Search recently published papers
from the CARSP Conference
The Role of Human Factors in Road Safety Audits
Last Updated on May 25, 2022
Robert Dewar, Western Ergonomics Inc.
The road safety audit has come into use as a means of improving road safety. It involves a formal safety performance examination of either an existing or future road by an audit team. This team is usually small (2-5) with experts in road safety, road design, and traffic operations. Other expertise might include human factors, enforcement (police), maintenance and pedestrian/bicycle safety.
Although audits for road safety are typically carried out by those with engineering expertise, there is much that can be contributed by experts who specialize in measuring impacts of human factors. These factors include sight distance, visibility of hazards, readability of the road, design, placement and size of signs, availability of pedestrian and bike crossings, warnings of curves, etc., adequacy of pavement markings, lighting at night, consistency of design (exit placement, curves, etc.), potential glare from vehicle headlights, roadside distractions and violation of driver expectancy.
The specialties involved depend on the issues being examined. The objective is to identify potential safety issues and suggest mitigation measures. It is essential to consider all road users – vehicle drivers, pedestrians, the elderly, children, cyclists, commercial and agricultural vehicles – in the evaluation process. Considerations must be given to day and night operations, weather, as well as traffic speed and volume. Data on crashes traffic volumes and traffic mix (including pedestrians and cyclists) are useful, especially if there is a high accident area on an existing roadway.
For a synthesis of these approaches see Wilson and Lipinski (2004). A site visit would normally be part of the RSA. Ut would include driving, cycling, and/or walking (where pedestrians would be present) the relevant road sections.
Potential Human Factors in Road Safety Audits
Human factor contributions to an audit might involve evaluation of the design plan, or current configuration of an existing roadway to determine whether driver expectancies are met, sight distances are appropriate, there is design consistency, and the appropriateness, placement, and visibility of traffic control devices such as signs. Of particular concern are complex locations (e.g., intersections) where unfamiliar drivers may become confused or overloaded with information. These situations can lead to stress and driver error.
An audit of a proposed or reconstructed road involves examining one or more sets of drawings of the road, one section at a time, for the team to provide feedback on the basis of preliminary design plans. This initial evaluation is then commented on by the client (road owner or designer) and another set of drawings produced for further evaluation by the RSA team. Following implementation of the recommendations that are accepted the road is then built or modified and is inspected by the team, which then provides additional comments and suggestions. Any further changes made may be inspected by the team leader as a final step in the RSA process.
The following issues might be relevant to the human factors expert:
- sight distance of road hazards such as curves and intersections as well as signs
- visibility of hazards
- consistency of design (exit placement, curves, etc.)
- readability of the road
- placement and size of signs
- availability of pedestrian and bicycle crossings
- warnings of intersections, curves, etc.
- adequacy of pavement markings
- lighting at night
- length of tapers (merge and deceleration/exit lanes)
- application of rumble strips
- walking speed of pedestrians
- distractions (e.g., billboards, pedestrians, entering traffic, parking)
- violation of driver expectancy
Knowledge of human limitations in information processing, and human reliance on previous experience to compensate for this limitation, has led to the “positive guidance” approach to highway design. (Alexander and Lunenfeld, 1975). This approach is based on a combination of human factors and traffic engineering, The central tenet of the positive guidance approach is that design according to driver limitations and expectations increases the likelihood of drivers responding to situations and information correctly and quickly. When drivers are not provided with information in a timely fashion, or are overloaded with information, or are surprised because their expectations are violated, slowed responses and errors may occur.
With respect to road design, the positive guidance approach emphasizes the following:
Expectancy – Design should conform to long-term expectancies (e.g., there are no traffic signals on freeways, freeway exits are on the right) as well as short-term expectancies (e.g., all curves on this of road are gradual).
Primacy – determine the placement of signs according to the importance of their information (e.g., warning signs are more important that tourist information signs), and in such a way as to avoid presenting the driver with information when and where it is not essential, or at locations with high demand on driver attention (e.g., intersections).
Redundancy – say the same thing in more than one way. The same information may also be given with two devices (e.g., “no passing” indicated with a sign and pavement markings, STOP signs on both sides of a wide intersection.
Spreading – where all the information required by the driver cannot be placed on one sign or on a number of signs at one location, spread it out along the road so that information is given in small chunks to reduce the information load on the driver.
Coding – where possible organize pieces of information into larger units. Color and shape coding of traffic signs accomplish this by representing specific information about the message based on the color of the sign background and the shape of the sign panel.
Perception of the road
An appreciation of the road environment, especially its hazards, can influence driving performance. However, we know relatively little about what goes on inside the driver’s head when he/she looks at and assesses the road ahead. Drivers’ ability to “read the road” is an important consideration in traffic safety. The term “self-explaining roads” has been used in Europe to refer to the predictability of the road ahead, which depends partly on consistency of the road design. Designs need to communicate to drivers what type of road they are on or that can be easily categorized by drivers as requiring specific kinds of driving behavior. Ideally, the road ahead should be clearly visible for a considerable distance, but this is often not possible, as roads have hills, curves, and nearby vegetation.
Certain highway environments give us pleasure, while others put fear into us. Roads with greater collision rates tend to have less visual stimulation and score lower on activation and perceptual variation, which the author considers indicative of something like highway hypnosis due to sub-optimal attention and/or activation. This agrees with the psychological phenomenon of “optimal level of arousal” which has been found to be an inverted U-shaped function, with either too little or too much arousal leading to poorer task performance.
Drivers can sometimes be misled into exiting a main road inadvertently onto a minor road when that road is tangent to the main road at a curve. This has been referred to as a “visual trap”, as it traps the driver into entering the wrong road. When the minor road is cut through a forest or has a line of utility poles the effect can be more powerful.
Driving is essentially a perceptual-motor task with high demands on driver attention. These demands vary depending on several factors, including road geometric features. Workload refers to the effort and attention required to do a task. It is a measure of the information processing demands imposed on a driver by roadway geometry.
As demands of the task (and associated arousal levels) increase, so does performance, up to an optimal point, after which performance begins to deteriorate. Drivers will generally try to keep the workload within some range that is comfortable for them. If the situation becomes too demanding (e.g., unusual road geometry, heavy traffic, bad weather) the driver will pay greater attention and possibly reduce speed, but if it is too boring (long straight stretches of highway with little traffic) the driver may become inattentive and engage in activities irrelevant to the driving task, such as conversation, using a cell phone, looking at the scenery or smoking. The optimal level for an individual driver depends upon his/her skill at the driving task and familiarity with the roadway and its features. Workload will typically increase for the novice driver or one who is unfamiliar with the vehicle or the road and can increase with the appearance of an unexpected roadway feature.
The extent of driver workload is determined by factors such as time available to respond to a situation, sight distance, operating speed, traffic volume, and road features recently encountered. Examples of features rated as inducing high workload are narrow bridges, reverse curves and a reduction in the number of lanes. At specific locations, large changes in workload over a short distance have been found to be associated with more accidents.
Two main components of driver information overload are Information Search Demand (ISD) and Driving Task Demand (DTD). These are two sources of demand on the driver's attention. The first involves dealing with the information displays confronting the driver, while DTD comes from dealing with traffic and the roadway while trying to navigate. ISD is the demand on the driver by the effort of searching the relevant information sources along the roadway and processing the information so as to support navigation decision-making.
When the driver becomes overloaded with information the following may occur:
- stress, frustration or confusion
- getting lost
- erratic maneuvers at freeway exits
- slowing to read navigation signs
- missing an exit or turn
- missing traffic signs or signals
- failing to monitor other road users, including pedestrians crossing with the right-of-way.
Design consistency refers to similar road configurations for specific features, such as curve radii, distance between exits or intersections, and the conformance of a highway’s geometry with driver expectancy. Drivers make fewer errors in the vicinity of geometric features that conform with their expectations. Techniques to evaluate the consistency of a design include alignment features, speed distribution measures, and driver workload. Alignment features are quantitative measures of the general character of a roadway segment’s alignment. Potential indicators of geometric inconsistency include a large increase in the magnitude of the alignment indices for a successive roadway segment or feature or a high rate of change occurring over some length of road. Speed distribution measures - including variance, standard deviation, coefficient of variation, and coefficient of skewness – are potential candidates for a consistency rating method.
The most promising measures of design consistency are:
- average radius of horizontal curvature on a roadway section
- predicted speed reduction by drivers on a horizontal curve relative to the preceding curve
- or tangent
- average rate of vertical curvature on a roadway section
- ratio of an individual curve radius to the average radius for that roadway section as whole.
Highway work zones present potential hazards because motorists are confronted with unexpected and often confusing conditions. Many present an abnormal and disruptive situation to the motorist who is accustomed to a clear and unobstructed roadway. This problem is exacerbated by the fact that in recent years there has been a shift from construction of new highway facilities to rehabilitation and improvement of existing roads as they age. At construction and maintenance areas crashes often increase in frequency and severity, due in part to inappropriate application of traffic control devices (TCDs), inadequate layout of the work zone, poor traffic management. Failure of drivers to attend to and respect TCDs and a general misunderstanding of the unique problems associated with work zones can lead to crashes. In addition, glare is a cause of nighttime crashes in work zones. Lack of enforcement is also a problem. Safety concerns here involve not only the well-being of vehicle occupants, but also that of pedestrians and workers.
A growing concern about work zone safety has prompted the establishment of the US National Work Zone Safety Information Clearinghouse, located at the Texas Transportation Institute. Their website (http://wzsafety.tamu.edu/) can be very helpful in finding information on work zone safety. Its objective is to make available research reports, information on law enforcement and public information campaigns, and to provide data on safety as well as “best practice” information on work zone design. In addition, the US Federal Highway Administration has on its website (www.fhwa.dot.gov) a section on work zone accidents and best practices.
Audits tend to emphasize motor vehicle travel. However, pedestrian facilities must also meet standards and guidelines for pedestrian safety audits are now available. Nabors et al. (2007) have analyzed pedestrian safety issues of concern in the design of the road system and propose guidelines for pedestrian safety audits. These authors provide a prompt list for use, including visibility, lighting, connectivity access management and traffic control devices. Each of the prompts can be applied to streets, street crossings, parking and transit areas. Consideration is given to pedestrian needs, traffic speeds, presence of schools and work zone, as well as pedestrian and motorist behavior. Pedestrian audit prompts include:
- Ensure no obstacles block the view of pedestrians
- Provide sidewalks on both sides of the street
- Pedestrians must be visible at night
- Avoid drainage grates and potholes where pedestrians walk
- No utility poles or mail boxes should impede pedestrians
- Curb ramps must be adequate for those in wheel chairs
- Pedestrians must be properly directed through work zones
- Walk signals must be long enough for slow pedestrians to cross
It is important to consider bicyclists in safety audits. Research in the US has found that cyclists and drivers are about equally at fault in bicycle crashes. Collisions while crossing each other’s path are more common that when both are traveling parallel.
The following conditions may be a safety concern for cyclists:
- Free flow/continuous right-turn lanes.
- Acceleration/deceleration lanes.
- Lane drops.
- Through lanes that become turn lanes.
- Shoulder drops to accommodate a turn lane.
- Bus stops near intersections
Alexander, G. and Lunenfeld, H. Positive guidance in traffic control. Washington, D.C.: Federal Highway Administration. 1975.
Federal Highway Administration (2006). FHWA Road Safety Audit Guidelines. Publication No. FHWA-SA-06-06.
Federal Highway Administration (2012). Bicycle Road Safety Audit Guidelines and Prompt Lists. Publication No. FHWA-SA-12-018.
Nabors, D., Gibbs, M., Sandt, L., Rocchi, S., Wilson, E. and Lipinski, M. (2007). Pedestrian road safety audit guidelines and prompt lists. FHWA Report FHWA-SA-07-007, Federal Highway Administration, Washington, DC.
Wilson, E. M. and Lipinski, M. E. (2004). Road Safety Audits: A Synthesis of Highway Practice. NCHRP Synthesis 336. Transportation Research Board, Washington, DC.