How to optimise your supply chain with AI?
To significantly reduce its carbon footprint and improve its supply chain, TEREOS enlisted JEMS to implement a data platform and a machine learning algorithm. An irrefutable result.
- Published on
In a few figures...
Over 10,000
lorries a day
circulate daily between the fields and factories during the harvest
Second floor
World Sugar Group
A major industrial player, with massive logistical challenges
12K
French farmer co-operators
A broad ecosystem, requiring fine coordination of operations
1+1
a data platform + a machine learning algorithm
put in place to manage and optimise collection flows
The project
Our approach
JEMS supported TEREOS in both the definition of the data roadmap and the execution of the project. The approach consisted of building a cloud data platform capable of centralising, structuring and exploiting logistics data, and then developing a machine learning algorithm to improve the management of truck flows during harvest periods.
The diagnosis
During the sugar beet campaign, TEREOS must orchestrate very significant logistical flows between the fields and the factories every day. This operational intensity naturally creates points of tension in the supply chain, particularly when truck arrivals are not sufficiently anticipated or spread out. Waiting times then increase, congestion multiplies, and collection becomes less efficient. In this context, TEREOS's challenge was to better utilise data to improve operational management, streamline flows, and reduce operational and environmental impacts.
The key deliverables
- Definition of the data roadmap
- Setting up a data platform on Microsoft Azure
- Data integration and organisation with Talend
- Structuring data storage and operations with Snowflake and DataStax
- Creating visualisation dashboards with Tableau
- Development of a machine learning algorithm to optimise truck waiting times
Comment optimiser in the supply chain with the'AI ?
Tereos benefits
A smoother supply chain
Optimising flows allows for better distribution of truck arrivals and limits saturation phenomena.
Reduced waiting times
The use of machine learning improves the organisation of rotations and reduces wasted time in the field.
A reduced carbon footprint
By reducing congestion and downtime, TEREOS also acts on its environmental footprint.
Improved handling
The data platform gives teams a more structured and actionable view of logistics operations.
