The Client
A European logistics company (road transport and contract logistics) with a distributed IT infrastructure, TMS/WMS systems, and growing ambitions in the area of data-driven operations.
The challenge
The organization planned to implement AI/ML solutions to optimize routes and forecast shipment volumes but lacked data science and MLOps capabilities. The IT department was focused mainly on maintaining operational systems, while recruiting AI specialists was difficult due to a highly competitive market and the company’s relatively low recognition as a technology employer.
The lack of resources delayed optimization projects, increased fuel costs, and limited the company’s ability to scale operations during peak periods.
The solution
We proposed a model combining recruitment and outsourcing of an AI/ML team. Within 8 weeks, we delivered a dedicated team in a staff augmentation model consisting of:
- ML Lead / Architect
- 2 Data Scientists
- ML Engineer
- Data Engineer
The team was responsible for:
- building a data pipeline integrating TMS, GPS, and ERP systems,
- developing a volume forecasting model,
- creating a route optimization algorithm considering fuel costs and SLA requirements,
- implementing an MLOps environment and model monitoring.
The project was carried out in close collaboration with the client’s operations and IT departments, with simultaneous knowledge transfer to the internal team.
The conclusions
- 12–15% reduction in fuel costs due to route optimization.
- 20% improvement in volume forecast accuracy.
- 30% reduction in operational planning time.
- Rapid access to AI capabilities without a lengthy recruitment process and the risk of an unsuccessful hire.
The client gained a scalable AI environment and a foundation for further automation of logistics processes.
