Next Article in Journal
Towards the Validation of an Observational Tool to Detect Impaired Drivers—An Online Video Study
Previous Article in Journal
Non-Ionic Surfactant Recovery in Surfactant Enhancement Aquifer Remediation Effluent with Chlorobenzenes by Semivolatile Chlorinated Organic Compounds Volatilization
Previous Article in Special Issue
Influence of Vision on Drivers: A Pilot Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 19
Issue 12
10.3390/ijerph19127549
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Improvements of Warning Signs for Black Ice Based on Driving Simulator Experiments

by
Ghangshin Lee
1,
Sooncheon Hwang
1 and
Dongmin Lee
2,*
1
Department of Smart Cities in Graduate School, University of Seoul, Seoul 02504, Korea
2
Department of Transportation Engineering & Smart Cities, University of Seoul, Seoul 02504, Korea
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2022, 19(12), 7549; https://doi.org/10.3390/ijerph19127549
Submission received: 28 April 2022 / Revised: 11 June 2022 / Accepted: 17 June 2022 / Published: 20 June 2022
(This article belongs to the Special Issue Vision and Driving Safety)
Browse Figures
Review Reports Versions Notes

Abstract

:
Black ice is one of the main causes of traffic accidents in winter, and warning signs for black ice are generally ineffective because of the lack of credible information. To overcome this limitation, new warning signs for black ice were developed using materials that change color in response to different temperatures. The performance and effects of the new signs were investigated by conducting driver behavior analysis. To this end, driving simulator experiments were conducted with 37 participants for two different rural highway sections, i.e., a curve and a tangent. The analysis results of the driving behavior and visual behavior experiments showed that the conventional signs had insufficient performance in terms of inducing changes in driving behavior for safety. Meanwhile, the new signs actuated by weather conditions offered a statistically significant performance improvement. Typically, driver showed two times higher speed deceleration when they fixed eyes on the new weather-actuated warning sign (12.80 km/h) compared to the conventional old warning sign (6.84 km/h) in the curve segment. Accordingly, this study concluded that the new weather-actuated warning signs for black ice are more effective than the conventional ones for accident reduction during winters.

1. Introduction

1.1. Background

During winter, black ice is formed on many roadways, and this natural phenomenon introduces many risks. Black ice is a transparent coat of ice on roadways or other transportation surfaces, and it cannot be seen easily. The occurrence time and location of this natural phenomenon are extremely difficult to predict because rapid transitions between dry, wet, snow, and ice are common on such surfaces. This recurrent condition caused by water, ice, and snow causes many accidents worldwide [1,2,3,4,5]. Given the terrain in Korea, many roadways pass through mountainous areas [6]. For numerous reasons, the rate of casualties in black ice-related accidents is 66.2% higher than that in general traffic accidents in Korea [7]. This can be ascribed to the fact that it is difficult for drivers to recognize road freezing. Moreover, the coefficient of friction under such conditions is lower than that under dry and wet surface conditions, which affects the stability of the vehicles and the probability of traffic accidents. The limitation of static conventional black ice warning signs in terms of contributing to drivers’ perception of hazards must be addressed. To evaluate whether the existing warning signs have positive effects on traffic safety, it is essential to know whether they serve their designed purposes. Of all the different components of driving skills, hazard perception alone has been found to be related to accident involvement in several studies [8,9,10,11]. Hazard perception in hazardous driving environments ahead has been studied across different age groups [12,13,14]. In these studies, it was assumed that the conventional static warning signs for black ice do not contribute to hazard perception among drivers of all age and gender groups. Therefore, more effective signs to warn drivers about black ice are needed.

1.2. Objective

The primary objective of this study was to investigate the effectiveness of the conventional static warning signs for black ice and develop new warning signs that are actuated by weather conditions. To these ends, driving simulator experiments were conducted using an eye-tracking device. As further highlighted in Section 3.2, the new weather-actuated warning sign for black ice and its final design selected were replicated to the greatest extent possible in terms of realism and detail in the driving simulator, thus allowing participants to provide natural behavioral datasets. More detailed insights into the methodological design of this study are provided later.

2. Literature Review

2.1. Impact of Warning Signs on Safety

To assess the impact of warning signs on safety, we reviewed related studies on warning signs for black ice. We determined the variables which studies used to evaluate the effectiveness of warning signs for speed reduction. Next, we have reviewed studies of visual behaviors for warning sign detection. Hazard perception in hazardous environments holds great importance for safe driving [9,15]. An understanding of hazards during the early development of driving prevents drivers from being involved in collisions [8,16], However, the conventional static warning signs for occasional hazards may be ineffective because, most of the time, drivers do not perceive any real threat from such signs. In the literature, many operating traffic devices have been evaluated for their effectiveness in terms of hazard warning, but only marginal attention has been dedicated to the effectiveness of conventional static warning signs for black ice. Several studies based on driving simulator experiments have suggested that drivers may become complacent about the importance or meaning of conventional static warning signs [17,18,19,20]. Despite the broad use of conventional static warning signs in the highway system, the effectiveness of these signs in improving drivers’ hazard perception has hardly been investigated. Most studies that investigated the effectiveness of warning signs considered nonconventional warning devices, such as curve warning signs [21,22], flashing LED stop signs and optical speed bars [23], rumble strips [24,25], retro-reflective fluorescent signs, and various road markings as indications of speed limits [20,26,27]. Most researchers examined the effectiveness of nonconventional warning devices relative to that of conventional warning devices by conducting travel speed, acceleration, deceleration, and gas/brake pedal power analysis. Accordingly, more research on conventional static warning signs is necessary.
Generally, many studies on the performance of warning signs have been conducted to determine the effect of these signs on reducing vehicle speed and enhancing road safety in different conditions [28,29,30,31,32]. In terms of curve advisory speeds, many studies have examined the effectiveness of curve warning signs [17,33,34,35]. In terms of the variation of curve warning signs, regardless of their immediate purpose, the literature review showed that warning signs reduce operating speeds in hazardous roadway sections and, thus, affect safety. Moreover, it suggested that typical warning signs are often overlooked owing to their frequent use and, consequently, drivers experience hazard perception latency [36].
For the impact of signs in winter weather conditions, the literature review revealed limited studies. However, all related published reports provided importance of further investigation for the winter weather conditions to prevent black ice-correlated fatalities and injuries [37,38,39,40]. One study found that flashing signals with a reduced speed limit once the sensor system detects ice formation are more effective than static warning signs. The yellow lights start to flash to inform drivers of black ice ahead, and red lights can be used by authorized officers to close lanes or road sections in case of emergency [39]. This study addressed the high cost of black ice-detecting devices as a limitation for application. As mentioned previously, this study analyzed drivers’ speed data obtained from the simulator experiments to determine the effectiveness of static warning signs in terms of reducing vehicle speed and enhancing road safety. After a review of warning signs and driving safety, the travel speed and gas pedal power for acceleration before signs were determined as a measurement of the effectiveness of conventional static warning signs on drivers’ behaviors.

2.2. Warning Signs and Visual Behavior

When drivers read signs, they move their eyes from the road to the signs through an alternating pattern of fixations (points at which the eyes are stationary and focused on an object) and saccades (points at which the eyes are moving between fixations). The basic premise of visual behavior is that increased processing demands are related with increased processing time or a change in pattern of fixations [41]. Increased processing time may be indicated by longer duration fixations or a larger number of fixations.
Previous studies have revealed many details about the differences in visual behavior between novice and expert drivers [16,42] and in different visual clutter conditions [43,44]. Even so, considering the limitations of static warning signs for black ice, the research attempted to evaluate how effective the signs were in providing warning information regarding hazards for approaching drivers through measuring visual behavior. In the literature, many studies have attempted to set the hazard perception parameter in terms of standard response time and eye movement analyses. One study calculated braking responses, number of fixations (N), and fixation duration (milliseconds) among experienced and inexperienced drivers. The results indicated that hazard perception was faster among experienced drivers than among inexperienced drivers [45].
In terms of evaluation, many related studies used the number of fixations (N) and fixation duration (milliseconds) as a measurement of the effect of warning signs, along with speed and gas pedal power during acceleration as driver performance variables. According to the literature review, fixation number is a measurement of visual behavior that is strongly correlated with reaction time and, therefore, used as a measure of search efficiency [46]. Average fixation duration is a measure of the amount of time necessary for information processing at each fixation.

3. Materials and Methods

3.1. Apparatus

The experiment data were collected, including operating performance data; the variables of the fixed-base driving simulator used in this experiment were acceleration, deceleration, and lateral position. The values of gas and brake pedal use, representing the pressing intensities of these two pedals, ranged from 0 to 1. For instance, “0” denoted that the pedal was not pressed, while “1” denoted that the pedal was pressed to its full depth. The driving speed and gas pedal power were collected every 10 m before the sign, and that the values were averaged for all participants.

3.1.1. Eye Tracker

In the experiment setup, a pair of Tobii Pro Glasses 3, consisting of a head-worn eye tracker connected to the recording device holding the saved data of the Tobii Pro Glasses 3, was used. This is a binocular eye-tracking setup that uses two cameras and six glints per eye for gaze tracking, and it was selected because it is known to be the most accurate eye-tracking unit [47,48,49].
The Tobii Pro Glasses recording device was connected to a desktop running the Tobii Glasses Controller software (version 1.11.4) (Tobii Technology, Stockholm, Sweden) using an ethernet cable. The system was set to the 50 Hz mode and calibrated with the one-point calibration procedure using the marker provided with the eye-tracking setup. The calibration consisted of (a) fixing the marker to the center of the stimulus grid, (b) instructing the participant to fixate on the marker’s center, and (c) entering the calibration mode in the Tobii software (Tobii Technology, Stockholm, Sweden), after which the process was completed automatically. The front-facing cameras recorded a video stream at 25 Hz with a resolution of 1920 × 1080 px, and the four eye cameras recorded a stream at 50 Hz containing four eye images (two views of each eye) with a combined resolution of 240 × 960 px. The study followed the Tobii Manual definition of fixation, which lasts more than 60 ms [50]. Fixation was defined as the time between the end of one saccade and the beginning of the next saccade.

3.1.2. Driving Simulator

The UC-win/Road program was initially used for interactive virtual reality (VR) modeling in construction planning, urban planning, civil engineering, and traffic modeling [51,52,53]. By combining the program with a driving simulator, several driving situations in diverse environments can potentially be emulated. In this study, the experiment employed an automatic car comprising a seat, a seat belt, a steering wheel, a speedometer, an ignition and key unit, an accelerator, and a brake pedal. The visual display system consisted of three-color monitors spanning a 180° field of view synchronized to display a realistic view of a computer-generated road environment, as shown on Figure 1. Moreover, the simulator included an audio system to provide realistic traffic sounds and instructions.
The route comprised two main sections, i.e., curved and tangential. The driving experiment was performed in daylight under gloomy weather conditions.

3.2. Experimental Stimulus

Before the main simulator experiment, a preference survey of six static warning sign was conducted. This was an online survey in which 124 participants answered which of the six design concepts was the most preferred as a black ice warning sign. The more preferred design for the sign, which is shown in Figure 2a (top left), was selected [54].
After the preference survey of the sign design, the sign used in the experiment was redesigned using the same concept. The final design was modified to be actuated under freezing temperatures to show text messages with a pictogram for black ice. The pictogram was also revised to display the black ice warning even without the snowflake pictogram, as shown in Figure 2b. Detailed designs of all the signs mentioned in the study are shown in Figure 2.

3.3. Scenarios

The roadway geometry depicted in the simulations was a virtual-reality reconstruction of a real rural two-lane road, including horizontal curve and tangent segments, in Gangwon-do province. The roadway segment was a 5.56 km section of South Korea Route 28. This reconstruction was based on the topographic layout, which consisted of a straight road. The lane widths, road markings, sight distances, and other road engineering characteristics of the roadway segment, including a 3.6 m lane width and 140 m curve radius, were incorporated into the simulation to provide a similar road perception. Several different scenarios were designed with several combinations of experimental conditions and implemented in virtual reality. Two scenarios for geometric conditions, including a curved segment and tangent segment, and four scenarios for warning signs for black ice were used for the experiments. In each case, the traffic was displayed in random order. The signs were located 22 m after the end of curve segment and at the end of the 200 m tangent segment. Road markings were yellow and white, with a width of 10 cm. The length of road influenced by signs on the straight segment was 200 m. The corresponding length of road on the curved segment was 75 m.
In the four experiments involving signs, i.e., no sign, conventional static warning sign for black ice, new warning sign not actuated by weather conditions, and new warning sign actuated by weather conditions, as shown in Figure 3, two typical sign locations, i.e., at the ending of a curve section and at a point in a tangent section, were considered to investigate their effects on driving behavior, as illustrated in Figure 4.

3.4. Procedure of the Experiments and Participants

Each participant was tested individually in the experiment. As a first step, upon their arrival, the participants were briefed on the requirements of the experiment. All participants read and signed an informed consent document and were asked to complete a short questionnaire to provide basic information (i.e., age, gender, etc.). To collect concrete data for improving static warning signs for black ice, the participants were informed how the signs signify a freezing temperature with a red pictogram when actuated, as described in Section 3.2. As a second step, the participants were introduced to the driving simulator and provided an overview of its functions. Then, they were familiarized with the vehicle in a neutral condition. In this condition, each participant drove through a reconstruction of scenario 1 in reverse, which allowed for familiarization with the torque-based tactile feedback provided through the steering wheel and pedal control.
The participants were informed that they were to drive on a mountainous road until asked to stop. The participants were asked to adapt their speed to the driving conditions as follows: “Please drive as you would drive in the same situation in the real world until you are advised to stop. The speed limit on this roadway is 60 km/h”. The participants were informed that some disturbances could occur during the simulation and that they could stop the experiment at any time.
Next, the participants started to drive using the simulator for four separated runs in random order after completing calibration of the eye tracker. In each run, they encountered randomly assigned sign conditions. These experimental sign conditions included no signs, conventional static warning signs, new warning signs not actuated by weather conditions, and new warning signs actuated by weather conditions. All the participants drove in all situations while wearing the eye tracker.
After the fourth run, the participants were asked to complete a questionnaire concerning the effectiveness of each sign in terms of hazard perception in black ice zones. The participants took an average of 50 min to complete the experiments, with the driving phase lasting an average of 30 min
In this study, the experiment recruited 37 healthy subjects from our panel who occasionally joined our simulation experiment to participate in a driving simulator experiment with visual perception of traffic scenes. All subjects had a valid driver’s license and normal or corrected-to-normal vision (using contact lenses or glasses). Those who showed any signs of simulator sickness were excluded from the experiment. All subjects reported driving daily and holding a driver’s license for at least 3 years. Table 1 summarizes the gender and age variation of the subjects.

3.5. Data Analysis Methodology

3.5.1. Determination of Location at Which Driver Behavior Starts to Change

To find the distance from the signs where drivers changed their behavior due to perception of the signs, vehicle speed and gas pedal power were used to identify significant differences between the group that drove with no signs and the group that drove with signs, including conventional static warning signs and new warning signs not actuated and actuated by weather conditions. To determine the influence of traffic signs, the study compared the datasets of the two groups. The study analyzed the behavior of drivers approaching the signs, during the approach, and in two geometrical sections. The indicators selected for these analyses included the vehicle speed and deceleration rate measured using the pressure applied to the gas pedal. The data were input into a connected computer for analysis.

3.5.2. Correlation between Drivers’ Behavior Using Deceleration and Visual Data

The study compared the deceleration behaviors of drivers between the two groups, separated by visual data, fixation count, and duration data, i.e., the group that fixed eyes on signs and the group that did not. The fixation visit data were used to distinguish whether the subjects had looked at the signs. As addressed in Section 3.1.1, fixation was defined as points at which the eyes were stationary and focused on an object for longer than 60 ms. This study did not intend to find statistically significant differences during test between the two groups but intended to observe the distinction between groups regarding differences in speed deceleration as a function of the sign.
This study aimed to investigate the effectiveness of two different types of signs. The t-test was utilized to find significant differences between each sign and driving without any sign.

4. Results

The study first determined the location at which the drivers started to significantly change their speeds and gas pedal power. The results provided excellent indications of where the signs influenced the drivers. As shown in Figure 5, the locations where the drivers’ behaviors started to change signified the distance from the signs where the drivers began to exhibit significantly different driving behavior considering the presence of signs. Thereafter, speed differences at these locations among the participants were computed to determine the effectiveness of warning signs in indicating an upcoming black ice hazard.

4.1. Determination of Location at Which Driver Behavior Started to Change

The distance from the signs where the drivers’ behavior showed significant change could be identified on the basis of the agreement between the drivers’ behaviors and the presence of signs. As the data indicate, in each case, promising results were obtained in terms of the location at which driver behavior started to change in terms of vehicle speed and gas pedal power. In case of the curve segment, at 90 m before the signs, drivers drove at lower speeds (M = 45.88, SD = 11.26) than those who traversed the same roadway without any speed warning signs for black ice (M = 49.91, SD = 7.55), t (147) = 2.6, p = 0.012. Moreover, the gas pedal power of the drivers presented with a sign (M = 0.23, SD = 0.19) was lower than that of the drivers who were not presented with a sign (M = 0.3, SD = 0.17), t (147) = 2.4, p = 0.02, as summarized in Table 2. Meanwhile, in the tangent segment, at 160 m before the signs, the drivers who were presented with a sign drove at lower speeds (M = 55.84, SD = 7.93) than those who were not presented with any speed warning sign for black ice (M = 59.15, SD = 6.33), t (147) = 2.325, p = 0.02. Moreover, the drivers who were presented with a sign used less gas pedal power (M = 0.6, SD = 0.25) than those who were not presented with a sign (M = 0.72, SD = 0.17), t (147) = 3.282, p = 0.002.
Subsequently, speed and gas pedal power allowed us to distinguish the points to be analyzed to evaluate the effectiveness of warning signs in each case. On the basis of these comparisons, the location at which driver behaviors started to change for each sign were determined to be 90 m in case of the curve segment (p = 0.012) and 160 m in case of the tangent segment (p = 0.02).

4.2. Driving Behavior: Speed Deceleration

During determination of the location at which driver behaviors started to change in response to a sign, the maximum and minimum vehicle speeds were extracted to represent the effects of signs on drivers’ deceleration behaviors. Both indicators measured not only the intensity of drivers’ deceleration but also the overall behavioral shift in response to the signs. The deceleration value of each participant was computed by subtracting the minimum vehicle speed from the maximum vehicle speed. The means of drivers’ deceleration behaviors are summarized in Section 4.2.1 and Section 4.2.2.

4.2.1. Conventional Static Warning Signs for Black Ice

Table 3 shows the results of multiple comparisons of average drivers’ deceleration behaviors in response to each sign. As a result, it was found that the conventional static warning signs did not lead to a significant difference in drivers’ deceleration behaviors in all cases (curve: M = 6.48, SD = 4.70; tangent: M = 7.54, SD = 5.50) relative to the no-sign case (curve: M = 7.83, SD = 4.94; tangent: M = 6.96, SD = 3.65). These results indicate that the conventional static warning signs did not significantly affect drivers’ deceleration behaviors, regardless of road geometry. Conclusively, the drivers were not significantly influenced by the conventional static warning signs for black ice.
According to Figure 6 and Figure 7, the values of vehicle speeds and gas pedal power were consistent between the cases where drivers were provided conventional static warning signs for black ice (old sign) and no sign at all. This result is in line with earlier predictions, i.e., static warning signs failed to provide a hazardous alert to drivers such that they could perceive the urgency to decelerate the vehicle. Therefore, there were no significant differences in drivers’ deceleration behaviors upon the provision of static warning signs for black ice.

4.2.2. New Warning Signs Actuated by Weather Conditions

As can be inferred from Table 3, drivers exhibited enhanced deceleration behaviors when the new warning signs for black ice were actuated (not actuated/actuated curve; segment: M = 11.35/11.19, SD = 5.64/7.19; tangent segment: M = 14.72/11.11, SD = 7.95/7.64) than those when presented with conventional static warning signs for black ice (curve: M = 6.48, SD = 4.70; tangent: M = 7.54, SD = 5.5). This is because the preceding vehicles were affected only by the new warning signs actuated by weather and their deceleration behaviors changed.
According to Figure 4 and Figure 5, the speed and gas pedal power of the drivers presented with the new warning signs for black ice actuated by weather were lower, meaning that the new warning signs significantly influenced drivers’ deceleration behaviors. An interesting result was that the speed and gas pedal power values of the drivers presented with the new weather-actuated warning signs black ice were lower than those of the drivers presented with the new warning signs for black ice not actuated by weather. Thus, the warning information received by the drivers from the actuated signs had more significant effects on their vehicle deceleration behaviors.
By contrast, the results of all the new actuated warning signs for black ice in cases of the curve and tangent segment, summarized in Table 3, indicate that the vehicle had statistically significant differences in speed deceleration from the other signs, including the old sign and non-actuated sign, as well as no sign. The results in Figure 6 show that the average speeds associated with the new weather-actuated warning signs for black ice on the curve segment, in which the design of the new signs was similar to that shown in Figure 1b, were relatively lower than the average speeds associated with the new warning signs for black ice not actuated by weather on the curve segment. Moreover, in the case of the tangent segment, where the drivers were able to identify signs from 160 m, the gap between the vehicle speeds associated with the new warning signs for black ice actuated and not actuated by weather was smaller. This was because the preceding vehicle was affected by the warning information in the tangent segment, but the subject vehicle was affected by both the warning information and the change in maneuvering behavior in the curve segment.
Conclusively, the new warning signs provided the greatest safety benefit in terms of speed reduction. More detailed meaningful findings were identified by analyzing the results obtained with the two road geometries. First, the new warning signs for black ice were the most effective when they were actuated by weather, regardless of the road geometry. Second, geometrical differences had a positive effect on drivers’ speed reduction behaviors, regardless of whether they received warning information from different signs. That is, when the preceding vehicle was driving on the curved segment, the subject vehicle was affected by both the warning information and the change in maneuvering behavior in the curved segment.

4.3. Visual Behavior: Total Fixation Visit and Average Fixation Duration

The fixation visits, i.e., the number of driver eye fixations on warning signs, were measured as shown in Table 4. Fixation visits may represent the extent of an increased processing attempt for information [41]. The results show that the conventional static warning signs showed the lowest number of fixation visits in case of the curved segment (25 visits) and the tangent segment (65 visits). The new non-actuated warning signs showed 67 fixation visits in the curved segment and 93 fixation visits in the tangent segment. The new weather-actuated warning signs showed 62 fixation visits in the curved segment and 133 fixation visits in the tangent segment. These results indicate that the new weather-actuated warning signs had the most increased processing time, regardless of the road geometry.
Meanwhile, fixation duration, which indicates the processing time for which drivers’ gaze is fixed at one location, may also represent the increased processing demands for driver comprehension of each sign [41]. Fixation duration of the new warning signs was the longest in the curved segment (352.03 ms) and in the tangent segment (621.51 ms).

4.4. Correlations between Variables

Table 5 summarizes the drivers’ deceleration measure, standard deviation, and gas pedal power between two groups, i.e., the group that was presented with warning signs and the group that was not presented with any warning signs. The deceleration measure indicates the vehicle speed reduction between the starting point of change in drivers’ behavior and the point at which the signs were installed. The standard deviation shows the diameter of driver’s deceleration measure for each sign. The gas pedal power shows the pressure applied by the driver between the start of change in driver behavior and the point at which signs were installed.
  • Conventional static warning signs for black ice
The result of the speed deceleration comparison between the groups who were presented with the conventional static warning signs for black ice and no sign show the ineffectiveness of the conventional static warning signs in terms of reducing vehicle speed for safety, regardless of sign visibility. This means that the drivers who saw the conventional static warning signs for black ice drove the same as the drivers who did not see the signs. An interesting result was that the speed deceleration value in the tangent segment was higher than that in the curved segment, regardless of whether the drivers saw the signs.
2.
New warning signs not actuated by weather
The results in Table 5 for the new non-actuated warning signs indicate that the new warning signs were more effective than the conventional static warning signs in terms of slowing down vehicles, regardless of whether they were actuated. An interesting result was that the speed deceleration value of new non-actuated warning signs in the tangent segment was higher (not seen: 13.30 km/h; seen: 15.99 km/h) than that of the new weather-actuated warning signs (not seen: 11.36 km/h; seen: 10.95 km/h). This is because the preceding vehicle was affected only by the new warning signs from the location at which drivers’ behavior started to change, and the distance between the sign and the location was 160 m, as shown in Table 2. The result indicated that drivers did not show a consistent pattern for the new non-actuated warning signs.
3.
New weather-actuated warning signs
Previous studies on roadside signs have suggested that horizontal curves are among the most hazardous situations for drivers [17,18,19,20,21]. In most studies, drivers who are regularly unaware of approaching changes in roadway geometry or do not sufficiently reduce their operating speed when negotiating these geometric changes were analyzed [38,39,40]. Therefore, the new weather-actuated warning signs for black ice must demonstrably provide warning information to driver such that they can take appropriate actions when faced with a hazardous curve.
The results obtained using the new weather-actuated warning signs, summarized in Table 5, indicate that, in terms of slowing down vehicles or warning drivers about hazards in curved segment, these warning signs were the most effective when actuated. The speed deceleration due to the new weather-actuated warning signs increased by 33.7% when drivers looked at the signs (12.80 km/h) compared to when they did not look at the signs (9.57 km/h). Moreover, among all signs, the drivers who saw the new weather-actuated warning signs had the lowest gas pedal power (0.37/0.33) in the curved and tangent segments. This is because the preceding vehicle perceived the hazard warning from the new weather-actuated warning sign, which caused the driver of that vehicle to reduce the pressure on the gas pedal in response.

5. Discussion

Black ice is one of the main causes of winter traffic accidents, and warning signs for black ice are generally used to prevent those accidents. However, these signs provide warnings about black ice to drivers throughout the year, even in summer when the possibility of black ice is extremely low, because the sign is a static sign. Thus, drivers grow accustomed to the idea that the presence of the conventional static warning sign does necessarily indicate the presence of a hazardous condition ahead, causing drivers to ignore warning signs. To overcome this limitation, new warning signs for black ice were developed in a study under the Ministry of Land, Infrastructure, and Transport of Republic of Korea [55]. In that study, various technologies were used to change the color of the sign in response to different temperatures, and our research team investigated the performance and effectiveness of the new warning signs in terms of driver behaviors and deduced guidelines for installation of the developed signs.
A driving simulator was used to evaluate the effects of the conventional static warning signs and new weather-actuated signs for black ice for two different road geometries. The participants encountered a curve in a simulated road that offered a clear sight of the roadside. It was hypothesized that, upon provision of the new weather-actuated warning signs, the drivers would be cautious and decelerate to drive safely. Drivers’ behaviors were recorded after computing vehicle speed and gas pedal power data to determine where the signs had an impact, regardless of the geometrical aspect. The results indicated that the conventional static warning signs for black ice did not appropriately alert drivers. These results are consistent with the results of previous studies, which suggested that the static warning signs for occasional hazards may lose their effectiveness because no real threat is perceived by drivers most of the time [12]. This result is in line with those of previous studies in which static warning signs performed inadequately [12,13].
We systematically evaluated the effectiveness of warning signs for black ice by analyzing the behavioral changes of participants in different geometrical cases and the interactions between the participants and the installed signs in terms of speed deceleration values and gas pedal power. Furthermore, we investigated drivers’ behavior within visual correlation between the drivers who fixed and did not fix eyes on the signs. The results indicated that the existence of warning signs was significantly associated with drivers’ behaviors such as speed reduction. Moreover, this finding is consistent with the reviewed previous studies in which it was concluded that visual traffic warning devices are correlated with speed reduction [16,17,25,28,31].
This finding reinforces how drivers’ speed reduction suggests the effectiveness of warning signs. The conventional static warning signs, which had the lowest number of fixation visits and shortest duration, showed that drivers did not reduce their speed. Meanwhile, the new weather-actuated warning signs, which had the greatest number of fixation visits and longest duration, showed the most significant speed reduction. The increase in fixation visits and duration of new weather-actuated warning signs may represent that the signs warned drivers to become cautious, causing them to increase processing demands with respect to the signs [56,57]. Through the study, it was also found that when drivers were more fixed on the signs, they mostly presented greater speed reduction. This finding implies that traffic warning systems including warning signs must draw drivers’ eyes as a fixation maneuver. However, duration of fixation is related to ease of comprehension; thus, a higher duration of fixation may denote a lower efficiency of the signs and less safe conditions.
There are some limitations of this research to be improved through further studies. First, a study using a driving simulator can always have problems related to completely representing real driving situations, despite the many strengths of driving simulator experiments. The simulator was designed to recreate the characteristics of different driving conditions as precisely as possible, but it cannot capture the dynamism of real-life driving. Nevertheless, this simulator has been used extensively [31,32]. Second, Because South Korea does not permit a driver’s license for color-blind individuals, the experiment did not consider drivers with color blindness to analyze the effectiveness of black ice warning signs design. Third, to obtain more universal results, the sample size from the experiment may not be enough, and there was no consideration of population distribution including age and gender in Korea. Therefore, further research with larger samples, different ages, and similar proportions of individuals from different genders is needed to confirm the findings of this study [8,54,58,59] Fourth, as addressed in Section 3.1.1, fixation was defined as the point at which the eyes were stationary and focused on an object for longer than 60 ms [41]. Therefore, drivers who looked at warning signs for less than 60 ms were included in the group that did not fix eyes on signs. Further research with more appropriate regulation with correlated references for the definition of fixation are needed to confirm the findings of this study. Fifth, for the purpose of the study, participants learned the meaning of the new sign and mechanism of the actuated sign before beginning the driving simulator experiment. This may have brought some distortion of results. In further study, this issue should be considered carefully. Further investigations are also needed to determine the effect of such sign during nighttime, snow, fog, and rainy conditions.

6. Conclusions

Conventional static warning signs for black ice are commonly used in traffic environments worldwide [12]. Studies have demonstrated that many static warning signs have not been adequately evaluated to judge their effectiveness [12,13]. Even though the availability of warning signs is significantly associated with drivers’ deceleration behavior, these finding seem to suggest that the conventional static warning signs for black ice fail to clearly convey to a driver the hazardous conditions ahead with the lowest number of fixations and shortest duration among all signs. Lastly, the results of the experiments suggest that the static warning signs for black ice are statistically insignificant in terms of their effectiveness in providing hazard perception to drivers to reduce speed, with the highest number of fixations and longest duration time.
By contrast, the new weather-actuated warning signs successfully provided hazard perception to drivers, who reduced vehicle speed significantly in response. The new warning signs led to the greatest deceleration, which signifies their effectiveness in terms of clearly conveying hazardous conditions ahead to drivers. Moreover, the new warning signs were the most effective in terms of reducing vehicle speed when they were actuated by freezing weather. The number of fixation visits was the highest among all other signs. However, due to the unfamiliar new weather-actuated design, the sign showed the longest fixation duration [55,56]. Nevertheless, the new warning signs were associated with the lowest gas pedal power when they were actuated by weather.
This study not only provided valid analytic data about the effectiveness of black ice warning signs, which suggest that static conventional warning signs are inadequate while the new weather-actuated warning signs are adequate, but also exploited the benefits of the driving simulator and eye-tracking technologies, which could contribute to future evaluations of the effects of warning signs on drivers’ speed-reducing behaviors and visual attention toward signs.

Author Contributions

Conceptualization, D.L. and S.H.; methodology, G.L., D.L. and S.H.; software, G.L.; validation, G.L., D.L. and S.H.; formal analysis, G.L. and S.H.; investigation, G.L. and S.H.; resources, G.L. and S.H.; data curation, G.L. and D.L.; writing—original draft preparation, G.L. and D.L.; writing—review and editing, D.L.; visualization, G.L.; supervision, D.L.; project administration, D.L.; funding acquisition, D.L. All authors have read and agreed to the published version of the manuscript.

Funding

The authors would like to express their gratitude for the financial support received from the Korea Agency for Infrastructure Technology Advancement, project “Development of road traffic safety improvement technology considering characteristics of external stimuli and traffic vehicles, grant number KAIA22POQW-B152745-04”.

Institutional Review Board Statement

Ethical review and approval were waived for this study because personal identification information was not collected and used for any analysis in this study as defined by Korean Bioethics and Safety ACT Enforcement Regulation (773-13) and due to restrictions arising from the COVID-19 pandemic.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data used to support the findings of this study are available from the corresponding author upon request.

Acknowledgments

This work was supported by the Korea Agency for Infrastructure Technology (KAIA22POQW-B152745-04).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Andersson, A.K.; Chapman, L. The impact of climate change on winter road maintenance and traffic accidents in West Midlands, UK. Accid. Anal. Prev. 2011, 43, 284–289. [Google Scholar] [CrossRef] [PubMed]
  2. Andreescu, M.; Frost, D. Weather and traffic accidents in Montreal, Canada. Climate Research [Internet]. Inter-Res. Sci. Cent. 1998, 9, 225–230. [Google Scholar] [CrossRef] [Green Version]
  3. Chapman, L.; Thornes, J.E. What spatial resolution do we need for a route-based road weather decision support system? Theor. Appl. Climatol. 2011, 104, 551–559. [Google Scholar] [CrossRef]
  4. Norrman, J.; Eriksson, M.; Lindqvist, S. Relationships between road slipperiness, traffic accident risk and winter road maintenance activity. Clim. Res. 2000, 15, 185–193. [Google Scholar] [CrossRef] [Green Version]
  5. Usman, T.; Fu, L.; Miranda-Moreno, L.F. Accident prediction models for winter road safety. J. Transp. Res. Board 2011, 2237, 144–151. [Google Scholar] [CrossRef] [Green Version]
  6. Lee, H.S. Analyses on sunshine influence and surface freezing section of road using GIS. J. Korean Soc. Surv. Geod. Photogramm. Cartogr. 2005, 23, 293–301. [Google Scholar]
  7. Samsung Traffic Safety Research Institute. Analysis of the Relationship between Regional Climate Changes and Traffic Accidents; Samsung Traffic Safety Research Institute: Seoul, Korea, 2016. [Google Scholar]
  8. Wetton, M.; Horswill, M.; Hatherly, C.; Wood, J.; Pachana, N.; Anstey, K. The development and validation of two complementary measures of drivers’ hazard perception ability. Accid. Anal. Prev. 2010, 42, 1232–1239. [Google Scholar] [CrossRef] [Green Version]
  9. Wallis, T.S.A.; Horswill, M.S. Using fuzzy signal detection theory to determine why experienced and trained drivers respond faster than novices in a hazard perception test. Accid. Anal. Prev. 2007, 39, 1177–1185. [Google Scholar] [CrossRef]
  10. Horswill, M.S.; McKenna, F.P. Drivers’ hazard perception ability: Situation awareness on the road. In A Cognitive Approach to Situation Awareness; Banbury, S., Tremblay, S., Eds.; Ashgate: Aldershot, UK, 2004; pp. 155–175. [Google Scholar]
  11. Shahar, A.; Alberti, C.; Clarke, D.; Crundall, D. Hazard perception as a function of target location and the field of view. Accid. Anal. Prev. 2010, 42, 1577–1584. [Google Scholar] [CrossRef]
  12. Al-Kaisy, A.; Hardy, A.; Nemfakos, C. Static warning signs of occasional hazards: Do they work? ITE J. 2008, 6, 38–42. [Google Scholar]
  13. Scialfa, C.T.; Borkenhagen, D.; Lyon, J.; Deschênes, M.; Horswill, M.; Wetton, M. The effects of driving experience on responses to a static hazard perception test. Accid. Anal. Prev. 2012, 45, 547–553. [Google Scholar] [CrossRef] [PubMed]
  14. Carson, J.; Mannering, F. The effect of ice warning signs on ice-accident frequencies and severities. Accid. Anal. Prev. 2001, 33, 99–109. [Google Scholar] [CrossRef]
  15. Insurance Institute for Highway Safety. Fatality Facts 2008: Roadside Hazards. 2010. Available online: http://www.iihs.org/research/fatality_facts_2008/roadsidehazards.html (accessed on 17 February 2010).
  16. Pradhan, A.; Pollatsek, A.; Knodler, M.; Fisher, D. Can younger drivers be trained to scan for information that will reduce their risk in roadway scenarios that are hard to identify as hazardous? Ergonomics 2009, 52, 657–673. [Google Scholar] [CrossRef] [Green Version]
  17. Vest, A.; Stamatiadis, N. Use of warning signs and markings to reduce speeds on curves. In Proceedings of the 84th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 9–13 January 2005. [Google Scholar]
  18. Zwahlen, H.T. Advisory speed signs and curve signs and their effect on driver eye scanning and driving performance. Transp. Res. Rec. 1987, 1111, 110–120. [Google Scholar]
  19. Zador, P.; Stein, H.S.; Wright, P.; Hall, J. Effects of chevrons, post mounted delineators, and raised pavement markers on driver behaviour at roadway curves. Transp. Res. Rec. 1987, 1114, 1–10. [Google Scholar]
  20. Horberry, T.; Anderson, J.; Regan, M.A. The possible safety benefits of enhanced road markings: A driving simulator evaluation. Transp. Res. Part F Traffic Psychol. Behav. 2006, 9, 77–87. [Google Scholar] [CrossRef]
  21. Lyles, R.W. An Evaluation of Warning and Regulatory Signs for Curves on Rural Roads (Report No. FHWA/RD-80/009); FWHA, US Department of Transportation: Washington, DC, USA, 1980.
  22. Porter, R.J.; Donnell, T.E.; Mahoney, K.M. Evaluation of effects of centreline rumble strips on lateral vehicle placement and speed. Transp. Res. Rec. 2004, 1862, 10–16. [Google Scholar] [CrossRef]
  23. Arnold, E.D.; Lantz, K.E. Evaluation of Best Practices in Traffic Operations and Safety: Phase I: Flashing LED Stop Sign and Optical Speed Bars; Report FHWA/VTRC 07-R34; Virginia Transportation Research Council: Charlottesville, VA, USA, 2007. [Google Scholar]
  24. Thompson, M.B.; Carlson, P.J. Speed changes due to transverse rumble strips on approaches to high-speed stop-controlled intersections. J. Transp. Res. Board 2006, 1973, 1–9. [Google Scholar] [CrossRef]
  25. Katz, J.; Duke, D.E.; Rakha, H.A. Design and evaluation of peripheral transverse bars to reduce vehicle speeds. In Proceedings of the 85th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 22–26 January 2006; p. 14. [Google Scholar]
  26. Krull, K.; Hummer, J. The Effect of Fluorescent Yellow Warning Signs at Hazardous Locations; Final Report; North Carolina State University: Raleigh, NC, USA, 2000. [Google Scholar]
  27. Al-Ghamdi, A.S.; Al-Gadhi, S.A. Warning signs as countermeasures to camel-vehicle collisions in Saudi Arabia. Accid. Anal. Prev. 2004, 36, 749–760. [Google Scholar] [CrossRef]
  28. Montella, A.; D’Ambrosio, A.; Aria, M.; Galante, F.; Mauriello, F.; Pernetti, M. Perceptual measures to influence operating speeds and reduce crashes at rural intersections: Driving simulator experiment. Transp. Res. Rec. 2010, 2149, 11–20. [Google Scholar] [CrossRef]
  29. Dutta, A.; Fisher, D.L.; Noyce, D.A. Use of a driving simulator to evaluate and optimize factors affecting understandability of variable message signs. Transp. Res. Part F Traffic Psychol. Behav. 2004, 7, 209–227. [Google Scholar] [CrossRef]
  30. Jamson, A.H.; Pyne, H.C.; Carste, O.M.J. Evaluation of traffic calming measures using the Leeds Driving Simulator. In Proceedings of the Driving Simulation Conference, Paris, France, 7–8 July 1999. [Google Scholar]
  31. Katz, B.J.; Molino, J.A.; Rakha, H.A. Evaluation of design alternatives of peripheral transverse bars to reduce vehicle speeds and center line encroachment in a driving simulator. In Proceedings of the 87th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 13–17 January 2008. [Google Scholar]
  32. Godley, S.; Triggs, T.J.; Fildes, B.N. Driving simulator validation for speed results. Accid. Anal. Prev. 1997, 34, 589–600. [Google Scholar] [CrossRef]
  33. Brenac, T. Safety at curves and road geometry standards in some European countries. Transp. Res. Rec. 1996, 1523, 99–106. [Google Scholar] [CrossRef]
  34. Charlton, S.G. Perceptual and attentional effects on drivers’ speed selection at curves. Accid. Anal. Prev. 2004, 36, 877–884. [Google Scholar] [CrossRef]
  35. Charlton, S.G. The role of attention in horizontal curves: A comparison of advance warning, delineation, and road marking treatments. Accid. Anal. Prev. 2007, 39, 873–885. [Google Scholar] [CrossRef] [PubMed]
  36. Srinivasan, R.; Baek, J.; Carter, D.L.; Persaud, B.; Lyon, C.; Eccles, K.A.; Gross, F.B.; Lefler, N. Safety Evaluation of Improved Curve Delineation; Publication FHWA-HRT-09-045; US Department of Transportation: Washington, DC, USA, 2009.
  37. Veneziano, Y.Z.; Turnbull, I. Speed impacts of an icy curve warning system. IET Intell. Transp. Syst. 2014, 8, 93–101. [Google Scholar] [CrossRef] [Green Version]
  38. Lindgren, R.; Clair, S.S. Butte creek ice warning system. In Oregon Department of Transportation Research Report; Oregon Institute of Technology: Salem, MA, USA, 2009. [Google Scholar]
  39. Liu, T.; Wang, N.; Yu, H.; Basara, J.; Hong, Y.E.; Bukkapatnam, S. Black Ice Detection and Road Closure Control System for Oklahoma; Department of Transportation, Planning and Research Division: Oklahoma City, OK, USA, 2014.
  40. Michigan Department of Transportation. Bridge Deck Warning System, Roundabouts Highlight a Year of MDOT Safety Projects; Michigan Department of Transportation: Traverse City, MI, USA, 2019.
  41. Raney, E.G.; Campbell, S.J.; Bovee, J.C. Using eye movements to evaluate the cognitive processes involved in text comprehension. J. Vis. Exp. 2014, 83, e50780. [Google Scholar] [CrossRef]
  42. Pradhan, A.; Hammel, K.; DeRamus, R.; Pollatsek, A.; Noyce, D.; Fisher, D. The use of eye movements to evaluate the effects of driver age on risk perception in an advanced driving simulator. Hum. Factors 2005, 47, 840–852. [Google Scholar] [CrossRef]
  43. Cathcart, J.M.; Doll, T.J.; Schmieder, D.E. Target detection in urban clutter. IEEE Trans. Syst. Man Cybern. 1989, 19, 1242–1250. [Google Scholar] [CrossRef]
  44. Ho, G.; Scialfa, C.; Caird, J.; Graw, T. Visual search for traffic signs: The effects of clutter, luminance, and aging. Hum. Factors 2001, 43, 194–207. [Google Scholar] [CrossRef]
  45. Huestegge, L.; Skottke, E.M.; Anders, S.; Müsseler, J.; Debus, G. The development of hazard perception: Dissociation of visual orientation and hazard processing. Transp. Res. Part F Traffic Psychol. Behav. 2010, 13, 1–8. [Google Scholar] [CrossRef]
  46. Torbjörn, F.; Gregersen, N.P. A comparison of eye movement behavior of inexperienced and experienced drivers in real traffic environments. Optom. Vis. Sci. 2005, 82, 732–739. [Google Scholar]
  47. Niehorster, D.C.; Santini, T.; Hessels, R.S.; Hooge, I.T.; Kasneci, E.; Nyström, M. The impact of slippage on the data quality of head-worn eye trackers. Behav. Res. Methods 2020, 52, 1140–1160. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  48. Marín-Conesa, E.; Sánchez-Ferrer, F.; Grima-Murcia, M.D.; Sánchez-Ferrer, M.L. The application of a system of eye tracking in laparoscopic surgery: A new didactic tool to visual instructions. Front. Surg. 2021, 8, 643611. [Google Scholar] [CrossRef]
  49. Othman, N.; Fairuz, I.R. Mental workload evaluation of pilots using pupil dilation. Int. Rev. Aerosp. Eng. 2016, 9, 80–84. [Google Scholar] [CrossRef]
  50. Rayner, K.; Inhoff, A.W.; Morrison, R.E.; Slowiaczek, M.L.; Bertera, J.H. Masking of foveal and parafoveal vision during eye fixations in reading. J. Exp. Psychol. Hum. Percept. Perform. 1981, 7, 167. [Google Scholar] [CrossRef] [PubMed]
  51. Deng, H.; Zheng, C.Q.; Shao, C.F.; Li, Y.H. Construction and application of 3D traffic environment build platform based on UC-win/Road. In Applied Mechanics and Materials; Trans Tech Publications Ltd.: Stafa-Zurich, Switzerland, 2014; Volume 505, pp. 1225–1229. [Google Scholar]
  52. Fukuda, T.; Hitoshi, T. Development of use flow of 3D CAD/VR software for citizens who are non-specialists in city design. In Proceedings of the 15th International Conference on Computer Aided Architectural Design Research in Asia, Hong Kong, China, 7–10 April 2010; pp. 521–530. [Google Scholar]
  53. Liang, G.; Wang, F.; Shi, Q.; Li, R.; Ji, Z.; Dong, C. Research on the prepositive distance of crosswalk warning markings for unsignalized road section. J. Adv. Transp. 2021, 2021, 8855970. [Google Scholar] [CrossRef]
  54. Lee, G.S.; Lee, D.M.; Hwang, S.C.; Kwon, W.T. Analyzing driving behavior, road sign attentiveness and recognition with eye tracking data. J. Korea Inst. Intell. Transp. Syst. 2021, 20, 117–132. [Google Scholar] [CrossRef]
  55. Korea Agency for Infrastructure Technology Advancement. Development of Road Traffic Safety Improvement Technology Considering Characteristics of External Stimuli and Traffic Vehicles; Korea Agency for Infrastructure Technology Advancement: Anyang, Gyeonggi-do, Korea, 2018; Volume 20, pp. 42–47. [Google Scholar]
  56. Babić, D.; Dijanić, H.; Jakob, L.; Babić, D.; Garcia-Garzon, E. Driver eye movements in relation to unfamiliar traffic signs: An eye tracking study. Appl. Ergon. 2020, 89, 103191. [Google Scholar] [CrossRef]
  57. Horswill, M.; Kemala, C.; Wetton, M.; Scialfa, C.; Pachana, N. Improving older drivers’ hazard perception ability. Psychol. Aging 2010, 25, 464–469. [Google Scholar] [CrossRef] [Green Version]
  58. Horswill, M.S.; Pachana, N.A.; Wood, J.; Marrington, S.A.; McWilliam, J.; McCollough, C.M. A comparison of the hazard perception ability of matched groups of healthy drivers aged 35–55, 65–74, and 75–84 years. J. Int. Neuropsychol. Soc. 2009, 15, 799–802. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  59. Horswill, M.S.; Anstey, K.J.; Hatherly, C.G.; Wood, J.M. The crash involvement of older drivers is associated with their hazard perception latencies. J. Int. Neuropsychol. Soc. 2010, 16, 939–944. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Ijerph 19 07549 g001 550
Figure 1. Location of study sites and the signs on the segments for experiments: (a) driving simulator; (b) driving simulator during experiment.
Figure 1. Location of study sites and the signs on the segments for experiments: (a) driving simulator; (b) driving simulator during experiment.
Ijerph 19 07549 g001
Ijerph 19 07549 g002 550
Figure 2. Study signs for each experiment step: (a) six versions of new warning signs actuated by weather conditions; (b) final selected design for new warning sign before actuation (top) and after actuation by freezing temperatures (bottom); The Korean sign means “Caution on Black Ice”.
Figure 2. Study signs for each experiment step: (a) six versions of new warning signs actuated by weather conditions; (b) final selected design for new warning sign before actuation (top) and after actuation by freezing temperatures (bottom); The Korean sign means “Caution on Black Ice”.
Ijerph 19 07549 g002
Ijerph 19 07549 g003 550
Figure 3. The signs used for VR simulator experiments: (a) conventional static warning sign for black ice; (b) new warning sign not actuated by weather conditions; (c) new warning sign actuated by weather conditions. The Korean sign means “Caution on Black Ice”.
Figure 3. The signs used for VR simulator experiments: (a) conventional static warning sign for black ice; (b) new warning sign not actuated by weather conditions; (c) new warning sign actuated by weather conditions. The Korean sign means “Caution on Black Ice”.
Ijerph 19 07549 g003
Ijerph 19 07549 g004 550
Figure 4. Location of study sites and the signs on the segments for experiments: (a) curve segment study site; (b) tangent segment study site; (c) curve segment in experiments; (d) tangent segment in experiments.
Figure 4. Location of study sites and the signs on the segments for experiments: (a) curve segment study site; (b) tangent segment study site; (c) curve segment in experiments; (d) tangent segment in experiments.
Ijerph 19 07549 g004
Ijerph 19 07549 g005 550
Figure 5. Speed and gas pedal power comparison for no-sign conditions and with signs in curve segment (left) and tangent segment (right).
Figure 5. Speed and gas pedal power comparison for no-sign conditions and with signs in curve segment (left) and tangent segment (right).
Ijerph 19 07549 g005
Ijerph 19 07549 g006 550
Figure 6. Comparison between no-sign conditions and conventional static warning signs for black ice.
Figure 6. Comparison between no-sign conditions and conventional static warning signs for black ice.
Ijerph 19 07549 g006
Ijerph 19 07549 g007 550
Figure 7. Speed and gas pedal power comparison between no-sign conditions and each sign, including conventional static warning sign and new warning signs for black ice actuated and not actuated by weather.
Figure 7. Speed and gas pedal power comparison between no-sign conditions and each sign, including conventional static warning sign and new warning signs for black ice actuated and not actuated by weather.
Ijerph 19 07549 g007
Table
Table 1. Participant age groups by gender.
Table 1. Participant age groups by gender.
Age20s30s40s50sTotal
Male4136629
Female22318
Total6159737
Table
Table 2. Speed and gas pedal power comparison between no-sign conditions and signs in t-test for two cases.
Table 2. Speed and gas pedal power comparison between no-sign conditions and signs in t-test for two cases.
Distance From Sign (m)Mean (Standard Deviation)
No Sign/Signs		t-Value
(p-Value)
SpeedGas Pedal PowerSpeedGas Pedal Power
Curve segment12051.74/49.17 (8.74/11.37)0.12/0.15 (0.15/0.18)1.482(0.143)−1.083 (0.283)
11050.55/47.68 (8.55/11.51)0.19/0.19 (0.16/0.19)1.4 (0.163)0.053 (0.958)
10050.05/46.74 (8.10/11.42)0.25/0.19 (0.17/0.19)2.001 * (0.049)1.653 (0.104)
9049.91/45.88 (7.55/11.26)0.3/0.23 (0.17/0.19)2.565 * (0.012)2.395 * (0.02)
Tangent segment19056.89/56.96 (6.06/7.31)0.75/0.67 (0.13/0.18)1.463 (0.145)2.627 ** (0.009)
18057.70/55.37 (6.13/7.48)0.76/0.66 (0.12/0.2)1.725 (0.086)2.888 ** (0.004)
17058.47/55.69 (6.23/7.64)0.75/0.63 (0.12/0.23)2.019 * (0.045)4.257 ** (0.000)
16059.15/55.84 (6.33/7.93)0.72/0.6 (0.17/0.25)2.325 * (0.021)3.282 ** (0.002)
* p < 0.05, ** p < 0.01.
Table
Table 3. Comparison between each of the signs in terms of speed deceleration values in two cases.
Table 3. Comparison between each of the signs in terms of speed deceleration values in two cases.
Sign Conditiont-Value (p-Value)
No SignOld SignNot ActuatedActuated
Curve segmentNo sign----
Old sign−0.048 (0.962)---
Not actuated−0.426 (0.672)−0.360 (0.72)--
Actuated−3.317 (0.001 **)−3.125 (0.003 **)−2.683 (0.009 **)-
Tangent segmentNo sign----
Old sign−1.220 (0.227)---
Not actuated−0.192 (0.848)1.103 (0.274)--
Actuated−2.978 (0.004 **)−1.878 (0.065)−2.934 (0.005 **)-
** p < 0.01.
Table
Table 4. Fixation visits and fixation duration for each sign in two cases.
Table 4. Fixation visits and fixation duration for each sign in two cases.
Number of Participants Who Fixed Eyes on Sign (%)Total Number of Fixation Visits (N)Mean of Fixation Duration (SD) (ms)
Curve segmentOld sign12 (32%)25252.36 (222.9)
Not actuated19 (51%)67247.27 (162.5)
Actuated19 (51%)62352.03 (239.0)
Tangent segmentOld sign16 (43%)65499.44 (594.9)
Not actuated19 (51%)93621.51 (782.7)
Actuated23 (62%)133606.67 (715.6)
Table
Table 5. Comparison between groups who fixated or did not fixate on signs.
Table 5. Comparison between groups who fixated or did not fixate on signs.
Eye Fixation on Signs/No Eye Fixation
Old SignNot ActuatedActuated
Curve
Segment		Speed deceleration (km/h)6.84/6.3210.83/11.9212.80/9.57
Speed standard deviation (km/h)2.4/2.214.08/4.444.49/3.34
Gas pedal power0.38/0.430.50/0.480.37/0.33
Tangent
Segment		Speed deceleration (km/h)8.15/7.0415.99/13.3010.95/11.36
Speed standard deviation (km/h)2.84/2.604.55/4.313.93/3.97
Gas pedal power0.62/0.560.64/0.580.53/0.5
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Lee, G.; Hwang, S.; Lee, D. Improvements of Warning Signs for Black Ice Based on Driving Simulator Experiments. Int. J. Environ. Res. Public Health 2022, 19, 7549. https://doi.org/10.3390/ijerph19127549

AMA Style

Lee G, Hwang S, Lee D. Improvements of Warning Signs for Black Ice Based on Driving Simulator Experiments. International Journal of Environmental Research and Public Health. 2022; 19(12):7549. https://doi.org/10.3390/ijerph19127549

Chicago/Turabian Style

Lee, Ghangshin, Sooncheon Hwang, and Dongmin Lee. 2022. "Improvements of Warning Signs for Black Ice Based on Driving Simulator Experiments" International Journal of Environmental Research and Public Health 19, no. 12: 7549. https://doi.org/10.3390/ijerph19127549

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop