School of Engineering

Event items

Vehicle emissions research using telematics data

Speaker: James Tate (Leeds University)

Date/Time: 13 December 2017; 13:00-14:00

Venue: Cassie Building 2.32

High resolution, high quality telematics data is now routinely collected from a growing proportion of passenger cars and commercial vehicles. Telematics providers are not only collecting data for vehicle logistics and security services, but also to support the motor insurance industry develop products tailored to a customer’s use of a vehicle. Telematics data is therefore moving away from being biased by a limited sample of commercial fleets with professional drivers, towards pervasively surveying the real traffic movements and driving behaviour of a growing share of the general passenger car fleet.

Telematics data offers the opportunity to move away from modelling average traffic situations (e.g. weekday AM and PM peak periods) with aggregated emissions factors, to systematically studying the impact of the real-life variations in traffic flow and congestion over complex networks due to incidents, special events, changes in weather conditions and fluctuations in traffic demand.

A sample of over 56,000 kms of quality checked second-by-second real driving trajectory records has been recorded through a calendar year for a study area 1km west of Leeds City centre. The study area has been identified as the only urban area in the North of England (UK0004) to fail to achieve the EU nitrogen dioxide (NO2) annual average directive in 2020.

The links of most concern are the A58 and A58M (two and three-lane urban motorway). The road network incorporates several multi-lane and multi-level intersections which are frequently heavily congested inside and outside AM and PM peak periods. Traditional traffic and coupled emission modelling methods would struggle to model this complex environment with any degree of certainty.

An automatic number plate recognition (ANPR) survey was also conducted over the course of the week commencing 16 April on the main A58(M) road. These vehicle number plates were cross-referenced with the UK registration database so the local vehicle fleet mix in different time periods was known. The raw vehicle activity data has been supplemented with predictions of tail-pipe emissions (CO2, NOX, NO2, PM, PN, CO, HC) from the European instantaneous emission model PHEM (Hausberger et al, 2014), scaled according to the local fleet mix and total traffic flow (measured by in-situ automatic traffic counters).

The seminar will describe the methods and visualise the spatial and temporal variations in traffic flows, speeds and associated emissions. Work in progress is assessing the impact of different CAZ scenarios.