The Grand Lyon Opticities pilot project aimed to demonstrate that a prediction tool can help traffic operators to anticipate and mitigate peak congestion by forecasting the future network flow patterns that would result from various traffic management actions, providing accurate strategy comparison and thus implementing the best strategy in the field.
November 2013 – October 2016
Opticities is a 3-year project to test ITS innovations in urban contexts. Coordinated by Grand Lyon and supported by the European Commission Directorate-General for Research and Innovation, one of the working projects is to test the integration of traffic prediction tools into the existing traffic control systems of Grand Lyon.
Aimsun Live’s dynamic, high-speed simulation of large areas allows the traffic operators in Lyon’s traffic management center (CRITER) to visualize traffic conditions before they actually unfold. 3-4 minutes is all it takes to produce comprehensive traffic predictions for the next hour.
Aimsun Live continuously processes live field data, simulating vehicle movement inside the Lyon study area, which covers approximately 1,400 km of roads. By combining live traffic data feeds and high-speed simulations with emulation of congestion mitigation strategies, Aimsun Live can accurately forecast the future network flow patterns that will result from a particular traffic management strategy.
Operators can simulate different scenarios, according to different strategies / travel policies (based on control plan configurations), to assess their relative impact on the network.
Scenario results are ranked according to indicators initially defined. For each simulated scenario, the traffic state is displayed and then operators can apply simulated actions according to these results.
In order to compare scenarios five indicators were chosen: global fluidity; dynamic congestion; road level hierarchy; and indicators for pedestrian data. These indicators help traffic operators select the best strategy to apply not only to recurring congestion but also to unplanned incidents and events. This will enable operators to target specific areas of intervention to minimise personal journey times, to analyse the impact and effectiveness of strategies deployed and to build and refine a library of interventions/strategies for future use.
To provide precise predictions, the modeling team has to integrate the following data in to the Aimsun Live model of Lyon: static model, public transport data, the control plan for all traffic light controllers, traffic demand data, history of traffic data and definition of events and response scenarios. The historical traffic data is used to generate patterns for real time simulation. And finally, definition of events and response scenario is used to execute predictive simulations when special events occur.