1. Introduction
Pedestrians are the most vulnerable group in road traffic accidents, and collisions between vehicles and pedestrians are recognized as a serious problem in numerous countries. According to the World Health Organization, the pedestrian fatality rate accounts for ~23% of the entire fatality rate in road transportation [
1]. In Korea, as a result of continuous efforts to reduce pedestrian fatalities, the number of such fatalities decreased by an average of 6.7% over the past 10 years (2011–2020) [
2]. However, compared with OECD countries, in terms of the number of pedestrian deaths in 2018, Korea ranked 30th out of 31 countries [
3], and there are still more than 1000 deaths occurring a year [
2].
In general, human factors, vehicle factors, and road/environmental factors influence the occurrence of traffic accidents [
4]. The risk of accidents due to vehicle factors and road/environmental factors can be reduced by developing vehicle technology and through advances in traffic operation technology [
5]. However, human behavior, which is a human factor, has limitations, in that the speed of change is slow and difficult, unlike technological development; thus, most traffic accidents are related to the behavior of pedestrians and drivers [
5]. The major risk behaviors that have been observed recently include the use of mobile phones or related devices while on the move, and many studies suggest that these behaviors increase the occurrence or risk of accidents [
6,
7,
8,
9]. In Japan, about 35% of fatal pedestrian accidents are caused by driver distraction [
10], constituting the main cause of vehicle and pedestrian collisions [
11].
Among pedestrian traffic accidents in Korea, accidents that occurred while crossing the road accounted for 52.5% of the total in 2020, which is a larger proportion than other types [
2]. In particular, the fatal accident of a child pedestrian on a crosswalk by a right-turning vehicle in 2021 became a significant social issue [
12,
13]. From 2018 to 2020, the percentage of pedestrian casualties caused by right-turning vehicles comprised 9.9% of all traffic accidents, and it was found that the drivers could not recognize pedestrians due to the blind spot when turning right [
14]. In other words, to secure the safety of crossing pedestrians, it is necessary for both drivers and crossing pedestrians to pay close attention to the risk of a collision, with drivers’ attentive behavior emphasized for the safety of pedestrians.
In the past, to reduce road traffic accidents, a method to forcibly reduce the speed of a vehicle by installing a physical facility on the road was applied based on a study on the severity and risk of pedestrian accidents according to vehicle speed [
15,
16,
17,
18,
19]. However, with the development of science and technology, voluntary speed reduction is induced as a supplement. Representative examples include lighting facilities around crosswalks and curbs using object-recognition technology [
20,
21,
22,
23] and flashing safety signs such as VMS (variable message sign) information [
22,
24,
25,
26,
27,
28,
29,
30,
31]. In addition, a number of studies analyzed the effect of speed reduction through empirical studies to improve traffic safety [
20,
21,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31].
This study analyzed the effect of a pedestrian-crossing information notification system provided in the blind spot of the right turn in a children’s protection zone (School Zone), which is a low-speed section of Jeollabuk-do. In the process of installing the system, a briefing session was conducted for local residents. This study is meaningful because it does not simply compare the situations before and after system installation but examines changes in driver behavior depending on whether information is provided in a situation where the driver is fully aware of the system. As for the indicators of the analysis, as presented in previous studies, an analysis was performed on speed changes in vehicles, and drivers further reviewed their compliance with the speed limit, one of the main causes of traffic accidents [
32].
2. Literature Review
Considering previous empirical studies that applied detection technology, Hakkert et al. (2002) [
20] installed sensors on both sides of the crosswalk to detect pedestrians in the crosswalk waiting section and provide information to vehicles through floor warning lights. As a result, vehicle speed decreased by 2 to 5 km/h. Furthermore, it was suggested that the rate of yielding the right-of-way to pedestrians increased. Costa et al. (2020) [
21] evaluated an integrated lighting warning system in non-signal crosswalks at nighttime and compared and analyzed the effects under seven different conditions. They found that the yield rate of vehicles improved the most when drivers were provided with information that enabled all standard road lighting plus enhanced dedicated lighting, flashing beacons, and curb LED lighting. Hong et al. (2021) [
22] analyzed the effect of an information-providing system using VMS and floor warning lights through vehicle and pedestrian detection information collected from radar sensors and thermal imaging cameras. As a result of analyzing the day and nighttime zones separately for a children’s protection zone and an elderly protection zone, it was found that the approaching speed of vehicles at the crosswalk decreased. Jin and Lee (2016) [
23] performed an analysis of the effects of a cross-safety support system that provides information to pedestrians and drivers using VMS, LED lights in crosswalks, text-display devices, and voice devices. By analyzing vehicle driving speed and pedestrian behavior during the day and nighttime for the crosswalk in front of an elementary school, they found that the vehicle speed decreased, and, in terms of pedestrian behavior, the frequency of looking left and right before and during crossing also increased by about 1.63–2.25 times depending on the time of the day. Hoye and Laureshyn (2019) [
29] analyzed the effect of a pedestrian crossing warning system with an automatic pedestrian-detection function and showed that the yielding behavior of drivers improved without negatively affecting the behavior of pedestrians.
Regarding studies related to rectangular rapid flashing beacons (RRFBs), Van Houten et al. (2010) [
30] compared and analyzed the effects of installing additional beacons in a median zone (or refuge island) and the side of the road to improve the visibility of RRFBs. They showed that the yielding behavior of the driver was effectively improved due to RRFBs, and when RRFBs were additionally installed in the median zone (or refuge island), the efficiency was further increased. In addition, this increase in the driver yield rate continued for two years without a reduction effect. Similar studies [
24,
25,
26,
27,
28,
31] on RRFBs have been conducted, and all studies suggest that the yield rate of drivers improved. In particular, Shurbutt and Van Houten (2008) [
31] reported a reduction in collisions between drivers and pedestrians and the proportion of pedestrians trapped in crosswalks in the middle of the road through light-emitting diode (LED) blinkers with irregular flash patterns. Ross et al. (2011) [
24] recommended the installation of an RRFB in a speed section of 35 mph or more, while Fitzpatrick et al. (2015) [
27] analyzed the difference between RRFBs and circular rapid-flashing beacons (CRFBs) to show that there was no difference in the effect of the beacon type. Fitzpatrick and Park (2021) [
28] found that RRFBs are more effective at night, whereas LED-embedded (LED-Em) crossing signs are more effective during the day. Pedestrian hybrid beacons (PHBs), however, suggested that there was no difference in driver yield between day and night.
5. Conclusions
In this study, we conducted an analysis of a pedestrian information provision system introduced to improve sustainable traffic safety. The system provides pedestrian information in advance to a driver entering an intersection when a pedestrian crosses a crosswalk in the blind spot of a right turn. For the analysis data, the speed and image information of individual vehicles was collected using radar equipment, and the image information of pedestrians and vehicles at crosswalks was collected using video equipment.
For the average point speed of the vehicle, the statistical test (t-test) results showed no significant results regardless of the presence or absence of pedestrians. However, when pedestrians were present (when information was provided), the average point speed of the vehicles showed a slight decrease compared with that before system installation, and when there were no pedestrians (when information was not provided), the average point speed increased slightly. The speed limit compliance rate of vehicles increased compared with that before installation when pedestrians were crossing, whereas, when there were no pedestrians (after installation, the system does not provide information), the speed limit compliance rate of vehicles decreased compared with that before installation. In addition, as a result of the chi-square test, it was found that, in situations where pedestrians exist (when information is provided), there is a difference in the speed limit compliance rate before and after system installation. However, it was found that there was no difference in situations where pedestrians did not exist (when information was not provided).
As shown in the analysis results of this study, even if a pedestrian information provision system is installed in a low-speed section, such as a child protection zone (school zone), the deceleration effect of the vehicle is insignificant. However, by providing pedestrian crossing information, the driver can see that their driving behavior should change to comply with the speed limit (improving the speed limit compliance rate).
However, as our study was conducted on a single point, experiments in various similar sections (low-speed sections) are required in the future. In particular, in Korea, in addition to child protection zones (school zones), there are elderly protection zones (silver zones) and village resident protection zones, depending on the characteristics of the pedestrians. Therefore, it seems necessary to conduct research on these sections.
In this regard, we have two points of discussion.
(i) In terms of traffic safety, the change in vehicle speed is a major indicator, but is it an appropriate evaluation indicator even in a low-speed section?
The speed of the vehicle is highly correlated with the degree of the accident (fatal traffic accident) in the event of a traffic accident. The effect of reducing the speed of a vehicle is an important indicator because it can induce a relatively low-level accident in the event of a sudden stop or an accident caused by a vehicle in a dangerous situation. However, it is necessary to discuss whether it is appropriate to expect this effect even in low-speed driving sections (30 km/h or less), as in our analysis. As an example, the measure of effectiveness presented in the Highway Capacity Manual applied in Korea applies different indicators depending on expressways (density), multi-lane roads (V/C, average travel speed), and two-lane roads (total delay rate). In this way, when evaluating the effectiveness of a traffic safety system (or pedestrian safety system), it is necessary to develop measures of effectiveness according to the speed limit or section.
(ii) Would providing information only in situations where attention is needed satisfy both traffic safety and traffic flow requirements?
In general, in order to improve traffic safety (especially pedestrian safety), the installation of a system of facilities to physically reduce the speed of vehicles has been improved in Korea. However, in the pedestrian information provision system of this study, there was no difference in the average speed even when information was provided to the driver in advance. This means that the travel time does not change significantly from the viewpoint of vehicle traffic. However, by providing pedestrian caution information, it can be seen that the number of speeding vehicles (vehicles violating the speed limit) has decreased. In this respect, we need to look at the intelligent transportation system (ITS) projects introduced in the 1990s. ITSes are built to ease road congestion and improve travel efficiency by providing various types of information to vehicles (drivers) driving on the road [
34,
35,
36,
37,
38,
39]. As such, the pedestrian information provision system of this study is similar to an ITS in the sense that it provides information in advance, and it is judged that it is necessary to discuss efficient operation methods.
Therefore, in the future, related guidelines should be continuously developed to introduce a sustainable traffic safety system (or pedestrian safety system) in relation to the above, and research should be conducted to find ways to maximize the effectiveness of the system, as mentioned in a previous study [
29].