Toyota Partners on Smart Taxi Service in Tokyo

Toyota Motor Corp. has developed a software tool to help taxi drivers and fleets anticipate demand in different locations based on real-time and historical data.

The carmaker is working with JapanTaxi Co., KDDI Corp. and Accenture plc on the program. During initial pilot tests last month in Tokyo, drivers who used the system posted 20% higher sales than the same period last year vs. a 9.4% overall average increase for the period, according to the partners. They claim the system had a 94% accuracy rate during the tests.

The companies integrate taxi service log data and demographic information from smartphones with other factors that affect taxi demand, such as weather, public transport availability and local events. Artificial intelligence-based learning models then are applied to predict taxi demand throughout the city.

Participating drivers have a PC-tablet that shows the predicted number of occupied taxis and the number of unoccupied taxis around an area. This allows drivers to position their vehicles based on supply and demand. Drivers also can receive supporting information that shows the routes where they are likely to find passengers.

Toyota processes and analyzes data and provides it to JapanTaxi, which is developing a demand prediction app for taxi drivers. KDDI is working on the demographic prediction technology, and Accenture is responsible for the AI-based analytics algorithm.

A wide-scale rollout of the system is planned for later this year. In the future, the partners also plan to study correlations between an analysis of driving images collected from Toyota’s data-transmission driving recorder that is being installed on an increasing number of taxis.​​​​​