One of the platform’ s key tools, Product Insights, helps diagnose the reasons behind unusual spikes or dips in supply and demand. These insights let planners make informed decisions when demand suddenly shifts, whether that’ s because of seasonality or a social media trend sending a product viral.
“ Our automation,” Toni says,“ allows us to scale up or down to respond quickly to spikes in demand. The goal is to shift from reactivity to proactivity by identifying and preventing upcoming stockouts.”
Ocado’ s forecasting system uses deep learning models, designed to mimic human information processing. It trains on billions of data points from the Ocado Smart Platform across multiple retailers, ensuring accurate demand forecasting for new items without long sales histories. The system continuously learns from customer behaviour and updates recommendations with fresh data.
Toni explains:“ Our demand forecasting system helps retailers make smarter decisions on waste reduction and product availability. It runs on advanced models trained on billions of sales points across regions.”
This enables retailers to anticipate demand without relying on extensive sales histories, even for new products, while continuously adapting to shifting shopping behaviors and drawing on collective industry insights. During peak periods such as Christmas, customers frequently reserve delivery slots weeks in advance but continue modifying their baskets until the last moment.
supplychaindigital. com 77