After 36 months of close collaboration, the European research project SETA has drawn to a close. Aimsun has been part of the consortium of 15 partners from five countries, working to improve how mobility is monitored, understood and managed in large metropolitan areas.
The team has now reported and realised its major milestones in the attempt to answer the following questions:
• Which technologies and methodologies can change the way mobility is organised, monitored and planned in large metropolitan areas?
• How can these tools support decision makers in evidence-based decisions to improve town planning and infrastructure?
• How can we provide support for individual citizens to plan their journeys in a more efficient and sustainable way?
The Aimsun research team worked particularly closely with TU Delft on exploring and developing mobility prediction models and the use of big data in simulation-based models, including:
Improving calibration techniques by adding new data sources
Extension of the state-of-the-art of demand estimation and calibration
Developments in advancing network representation
Travel demand prediction ( i.e., origin-destination flows, isolated local flows)
Traffic condition prediction (speeds, travel times).
Developments incorporated different techniques, including transfer learning, shapes models, multi-task learning, network science predictors, and meta-models—to obtain meaningful and scalable models that can be used in mobility planning and management.
The cities of Birmingham (UK), Turin (Italy), and Santander (Spain) provided real traffic data, including smartcard data and traffic counts inferred from traffic cameras, which were used to make significant contributions to the state-of-the-art in demand and traffic predictions for both car traffic and public transport networks. This research has influenced the development of Aimsun’s own software and will contribute to new features in upcoming versions of both Aimsun Next and Aimsun Live mobility modeling software.
Within SETA, Aimsun aimed to minimize the amount of manual editing required for development of the network model, relying instead on maximizing task automation to get a first stage network model, which could then be updated, calibrated and validated with direct traffic observations. The network graph was developed based on a high-quality network representation from OpenStreetMap (OSM) and other sources provided by Birmingham City Council, in an advanced Aimsun Next traffic simulation model. The team was able to calibrate and validate the network model using a system provided by The Floow, that processes floating car data with various built-in capabilities to derive estimates of the average hourly traffic flow and demand within 24 hours, and traffic count data provided by Birmingham City Council.
The SETA research project aimed to create an open, sustainable, ubiquitous data and service ecosystem for efficient, effective, safe, resilient mobility in metropolitan areas, using big data to understand and model mobility with a precision and granularity that has been impossible up until now.
SETA is funded under the Horizon 2020 framework program of the European Commission with a budget of 5,5 million euros, for the period 1 February 2016 – 1 February 2019.
Comune di Torino (IT)
The Floow Ltd (UK)
Scyfer B.V. (NL)
Knowledge Now Ltd (UK)