Centre for Automotive Safety Research reports
Permanent URI for this collection
Browse
Recent Submissions
Item Metadata only Survey of cyclist travel and crash experiences in the Australian Capital Territory(Centre for Automotive Safety Research, 2024) Thompson, J.; Ponte, G.; Australian Capital Territory Road Safety FundCycling is a popular means of transportation with immense benefits to both individuals (e.g., health) and society (e.g., environmental). The proportion of the population who cycle increased in all states and territories of Australia between 2001 and 2010, with the Australian Capital Territory (ACT) having the highest proportion of cyclists. However, safety issues such as less than optimal infrastructure, near-misses, collisions with vehicles, and severe injury outcomes, could be holding back further increases in participation. This project investigated the daily travel and safety of cyclists in the ACT. Three hundred and twenty-one cyclists completed a survey about their travel, crashes, injuries, and perceived safety on the ACT road network. Participants were a broad cross-section of cyclists in age, gender, and home residence across the ACT. They regularly rode (81% rode 2 to 3 days per week or more) resulting in high average distances (average of 82 kilometres per week). Almost two-thirds (60%) identified as confident riders. Overall, 80% thought that Canberra is a safe place to ride. However, there were safety issues that they encountered on the ACT road system. They did not feel comfortable riding on many common road situations in the ACT, they reported a large number (272 locations) of infrastructure issues with the road network that affect cyclists, half (53%) had been involved in a crash, and two-thirds (68%) had experienced a near miss. These are likely to be substantial cycling barriers to less experienced cyclists, as well as individuals considering taking up cycling. Cycling-focussed infrastructure and facilities should be provided to encourage cycling and improve safety. It is hoped that the perspectives of the cyclists surveyed for this project can contribute to creating a safer environment for cyclists in the ACT.Item Metadata only Update to 2013 report: Driver incentive and reward schemes(Centre for Automotive Safety Research, 2023) Bailey, T.; This research was funded via a deed with the South Australian GovernmentThere is some evidence in Bailey (2013) that rewards can be effective in encouraging safer driving if they are specifically tailored to groups such as young drivers, some work road safety settings, and/or specific driving behaviours such as speeding. However, very few schemes are evaluated for effectiveness. Experimental studies of schemes since 2013, particularly those that reward specific driving behaviours and/or of specific categories of driver types generally afford favourable results, along with a more advanced understanding of when, where, how, and why reward schemes work well (or not). This research has been substantially assisted by sophisticated driver monitoring and feedback technologies such as telematics. However, reward schemes must be seen to use clear evaluation processes of drivers’ performances, as unfair systems may well lead to complaints and repudiation of the schemes. Moreover, not all drivers respond in the same way or to the same extent to reward schemes, thus making wide-scale reward schemes of questionable cost-benefit effectiveness. Some drivers, particularly some young drivers, are motivated more by a sense of thrill when engaging in risky behaviours, than in any tangible reward for driving safely; penalties are likely to be more effective for habitually risky drivers. Effective reward schemes in fleet settings are best embedded in a company’s safety culture. Fleet reward schemes can improve both eco-friendly behaviours and safe driving. Small financial rewards consistently given are likely to be more effective in improving safe behaviours than large rewards given occasionally. Reward schemes considered likely to bring most benefit include: (i) rewarding offence-free driving periods by provisional licence drivers; (ii) telematics-based monitoring of driver performance by insurance companies; and (iii) reward schemes for drivers in fleet settings. Importantly though, any establishment of new reward schemes or support for existing ones, should be accompanied by evaluations of those schemes’ effectiveness in improving driver behaviour, along with commitment to modify or abandon the schemes as needs be.Item Metadata only Mode choice and safety: How might a shift from car trips to walking, cycling, and public transport affect road safety?(Centre for Automotive Safety Research, 2023) Raftery, S.; This research was funded via a deed with the South Australian GovernmentIncreasing walking, cycling, and use of public transport is identified in South Australia’s strategic plans (Carbon Neutral Action Plan 2016-2020 and Operation Moving Traffic) to reduce the state’s carbon footprint and improve the operation of the transport system. The aim of this report is to investigate the road safety implications of a shift in travel modes from motor vehicles to active, yet vulnerable modes. To address this topic, a literature review was undertaken to address two key questions: What is the safest mode of travel? What happens when there are changes in travel mode? Studies of road user risk using a variety of exposure variables (e.g., number of trips, per kilometre travelled, or per time spent travelling) reveal a general hierarchy of greatest to lowest risk of motor cycle/scooter/moped > bicycle > pedestrian > passenger vehicle > bus. Factors contributing to risk among pedestrians, cyclists, and public transport passengers are examined. Further evidence regarding the reduction of passenger vehicles, the Safety in Numbers effect for pedestrians and cyclists, and the effects of mode shift is also reviewed. While an increase in injuries among vulnerable users can be expected, the net effect on road safety will be an overall reduction in total casualties, largely driven by a reduction in car related casualties. Improving safety for vulnerable road users will be important: a traffic system made safe for vulnerable road users will be safe for all road users.Item Metadata only An assessment of ACT road infrastructure for compatibility with Advanced Driver Assistance Systems(Centre for Automotive Safety Resarch, 2024) Mackenzie, J.; van den Berg, A.; Ponte, G.; ACT Road Safety FundTo explore the compatibility of the ACT road network with modern vehicle ADAS, an instrumented vehicle was driven throughout the Territory to collect data over a period of five days. Feedback from the consultation of ACT road infrastructure stakeholders was used to assist in the selection of roads for data collection, which included all main highways as well as significant proportions of the urban and rural arterial network. The instrumented vehicle was fitted with a Mobileye dev-kit and Video Vbox HD2 system which provided the capability to collect details about what a commercial-grade ADAS is able to “see” while travelling through the road network. There were 759,772 points of data collected over 1,349 km of roadway during the study along with the detection of 1,963 speed limit signs. This dataset was then analysed to investigate what details regarding line markings and speed limit signs the Mobileye was able to detect. These analyses were also augmented with additional data obtained from the Open Street Map road network. Based on the analyses, high-resolution maps were generated that show ADAS is likely to have a good compatibility with the ACT road network in general. Geographic datasets were also generated as an output, providing an opportunity for further analyses.Item Metadata only Collection and analysis of EDR data from crash-involved vehicles: 2021 summary report(Centre for Automotive Safety Research, 2023) Elsegood, M.; Doecke, S.; Ponte, G.; This research was partly funded via a deed with the South Australian Government and by the Victorian Transport Accident CommissionModern vehicles are fitted with Event Data Recorders (EDRs) that constantly record variables such as speed, seatbelt usage, accelerator/brake pedal position, and steering wheel angle. When a crash occurs, a snapshot of the final few seconds of these variables are saved on the EDR in addition to crash related variables such as change in velocity (delta-v) and safety system deployments. In 2017, CASR established a data collection process whereby a large number of crash-involved vehicles could be accessed regularly from a single location (an auction yard) and the EDR data downloaded. Additionally, the South Australian Police Major Crash unit provided EDR data to CASR, downloaded from vehicles involved in investigated serious crashes. In 2021, CASR successfully retrieved EDR data from 171 crashed vehicles, of which 143 (83.6%) had associated police vehicle collision reports. This collection has contributed to a current total of 827 EDR records with 639 matched to police reports and 171 injured occupants matched to hospital injury data. In the sample of cases collected by CASR from the auction yard, 27.1% of bullet (striking) vehicles, and 40.3% of free-speed vehicles were found to be speeding. The rate of seatbelt wearing for front seat occupants in the sample was 96.9%.Item Metadata only The contribution of driver fatigue to casualty and fatal crashes in South Australia(Centre for Automotive Safety Research, 2023) Thompson, J.; Wundersitz, L.; This research was funded via a deed with the South Australian GovernmentDetermining whether fatigue (referring to a person falling asleep or experiencing deficits in functioning due to being tired, drowsy, or sleepy) is involved in a motor vehicle crash is difficult. Consequently, there is high variability in estimates of the contribution of fatigue to crashes. This study explored the contribution of driver fatigue to casualty and fatal crashes in South Australia, as well as the role that fatigue played in the crashes, the risk factors that led to the fatigue and the countermeasures that could have prevented the crashes. Data from two sources were examined: in-depth at-the-scene investigations of casualty crashes by the Centre for Automotive Safety Research (2014 to 2019) and Coroner reports on fatal crashes (2014 to 2015). Fatigue-related crashes were identified through evaluation of various forms of evidence (e.g., from crash participant/witness interviews, events preceding the crash). Fatigue contributed to 4.3% of casualty crashes and 11.5% of fatal crashes. Most fatigue crashes occurred during daylight hours (72.4%), on high-speed roads (86.2%), and at midblock locations (89.7%), and most (82.8%) involved the driver falling asleep, leading to the vehicle departing its lane, and either rolling over or colliding with a roadside object or oncoming vehicles. The most common risk factors for fatigue were long distance driving (41.4% of crashes), no/reduced/broken sleep (27.6% of crashes), illicit drug use (17.2% of crashes), and abnormal work/sleep routines (17.2% of crashes). The countermeasures that could have prevented the most fatigue crashes were lane keep assist (could have prevented 79.3% of crashes), lane departure warnings (65.6%), drowsiness detection/warnings (55.2%), audio tactile centre lines (41.4%), and autonomous emergency braking (34.5%). Vehicle technologies, therefore, comprised four of the five most common countermeasures. This study was the first, of which the authors are aware, to estimate the contribution of driver fatigue to crashes within South Australia.Item Metadata only A novel low-cost Safe System-aligned treatment for regional and remote intersections(Centre for Automotive Safety Research, 2023) Mongiardini, M.; Stokes, C.; This project was funded through the Australian Government’s Road Safety Innovation Fund (RSIF) program, administered by the Office of Road Safety (Project Number RSIF2-74). In-kind support was also provided by the South Australian Department for Infrastructure and Transport (DIT).An intersection active warning system aiming to improve vehicle safety at intersections has been developed and trialled. This system, named RJAWS Lite, is an advisory variation of the existing mandatory-speed based RJAWS and provides two major visual warnings: (i) speed advisory on major approaches and (ii) run-through prevention on the minor approach. A controlled before-after analysis conducted on speed data from a trial at six rural intersections in South Australia indicates that RJAWS Lite can provide substantial safety benefits, albeit to a lesser degree than the full version of the RJAWS. Nonetheless, the RJAWS Lite involves considerably lower equipment and installation costs, combined with less restrictive regulations, than the mandatory full version of the RJAWS and so could be expected to be suitable for more widespread installation, which could therefore effectively increase road safety at a larger number of rural junctions. Future streamlining of the current large billboard-style RJAWS signage may likely contribute to improve driver compliance with the advised speed and therefore further increase the predicted safety benefits.Item Metadata only Injury crashes involving child cyclists in South Australia 2009-2018(Center for Automotive Safety Research, 2023) Elsegood, M.; Hutchinson, T.P.; This research was funded via a deed with the South Australian GovernmentThis report contains a straightforward statistical summary of child (aged under 16 years old) cyclists injured in road crashes reported in South Australia for the period of 2009-2018 obtained from the Traffic Accident Reporting System (TARS). A set of tables based on various characteristics of the crashes is presented, detailing the times of the crashes, the places, the site details and circumstances, the demographics of the cyclists, the demographics of the other vehicles and their drivers, and the outcomes of the crashes. The tables are also grouped by age groups of the cyclists, the postcodes of the crashes, and by injury severity of the cyclist. The purpose of this report is to stimulate further insight and investigation into child cyclist crashes.Item Metadata only Development and trial of a process to audit vehicle safety technologies in rural areas(Centre for Automotive Safety Research, 2023) Elsegood, M.; Mackenzie, J.; This research was funded via a deed with the South Australian GovernmentThe purpose of this pilot project was to devise and trial a method by which the safety technologies of a sample of the vehicle fleet in South Australia could be identified and compared between vehicles being used in metropolitan areas and rural areas. The trialled method proved successful in collecting vehicle data, associated with location. The collected data was analysed in several ways to showcase what would be possible with a large study. Alternative methods that may offer more accurate results were discussed.Item Metadata only Speed limit reductions in 2011 on South Australian high speed roads(Centre for Automotive Safety Research, 2023) Kloeden, C.; This research was funded via a deed with the South Australian GovernmentIn 2011 the speed limit on 864 km of South Australian rural roads was lowered from 110 km/h to 100 km/h. This study examined the number of injury and fatal crashes on the affected roads and on all other 110 km/h roads in the five years before the speed limit reduction and the five years after. On the roads where the speed limit was lowered, the raw number of injury crashes decreased by 6.9 per cent and the number of fatal crashes decreased by 20.0 per cent after the speed limit was changed. However, neither of these results was statistically significant. On 110 km/h roads where the speed limit was not changed, crashes over the same two time periods were examined. Injury crashes decreased by 16.4 per cent and fatal crashes increased by 14.6 per cent on these roads. If the changes on these roads are assumed to represent the general background changes on all roads then the net effect of the change in speed limit on the roads where the speed limit was lowered were a 9.5 per cent increase in injury crashes and a 34.6 per cent reduction in fatal crashes. However, neither of these results was statistically significant. More sophisticated modelling of yearly injury and fatal crash numbers for the roads where the speed limit was changed similarly failed to find any statistically significant effect of the speed limit change. Incorporating crash numbers on roads where the speed limit was not changed into these analyses also did not find any statistically significant effect of the speed limit change. These results do not imply that the change in speed limit did not affect the underlying risk of injury and fatal crashes on the roads where the speed limit was lowered. The change in speed limit would certainly have had some effect and the large body of research on the subject suggests that it would likely be to lower the underlying risk. In the current study, there are just too few crashes on the roads where the speed limit was lowered to be able to discern even a large underlying effect of lowering the speed limit on the risk of injury and fatal crashes. It is suggested that this study be repeated once 10 years of pre and post speed limit change crash data is available in 2022 when the 2021 crash data is finalised. Note that the analysis for this study was carried out in 2018 using the data available at that time.Item Metadata only An examination of trends in South Australian workers compensation claims(Centre for Automotive Safety Research, 2015) Kloeden, C.; Hutchinson, T.P.; Harrison, J.; SafeWork SAIn South Australia, the great majority of employees have their workers compensation claims reported to WorkCover SA. This study examined such claims for the financial years 2001/2002 to 2012/2013. Over this period, there was a 41% reduction in claim numbers. Three factors were identified as contributing to this change. The total number of hours worked increased by 14%. There was a movement away from high risk manufacturing jobs into lower risk community service and mining jobs (which would imply a 5% reduction in claims). The remaining underlying risk fell by 45% (5.4% per year). While this reduction is pleasing, it may not be fully representative of an actual improvement in safety. The more serious claims (11 or more days of work lost) accounting for the great majority of suffering and costs have only gone down by 15% over the same period, mostly between 2005/2006 and 2007/2008, and have even shown some increases in recent years. Clearly there is much left to do in improving worker safety in South Australia.Item Metadata only Crashes involving road users who have recently arrived in South Australia(Centre for automotive Safety Researchy, 2023) Thompson, J.; Lindsay, V.; Baldock, M.; This research was funded via a deed with the South Australian GovernmentThere have only been a limited number of previous road safety studies of recent arrivals (tourists and migrants) in Australia. The present report examined crashes in South Australia in which a road user (car driver, motorcycle rider, bicycle rider or pedestrian) who had recently arrived in Australia (i.e. country of birth was other than Australia and had been in Australia for one year or less) was injured and admitted to hospital. Records from the Royal Adelaide Hospital (RAH) in South Australia for 56 injured recently arrived road users from May 2014 to October 2017 were linked with police-reported crash data and results of forensic blood tests for alcohol and drugs. They comprised 2.7% of the total 2,037 crash-involved road users who were admitted to the RAH during this period, which indicates that they constitute only a small part of the road trauma in South Australia. Characteristics of the sample of recently arrived road users and their crashes were examined and comparisons made to 1,837 road users who were either born in Australia or had been in Australia for more than one year and who were admitted to the RAH over the same period. Recent arrivals were younger and more likely to be at-fault for their crashes. They did not differ from longer-term Australian road users in the likelihood of testing positive for alcohol and were less likely to test positive for a proscribed drug. Some of their crashes may have been due to unfamiliarity or confusion with either the road environment or the road rules in South Australia. Also, recently arrived car drivers were more likely to crash in rural areas and areas with high speed limits, to be involved in departed lane/off path type crashes, to experience longer delays in the arrival of medical care to the scene of their crashes, and to have been more severely injured. This report provided an important update on the road safety of a group of road users who have been largely overlooked by recent research. Countermeasures to improve their safety are discussed with reference to these findings.Item Metadata only Car drivers with an AIS2+ spine injury: Description of a sample from South Australia(Centre for Automotive Safety Research, 2023) Dutschke, J.K.; Lindsay, V.; Hutchinson, T.P.; Jones, C.; South Australian GovernmentCar drivers from a database of road users discharged from the Royal Adelaide Hospital (RAH) between 1 July 2015 and 30 December 2017 following a road crash were analysed in this study. The hospital information was supplemented with details of the crash, obtained from the Police and the road authority. This study focused on those who had a spine injury coded at least 2 on the Abbreviated Injury Scale (AIS2+). This sample is from a single hospital, and there are several likely biases. Results. There were 518 car drivers with one or more AIS2+ injuries, of whom 152 had one or more AIS2+ spine injuries, mostly vertebral fractures. Of these 152, the maximum spine AIS score was 2 for 87%, 3 for 12%, and >3 for 1%. Tables and Figures are provided for characteristics of the spine-injured drivers and their accidents: gender, age group, crash location, speed limit, crash type, impact type, vehicle year, seatbelt use, Injury Severity Score, and days in hospital. Discussion. The biases in the dataset mean that there is no suitable comparison group. Nevertheless, data is given for three groups that provide some context: car drivers with an AIS2+ injury who did not have an AIS2+ spine injury, car drivers who did not have an AIS2+ injury, and a sample from the TARS database. (TARS refers to Traffic Accident Reporting System, that is, the accident reports that originate with the Police.) The hospitalisation characteristics show that this cohort/sample AIS2+ spine injury group is more seriously injured than other AIS2+ injured car drivers.Item Metadata only Vehicle speeds in South Australia 2020(Centre for Automotive Safety Research, 2023) Kloeden, C.; Woolley, J.; South Australian GovernmentA systematic and ongoing method of measuring vehicle speeds was introduced in South Australia in 2007 in order to assess the effects of speed reduction countermeasures and to monitor the speed behaviour of South Australian motorists over time. More than 100 sites around South Australia have speed measurements taken for a one week period at the same time each survey year. This Report summarises the data collected in 2020 and makes comparisons with previous surveys and partial surveys dating back to 2002. The following general observations are apparent when considering all the surveys and all the speed measurements: vehicle speeds have been trending down on all road types over the course of the surveys; the percentage of vehicles obeying the speed limit has been increasing; reductions in high level speeding are more pronounced than those for low level speeding; vehicle speeds appear to be stabilising in recent years on 50-80 km/h roads while continuing to decline on 100-110 km/h roads; speed limit compliance is lowest on Adelaide 50 km/h collector roads and rural 100 km/h arterial roads; and vehicle speeds in 2020 appeared to generally increase from 2018 especially on 50 and 110 km/h roads (this may have been COVID-19 related).Item Metadata only Motorcycle protective clothing in the ACT(Centre for Automotive Safety Research, 2022) Baldock, M.; Thompson, J.; 2021 ACT Road Safety Community Grants ProgramThis study involved an investigation of the use of protective clothing by motorcyclists in the ACT, encompassing both commuting (n=33) and recreational (n=45) riders. Rider knowledge about MotoCAP, the star rating scheme for protective clothing that operates in Australia and New Zealand, was also examined. Rates of wearing of protective clothing among ACT motorcyclists was found to be high, especially for jackets and gloves. Gains could still be made through the promotion of the importance of wearing protective motorcycle pants, especially garments fitted with armour. Only a minority of riders were aware of MotoCAP and only a small proportion had ever visited the website. The high importance placed by riders on physical protection when riding suggests that there is a big market for the resources provided by MotoCAP and expanding publication of star ratings beyond the website may assist with reaching them, something which MotoCAP is currently pursuing.Item Metadata only A review of SA line markings for suitability with lane support systems: a video analysis trial(Centre for Automotive Safety Research, 2023) Elsegood, M.; Mackenzie, J.; South Australian GovernmentCustom video analysis software was used to determine when the lane support systems of two vehicles could identify the lane line markings along Victor Harbor Road, South Australia. The video analysis software proved successful and produced an average line marking detection rate of 97.8% between the two test vehicles. The locations and a heat map of areas where line marking non-detections occurred was generated. The situations in which non-detections occurred were potentially attributed to shadows over the line markings, newly formed lanes, uncommon lane dividing markings and absences of line markings. Potential solutions to overcome the non-detections along Victor Harbor Road are discussed.Item Metadata only Analysis of treatments on the Duke's Highway in South Australia(Centre for Automotive Safety Research, 2022) Kloeden, C.; South Australian GovernmentSince 2001 there have been a number of treatment programs applied to the Duke’s Highway in South Australia with the aim of improving the safety of the road. Between 2001-2017 the shoulders of the road have been sealed, overtaking lanes have been installed and audio tactile line marking has been installed. Since 2009, rest areas have been installed, more overtaking lanes have been installed, clear zones have been made and roadside hazards have been removed or protected. Starting in 2011, a wide centre line is being painted along the length of the road. Since 2013, the road has been resurfaced, intersections have been treated and signage has been added. Average speed cameras were installed in 2013 and activated in 2014. Taken as a group, these treatments would be expected to reduce the number of injury crashes on the Duke’s Highway beyond any pre-existing or general trend. Between 1983 and 2016 there were 742 injury crashes recorded on the Duke’s Highway in 100 and 110 km/h speed zones. Over that time period the number of injury crashes per year has been generally declining. For the 1983-2000 period the number of injury crashes declined by about 1.7% per year. For the 2001-2016 period the number of injury crashes declined by about 6.8% per year. The trends for South Australian injury crashes on all 100 and 110 km/h roads were a 1.9% reduction for 1983-2000 and a 3.0% reduction for 2001-2016. The observed reduction on the Duke’s Highway for 2001-2016 was greater than the previous trend on that road and the general South Australian trend during the same time period. This is highly suggestive of the Duke’s Highway treatments having resulted in fewer injury crashes on that road than if no treatments had been applied.Item Metadata only Profiling head on crashes in South Australia(Centre for Automotive Safety Research, 2023) Doecke, S.; Thompson, J.P.; Elsegood, M.E.; Ponte, G.; South Australian GovernmentThis report provides a profile of head on crashes in South Australia using police-reported crash data from the Traffic Accident Reporting System (TARS) and data from the Centre for Automotive Safety Research (CASR) at-scene in-depth crash investigations. TARS data revealed that head on crashes are: ten times more likely to be fatal, about 2.5 times more likely to occur in a 100 or 110 km/h zones, four times more likely to occur on a curve, and more likely to involve a truck or a motorcycle than all crashes. Only 5% of head on crashes occurred as a result of a vehicle overtaking. CASR’s at-scene in-depth crash investigations of 24 head on crashes revealed further insights. These including identifying the top contributing factors as a medical condition, fatigue, drugs and speed; the top mitigating interventions as centre barriers, speed limit reductions and wide centre medians; and the top prevention interventions as lane keep assist, wide centre medians and ESC. The higher injury severity associated with head on crashes highlights the need to better understand these crashes and to develop and implement effective interventions to mitigate and prevent these crash types and associated injuries.Item Metadata only Options for mitigating filter right turn crashes at signalised intersections based on video footage of crashes(Centre for Automotive Safety Research, 2022) Stokes, C.; Woolley, J.; South Australian GovernmentSignalised intersections represent locations along the road network where traffic managers exert the greatest control on road users. However, crashes do occur at signalised intersections and because of the combination of high speeds and right angle crash configurations, they are likely to cause harm to those involved. Filter right turns present substantial danger to road users due to the complexity of decision making, risk of severe outcomes and the lack of control being placed over road user movements. The main aim of this research is to use footage of filter right turn crashes at signalised intersections to identify and analyse crash mechanisms associated with these crashes, and to develop recommendations that can lessen or eliminate the risk of filter right turn crashes should full right turn control not be able to be employed. A number of recommendations are made and include fully controlling right turns or where filter turns are allowed, reducing through vehicle speeds and control right turns at off-peak times when volumes do not warrant filter turns. It is also recommended that this study be revisited on a frequent basis and that further research be undertaken to better understand specific issues of dynamic visual obstruction and drivers’ abilities to judge speed and distance at night.Item Metadata only A comparison of uncontrolled, give way sign controlled and stop sign controlled intersections in South Australia(Centre for Automotive Safety Research, 2022) Stokes, C.; Woolley, J.E.; This research was funded via a deed with the South Australian GovernmentThe Centre for Automotive Safety Research (CASR) has undertaken study to compare uncontrolled, give way and stop sign controlled intersections as used in South Australia. The answers to two basic questions are desired: is more control justified if safety is impaired; and do road users differentiate by the type of control employed? The use of different levels of control was investigated through a literature review, analysis of in-depth crash investigation files and a survey of road users in South Australia. There appears to be some benefit to the use of greater control but the reasons behind this are unknown. The survey results also suggest a residual of road users are confused and prone to error at controlled and uncontrolled intersections. Recommendations to help remedy this issue include adding redundancy at intersections through the use of Safe System design principles, controlling all uncontrolled T-junctions and upgrading superseded intersection advanced warning signs.