Concept of data-driven supply chain management
With the help of sales forecasts and demand forecasting, the entire supply chain is to be optimised for a beverage company.
Significant reduction of the whip effect, which increases as the level in the supply chain rises.
Reduction of out-of-stock situations despite simultaneous reduction of stocks
Central calculation of order proposals enables transition to vendor-managed inventory: "Delivery is already waiting at the door when the order is actually only supposed to be placed".
Challenge
- Out-of-stock situations and excess inventory exist at all levels of the supply chain and need to be reduced
- Whip effects are to be identified in the historical sales data and countermeasures implemented
- Increasing product diversity should be managed without the need for manual intervention
Solution
- Breaking down the data silos of the various stages and establishing an overarching data lake for centralised control of the entire supply chain
- Implement automated sales forecasts for all stages of the supply chain with machine learning algorithms based on historical sales data.
- Forecasting the unknown demand by modelling the demand data from the historical sales data
Result
- A universal, automated forecasting environment that can be flexibly deployed at all stages of the supply chain
- Reduction of the whip effect through shorter response times and reliable forecasts at all levels
Are you interested in your own use cases?