Skip to content | Change text size
 

Honours Supervisors, the Research Teams and Project Topics
    More projects to follow

Honours Projects 2010

Psychology

Human Factors Group – Michael Lenné

Project Title: The effects of music player use on the driving performance of young novice relative to experienced drivers

The project will use a desktop driving simulator to explore the effects on driving performance of interacting with a music or mp3 player while driving. Of particular interest here are the effects on young, newly licensed drivers relative to older, more experienced drivers as young drivers have a relatively higher risk of crash involvement and are more likely to bring into the vehicle music players, such as the iPod, than other driver groups. The outcomes of this work would have implications for raising public awareness of the issues associated with music or mp3 player use while driving, as well as for music player interface design, driver licensing and training.

Project Title: A simulator study of the effects of singing on driving behaviour.

Listening to music is one of the most prevalent activities in which drivers engage. Although research in the area is limited, there is some evidence to suggest that certain aspects of listening to music (e.g., genre, tempo) can affect driving behaviour, and it is often cited as one of the major driving distractions, especially for young, novice drivers. Activities that involve a substantial cognitive component have been demonstrated to distract drivers and impair driving performance. The present study will investigate whether singing along to music while driving, which is assumed to involve a greater cognitive component than listening to music alone, impairs driving performance.

Project Title: Driver behaviour at level crossings and comparison with signalised intersections

Recent MUARC research has used the simulator to study driving behaviour at level crossings with different controls (e.g., stop signs, flashing lights, etc). Following this work, this project will further explore these issues by using MUARC's advanced driving simulator to study driver behaviour on approach to level crossings and to make comparisons with behaviour at normal road intersections. The outcomes of this work will contribute to debate about the operation of controls at level crossings to improve safety.

Using Cognitive Work Analysis for railway level crossings collision countermeasure specification

Across Australia in 2007 there were 57 collisions between trains and vehicles at level crossings (Australian Transport Safety Bureau, 2008). An understanding of road user behaviour at level crossings is required to inform the development of appropriate in-vehicle technologies (e.g. Intelligent Transport Systems) and road infrastructure modifications (e.g. signage, warnings, traffic lights) to reduce such incidents. This project involves the application of the Cognitive Work Analysis framework (Vicente, 1999) to model the rail level crossing system. The CWA framework comprises five phases, each modelling different constraint sets: Work Domain Analysis (WDA), control task analysis (CTA), strategies analysis, Social Organisation and Co-operation Analysis (SOCA), and worker competencies analysis. The outputs of the analyses developed will be used to propose countermeasures for preventing collisions at railway level crossings.

Behavioural Safety Science – Judith Charlton

Project Title: Driving Behaviours of Young Drivers with Attention Deficit and Hyperactivity Disorder (ADHD)

Supervisors: Dr Judith Charlton, Sjaanie Koppell and Carlyn Muir

Discipline suitable to undertake project: Psychology; Behavioural Neuroscience; Occupational Therapy

In Australia , young drivers (aged between 17-24) have a higher rate of crash involvement per 100,000 population or per licensed drivers compared to any other age group (OECD, 2006).

Some of the reasons that have been proposed to explain young peoples' heightened crash risk include the fact that they are physically and psychologically less mature, their brain development is incomplete, they are less able to assess risk than more mature and experienced drivers, and more likely to engage in risk-taking driving behaviours (e.g. speeding, driving under the influence of alcohol or drugs and not wearing seatbelts). Additionally, medical conditions affecting mental health in young people may further contribute to this group's risk. This project will focus on one such condition - Attention Deficit Hyperactivity Disorder (ADHD) – and will examine driving behaviour amongst young novice drivers with ADHD. ADHD is characterised by a range of impairments including difficulty in planning; difficulty sustaining focus, shifting focus from one task to another, or filtering out distractions; impaired processing speed; difficulty utilising working memory and impulsivity - all of which can interfere with the safe operation of a motor vehicle. The ability to selectively attend to appropriate visual cues and sustain focussed attention in a traffic environment is essential for safe driving; yet despite the critical role of vision in driving, few studies have examined the visual scanning patterns of young drivers with ADHD. This study will use a traffic hazard perception task and eye tracking technologies to examine whether these abilities are compromised in ADHD.

Engineering

Road engineering - David Logan and Bruce Corben

Intersection Conflict Analysis

In this project, one or several intersections will be studied where a recognised traffic/vehicle conflict situation exists. For example, one intersection which has been identified involves a relatively high number of crashes, particularly between passenger vehicles and heavy vehicles. A potential solution would be to implement better timed signalling at the intersection, such as integrating right-turning arrows. However, the exact source of the conflict first needs to be identified in order to make any recommendations. This project will involve analysing one or more such intersections for crashes, near-missed crashes and other conflict situations and the circumstances under which they occur. The data will be analysed in order to find any trends in these situations.

IT

Vehicle safety - David Logan and Bruce Corben

Electronic collection of data for real-world crash investigation

This project will involve the development of an electronic data collection system for the in-depth data MUARC collects on real-world crashes. This data is comprised of a vehicle inspection, crash site inspection, medical reports, a patient interview and a police report. Currently, the data is partially electronically stored but also involves records on hardcopy case files. In this project, the student will explore the most efficient method in which to collect and store the data electronically, potentially how to link the data and then implement some or all of these methods.

Statistics

Injury Analysis and Data – Stuart Newstead

Statistical methodology for vehicle safety research (two projects)

For many years, the IAD team has undertaken a programme of research rating the relative safety of vehicles in protecting their occupants in the event of a crash. The primary product is an estimated set of crashworthiness ratings for both individual makes and models of vehicles as well as by broad market group. The crashworthiness rating is an estimate of the risk of death or serious injury to the driver of a vehicle given involvement in a crash where either someone is injured or a vehicle is towed from the crash scene. It is defined as the product of two conditional probabilities, namely the risk of sustaining an injury of any severity given involvement in a tow-away crash (injury risk) and the risk of death or serious injury given some level of injury is sustained (injury severity). To date, each of the crashworthiness component measures has been estimated using logistic regression analysis.

Project 1

Of interest is to see whether generalised estimating equations (GEEs) can be applied to estimate each component of the crashworthiness measure. Use of a Generalised Estimating Equation approach offers the advantage of taking into account potential correlations within the crash data used to calculate each of the injury risk and injury severity components. The objective of this project is to assess how the use of GEEs affects the estimated crashworthiness ratings and whether it reduces the variance of the estimates.

Project 2

One of the weaknesses in the current analysis approach used to estimate the crashworthiness rating lies in the estimation of its variance. As described, the crashworthiness rating is not estimated directly but is a product of two components. To estimate the variance on the crashworthiness rating, it is first assumed that the two component measures are independent and then an asymptotic formula is applied to estimate the crashworthiness variance from the variance estimates of the component measures. The aim of this project is to apply bootstrapping techniques to calculate the variances of the crashworthiness ratings and to compare this with estimates derived using the asymptotic formula.

Statistical methodology for road safety programme evaluation

The evaluation of road safety programmes is vital to ensure countermeasures are achieving their desired outcomes, and providing the best possible outcomes for the investment in them. The IAD team has been involved in various evaluations of major road safety programmes relating to areas such as infrastructure, blackspots and speed cameras. In particular, the programme evaluation of road safety interventions often involves the use of time series to model crash-based data. The IAD team has also used time series modelling for the examination of factors influencing road crash and injury trends and the projection of future trends in key road safety target areas and road user groups.

Various techniques are available for the analysis of time series. These include classical linear regression, generalised linear models (GLMs) and nonlinear models. In particular, GLMs and non-linear models can be used to overcome some of the restrictions of classical linear regression however they do not take into account time dependencies between consecutive observations in a time series. To deal with this issue, alternative time series techniques such as Generalised Estimating Equations (GEEs) and ARMA (Auto-Regressive Moving Average) could be used. The objective of this project is to examine and compare the use GEE and ARIMA techniques with traditional Generalised Linear Models using crash-based data including an assessment of the extent to which time dependencies or correlation exists within such data.

An honours project in MUARC will compliment undergraduate and Honours coursework in many discipline areas. For further information please contact :
Dr Melanie Franklyn, Honours Co-ordinator, Monash University Accident Research Centre
Tel: (03) 9905 4684, Email: Melanie.Franklyn@muarc.monash.edu.au