Within Retail there are thousands of products selling daily in various shop locations.
To streamline & plan Supply Chain it's key a retailer performs demand forecasting to plan ahead their purchasing, production and logistics. Most organizations do (semi-)manual demand forecasting where planners manually plan thousands of products across too many stores.
The objective of this project was to streamline the Supply Chain by firstly forecasting demand smartly based on historical sales and secondly using this forecast to run a dynamic distribution requirement planning showing when to order and deliver what and where. We leverage 3 years of historical data and defined a predictive approach where we could predict with good accuracy sales up to 100 days ahead taking into account weather, promotions, seasons and special events (holidays, festivals, etc.).
Using this data-driven approach, we facilitated and accelerated Supply Chain planning and increased the forecasting accuracy.