GermanyLogistics and transport2020 – ongoingHamburger Hafen und Logistik (HHLA)
Machine learning to predict container dwell time and outbound mode
HHLA used machine learning at two Hamburg terminals to store containers by predicted pickup time
Problem
Containers stacked without knowing when they will leave have to be moved again to reach those picked up earlier. Each extra move costs yard capacity and time.
Approach
HHLA teams and its consulting subsidiary HPC Hamburg Port Consulting built models that learn from historic data. At the Altenwerder terminal (CTA) the model forecasts each container's pickup time for the automated block storage. At the Burchardkai terminal (CTB) it also predicts whether a box leaves by truck, train or ship, and helps assign yard slots.
Results
No published figures
Takeaway
Predicting when and how a container leaves lets a terminal place it correctly the first time, which reduces rehandling.
Categories
- Size
- Large enterprise
- Industry
- Logistics and transport
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
HHLA announced that machine-learning models were running at CTA and CTB to predict dwell time and outbound transport mode. It reported a clear positive effect because containers need fewer moves. The release gives no figures.
Full details are in the original sources.
