Transport modelling based on PNT data

Client Brief

Somerset County Council and Sedgemoor District Council required a cost-effective mechanism to test freight scheduling options on key routes around Sedgemoor. Understanding the key impacts of HGV movements was essential to ensuring the wider function of the transport network. The client wanted a mechanism to understand this without building a large-scale traditional transport model.


City Science Response

Working with Somerset and Sedgemoor, City Science secured funding from the UK Space Agency Space for Smarter Government Programme to develop a Digital Twin of the routes by pairing a model with satellite position, navigation and timing (PNT) data. By combining GPS data, vehicle counters and loops with detailed junction performance models, the team developed a ‘What-if’ analysis tool and the ability to provide ongoing checks to transport models. This enabled the user to explore the impact of different HGV schedules, understanding the effect of absolute numbers and scheduling on the road network.


As an output to the project, we developed an interactive notebook to test and understand a range of different scenarios. This was delivered at a fraction of the cost of a traditional transport model. The UK Space Agency was provided with a feasibility report, demonstrating the potential benefits of PNT data across a range transport modelling use-cases.
"The innovation shown by the City Science team, and the great work they’ve done in partnership with their public sector end-users has been fantastic. Not only have the team shown that they’re able to produce quick and low-cost traffic models from PNT data, they’ve also identified many significant benefits this capability could offer over traditional processes. Applications such as this could be helpful right across the UK, and I look forward to seeing how it develops."
- Iain Hughes, SSGP Projects Manager

Total length of queuing traffic at each junction in Bridgewater shown in City Science's Cadence software. Click to explore a demo model.