Human Factors of Transport Safety and Environmental Issues

Driver Monitoring and Feedback
Project Aims and Objectives
To develop a driver monitoring and feedback system which will inform fleet operators and their drivers of driving styles and behaviours that are undesirable for safety or environmental reasons, in order to allow the prevalence of such behaviours to be reduced. The project will identify the most effective method of delivering such a system through feedback, training, and a combination of both.

Summary of Project
Previous studies (e.g. DRIVE2, SAMOVAR) have shown that the monitoring of drivers can have a positive effect on their subsequent driving behaviour. It is also widely viewed that providing individuals with feedback on performance is essential for the acquisition of complex skills such as those required during driving. Although the DRIVE2 and SAMOVAR studies demonstrated the positive benefits that could be achieved by monitoring aspects of driving behaviour, the data recording units only measured accident involvement. Additionally, whilst drivers were made aware of the presence of the data recorders installed on their vehicles, they received no feedback based on the recorded information.
This study aims to investigate the feasibility and effectiveness of monitoring more complex aspects of drivers' behaviour, and providing feedback to drivers based upon the recorded information. The study also hopes to provide information regarding the practical considerations and issues faced when attempting to implement such an initiative within a fleet environment.
In order to achieve these objectives, the study comprises two major aspects. First, the investigation of the parameters related to safe and efficient driving styles, which could be recorded by a commercially available data recorder unit. Second, how this information could best be provided to drivers within the constraints of the existing fleet management systems and its effectiveness investigated. The experimental design for the second of these aspects is discussed below.

Experimental Design
• Sample size Siemens VDO FM200 journey data recorder units have been installed in 50 RAC patrol vehicles. The 50 patrol vehicles were selected to be the same model (Ford Transits) and approximate age, and not to be due for replacement before the end of the current study. Each vehicle is driven by only one driver on a day to day basis. Therefore it has been assumed that the same 50 drivers will be responsible for these vehicles throughout the study. • Feedback content Within the first phase of the study, the parameters to be recorded by the FM200 units were determined. A selection of these has been chosen for the feedback information to be provided to the 50 patrol drivers taking part in the study. Two forms of feedback are being investigated within the current study. The two forms of feedback were determined based upon (a) the existing operating procedures within the fleet and the information that could practically be provided to the patrols taking part in the study, and (b) research knowledge of those forms of information that may hope to elicit changes in subsequent driver behaviour.
Feedback A: Monthly feedback information All 50-patrol drivers will receive this form of feedback. Every month, feedback information will be provided to each patrol outlining their performance on three parameters.
These parameters are:
i)Harsh Acceleration (time spent exceeding a pre-set threshold).
ii)Harsh Acceleration (time spent exceeding a pre-set threshold).
iii)Fuel Consumption (measured by a fuel flow meter).

In addition to absolute figures of the individuals' performance on each of the 3 parameters, the feedback sheet sent to each patrol discusses his/her performance in terms of how it compares with both other patrols in their group, and with all 50 patrols within the study. Comparisons will also be presented based on both his/her performance over the previous month, and his/her overall
performance since the start of the study.
Feedback B: Management Feedback
Of the 50 drivers taking part in the study, 25 were randomly selected to receive the second form of feedback. Feedback B is provided in addition to Feedback A. Once a month, and after Feedback A has been sent to the drivers taking part in the study, the 25 patrols will meet with their Supervisor. Within these meetings his/her Supervisor will go through a series of questions investigating both the patrols' perceptions of the information received as part of Feedback A, and whether they have attempted to change their driving in response to the information received regarding their driving performance. A standardised question sheet has been prepared by TRL and provided to the Supervisors in order to lead the informal discussion with the patrol driver.

Dependent Variables
Changes in drivers' performance will be measured in terms of the parameters recorded by FM200 units. Three months of data were collected by the units prior to the provision of any feedback information. This will be used to provide a simplistic baseline against which comparisons with data collected later in the study will be compared.


Lee Smith, Transport Research Laboratory, UK