2.1 The Driving Task
A driver is continuously balancing three main tasks: control, guidance, and navigation. Control refers to all activities involved in the driver's interaction with the vehicle. The driver receives control information from the vehicle's displays, from the "feel" of the vehicle, and from the roadway. Control also involves the driver maintaining a safe speed and keeping the vehicle on the proper path (4, 5).
Guidance of the vehicle is how the driver interacts with other vehicles, the roadway, and the surrounding environment to guide the vehicle safely down the roadway. As drivers travel along the roadway, they receive information (e.g., alignment, grade, hazards, traffic, traffic control devices, etc.) from various sources and use their judgment to adjust the vehicle's speed and path appropriately (4).
Navigation is how the driver is going to get from their origin to their destination. Navigation involves both pre-trip and in-trip decisions. Pre-trip navigation can include planning a route, receiving instructions, mapping a route, etc. In-trip navigation includes using route guidance and landmarks en-route to get from the origin to the destination (4). Drivers may also use electronic navigation devices or global positioning system (GPS) units to obtain navigation information en-route.
At any given time, a driver is receiving and processing information that contributes to decisions being made regarding one or more of these tasks. Drivers continuously use information from a number of real-time sources as well as their experiences and general expectations to make driving decisions. Drivers "juggle" these multiple sources of information by giving any one piece of information a short amount of attention before moving onto the next piece of information. Drivers should prioritize the relative importance of the information being received. Importance is assigned based on the consequence of not paying attention to an information source. Information with the highest importance should be addressed first. Obviously, among the main driving tasks, control is the most important, followed by guidance and navigation. Thus, under less than ideal conditions, such as wet weather, drivers tend to pay more attention to the control of the vehicle and less attention to less critical information (4).
2.2 Sight Distance
For the safe and efficient operation of vehicles, drivers need to be able to see the roadway ahead and any potential hazard soon enough to react to it. Sight distance is the length of roadway ahead that is visible to a driver. Sight distance is comprised of driver perception-reaction time and the time to complete the maneuver safely. The driver perception-reaction time includes the time to perceive the condition (e.g., an object in the roadway), time to complete cognitive functions to understand the condition (e.g., recognize the object and decide if and how to respond), and time to initiate the desired maneuver (e.g., take foot off accelerator and step on brake). The driver perception-reaction time varies based on the distance to and nature of the object, the visual performance of the driver, the cognitive capabilities of the driver, the atmospheric visibility, and the roadway environment. The time to safely complete the maneuver is primarily impacted by the roadway condition and the vehicle's performance capabilities, especially tire-pavement friction, road-surface conditions (e.g., wet), and downgrades (6).
In general, sight distance should be sufficient to allow a driver traveling at or near the design speed to stop before reaching a stationary object in the travel path. Stopping sight distance is the distance traveled from the instant the driver sees and recognizes a hazard to the instant the brakes are applied (brake reaction time) plus the distance actually needed to stop the vehicle (braking distance). A Policy on Geometric Design of Highways and Streets (2) provides a thorough explanation of stopping sight distance as well as the design values and stopping sight distances for various design speeds based on assumed conditions on level roadways. The design values are determined using values for a predetermined design driver and design vehicle under a predetermined set of circumstances. However, the stopping distance can vary based on factors, including (but not limited to) driver reaction times, vehicle deceleration rates (which are different for each type of vehicle), braking efficiency, coefficient of adhesion, and roadway grade. The coefficient of adhesion refers to the frictional force between the tires and the pavement. Thus, the coefficient of adhesion is lower under poor tire conditions and poor pavement conditions than it would be for good tire conditions on dry pavement. A decrease in the coefficient of adhesion will result in an increase in stopping distance. Using longer distances increases the margin for error for all drivers, especially during wet weather conditions. However, as discussed in Chapter 1, there are roadway environments where the operating speed is greater than the design speed and the safe stopping distance is greater than the available sight distance.
2.3 The Driver
In 2010, there were more than 210 million drivers in the United States. Approximately half of these drivers were male and half were female. Almost three-quarters of these drivers (71 percent) fall into the 25 to 64 age range, with younger drivers (24 or less) comprising 13 percent of the driving population and older drivers (at least 65) representing 16 percent of the driving population (7). However, it is expected that the older driving population will double over the next 30 years (8). The changing demographics of the driving population are important to recognize when considering the impacts of wet weather on the driving task. Typically, drivers with more experience are less risk tolerant than younger drivers. Driver experience and maturity impacts decision-making, especially under less than ideal conditions, such as wet weather. For example, in wet weather conditions, younger drivers may tend to devote less attention to control of the vehicle than they should. Brake reaction times can vary between drivers from as low as 0.64 seconds for alerted drivers to over 3.5 seconds for un-alerted drivers when simple signals are used in research tests (2).
2.4 The Vehicle
In 2010, an estimated 242 million motor vehicles were registered in the United States (7). These motor vehicles consist of passenger cars, light trucks and vans, motorcycles, recreational vehicles, buses, and commercial vehicles, all of which have different design and operational characteristics. Thus, the amount of control needed to drive a vehicle safely varies based on the type of vehicle. For example, it is less demanding to control a passenger vehicle with an automatic transmission than it is to control a commercial vehicle with multiple gears and clutches. Even so, most control activities, once mastered, are performed automatically with little conscious thought (4). A vehicle's design and maintenance also impact a driver's ability to control the vehicle, especially during wet weather. Anti-lock brakes reduce wheel lock-up under braking, which reduces the likelihood of skidding on slippery pavement. Vehicle maintenance items include, but are not limited to, properly working windshield wipers, proper tire pressures, adequate amount of tread on tires, and maintenance of the brake system. AASHTO suggests that although most drivers decelerate at rates greater than 14.8 ft/s2 when an unexpected object is encountered in the roadway, 90% of drivers decelerate at rates greater than 11.2 ft/s2, and thus 11.2 ft/s2 is used as a design value (2).
The previously mentioned stopping sight distances are typically based on passenger cars, so they do not explicitly consider commercial vehicles. In general, commercial vehicles need longer stopping sight distances than passenger cars. However, because commercial vehicle drivers are positioned higher above the pavement, they can generally see further down the road than a driver of a passenger car. Thus, commercial vehicle drivers typically have more time to perceive and react to a potential hazard, somewhat making up for the increased braking distances. So, separate stopping sight distances for different vehicle types are not typically used in highway design. One exception includes locations where horizontal sight restrictions exist on downgrades, especially at the end of long downgrades where commercial vehicle speeds are similar to or exceed passenger car speeds. Under such conditions, it is desirable to provide stopping sight distances longer than those provided for passenger cars (2). Another situation where commercial vehicle drivers may not be able to see further than passenger car drivers is during rain. Under these conditions, the intensity of the rain may limit the view of the roadway and surrounding environment of all drivers.
2.5 The Visual Scene
Wet weather reduces the visibility of potential hazards and traffic control devices, making them more difficult for drivers to detect. The distance at which a driver can see a non-illuminated, non-reflective object is dependent upon a variety of factors, including the vehicle's headlights, the driver's visual capabilities, and the driver's expectation of seeing a hazard. While many traffic control devices are retroreflective, rain can reduce the retroreflective materials' ability to perform and decrease the visibility of the device. Under wet weather conditions, internally illuminated traffic control devices may provide better visibility.
As discussed, under normal driving situations drivers have the ability to process information from multiple sources, but as conditions deteriorate, they begin shedding information sources in order to attend to the primary driving functions of control and guidance. Traffic control devices (e.g., speed limit signs) have to compete for the driver's attention; therefore, it is critical that they be placed where drivers can see them and quickly obtain information.
2.6 The Pavement
As mentioned, the time to safely stop once a hazard is detected and the driver applies the brakes is dependent upon road surface conditions and available tire-pavement friction. Pavement conditions that may contribute to an increase in wet weather crashes include (9):
- Rutting or wheel path channelization.
- Build-up on shoulder edges that causes ponding on the road surface.
- Bleeding of pavements.
- Drainage issues that result in water on the pavement.
- Geometrics (grades and curvature).
Water on the pavement can significantly reduce tire-pavement friction. Even as little as 0.002 inches of water on the pavement can reduce the coefficient of friction by 20 to 30 percent (10).
Rutting, build-up on the shoulder, or drainage issues can result in relatively thick water films on the pavement surface. The coefficient of friction for a vehicle tire sliding on wet pavement decreases exponentially as the water film thickness increases. The effect of the water film thickness is influenced by tire design and condition, with worn, smooth, and "bald" tires being most sensitive. At low speeds (less than 20 mph), the effect of water film thickness on friction is minimal; however, at higher speeds (greater than 40 mph), the effect is more pronounced. When relatively thick water films are present and vehicles are traveling at higher speeds, hydroplaning can occur (9).
Obviously, the deceleration maneuver is impacted by wet pavement. A Policy on Geometric Design of Highways and Streets (2) recommends 11.2 ft/s2, a comfortable deceleration rate for most drivers, as the deceleration threshold for determining adequate stopping sight distance. Such deceleration corresponds to the braking friction on packed snow and thus is within the driver's capability to maintain control of the vehicle during braking maneuvers on wet pavement. Also, most vehicle braking systems and the tire-pavement friction levels on most roadways, even under wet conditions, are capable of providing a deceleration rate of at least 11.2 ft/s2.
Previous research has shown that under emergency conditions, mean deceleration rates are the same between good and poor visibility conditions, but differ among good and poor traction conditions (17.7 ft/s2 and 13.8 ft/s2, respectively). However, the 85th percentile values are the same (12.1 ft/s2). On wet pavement with vehicles without an anti-lock braking system, the mean and 85th percentile deceleration rates are about 13.8 ft/s2 and 12.2 ft/s2, respectively (54% and 47% of the pavement's coefficient of friction, respectively). On wet pavement with anti-lock brakes, the mean and 85th percentile deceleration rates are about 17.1 ft/s2 and 14.5 ft/s2, respectively (66% and 56% of the pavement's coefficient of friction, respectively) (6).
2.7 Speed Zoning
All States follow the same basic approach to formulate speed regulations. This approach specifies that drivers should not exceed speeds that are safe and prudent for existing conditions, even if the posted speed limit indicates that a higher speed is allowable. Thus, a driver is responsible for considering the roadway environment and potential hazards when selecting a speed. This includes weather, visibility, traffic, and roadway characteristics (5).
A safe and reasonable speed limit is typically based on a statutory speed limit or traffic engineering study. Speed zones established based on these studies may change the basic speed limits set by law. Speed zones can either be regulatory or advisory. Regulatory speed limits are enforceable, whereas advisory speed limits are generally not enforceable, although drivers could be cited for violating the basic speed rule that requires drivers to drive at an appropriate reduced speed when necessary (5).
In most cases, the establishment of a speed zone is predicated on the assumption that most drivers operate their vehicles in a safe, reasonable, and prudent manner. The speeds that the majority of drivers choose to travel on a given roadway segment are considered an indication of safe and reasonable speeds. The 85th percentile speed of free flowing vehicles, the speed at which 85% of drivers travel at or below at a given point on the roadway, is commonly taken to determine the posted speed for that segment. This driver-defined maximum safe speed can then be adjusted slightly, if necessary, based on site-specific factors, including sight distance restrictions, pavement conditions, etc. A speed limit to the nearest 5 mph increment of that maximum safe speed is then typically posted (5, 11).
Even though the speed limit may be based on the 85th percentile speed, many studies have reported that the posted speed limit is usually significantly lower than the measured 85th percentile value. For example, the Institute of Transportation Engineers (ITE) found that for roadways with posted speed limits of 45 mph and below, most of the measured speeds are higher than the posted speed limit (12). When the posted speed limit is 55 mph or more, about half of the measured speeds are above the posted speed limit. This may be due in part to the difficulty in predicting operating speeds (and thus the speed limit) based on the roadway geometry and roadside features.
As discussed previously, all basic State speed laws typically require drivers to adjust their speed to existing road conditions; thus, the primary responsibility for adjusting speed to adverse conditions, such as wet weather, rests with the driver. Nevertheless, in an attempt to improve safety, some jurisdictions have decided to reduce speed limits at specific locations during adverse weather conditions.
A VSL is one that changes with changing conditions. The use of this operational strategy is by no means new; dating back more than 60 years, agencies have modified speed limits on a temporary basis for a variety of reasons. School zones and work zones are some of the more common applications of VSLs. Modern VSL systems use technology to adjust speed limits in real-time. Speeds may be modified based on traffic conditions, adverse weather conditions, road surface conditions, or work zones to better control operating speeds within the monitored section of roadway. VSLs not only work to maintain safe driving speeds, but also to warn drivers of changing road conditions. However, when wet weather conditions get too bad, the speed limit displayed on the VSL could still be "faster" than what drivers feel is a speed at which they can comfortably control and guide their vehicle. This can be problematic because drivers trust VSL systems to provide them with the maximum safe speed for the conditions
As noted, driver behavior varies during adverse weather events based on experience and risk tolerance. Therefore, some drivers may slow down when they encounter wet weather, but others may maintain or even increase their speeds. The result of this varying behavior may be a significant speed differential between slower and faster vehicles on the roadway. It is generally accepted that an increase in speed differential increases the potential for conflicts between vehicles, which in turn increases the potential for crashes. VSL systems based on weather conditions can work to reduce the speed differential under unpredictable events and encourage more uniform speeds. However, there is still the potential for speed differentials to occur immediately upstream of the rain event and reduced speed limit (i.e., as drivers enter the reduced speed limit zone).