Quintessa

Modelling Traffic Flow in Henley

Placement student Adam Hazell has recently presented his work on traffic flow modelling to Quintessa employees, rounding off a year long project involving data analysis and software development.

In addition to the internal and client facing work they undertake during their twelve month work placements at Quintessa, students from the University of Bath are assigned a personal project to carry out. This allows them to develop their technical skills and research new areas that of interest.

Adam Hazell investigated methods of modelling traffic flow, with the aim of finding alternative traffic management systems for Henley-on-Thames that might ease the current level of congestion. Using data collected by the Department for Transport, he developed a software solution to simulate the town centre (Figure 1) utilising an adapted version of the Nagel-Schreckenberg model for traffic flow modelling.

The software implemented a novel way to model traffic, by allowing users to create and visualise an interlinked network of Nagel-Schrekenberg models, and run them at scale to produce relevant traffic flow data. It also allowed for the addition of road specific features such as traffic lights and right of way priority for a given road.

A sufficiently accurate model of Henley's current traffic management system was developed, and was then run with various alternative management systems in order to compare their effects on the resulting levels of congestion. Illustrative alternatives included the reversal of the one-way system in the town centre, and the construction of a new bridge across the River Thames.

The results from the first alternative model illustrated that the current implementation of the one-way system in Henley seems to have had a positive effect, as its removal resulted in a substantial increase in congestion throughout the day (Figure 2). The effect of the construction of a new bridge over the Thames was shown to depend greatly on its position relative to the town. If it were built to the south, the reduction in congestion appeared to be far less than if it were built to the north (Figure 3 and Figure 4) since town centre congestion is significantly impacted by the number of vehicles travelling into Henley from the north along the A4130.

An animated gif of simulated cars moving through a traffic network
Figure 1: The software simulating Henley town centre
A graph comparing traffic levels of the one and two-way simulations. The one-way system is much less congested throughout the day
Figure 2: A comparison of traffic levels between the one and two-way road systems
A graph comparing traffic levels resulting from reduced traffic volume from the Berkshire side of Henley by 0, 15, and 30%. There is not a significant reduction in congestion from these reductions in volume
Figure 3: A comparison of traffic levels resulting from reduced numbers of vehicles travelling to Henley from the south
A graph comparing traffic levels resulting from reduced traffic volume from the Oxfordshire side of Henley by 0, 15, and 30%. There are significant reductions in congestion from both reductions in volume
Figure 4: A comparison of traffic levels resulting from reduced numbers of vehicles travelling to Henley from the north