Japanese company Fujitsu is working with researchers at Carnegie Mellon University to develop social digital twin technology for use by smart cities.

A social digital twin digitally reproduces the relationships and connections between people, goods, the economy and society to offer a simulation, prediction and decision-making environment in which to solve diverse and complex social problems. This research is the first attempt between Fujitsu and CMU to explore future applications of social digital twins in global communities.

A project through CMU’s mobility data analytics centre will leverage real world data, including input of traffic regulations and the movement of vehicles, to evaluate the effectiveness of measures designed to estimate and control traffic flow dynamically.

Another project with the CMU’s computational behaviour lab in the School of Computer Science’s Robotics Institute will extend current capabilities in 3D modelling of pedestrians and forecasting their behaviour over time in urban environments. This technology can be used to monitor activity on streets and determine where issues or accidents may be taking place.

CMU’s efforts will be led by Laszlo Jeni, director of the computational behaviour lab, and Sean Qian, director of the mobility data analytics centre. Fujitsu and CMU will draw on the findings of these projects to create foundational technologies for social digital twins that will simulate traffic networks and movement patterns of people in real time. That work will build off of the deployment of the researchers’ projects with CMU’s transportation research institute, Traffic21.

The researchers anticipate that the social digital twin technology will play an active role in improving efforts to ease congestion, positively influence travel behaviour and ultimately help to realise more sustainable and safe cities.

Within this project, Fujitsu and CMU leveraged so-called converging technologies that combine computer sciences and knowledge from the humanities and social sciences, aiming to solve diverse and complex problems faced by cities working towards the realisation of a sustainable society. The researchers aim to develop a platform that delivers a broad set of services for various social issues based on accurate simulations of the movements of people and vehicles and the ability to visualise and predict future actions and possible risks based on human behavioural characteristics.

By using the social digital twin platform to analyse and predict the behaviour of people and movements of vehicles, the effects and potential risks of interventions can be reflected in advance to optimise outcomes of urban planning and policy.

Fujitsu and CMU’s research will initially focus on developing sensing technology to understand people’s movements, improve behaviour forecasting through artificial intelligence, and create social digital twin models to simulate how people interact with goods, the economy and society.

The research will include a social digital twin model based on real-time traffic data from road networks that can dynamically understand the daily changing traffic demand of the city. Researchers will be able to use the digital models to test urban traffic technology to adjust traffic regulations and toll systems in accordance with the efficient traffic flow.

In addition to the analysis of traffic congestion and ways to deliver economic efficiency, Fujitsu and CMU will further leverage the social digital twin platform to promote the verification of detailed measures to solve environmental problems, including the reduction of CO₂ emissions and improving urban transportation networks.

The researchers will additionally continue their efforts to contribute to the realisation of safe and sustainable next-generation smart cities by promoting measures to mitigate pandemics and ensure the flexible, efficient allocation of medical resources while driving economic growth.