After three years of research and development, the KIBA project – Artificial Intelligence and Discrete Load Optimization Models to Increase Utilization in Combined Transport – was successfully completed.
Under the leadership of Kombiverkehr KG, the Deutsche Umschlaggesellschaft Schiene - Straße (DUSS), Goethe University Frankfurt am Main, Inform, KombiConsult, the Technical University of Darmstadt and VTG have developed a demonstrator for a network capacity-control and train-loading-optimization system.
The project was funded by the Federal Ministry for Digitalization and State Modernization (formerly the Federal Ministry for Digital and Transport). Kombiverkehr Managing Director: Heiko Krebs:
“With KIBA we have shown how AI can make rail freight transport more efficient. The developed prototypes help to better utilize train capacities, use resources more efficiently, and make
Combined Transport more attractive. This is an important contribution to shifting from road to rail and thus also to climate protection.”
The prototype developed within the project is now being prepared for near-term use by terminal operators and operators in rail freight transport.
Maximize train loading capacity
By means of artificial intelligence (AI) and mathematical optimization, efficiency, safety and sustainability in rail freight transport were to be increased. Supported by a central master data database, methods for network optimization and for train-loading planning could initially be developed. These results were prepared for users in a web-based visualization in an understandable way.
According to the operator, the developed models for train-loading planning ensure that the capacities of the trains in terms of
loading weight and length are utilized as fully as possible, crane routes and transshipment processes are reduced, and numerous variables are automated and simultaneously taken into account.
Short transit times and few transfers
The network planning combines AI-based demand forecasts with mathematical optimization to distribute loading units so that trains are optimally utilized and transports reach their destination with short transit times and few transfers. Dr. Rafael Velásquez, Director of Optimization & Integration at Inform:
“The combination of AI and optimization opens up entirely new possibilities in combined transport. Forecasts of bookings for the transport of loading units can be directly translated into optimization procedures, enabling trains to be efficiently loaded and networks to be controlled more stably. This results in
a practical approach that directly supports operating systems and increases the performance of rail freight transport.”
KIBA not only represents progress for Combined Transport, but also symbolizes a bridge between theory and practice. Krebs:
“The close collaboration between Kombiverkehr, INFORM, DUSS, KombiConsult and the two universities was the decisive success factor for the project. Only by pooling the different competencies could such an innovative result be achieved.”
With the project completion, an important foundation has been laid to further test the developed solutions in practice and to integrate them into existing systems. For productive deployment, further steps are necessary, including ensuring data quality, automating the exchange of information, and conducting live tests with the respective production systems, according to the project
