by Oliver Schmitz, Research Associate, and Emin Nakilcioğlu, Research Associate, Fraunhofer Center for Maritime Logistics and Services (CML)

As container logistics grows increasingly complex, the ability to anticipate and respond to disruptions becomes a defining factor in terminal performance.

The KILOG project (Künstliche Intelligenz für Logistikoptimierung in deutschen Häfen/Artificial intelligence for logistics optimization in German ports) explores how artificial intelligence (AI) can strengthen operational precision and resilience at the Altenwerder (CTA) and Burchardkai (CTB) container terminals of HHLA.

Funded under the IHATEC II program (Innovative Hafentechnologien/Innovative port technologies), the project addresses intermodal use cases such as rail slot optimization, predictive maintenance, and container availability forecasts.

This article focuses on two of those initiatives: a container flow forecast that predicts yard blockages before they occur and on a pipeline, based on a large language model (LLM) that consolidates ship arrival information from heterogeneous sources.

Together, they reduce manual effort, enhance planning reliability, and provide the foundation for scalable, data-driven decision-making across terminal operations.

Download PDF