Toyota, Accenture tackle traffic woes with AI taxi fleet system

AI applied to taxi services in trials have predicted demand with a 94.1 percent accuracy rate.
Written by Charlie Osborne, Contributing Writer

Toyota, JapanTaxi, KDDI, and Accenture have announced plans to pilot an artificial intelligence (AI) taxi dispatch system in Japan.

Congestion and traffic in urban areas and major cities are common. When the general public needs to order a ride, traffic, weather events, local activity and more can cause delays and a lack of availability.

While you expect delays at some times of the day or night, higher rates during times of demand, and disruption when traffic incidents or poor weather impacts services, the potential loss of revenue and custom can be a serious issue for taxi companies.

Japan's rush hour traffic is almost legendary, with congestion and peak traffic often carrying on after 10 pm on both roads and train lines.

AI has been billed as a way to reduce the impact of disruption and to better predict how much demand there may be for services, which can not only improve customer service but boost revenue for firms themselves.

Toyota, JapanTaxi, KDDI, and Accenture hope to capitalize on this through the development of a taxi dispatch support system which can predict demand for taxi services.

The system combines historical data, weather reports, local event schedules and public transport availability before using AI learning models and algorithms to predict the demand for hired transport.

"The system uses artificial intelligence (AI) to predict the number of occupied taxis in Tokyo using 500-meter, mesh-based parameters every 30 minutes," the firms say. "As a result, more taxis can serve areas with higher demand and insufficient supply to reduce waiting time for passengers while taxi occupancy rate is increased with optimal vehicle-dispatch."

Toyota is providing the systems to process and analyze data through its Mobility Services Platform (MSPF), JapanTaxi will be developing an application which offers taxi demand prediction data to taxi drivers, while KDDI has shouldered the task of creating a new demographic prediction system which monitors the flow of people in cities.

Accenture is collaborating with each company for the development and refinement of an analytics algorithm which powers the taxi demand prediction engine.

A pilot has already begun in the metropolitan area of Tokyo, Japan. According to the companies, the system has managed to predict consumer demand with an accuracy rate of 94.1 percent.

Due to this level of accuracy, more cars were provided at peak times, which is a benefit for passengers. However, drivers have benefitted too, by recording an average sales increase of 20.4 percent during the month of February, in comparison to an average overall increase of 9.4 percent throughout the trial.

The trial will be rolled out to new taxi fleets later this year.

These companies are not the only players on the field which are turning towards AI to tackle traffic and commuting challenges.

In February, Sony announced a partnership with six Japanese taxi companies -- Green Cab, International Automobile, Kusumi Transportation, Daiwa Motor Transportation, Checker Cab Radio Cooperative Association, and Hinomaru Transportation -- to develop and launch an AI-based taxi service.

See also: Ford's self-driving cars to become Miami's new pizza delivery guy

The firms will establish a joint venture that will focus on creating dispatch services through AI technologies that are capable of consumer demand forecasts.

Sony says that AI can improve customer services by fine-tuning demand predictions and "prepar[ing] the necessary number of taxis when necessary."

Nanowires, silver, and AI: The future of our smartphones (in pictures)

Previous and related coverage

Editorial standards