• PDF version [PDF: 1.8MB]

East Nippon Expressway Co., Ltd. (hereinafter referred to as NEXCO EAST) and NTT DoCoMo, Inc. (hereinafter, DoCoMo) is, CA Tokyo Wan Aqua-Line Expressway (hereinafter referred to as aqua line) "AI traffic jam prediction" in the demonstration experiments in ^{※ 1} of the 2019 12 month 20 days (gold) from E17 Kan-Etsu Expressway (below, Kan-Etsu Expressway) was applied, the same day 14 hours, 00 minutes to NEXCO EAST WEB site "of DraPla" ^{※ 2} in the forecast and the predicted required time for each 30-minute traffic demand ^{※ 3} Will be delivered.
Kan-Etsu Expressway is NEXCO EAST jurisdiction, and most of it occurs in the section from the Numata interchange (hereinafter referred to as IC) to the Nerima IC.
In "AI Congestion Prediction" on the Kan-Etsu Expressway, we will show you the forecast required time and forecast traffic demand according ^{to the start point and end point * 4} selected by the customer between the Numata IC and Nerima IC on the In-bound
Kan-Etsu Expressway should consider using it to avoid traffic jams, such as adjusting departure times, adding drop-in locations, and changing boarding / alighting ICs, based on the forecast information of "AI Traffic Congestion Prediction." I am.

Usage image

This experiment is based on the real-time version of mobile space statistics ^{* 5} (hereinafter referred to as "population statistics") created using the mechanism of the mobile phone network and the past results of traffic volume, traffic congestion, regulations, etc. owned by NEXCO EAST, By multiplying the "AI traffic jam prediction" technology developed by Docomo NEXCO EAST we can predict the required time and traffic demand after 14:00 from the number of people on the day. ..
Many Expressway, including the Kan-Etsu Expressway Expressway and traffic volume and congestion conditions change due to various factors such as seasonal changes in destinations, but Aqualine By expanding the "AI traffic jam prediction" technology being tested in Japan and newly establishing the following two technologies, it has become possible to apply AI traffic jam prediction to Expressway

1. Technology that predicts traffic demand at each location based on actual crowds that change depending on the season and weather
2. Technology that predicts travel time considering differences in traffic demand at each location

When comparing the required time prediction results of "AI traffic jam prediction" and the conventional traffic jam forecast calendar (traffic jam prediction), one-fifth (11% → 1.9%) of days have a maximum error of 30 minutes or more I was able to confirm a significant improvement.

• Travel time between Numata IC and Nerima IC at legal speed: approx. 79 minutes
• Evaluation target: Wednesday, April 15, 2015-Friday, May 31, 2019 (785 days excluding the date of accidents and regulations)
• Improvement rate: (Number of days of error in conventional prediction-Number of days of error in AI traffic jam prediction) / Number of days of error in conventional prediction

In addition, according to the customer questionnaire results of the "AI traffic congestion prediction" demonstration experiment on the Aqua Line, we were able to confirm that more than 90% of the customers have the intention to continue using it and have a high level of satisfaction. In addition, more than half of the customers who used "AI traffic jam prediction" were able to take actions to avoid traffic jams, such as delaying the usage time, confirming the effect of behavior change by "AI traffic jam prediction". I made it. In addition, Kan-Etsu Expressway was selected the most as a hope for future development destinations.

The demonstration experiment on Kan-Etsu Expressway will be conducted until the end of March 2020, and based on the verification of the effect, we plan to proceed with consideration for full-scale introduction in 2020, including expansion to other routes.

NEXCO EAST is developing a "safe, secure, comfortable and convenient Expressway service" in its medium-term management plan. In addition, DOCOMO is working on solving social issues together with its partners in the medium-term strategy 2020 “beyond declaration”.

Both companies will continue to study the further utilization of "AI traffic jam prediction" and work on solving traffic problems.

1. AI traffic jam prediction is a technology that constitutes the NTT Group's AI "corevo®".
2. [NEXCO EAST | NTT DoCoMo] AI traffic jam prediction (demonstration experiment)
3. Traffic demand is the number of vehicles that potentially pass through the Expressway at each time of day, and corresponds to the amount of traffic when there is no limit to the amount of traffic (traffic capacity) that can flow on the road.
4. [Starting point] Numata IC, Shibukawa Ikaho IC, Fujioka JCT, Hanazono IC, Tsurugashima JCT, [Ending point] Tsurugashima JCT, Nerima IC
5. Domestic distribution statistics (real-time version), which is one of the lineup of mobile spatial statistics. This information indicates the number of people in a group for each area or attribute, and cannot identify individual customers. The demographics used in this experiment comply with the "Mobile Spatial Statistics Guidelines," which summarizes the basic items for creating and providing mobile spatial statistics, in order to strictly protect the privacy of our customers.
   Mobile Spatial Statistics Guidelines

• "corevo" is a registered trademark of Nippon Telegraph and Telephone Corporation.