Project increases accuracy of road network simulations
Alongside industrial collaborators TSS-Transport Simulation Systems, researchers from the Department of Computer Science have been awarded over £22,000 for their work on road network simulations by Department for Transport. The project completed at the end of January 2017 used new computing techniques to improve the speed and accuracy of road-use simulations by x33, which will allow for more informed investments, improved journey times and a better use of the existing road infrastructure.
Simulation is an essential tool used by transport planners for assessing proposed initiatives for congestion management and infrastructure design, but running large and detailed simulations requires a lot of computational power. This project applies novel University of Sheffield software (FLAME GPU) which utilises Graphics Processing Units (GPUs). With the addition of new algorithms and software techniques, an improvement in simulation runtimes and performance is expected.
Dr Paul Richmond, Project Lead, said: “Using GPU computing for transport microsimulation will provide a step change in computational performance allowing higher accuracy microsimulations to be used at greater scales to gain new insight within the planning and investment of infrastructure interventions.”
The commercial partner TSS-Transport Simulation Systems develops the well-known Aimsun transport microsimulation software and will ensure that the impact of GPU computing on traffic simulations is successfully and fairly measured.
This is a part of a larger investment of £2.5m made by the UK government in new concepts and technologies to future-proof the transport sector. The Transport Technology Research Innovation Grants (T-TRIG) cover all forms of transport and support high quality ideas, from concept to prototype.
John Hayes, Transport Minister, said: “This government is committed to delivering a much better transport system, and innovation is a vital part of that. We are already making headway and expect to see more smart ideas and technologies coming from this funding.”