The Opportunity
- A consumer goods company struggled, in some of their categories, to maintain desired case fill rates to customers. Often, “cuts” needed to be made to orders, resulting in supply misses and unhappy customers.
- The company theorized that forecast error was the key driver behind these cuts. But others suggested that operational factors, such as plant capacity, scheduling, and transportation issues, were to blame.
Our Approach
- As a proponent of data-driven decision making, we provided a formal hypothesis testing exercise to establish the role of forecast accuracy as a driver of case fill rate performance. We used regression modeling and decision tree learning models as part of our analysis.
The Impact
Some key findings of the study include:
- Forecast accuracy can indeed impact case fill percentage.
- However, accuracy accounts for a small amount of variance in cut percentage. Other, operational factors account for the remainder. This puts to rest the notion that forecasts are the key driver.
- This relationship varies substantially by brand. We were able to quantify the impact of a 1% improvement in accuracy for each brand. Based on these impacts, and the size of the business, heat maps show planners where they can make the most impact, if accuracy is a stronger factor.