Hamburger Hafen und Logistik AG (HHLA) is one of the first ports worldwide to develop solutions for its Hamburg container terminals that use machine learning (ML) to predict the dwell time of a container at the terminal.

The first two projects have now been successfully integrated and implemented into the IT landscape at Container Terminals Altenwerder (CTA) and Burchardkai (CTB).

Machine learning to advance operations

“Advancing digitalization is changing the logistics industry and our port business with it. Machine learning solutions provide us with many opportunities to increase productivity and capacity rates at the terminals,” said Angela Titzrath, chairwoman of the executive board of HHLA. 

The HHLA Chairwoman announced that further uses for ML were bound to be identified.

For instance, the productivity of automated block storage at CTA will be increased by means of an ML-based forecast to predict the precise pickup time of a container saying processes are substantially optimised when a steel box does not need to be unnecessarily restacked during its dwell time in the yard.

"When a container is stored in the yard, its pickup time is frequently still unknown. In future, the computer will calculate the probable container dwell time. It uses an algorithm based on historic data which continually optimises itself using state-of-the-art machine learning methods," HHLA said in a statement. 

A similar solution is applied at the CTB, where a conventional container yard is used alongside an automated one.

In this case, ML will support terminal steerage by allocating optimised container slots. In addition to the dwell time, the algorithm can help calculate the type of delivery.

Significant impact cited

It said the machine learning solutions can predict whether a container will be loaded onto a truck, the train, or a ship much more accurately than can be determined from the reported data.

HHLA said a significant positive effect can already be seen at both terminals since the containers are stored based on their predicted pickup time and must therefore be moved less frequently.