Dock operations analytics
Dock Vision AI
- Status
- Competition build
- Role
- Engineer
- Organisation
- IEEE Hackathon to commercial deployment
- Year
- 2024
The problem
A loading dock’s turnaround - gate, docked, unloading, gone - is tracked by eye and on clipboards, so a slow stage is only noticed after the truck has left and the detention charge has already landed.
Outcome
- Placement
- 1st placePlacementIEEE Hackathon 2024.
- Afterlife
- Paid deploymentAfterlifeConverted into a commercial installation.
Operations view: the live dock feed with people and vehicles detected, beside the turnaround event feed.
Approach
- 01
Built a real-time video-analytics system that watches the dock feed, detects the people and vehicles in frame, and calls each stage of a truck’s turnaround as it happens.
- 02
Wrapped it in an operations view, a supervisor view and an executive dashboard - a live event feed, turnaround-stage timing, and the cost of the delays it flags.
- 03
Took it from an IEEE hackathon build to a paid commercial deployment - the part most competition projects never reach.
Stack
- Python
- OpenCV
- Object detection
- Video analytics