Fujitisu Zinrai's Artificial Intelligence Handles Tokyo Port Management Better People

According to Japanese authorities, there are 280 accidents annually in the extremely busy Tokyo Bay. In order to prevent them, since last year Fujitisu Zinrai, a specialized maritime safety institute, has been working here.

The port AI became operational in December 2019, and the first phase of its testing was completed in March 2020. It turned out that the electronic brain had issued twice as many warnings and orders to change ships' course as human dispatchers usually do. Thus AI is not reinsured - speed of passage of ships through water area has in its settings the same priority as safety of movement. The AI's task is to prevent critical indicators for each of the port subsystems.

Fujitisu Zinrai continuously monitors the parameters of ships and simulates their routes and traffic characteristics. This enables the mutual position of all objects in the water area to be calculated in advance, risk zones to be predicted and ships to be deployed. On average, the AI predicts the creation of complex situations two minutes faster than a person, which gives enormous maneuvering time. The accuracy of the instructions given to individual ships also increases.

The final decisions are still made by the people, who are responsible for giving commands. The AI problem is reduced only to the analysis and forecasting of critical situations. And as a result of the tests, the port authorities have decided to leave the system and expand its area of operation. In the future, similar systems will appear in all ports of the country.

Post a Comment