In China, the Port of Qingdao reports that it was able to successfully berth a bulk carrier loaded with iron ore using an artificial intelligence (AI)-based system.
 
The port made use of its newly developed Berth Planning Agent, which is now being used to analyse elements such as vessel schedules and tide data, as well as other operational constraints, in real time. It takes but a few seconds for the system to identify the optimum berthing plan and relay all necessary information pertaining to that to operators of port equipment and terminal systems.
 
Previously, vessel berthing was one of the most labour-intensive steps in port operations. Its complexity was such that planners had to manually integrate 132 different vessel-related variables and additionally review plans several times a day.
 
Were a vessel’s schedule to be modified, this would mean that all calculations pertaining to it would have to be recalculated. The new system, which was initially put into operation in November of last year, is being constantly refined in line with real-world scenarios, something which affects each new vessel. According to the port, currently, AI berthing accuracy has reached around 80%.
 
AI USAGE SPREADS TO OTHER AREAS OF THE PORT
At Dagang, where general cargo and dry bulk operations are undertaken, security concerns have resulted in CCTV cameras now being linked to AI to improve security detection. Nowadays, each camera is aware of 208 protocols covering major port regulations and general rules. It therefore is able to monitor 99 potentially critical, high-risk scenarios, and so captures images and sends alerts as soon as a violation occurs. In other areas, AI monitors the conveyor belt system, reviewing video feeds to identify the presence of unwanted, foreign objects.
 
The port claims a 99.8% detection rate with minimal false alarms. On stacker-reclaimer operations, it uses combined camera and radar feed to accurately measure distances, avoid collisions, and handle material more intelligently. This has allowed automation in these operations of around 94%.