the needs of companies to calculate a dynamic safety stock while also using a non-standard forecast error measurement which can account for zeros in the demand history.
A Zero Value Forecast Accuracy Measurement
Companies that have a moderate to high percentage of their sales history that make it infeasible to use MAPE as an error calculation. Other error measures, that can handle zeros, are more suitable. However, if there is a desire to use dynamic safety stock it is of particular importance that the formula can incorporate the error measurement the company ends up using. Furthermore, the number of companies that have a moderate to high percentage of their sales history as zeros has greatly increased due in large part to significant increases in new product introductions. We cover this in the book Replenishment Triggers.
Most of these toothpaste containers contain essentially a similar to identical set of chemical compounds; however, marketing provides customers with different varieties of what is often the same product in order to promote purchases. Many of the claims are unfounded, but because there is very little regulation (in the US at least), one can say what they like regarding what the toothpaste will do for consumers. Whether promotional material on the packaging is true or not is barely mentioned, and anyone who might bring this up is considered hopelessly naïve as the primary focus is whether or not the claim will increase sales.
“Retailers are faced by increasing assortment. In grocery retail, product life cycles have been decreasing. As a consequence, it is increasingly difficult to forecast sales for an individual item in a particular store for tactical reasons, as time series tend to be short. Moreover, retail sales are faced with extensive promotion activities. Products are typically on promotion for a limited period of time, e.g. one week during which demand is usually substantially higher than during periods without promotion, and many stock-outs occur during promotions due to inaccurate forecasts.” – SKU Demand Forecasting in the Presence of Promotions Marketing
In an effort to justify its value to manufacturers, has greatly increased the number of products that are carried. However, marketing does not want to be held responsible for the naturally increased costs of producing so many different items or spreading similar demand levels over far more products. Therefore, it attempts to redirect the issue to being one that is squarely placed onto operations. Marketing could introduce fewer products, and actually increase the companies’ profitability, but it chooses not to for its own internal reasons that have nothing or, at least very little, to do with the benefits to the company as a whole. A primary reason for this is the motivation of marketing, which is based upon demonstrating its value through new product introduction, and or changing the terms of the sale of the existing items – generally referred to as promotions. The end result is that others must adjust around Marketing, regardless of the business outcomes.
Reviewing Intermittent/ Service Parts Error Measurements and Safety Stock Literature
We specifically searched for the use of dynamic safety stock with the types of error measurements that work with forecast errors that can handle zeros in the demand history like MASE & MAAPE, which we use in the Brightwork Forecast Explorer Error Calculator. We used academic search engine that reliably finds such articles. Ther is little there. Quite the opposite, the articles assume that the error measurement like MAPE that cannot be used with lots of zeros is applicable for most companies. Now, there are methods published for calculating safety stock dynamically in a service parts environment, but they are complicated. In reviewing them, we don’t think they would migrate well to a real environment. That is they are more academic curiosities, primarily used to show off math talent by the academic authors rather than to create a usable method for the industry.
This approach of combining the error measurement selection along with the customized safety stock is critical for getting a usable dynamic safety stock value is contained within the Brightwork Forecast Explorer. It is a combination of an error measurement for zero value periods combined with a customized dynamic safety stock formula.