Fujitsu Limited announced that its comprehensive traffic simulation system, developed under contract for the Ministry of Land, Infrastructure, Transport and Tourism (MLIT)-led regional transportation DX promotion project COMmmmONS, has been adopted for the Maebashi City Regional Public Transportation Plan, published by the local government on March 23, 2026. Analysis carried out by the system is included in the plan as scientific evidence supporting the policy to increase bus routes, one of the plan’s key measures.
Across municipalities nationwide, addressing the needs of transportation-disadvantaged residents and responding to carbon neutrality in the transportation sector have become urgent challenges, driving the need to advance and modernize public transportation systems.
In Maebashi City, challenges such as demographic changes, increasingly diverse mobility needs, and a shortage of bus drivers have emerged. As the city examined optimal bus route reorganization measures under the Maebashi City Regional Public Transportation Plan, it required robust and credible scientific evidence to substantiate the effectiveness of these measures.
Fujitsu was selected for the COMmmmONS project in April 2025 and developed a system capable of simulating both fixed-route and demand-responsive transportation, a first for Japan. The utility of the simulation results generated by this system was subsequently recognized, leading to its adoption in Maebashi City’s regional public transportation plan.
The comprehensive traffic simulation system leverages Fujitsu’s social digital twin technology to support the pre-verification of measures by simulating human and social behavior. It utilizes generally available statistical data on resident attributes, movement, and destinations, as well as ridership data obtainable from MaaS apps.
1. Evaluation of policy effects through high-precision simulation
The system utilizes unique Fujitsu technologies: artificial population technology, which generates resident data reflecting regional characteristics based on over 10 statistical anonymized datasets including the National Census; a behavioral selection model that uses AI to learn real-world travel data (i.e., travel times, costs for potential routes, residents’ ages, car ownership status, etc.) and reproduce the travel mode selection characteristics of local residents; and multi-agent simulation technology, capable of simulating multiple transportation modes with different characteristics. This enables the estimation of detailed resident travel demand and generates simulation results that closely reflect real-world travel conditions, even when actual travel data is insufficient, thereby accurately evaluating the expected effects of policies.
2. Streamlining plan formulation and accelerating consensus building
The system handles travel demand forecasting and modal split estimation as well as policy consideration and visualization of simulation results. Utilizing this system in regional transportation plan formulation can streamline the process, reducing the time required for consensus building with stakeholders, particularly transportation operators, by approximately 25%. The process previously could take between one and two years when outsourced to consulting firms.
3. Support for optimal plan formulation with multi-faceted evaluation indicators
The system provides a wide range of evaluation indicators, including usage status and service levels for each transportation mode, ride-sharing rates for demand-responsive transport, and overall project costs for measures. This enables a comprehensive evaluation of the impact of transportation policies on users, operators, and the entire region, supporting the formulation of optimal plans to resolve “transportation deserts” where securing public transportation is difficult, and to realize sustainable regional transportation.
Fujitsu plans to commercialize this system as a service by fiscal year 2026, developing it into a standard tool applicable across Japan. It will also promote collaboration with partners engaged in optimizing regional transportation, including local governments, consulting firms, and transportation operators. Through these efforts, Fujitsu aims to support the formulation of regional public transportation plans for local governments across Japan.
Fujitsu will continue to train the system using mobility data and other sources to establish it as an AI engine capable of accurately replicating the diverse behaviors of local residents so that it can contribute to urban development and community planning nationwide.
Under Uvance, Fujitsu‘s business model to address societal challenges, it will advance sustainable cities where everyone can live comfortably by enhancing regional transportation through data and AI.
