One of the largest packaged product producers and distributors in North America, faced low asset utilisation due to poor demand predictability, leading to suboptimal delivery schedules and in-store availability. To tackle this challenge, our team developed an AI-based forecasting model. This solution involved using weather data and other factors to predict which retail locations would require deliveries, as well as implementing operational improvements such as standardised equipment and centralised planning and dispatching. The results speak for themselves: we improved bottom-line benefits with €12.8M in a single year. But that’s not all: the forecast accuracy grew to 85%, with some regions reaching 90% while reducing zero deliveries below 4%.
