Traditional and innovative speed-reducing measures for curves: an investigation of driver behaviour using a driving simulator
Introduction
Based on accident modelling and empirical research, there is a strong relationship between speed and accidents (Salusjärvi, 1981, Finch et al.,1994). Increases in speed lead to increases in the risk of an accident occurring and accident severity. Different road classes have different accident rates associated with them. In the UK, rural, single-carriageway roads (A roads) have the highest rate of involvement in fatal accidents of any category of road (DETR, 1998) in the UK. They account for 1.1 deaths per 100 million vehicle kilometres, as opposed to 0.24 on motorways and 0.75 on urban A roads. Considering all injury accidents, an accident on a rural A road has the highest likelihood of being fatal: 2.2% of injury accidents resulting in death, against 1.4% on motorways and 0.7% on urban A roads.
Taylor and Barker (1992) found that 18.5% of accidents on UK rural single-carriageway A roads occurred on curves. A large proportion of these accidents were most likely caused by a driver travelling too quickly through a curve and either losing control of the vehicle, or being forced into a corner-cutting manoeuvre in order to maintain control of the vehicle. Such corner-cutting manoeuvres increase the likelihood of a collision with an oncoming vehicle. Rural curves are a safety problem as they are substantially older than many urban roads. New roads are designed for a particular design speed, considered to be “the highest continuous speed at which individual vehicles can travel with safety on the highway when weather conditions are favourable, traffic density is low and design features of the highway are the governing conditions for safety” (O'Flaherty, 1986). Many rural roads pre-date the design speed concept and tend to have a wide variation in the maximum speed at which different elements can be safely negotiated, despite the fact that the speed limit is constant. One of these elements are substandard horizontal curves.
There have been various attempts to decrease speeds on curves. A recent review (Comte et al., 1997) suggests that the provision of relevant and timely information to the driver is of paramount importance and the effectiveness of this information is enhanced if there is suggestion of enforcement. The provision of individual speed feedback, e.g. by displaying the numberplate of the speeding vehicle on a Variable Message Sign (VMS), has been found to be particularly effective (Helliar-Symons et al.,1984, Casey and Lund,1993, SWOV,1994, Garber and Patel.,1995). An underlying motivation to comply may exist whereby motorists believe that detection of their speed and sometimes their licence plate implies enforcement.
Traditional methods such as signing and traffic calming can be effective at reducing speed; however, their effects are localised in time and space. Advisory speed signs have been found to be more effective than general speed signs (Rutley,1972, Webb, 1980, Tenkink, 1990), especially if the reason for the advisory speed was apparent to the drivers. Overall, the literature on advisory speed signs is contradictory. It is unclear whether the provision of speed advice information encourages the driver to drive faster due to increased confidence, or whether it has the opposite effect of reducing driver speed (Donald, 1997).
Traffic-calming measures, e.g. transverse bars, have been found to be effective at reducing speed on curves (Hungerford and Rockwell.,1979, Agent,1980, Helliar-Symons,1981). Studies have identified that the accident rate for curves on rural roads is strongly related to the differences between the speed environment approaching a curve and the design speed of the curve (e.g. Koorey and Tate, 1997). The use of transverse bars was thought to be particularly relevant, therefore, in this study as the curves incorporated into the network were associated with considerably lower advisory speeds than the general speed limit around them. Previous research (e.g. Kaptein and Martens, 1998) has shown that providing drivers with audio feedback (using edgeline and centreline kamflex lines) some reductions in curve entry speed can be obtained. These were not included in the current study, however, due to design limitations.
Speed limiters (SLs) have been suggested as an alternative method of reducing speed and accidents. Two approaches are possible, the first being a blanket measure such that the system continually controls speed with reference to the external speed limit. Such a system has been trialled both in simulators (e.g. Comte, 1996) and on-road (Saad and Malaterre, 1982, Persson et al.,1993, Comte.,1999). Some of these studies have shown that behavioural adaptation arises in response to the system (e.g. the tendency to compensate for low speeds on stretches by driving faster in situations when the appropriate speed is lower than the current speed limit). The second approach involves using the SL only at accident blackspots, such as curves. This latter approach was used in the current study, whereby the SL was automatically activated on approach to curves in order to decelerate drivers to the advisory speed.
In this study drivers encountered curves in a simulated road network that were either treated with one of four implementations or untreated. The four systems (and their abbreviations) were:
- 1.
visual feedback using transverse bars with decreasing spacing (Bars);
- 2.
an in-car Liquid Crystal Display (LCD), located to the left of the instrumentation panel, displaying the advisory speed for the curve (In-car);
- 3.
a road-side VMS displaying the advisory speed and their numberplate (VMS); and
- 4.
a speed limiter that automatically reduced driver speed to the advisory speed (SL).
The evaluation of new technology should consider the effect of any additional demand it places on the driver that could lead to safety costs. A particular concern in this study was the effect of additional information (In-car display and VMS) on driver distraction. High visual workload has been shown to be associated with variations in driver performance measures such as reduced speed and increased lane deviation (Cnossen et al., 1997) and faster steering wheel reversals (Boer, 1996). As drivers were decelerating on approach to curves anyway, speed was not considered to be a valid measure. Instead, heading error was recorded as an indication of the amount by which drivers deviated from the correct path on the road. Another important aspect of new technology is driver acceptability. A questionnaire, developed by Van Der Laan et al. (1997) was used, allowing participants to express a preference between the different systems.
Section snippets
Experimental design
A two-factor (System×State) repeated measures design was used, whereby System consisted of five levels (VMS, In-car display, SL, Transverse Bars, Baseline), and State of two levels (System On, System Off). In the Baseline condition, there was no system activated and thus State had only one level (System Off) in this condition. Each participant completed five experimental routes, the ordering of which was balanced.
Participants
Thirty participants, 15 males and 15 females (mean age 32 and 30 years,
Results
The data were analysed using a multivariate analysis of variance. An analysis of variance revealed that curves of different radii (100 and 200 m) produced different results and they were thus analysed separately; however, there were no differences between left- and right-hand curves in terms of speed measurements so these data was combined in a single dataset.
Discussion
A driving simulator was used to evaluate the effects of speed management systems on driver speed choice. Drivers encountered curves in a simulated road network that were either treated with one of four implementations or untreated. The four systems employed ranged from an advice system to a fully automated SL. It was hypothesised that by providing information and speed advice to the driver, speed would be reduced on the treated curves. It was also hypothesised that the different systems would
Acknowledgments
Funds for the support of this research were provided by the EU Commission DGVII under the Transport RTD programme of the 4th Framework programme.
References (29)
- et al.
The effects of mobile roadside speedometers on traffic speeds
Accident Analysis and Prevention
(1993) - et al.
A simple procedure for the assessment of acceptance of Advanced Transport Telematics
Transportation Research C
(1997) Transverse pavement markings for speed control and accident reduction
Transportation Research Record
(1980)Steering Entropy in Dual Task Driving
(1996)- et al.
Traditional and raw task load index (TLX) correlations: are paired comparisons necessary?
- Carsten, O.M.J., Groeger, J.A., Blana, E., Jamson, A.H., 1997. Driver Performance in the Engineering and Physical...
- Cnossen, F., Brookhuis, K., Meijman, T., 1997. The effects of in-car information systems on mental workload: a driving...
- Comte, S.L., 1996. Response to automatic speed control in urban areas: a simulator study (ITS Working Paper, No. 477)....
External Vehicle Speed Control: On-road trials (Deliverable 10)
(1999)- Comte,S.L.,Várhelyi, A., Santos, J., 1997. The effects of ATT and non-ATT systems and treatments on driver speed...
Be Warned! A Review of Curve Warning Signs and Curve Advisory Speeds (ARRB Transport Ltd. Research Report No. 304)
Speed, speed limits and accidents (Project Report 58)
Control of vehicle speeds in temporary traffic control zones (work zones) using changeable message signs with radar
Transportation Research Record
Cited by (66)
Processing driving simulator data before statistical analysis by means of interpolation and an integral formula
2024, Transportmetrica A: Transport ScienceReviews and prospects of human factors research on curve driving
2023, Journal of Traffic and Transportation Engineering (English Edition)Evaluation of advanced curve speed warning system to prevent fire truck rollover crashes
2022, Journal of Safety ResearchDistractions by work-related activities: The impact of ride-hailing app and radio system on male taxi drivers
2022, Accident Analysis and PreventionEffect of chevron design on driver behaviour when encountering and passing through a dangerous curve
2022, Transportation Research Part F: Traffic Psychology and BehaviourEvaluation of advanced curve speed warning system for fire trucks
2021, Applied Ergonomics