Prerpints.org logo
Preprint
Article

Reconstructing Intersection Conflict Zone: Microsimulation-Based Analysis of Traffic Safety for Pedestrians

Altmetrics

Downloads

31

Views

24

Comments

0

A peer-reviewed article of this preprint also exists.

Submitted:

08 October 2024

Posted:

10 October 2024

You are already at the latest version

Alerts
Abstract

According to statistics from the World Health Organization traffic accidents are one of the leading causes of death among children and young people, and the statistical indicators are even worse for the population of elderly pedestrians. Preventive measures and a significant increase in pedestrian safety require a comprehensive approach that includes analysis of traffic infrastructure and regulations, as well as the behavior and interaction between road users. In this paper, a methodology based on traffic microsimulations was developed for selecting the optimal reconstruction solution of urban traffic infrastructure from the aspect of traffic safety. Comprehensive analyses of local traffic conditions at the location – infrastructural and those related to traffic users was proposed. Developed methodology was applied and tested at a selected unsignalized pedestrian crosswalk in the urban traffic network of the city of Osijek, Croatia. It enabled analyzes of possible solutions for improving the traffic safety for vulnerable pedestrian groups, taking into account the specificities of the chosen location in the residential area and traffic participants behavior measured at the filed. Through database analysis, parameters influencing the reaction time and crossing time of children and elderly pedestrians in conflict zones were identified. Using microsimulation traffic modeling (VISSIM) and statistical tools, an analysis was conducted on the incoming vehicle speeds for both the existing and reconstructed conflict zone solutions under different traffic conditions. The developed methodology allowed the selection of the optimal solution from the perspective of traffic safety, considering the actual and future traffic conditions of the location for all traffic users.

Keywords: 
Subject: Engineering  -   Civil Engineering

1. Introduction and Overview of Existing Research

Statistics related to road traffic safety published regularly by the European Commission [1] show a big difference in the level of road safety among EU countries. Sweden is the country with the best indicator in the number of deaths per 100,000 inhabitants with 22 deaths, but Croatia has in 2023 - 71 deaths per 100,000 inhabitants, which makes it one of the countries with the worst indicators within the EU and Europe in general.
According to the Bulletin on Road Traffic Safety in the Republic of Croatia from 2023 [2], in the last ten years, an average of 32,063 traffic accidents occurred on Croatian roads per year, of which almost a third of accidents resulted in casualties. In the same period, an average of 299 people died in traffic accidents per year, and it is estimated that around 5% of the victims, mostly young people, remained permanently disabled. If we compare casualties by type of traffic - in the EU car occupants suffer the most (45%), followed by pedestrians (18%), motorcyclists (16%) and cyclists (10%). Croatian statistics do not provide an overview of drivers by type of traffic, so the comparable data is the share of pedestrians killed in 2023, which is slightly lower than in the EU and amounts to 16.5%. In Croatia, pedestrians are killed to a significantly greater extent in settlements (66%), and in 2023, 44% of pedestrians killed were over 65 years old. The number of fatally injured child-pedestrians varies by year, in 2023, four children and young people under the age of 17 died in Croatia, which is about 9% of the total number of pedestrians killed during that year.
A factor that is extremely important when discussing a conflict between a vehicle and a pedestrian is the specific speed at which the conflict occurred (impact speed) because it directly affects the severity of pedestrian injuries. In the period from 1980. to 2017. studies were conducted in various countries of the world (UK, USA, Germany, Korea, China, Japan) in which traffic accidents involving pedestrians of different ages were analyzed, as well as the consequences and conditions in which they occurred [3,4,5] to determine the pedestrian fatality risk as a function of car impact speed. Analyzes of the results of these studies suggest that an impact speed of 30 km/h has on average a risk of fatality around 5%, the risk increases to 13% for an impact speed of 40 km/h and 29% at 50 km/h [5]. In Croatia, the regulated speed on city roads, unless otherwise specified, is 50 km/h, which is shown to be a speed that, in the event of a conflict between vehicles and pedestrians in a high proportion, can result in fatal consequences for pedestrians. Speed control on urban roads, which aims to reduce the number and consequences of traffic accidents, is achieved by one or a combination of several types of measures that include legal solutions (repressive measures) and traffic and/or infrastructure interventions, which depend on the location, measured operating speed, expected users and other factors. Reducing the permitted speed on a certain section of the road can be considered a mild measure of speed reduction. Studies that analyzed the influence between changes in speed limit and changes in average or operational speed found that when only the speed limit is decreased the effect on speed is moderate, for the decrease of 10 km/h average speed decreases approximately 2-3 km/h [6].
The analysis of the effects of panels indicating the speed was carried out in several studies in different countries and locations with different results. It has been shown that this measure has an effect if it is placed along the road directly at the location where there is a need to reduce the speed, but not on the stretch of the road where it is placed [7,8]. The effects of the introduction of Tempo-30 zones around school and playground zones in Calgary, Canada were analyzed on a sample of 27 locations to determine what affects the operational (V85) and average speed at these locations. Among the selected parameters (e.g. zone type, presence of children, roadway width, presence of speed monitoring device, road category, and type of control) the presence of a warning device in a location near schools had the greatest impact on speed reduction [9].
The paper [10] presents a study of the implementation of various traffic calming measures in several cities in the north of Spain. The impact of infrastructural solutions with a raised pedestrian crosswalk, narrowing of the traffic lane and the installation of a radar speed camera was analyzed based on 50-percentile and 85-percentile speed data. As a very effective traffic calming measure, the raised crosswalk was detected as it caused a speed decrease of approximately 9-10 km/h and a reduction effect was visible also outside the measurement area. Radar control, on the other hand, had an impact only on the place where it was placed.
The paper [11] analyzed various solutions implemented in Poland to reduce speed at the entrances to the Tempo-30 zones. As a measure of speed control, vertical treatments (speed table, raised junction) and horizontal (roundabout, mini roundabout, roundabout with offset or skewed approaches or both) were applied and analyzed. Changing of the road surface elevation proved to be the most effective measure of speed reduction among the 5 analyzed.
The influence of built environment, pedestrian infrastructure, and road infrastructure on pedestrian safety was analyzed on the basis of 20 roadways and a total of 315 pedestrian crashes in two Portuguese cities (Braga and Guimaraes) [12]. It was shown that longer distances between crosswalks significantly influence the increase in the number of traffic accidents while the presence of raised medians increases the vehicle-pedestrian crash frequency.
Studies investigating the use of the central pedestrian island as a measure to increase pedestrian traffic safety have yielded mixed results. The results of research conducted in Japan on 86,406 traffic accidents involving vehicle-pedestrian collisions [13] show that in intersection crashes, the installation of stop signs, medians, three light traffic signals, and innovative flashing and pedestrian-controlled traffic signals may reduce the fatality risk of pedestrians.
The results of the research conducted in Tehran show a statistically significant reduction in the mean speed of vehicles and the number of fatal accidents in vehicle-pedestrian collisions, after the application of central pedestrian islands [14]. Implementation of refuge islands at pedestrian crosswalks has reduced the number of fatalities for pedestrians by 64%, the research on irregular movements according to the type of crosswalk shows that crosswalks equipped with flashing amber lights, refuge islands, and traffic lights require a much more appropriate behavior from car drivers [15].
Different engineering countermeasures, aimed to increase conspicuity and visibility of pedestrian crosswalks at roundabouts, have been tested in order to assess their impact on road safety. These countermeasures included the installation of a median refuge island, displacement of zebra markings in advance of the intersection, and placement of “Yield here to pedestrians” vertical signs. The safety evaluation was performed by a before–after speed analysis and a driver’s eye movement analysis. The results showed that the reduction in arrival speed after the implementation of the measures was statistically significant, the zebra markings and the median refuge island resulted in the most glanced elements and the central island significantly lowered distractions in the gaze behavior of the driver [16].
Research of the impact of splitter-island on pedestrian safety at roundabouts using surrogate safety measures, conducted in Japan [17] showed that the application of splitter-island had significantly safer performance in all traffic flow directions at roundabouts. The significance of the geometric shape and the driver's vision is investigated by a study [18] and the results show that pedestrian refuges imposing symmetric lateral shift by 1 m which are not accompanied by street furniture items have no significant bearing on speed reduction in their vicinity, and this irrespective of their sitting along the stretch of road in the village and geometry of associated pavement markings. Conversely, an asymmetric lateral shift in the travel way alignment generated by the refuge island located on one side of the road centerline induces a considerable speed reduction, yet only when the driver sees residential buildings in close proximity of the road. The study [19] investigated the effect of various mid-block tools, such as refugee islands, speed tables, and raised pedestrian crosswalks to reduce speed. The findings were that the presence of the refugee island themselves does not significantly reduce the speed of vehicles.
A study conducted in Edinburgh [20] investigated pedestrian crossing behavior at a mid-block crosswalk with a refuge island, in an urban area with a high observed pedestrian accident frequency. The results show that the critical gap for crossing from the median to the curb is much shorter than that from the curb to the median. Pedestrians appear to be less cautious when crossing from the median to the curb as they are more likely to accept a shorter gap in traffic. Research conducted in Poland shows which elements affect the effectiveness of the central island as a traffic calming measure: free view, visibility of a pedestrian on the right-hand side of the island, and the refuge island surroundings [21].
The safety of pedestrian traffic depends also to a large extent on the behavior of pedestrians themselves and their ability to assess the traffic situation and the decisions they make in places where they are potentially subject to conflict with vehicles - pedestrian crosswalks [22-24]. Children and the elderly are particularly vulnerable groups of pedestrians due to their reduced ability to assess the situation and/or motor limitations [24,25]. Research conducted in developed countries shows that elderly people are at greater risk when crossing the street because they have a harder time evaluating their surroundings [26].
Previous research by the authors related to the behavior of children in conflict zones of pedestrian crossings show that the reaction time of younger children is statistically significantly longer than the reaction time of adults, only for children over 12 years of age, this difference is no longer statistically significant [27]. The use of mobile phones that occupies visual attention has been shown to be a statistically significant influencing factor on children’s [27], young people’s [28,29] and pedestrians in general [30] reaction time in real traffic conditions. It was also shown that age of the children as well as movement in the group has an influence on the behavior of children at pedestrian crosswalks with [31,32] and without traffic lights [33]. Of the infrastructural elements, children's behavior was significantly influenced by the length of the zone where they are potentially exposed to conflict with vehicles – conflict zone. When it comes to pedestrian crosswalks without traffic lights, preliminary results show that the approaching vehicle operating speed also has an impact on children's behavior [33].
Analyzes of the behavior and awareness of road traffic hazards among pedestrians in Poland for the age groups under 18 and over 65 years [34] was conducted using the survey on a sample of 265 participants under the age of 18 and 357 participants over the age of 65. The results show that older pedestrians less frequently exhibit dangerous behavior like crossing the road in an unauthorized place or using headphones on crosswalks but they tend to have less awareness of threats from other road users – they estimate for example pedestrian visibility at dust and drivers' reaction time less correct than respondents under 18 years of age, which is also confirmed by research [22-24]. Risk analysis related to pedestrian behavior in relation to their age, time gap, time of day, and vehicle speed [35] found that pedestrian decision on whether or not they will cross the road safely was made based on distance between them and approaching vehicle. The study found that elderly pedestrians estimated that they need the same average time to cross the street as young pedestrians even their measured walking speed was significantly lower. What was also found in this research is that pedestrians in general rely extensively on the estimation of a distance of the approaching vehicle while they are not sensitive to changes in the speed of the approaching vehicles.
The analysis of existing research shows that in the complex relationship between traffic participants, infrastructure and environmental conditions, it is necessary to look at the situation through the analysis of individual elements related to road traffic safety and through their interrelationship. As a tool for the analysis and comparison of possible solutions, the method of traffic microsimulation is used, as shown by the analysis of available research [36-38].
The aim of this paper is to define methodological steps for improving the conditions of pedestrian traffic safety based on the analysis of the actual micro-location through the analysis of the behavior of all traffic participants, and the impact of possible infrastructure solutions on the incoming vehicle speeds, critical safety parameter [3-5]. The proposed methodological steps were applied to the analysis of the selected conflict zone within the urban four-lane at-grade intersection at which pedestrian recording and automatic measurement of the number and speed of vehicles were performed in real traffic conditions. Statistically analyzed databases and statistically significant parameters of the impact on pedestrian traffic, including those related to motor traffic, were determined. Several solutions for the reconstruction of the intersection were proposed with the primary goal of reducing the vehicles’ speed and the effects of the proposed solutions were analyzed. Given that the development of the zone is expected, the impact of the increase in traffic load that can be realistically expected (from 100% to even 150% increase compared to the existing traffic) was analyzed. The increase in traffic load by 200% and 250% was done to theoretically analyze the relationship between traffic load and speed. For the analysis of the proposed solutions, the method of traffic microsimulations was used by creating models in the VISSIM software package, and for all analyzed variants, an analysis of the statistical significance of the speed difference was carried out. Application of the developed methodology on the described case study enabled conclusion about advantages and limitations.

2. Materials and Methods

The basic methodological steps proposed in this research are shown in the diagram in Figure 1.

2.1. Case Study Location Description

The observed pedestrian crosswalk which was analyzed by the proposed methodology is located on the two-lane eastern driveway of an unlit four-lane intersection consisting of the main street, Drinska Street (east-west) and the side street Krbavska Street (north-south) in the city of Osijek in Croatia (shown in Figure 2). The intersection is located in the immediate vicinity of the elementary school, which has the main entrance oriented to the east driveway, as well as a high school nearby. On the west driveway there is a Home for the Elderly and a Health Center. A secondary school, the Centre for Education for Children with Special Needs and the Cultural Centre of the Hungarian Minority, which includes a kindergarten, primary and secondary school, are oriented towards the southern driveway. The northern driveway has facilities for predominantly residential purposes. There is a large concentration of vulnerable traffic users at the observed pedestrian crosswalk.
The length of the vehicle-pedestrian conflict zone of the pedestrian crosswalk is 6.4 meters, and the width of the pedestrian crosswalk is 3.3 meters.

2.2. Database Formation

Two databases of field measurements and one database based on the results of microsimulation traffic modeling were formed (Figure 1).
The first database contains data on the behavior of pedestrians in the conflict zone. The selected dependent variables are the reaction time and crossing time of pedestrians. Both data were gathered from video recordings of pedestrians in real traffic conditions. Reaction time was established as the time from arrival of the pedestrian at the pedestrian crosswalk to the moment when checking the traffic situation was done. Crossing time was established as the time between going off the curb at a pedestrian crosswalk and climbing the curb on the opposite side of the road with both feet. The basic characteristics of pedestrian behavior that were collected in the field were used as input data for the formation of a microsimulation traffic model.
The second database contains data from the automatic traffic counter and includes traffic load, traffic structure and incoming vehicle speeds. The collected data were used to form microsimulation models of the existing conflict zone and variant solutions for reconstruction. By analyzing and comparing variant solutions according to flow indicators (delays and queue lengths) and safety parameters, the optimal variation solution for reconstruction [39] was selected, which served for a detailed analysis and comparison of incoming vehicle speeds for different traffic scenarios.
The third database contains the data on incoming vehicle speeds obtained using simulation traffic modelling. Microsimulation traffic modeling in VISSIM served as a tool for the analysis of incoming vehicle speeds for the existing conflict zone solution and the selected reconstruction solution. Different traffic loads of vehicles and pedestrians were analyzed in order to assess the impact of reconstruction and increase of traffic load on incoming vehicle speeds for different infrastructure solutions (before and after reconstruction).

2.2.1. Database – Pedestrians

The collection of field data on pedestrian behavior was made with a video camera on Thursday, June 14, 2022, and the results are presented in Table 1 and in more detail in [39]. The morning peak hour was chosen because then the traffic flow of pedestrians is the highest, due to the morning arrival of children to school.
By analyzing the collected data and the results of existing research [27,31,33], parameters that have an impact on the reaction time and crossing time of children and adult pedestrians were selected, and also include data about the infrastructural solution of the observed pedestrian crosswalk location.
The database is formed according to the input parameters shown in Table 1 and the selected dependent variables are pedestrian reaction time and pedestrian crossing time (Figure 1).
Out of a total of 36 observed variables, 21 are categorical variables. Variables 21-36 are constant for the observed location and the observed peak hour, so their influence on the dependent variables are not analyzed in this case.

2.2.2. Database of Measured Traffic Load and Vehicle Speed Data

Field measurements of vehicle traffic were carried out for the observed conflict zone of the crosswalk. Field data on the traffic load, the structure of the traffic flow and the speeds of incoming vehicles were collected using a video camera and an automatic traffic counter (SDR traffic data collector) during the morning peak hour in the same period when the data on pedestrian flows were collected.

2.2.3. Database of Incoming Speeds Obtained Using Microsimulations

Speed analysis using the VISSIM microsimulation model was made for the incoming speeds of the eastern driveway for the existing and reconstructed pedestrian crosswalk for counted traffic and traffic load increase from 50 to 250%. In the vicinity of the observed location of the pedestrian crosswalk, the construction of a new facility of the Clinical and Hospital Center Osijek is planned, so it is expected to increase the traffic flow of vehicles by 50%, 100%, and even up to 150% after construction. An increase in traffic of 200% and 250% at the observed location is not expected, but it served for the purposes of additional theoretical analysis of the impact of reconstruction and traffic load on the dynamic parameters of the traffic flow of vehicles.
For each project solution and each traffic load, an analysis of 10 different scenarios of vehicle encounters was made, and in order for the databases to be comparable for all traffic loads, the same 10 scenarios were worked with (the initial value of the random number generator is 42, and the default increment is 10). In order to have the same amount of data for analysis and comparison of incoming speeds, a comparison of average speeds was made every 60 seconds of simulation. For one hour of the simulation, 60 data of minute intervals of medium speeds per simulation were analyzed, i.e., for the analysis of 10 different scenarios of vehicle encounters, 600 data of minute speeds are obtained for each infrastructure solution and each traffic load.

3. Basic Traffic Analyses

3.1. Speed and Traffic Analyses

The results of field measurements of traffic load and traffic flow structure are shown in Table 2 and in more detail in [39].
Of the total number of vehicles, 89% are passenger cars, 9% are trucks, 1% are buses and 1% are motorcycles. Of the total number of pedestrians at the observed pedestrian crossing on the eastern approach, 82% are children and adolescents, and 18% are adult pedestrians. In the footage, 62 bicycles were spotted on the footpaths on both sides of the main street.
By measuring and comparing the incoming speeds in real conditions, the speeds of the eastern approach of the main road proved to be critical (Figure 2). Vehicles of the eastern driveway of the main traffic flow first encounter the pedestrian crosswalk (vehicle-pedestrian conflict zone), and only then the vehicle-vehicle conflict zone. For the western driveway, which does not have a pedestrian crosswalk, it is the other way around, so when vehicles encounter the pedestrian crosswalk, the vehicles pass the vehicle-vehicle conflict zone, which causes slowing down and homogenization of speeds. Within this paper, the results of the analysis of the critical flow of vehicles are presented, and the descriptive statistics for the measured speeds of the eastern approach are presented in Table 3.

3.2. Comparison of Traffic Parameters for Conflict Zone Reconstruction Solutions

To improve safety in the observed intersection zone, three solutions have been proposed:
Solution 1 - (Figure 3b.) of the reconstruction of the conflict zone envisages a pedestrian island and a narrowing of traffic lanes from 2x3.2m to 2x2.6m at the site of the observed pedestrian crosswalk.
Solution 2 - (Figure 3c.) of the reconstruction of the conflict zone envisages the central islands and the narrowing of the roadway to 2x2.8m. The central island ends before the conflict zone, so the length of the pedestrian crossing is 5.6m.
Solution 3 - (Figure 3d.) envisages the reconstruction of the driveway and combines the central islands with additional horizontal discontinuities and lane relocation, without narrowing the pavement, in more detail in [39].
To analyze the impact of the reconstruction and the increase in traffic load on the incoming vehicle speeds, four models were formed in VISSIM, the model for the existing pedestrian crosswalk (Figure 3a) and for the proposed infrastructural solutions (Figure 3 b-d).
The selected traffic flow indicators used to compare the reconstruction solution are the average QLen and maximum queue length QLenmax expressed in meters (m), the average vehicle delay VehDelay expressed in seconds per vehicle (s/veh), the number of vehicle stops at the STOP intersection (number) and the delay due to stopping VehDelaySTOP (s/veh), and the results are presented in detail in the paper [39]. Table 4 shows the results for the maximum QLenmax queue length and the average VehDelay vehicle delays. By applying traffic microsimulations, a comparison of arrival speeds (km/h) on the main route was made for all infrastructure solutions and different traffic loads (Table 5).
According to the presented results, the reconstruction solution 1 has the best traffic flow indicators in most traffic scenarios (Table 4), and together with the reconstruction solution 2, it also has the best dynamic indicators, i.e. the largest reduction in the incoming speed (Table 5). The chosen solution for the reconstruction is a solution with the central islands, because in addition to reducing speeds, it narrows the conflict zone and divides it into two independent conflict zones, which allow the pedestrian crossing in stages. This has significant benefits for pedestrians, especially for children who, up to a certain degree of cognitive development, do not have the ability to properly assess the incoming speed of a vehicle, have a longer reaction time [27] and a slower speed through the conflict zone [31,32]. A narrowed pavement means a shorter stay in the conflict zone, which is a particular advantage for pedestrians, especially for the older pedestrian population, which has motor limitations.
A detailed analysis and comparison of incoming vehicle speeds for the existing infrastructure solution of the conflict zone and the selected reconstruction solution (sedentary pedestrian island and narrowing of the conflict zone) is made in the continuation of the article.

4. Results and Discussion

In this section the results of the analyses of pedestrian and motorized traffic according to methodology presented at Figure 1 are presented and discussed.

4.1. Statistical Analysis of the Database – Pedestrians

In the Table 6 the results of the analyses of data on pedestrian behavior are presented. The results for children and young people (up to 18 years of age) and adult pedestrians (over 18 years of age) are analyzed separately, due to the possibility of comparison with earlier research results. Also, the influence of independent variables on the crossing time and on the reaction time was analyzed as earlier research [27,31,32] showed that the influence of the same independent variables on two different dependent variables is different. Data in Table 6 show the number of observed crossings (N), the mean of the dependent variables - crossing time and reaction time in seconds, standard deviation, median, minimum, and maximum values of the data sets for the observed dependent variables. An analysis of the distribution of data using the Anderson–Darling test was performed, and the null hypothesis is that the data follow a normal distribution, with a significance threshold of 0.05.
The results of the Anderson–Darling test show that for the dependent variable crossing time, the null-hypothesis cannot be discarded, i.e., that the data groups follow a normal distribution.
The results of this study (Table 6) show that children have a faster reaction time and a lower standard deviation than adult pedestrians, which differs from the results of previous studies [27] in which children of younger age have a statistically significant longer reaction time than adult. Shorter reaction time of children can be explained by the fact that there is a significant number of elderly pedestrians in the group of adult pedestrians at the analyzed location. The results also show a slightly shorter crossing time for children compared to adult pedestrians, we expected a more significant difference, but the results match previous research [31] because teenage children move more slowly through the conflict zone than younger children.
An analysis of the influence of independent variables (Table 1) on the dependent variables crossing time and reaction time was performed and the results are shown in Table 7. Given that a large number of influential (independent) variables are categorical, the non-parametric Spearman Rho test was chosen for correlation analysis. Compared to the Pearson correlation coefficient, the Spearman correlation does not require continuous-level data. The Pearson coefficient assumes a linear relationship between the two variables, whereas the Spearman coefficient works with monotonic relationships as well [40,41]. Table 4 shows the results for Spearman Rho correlation coefficient (SR) and p-value (p) for all data groups and both dependent variables.
In Table 7, statistically significant parameters are presented as bold. The age group of both children and adults affects both the crossing time and the reaction time, which is in line with previous research [27-35,42]. Adult gender, according to the results of this analysis, affects both crossing time and reaction time, and previous studies have given different results in the case of gender [31,42-44]. For adults, motor disorders were shown to affect crossing time, which was expected. The number of group members also affects the crossing time, which is in line with previous studies [31,32,42], but this study did not show an effect on reaction time, which does not coincide with existing research. [27]. Children’s running is negatively correlated with crossing time, and positively correlated with reaction time of children, which is in accordance with previous research [31,32,42]. Checking the traffic situation is positively correlated with crossing time and reaction time when children are considered, and also with adults’ reaction time. Approaching and stopping vehicles is positively correlated with the crossing time of children, which is expected, but does not have an impact on the crossing time of adults, which is expected if we take into account that adult pedestrians include a certain number of elderly pedestrians, who cannot walk faster. The total number of children at the pedestrian crosswalk extends both their crossing time and the reaction time but shortens the reaction time of adults. The total number of pedestrians increases the crossing time of children, but shortens the crossing time and reaction time of adult pedestrians. The results of the analysis of the influence of input parameters on the observed dependent variables, crossing time and reaction time, of this database are comparable to existing research, but having in mind the amount of data and only one location, the conclusions of this analysis should be taken as preliminary and related to specific site conditions.

4.2. Statistical Analysis of the Database of Incoming Vehicle Speeds Obtained by Microsimulation

A detailed analysis of the incoming speeds of the eastern driveway of the main direction in one-minute intervals of the peak hour and for ten different traffic scenarios was made for the existing solution of the conflict zone and the selected solution of reconstruction (presented at Figure 3c).
Table 8 shows the descriptive statistics of the database for the existing traffic solution of the pedestrian crosswalk, and Table 9 for the proposed reconstruction solution, for all analyzed traffic loads. For smaller traffic loads, there are one-minute intervals in which there were no vehicles (Table 8 and Table 9) and for the highest traffic load in the reconstructed conflict zone, one-minute intervals occur in which vehicles stand in a queue and the speed is 0 km/h (Table 9).
An analysis of the distribution of data using the Anderson–Darling test was performed, and the null hypothesis is that the data follow a normal distribution, with a significance threshold of 0.05. The results are shown in Table 8 and Table 9. The analysis of the distribution of data was made due to the selection of a statistical test for the evaluation of statistically significant differences between individual data sets.
A normalized cumulative speed diagram is shown for the existing infrastructure solution in Figure 4 and for the reconstructed one in Figure 5. In both diagrams, the operating speed of the V85 is marked and the reduction in operational speed for the reconstructed conflict zone can be clearly observed, which is expected. There is a noticeable reduction in operating speed for increased traffic load for both infrastructure solutions.
With the increase in traffic load, the operating speed (V85) decreases, as well as the mean speed and median speed (expected) and the standard deviation increases. That shows that it is still not a completely forced flow at all one-minute intervals, but a combination, so the vehicles periodically drive at the desired speed, the maximum speed decreases which is also expected. The minimum speed varies due to the impact of pedestrian flow and does not show sensitivity to the increase in traffic load within the analyzed data which is also expected.

4.2.1. Analysis of the Impact of Traffic Load Increase on Traffic Conditions and Incoming Vehicle Speed

According to the results of the Anderson–Darling test, the null hypothesis for all data sets can be rejected and it can be concluded that none of the data sets of the modeled velocities follows the normal distribution, so the nonparametric Bonett and Levene tests were used to evaluate the statistically significant differences between the data sets [31]. The null hypothesis is that the datasets statistically do not differ significantly, and the set significance threshold is 0.05. The results of the analysis of the increase in traffic load on the incoming vehicle speeds for the existing (Table 10) and reconstructed (Table 11) conflict zones are presented.
Table 10 presents the results of the analysis between the counted traffic and the increase in traffic, in order to determine how much traffic increase is needed for a statistically significant reduction in incoming speeds. For the existing pedestrian crosswalk and the counted traffic load, a statistically significant difference in incoming speeds is achieved for an increase in traffic load of 150% according to the results of both tests, and according to the result of the Levenna test for a traffic load increase of 100%. In the continuation of the analysis, a comparison of adjacent data groups was made to determine when the homogenization of speeds occurs due to traffic load. For traffic loads that increase by 150% or more, traffic slows down and there is no statistically significant difference in speeds in the data sets, as can be seen from Table 10.
The reconstructed pedestrian crosswalk shows greater sensitivity to the increase in traffic load, so according to the results of the Levenna test, the speed statistically differs significantly by an increase in traffic load of 50%, and according to the results of both tests, by a 100% increase. The reconstructed conflict zone also does not have a statistically significant difference in speeds when traffic increases by 150% or more, as can be seen from Table 11.
The results presented in Table 10 and Table 11 show that the impact of increasing traffic load on incoming vehicle speeds is sensitive to the infrastructural solution of the observed vehicle-pedestrian conflict zone. The reconstructed conflict zone, which slows down incoming speeds, shows greater sensitivity to increasing vehicle traffic load.

4.2.2. Analysis of the Impact of Reconstruction on Incoming Vehicle Speeds

An analysis of the impact of conflict zone reconstruction for all levels of traffic load was made and presented in Table 12, as well as the basic characteristics of the compared data groups, such as: mean speeds in km/h (Vmean), standard deviations (σ), variances (Varian) and results of statistical tests for the existing (Exist) and reconstructed (Recon) conflict zone.
In accordance with the results of the data distribution (Table 8 and Table 9), nonparametric tests Bonett and Levene were used, as in previous analyses. The null hypothesis is that the datasets do not differ statistically significantly, and the set significance threshold is 0.05.
The results of average speeds for all traffic scenarios and all traffic loads for the existing and reconstructed conflict zone are shown in Figure 6.
For all levels of traffic load, the reconstruction of the conflict zone statistically significantly slows down the incoming vehicle speeds. This is the expected confirmation of the success of the reconstruction in the observed case study, which aims to increase the safety of pedestrian flows, especially pedestrian flows of vulnerable transport users, such as children, young people and elderly pedestrians.

5. Discussion of Results and Conclusions

Planning and design of transport infrastructure should take into account the local specificities of users and the traffic infrastructure, as well as expected changes in traffic conditions. Within this paper, a methodology for the analysis of a selected critical segment of traffic infrastructure has been developed and applied with the aim to select solution that has positive impact on pedestrian safety. Based on direct measurements of pedestrian and motorized traffic indicators at the selected location and analysis of the impact of infrastructure solutions using the traffic microsimulation method, it was possible to come to an optimal solution.
The case study conducted based on the proposed methodology enabled the detailed analyses of two groups of pedestrians – young and elderly. The results of the pedestrian behavior analysis show that children have a faster reaction time than adult pedestrians, which differs from the previous results [25,27] and a higher speed through the conflict zone, which also disagrees with the results of previous studies [22,43], but is an expected result when considering that older pedestrians predominate among adult pedestrians, who, along with children, represent particularly vulnerable road users [23]. Both groups of pedestrians, children and adults, show a statistically significant influence of age on reaction time, comparable to [22,23] and on crossing speed, which agrees with the results of most studies conducted [22,23,25,43]. For adult pedestrians, the pedestrians’ gender is also a statistically significant parameter for both reaction time and the crossing speed [24], but not for children, which coincides with the results of previous studies done on signalized crosswalks in the same urban network [31,42]. Children’s gender impact happens to be locally dependent as studies from other locations have different conclusions. In some analyses, the influence of children’s gender was not a statistically significant variable, and in others it proved to be significant [44]. Both groups show statistically significant sensitivity to the total number of pedestrian-on-pedestrian crosswalk, and for the crossing speed, both groups show sensitivity to movement in the group, which agrees with previous research [31,33].
The analysis of risky behaviors at the observed location showed that running was a statistically significant variable for children, while adult subjects did not run across the road. The results of the analysis of the behavior of adult pedestrians show that they do not interact with the incoming vehicles’ flow neither checking the traffic situation, the approach of the vehicle, nor braking have an impact on the crossing speed. All of these variables have a statistically significant impact on children. These results indicate that the heterogeneous flow of pedestrians complicates the interaction with vehicle flow, which may prove to be extremely important for the safety of older pedestrians in conditions of increasing traffic load, as expected at the observed location.
The results of the analysis of the selected location show that the selected reconstruction solution gives a statistically significant slowdown for all levels of traffic load, which is the expected goal of spatial intervention.
The reduction of the speed of the traffic flow due to the increase in traffic load is a logical assumption that follows from the fundamental diagram, and with the influence of pedestrian flows and the infrastructural design of the conflict zone, the problem becomes complex. The results of microsimulations have shown that an increase in traffic load gives a statistically significant reduction in speed only for an increase in traffic of 150% (existing conflict zone) and 100% (reconstructed), which in this case is not a sufficiently reliable measure of increasing traffic safety, especially if the result is viewed in the context of previous results that show that the increase in traffic load has no impact on the behavior of older pedestrians. This part of the research should be investigated more in the future.
It is an interesting observation that the reconstructed conflict zone is more sensitive to an increase in traffic load (for a smaller increase in traffic, a statistically significant slowdown in the incoming flow of vehicles is given), it follows that there is a connection between the infrastructure solution, the increase in traffic and the decrease in the incoming speed [45,46].
The methodology used in the paper - the collection of data on motor and pedestrian traffic at the actual location, the analysis of influencing parameters on pedestrian behavior and the development of a microsimulation traffic model of the reconstruction of the zone with the aim of improving the safety of traffic flow in the conditions of increasing traffic load, enabled the analysis of the interrelations of all traffic users at the selected location. The planned reconstruction resulted in the calming of traffic both in the conditions of existing and increased traffic.
However, it was shown that children and adults (in this case, older) pedestrians are not affected by the incoming traffic flow in the same way. Children show greater sensitivity to incoming traffic than older pedestrians, which can put older pedestrians at greater risk in locations where a higher traffic load is expected, which is confirmed by the data on the number of killed elderly pedestrians in Croatia. This conclusion also raises the question of whether only its general impact on vehicle speed should be analyzed when analyzing the effectiveness of traffic calming measures, such as the one implemented in this paper. In cases where it is determined that the location is regularly used by older pedestrians, or whether the adequacy of the offered solutions from the aspect of the safety of older pedestrians should also be analyzed through certain criteria.
The limitation of this research is the application of the methodology to one conflict zone, so the results obtained should be viewed in this context. In the continuation of the research, it is necessary to analyze the behavior of pedestrians in various examples of pedestrian crosswalks without traffic lights, as well as the impact of different infrastructure solutions on reducing the speed of vehicle traffic flow. By applying expert systems and neural networks, we plan to develop prediction models that will be a useful tool in selecting optimal measures to increase traffic safety for special groups of vulnerable traffic users. Further research will also include analyzes of potential conflicts that can be valuable input in this kind of analyzes.

Author Contributions

Conceptualization, I.I.O. and A.D.T.; methodology, I.I.O; A.D.T.; Đ.Z.; M.Š.; software, I.I.O.; Đ.Z.; validation, A.A.D., and M.Š.; formal analysis, A.A.D. and M.Š.; investigation, I.I.O.; A.D.T.; Đ.Z. and M.Š.; data curation, I.I.O. and Đ.Z..; writing—original draft preparation, I.I.O. and Đ.Z. ; writing—review and editing, A.D.T. and M.Š.; visualization, I.I.O. and Đ.Z..; supervision, A.D.T. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

https://urn.nsk.hr/urn:nbn:hr:133:470433

Acknowledgments

This research is the result of the project “Transportation infrastructure in the function of the safety of vulnerable road users” (uniri-iskusni-tehnic-23-85) supported by the University of Rijeka.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Road safety statistics 2023 in more detail https://transport.ec.europa.eu/background/road-safety-statistics-2023_en.
  2. Bulletin on Road Traffic Safety in the Republic of Croatia from 2023 https://mup.gov.hr/UserDocsImages/statistika/2024/6/Bilten_o_sigurnosti_cestovnog_prometa_2023.pdf.
  3. Rosen, E.; Stigson, H.; Sander, U. Literature review of pedestrian fatality risk as a function of car impact speed. Accid. Anal. Prev., 2011., 43(1), 25-33. [CrossRef]
  4. Cuerden, R.; Richards, D.; Hill, J. Pedestrians and their survivability at different impact speeds. In Proceedings of the 20th International Technical Conference on the Enhanced Safety of Vehicles, Lyon, France, Paper No. 07-0440, (June, 2007.).
  5. Hussain, Q.; Feng, H.; Grzebieta, R.; Brijs, T.; Olivier, J. The relationship between impact speed and the probability of pedestrian fatality during a vehicle-pedestrian crash: A systematic review and meta-analysis. Accid. Anal. Prev., 2019., 129, 241-249.. [CrossRef]
  6. Thakur, S.; Biswas, S. Assessment of pedestrian-vehicle interaction on urban roads: a critical review. Arch. Transp, 2019. 51(3), 49-63. [CrossRef]
  7. Pérez-Acebo, H.; Otxoa-Muñoz, X.; Marquina-Llaguno, M.; Gonzalo-Orden, H. Evaluation of the Efficiency of Traffic Lights Turning Red in Case of Exceeding Speed Limit with Previous Panels Indicating the Speed. Transp. Res. Procedia, 2021, Vol.58, pp. 45-52. [CrossRef]
  8. Pérez-Acebo, H.; Ziolkowski, R.; Gonzalo-Orden, H. Evaluation of the Radar Speed Cameras and Panels Indicating the Vehicles’ Speed as Traffic Calming Measures (TCM) in Short Length Urban Areas Located along Rural Roads. Energies, 2021, 14, 8146. [CrossRef]
  9. Kattan, L.; Tay, R.; Acharjee, S. Managing speed at school and playground zones. Accid. Anal. Prev., 2011., 43(5), 1887-1891. [CrossRef]
  10. Gonzalo-Orden, H.; Pérez-Acebo, H.; Unamunzaga, A. L.; Rojo Arce, M. Effects of traffic calming measures in different urban areas, Transp. Res. Procedia, 2018, Vol. 33, pp. 83-90. [CrossRef]
  11. Vadeby, A.; Forsman, Å. Traffic safety effects of new speed limits in Sweden. Accid. Anal. Prev., 2018., 114, 34-39.. [CrossRef]
  12. Barón, L.; Otila da Costa, J.; Soares, F.; Faria, S.; Prudêncio Jacques, M. A.; Fraga de Freitas, E. Effect of built environment factors on pedestrian safety in Portuguese urban areas. Appl. syst. innov, 2021, 4(2), 28.. [CrossRef]
  13. Tanishita, M.; Sekiguchi, Y.; Sunaga, D. Impact analysis of road infrastructure and traffic control on severity of pedestrian–vehicle crashes at intersections and non-intersections using bias-reduced logistic regression, IATSS Res, 2023, V. 47 (2), pp. 233-239. [CrossRef]
  14. Sadrayi, A.; Boroujerdian, A. M. Assessment of Pedestrian Refuge Islands on Vehicle Speed Changes and Pedestrian Safety: Case Study in Tehran, J. traffic logist. eng. ,2016, Vol. 4, No. 1. [CrossRef]
  15. Makó, E.; Szakonyi, P. Evaluation of human behaviour at pedestrian crossings, Transp. Res. Procedia, 2016, Vol. 14, pp. 2121-2128. [CrossRef]
  16. Vignali, V.; Pazzini, M.; Ghasemi, N.; Lantieri, C.; Simone, A.; Dondi, G. The safety and conspicuity of pedestrian crossing at roundabouts: The effect of median refuge island and zebra markings, TRANSPORT RES F-TRAF, 2020, Vol. 68, pp. 94-104. [CrossRef]
  17. Karwand, Z.; Mokhtar, S.; Suzuki, K.; Oloruntobi, O.; Shah, M.Z.; Misnan, S.H. Impact of Splitter-Island on Pedestrian Safety at Roundabout Using Surrogate Safety Measures: A Comparative Study. Sustainability, 2023, 15, 5359. [CrossRef]
  18. Sołowczuk, A.; Kacprzak, D. Identification of Determinants of the Speed-Reducing Effect of Pedestrian Refuges in Villages Located on a Chosen Regional Road. Symmetry, 2019, 11, 597. [CrossRef]
  19. Kruszyna, M.; Matczuk-Pisarek, M. The Effectiveness of Selected Devices to Reduce the Speed of Vehicles on Pedestrian Crossings. Sustainability, 2021, 13, 9678. [CrossRef]
  20. Saleh, W.; Grigorova, M.; Elattar, S. Pedestrian Road Crossing at Uncontrolled Mid-Block Locations: Does the Refuge Island Increase Risk? Sustainability, 2020, 12, 4891. [CrossRef]
  21. Majer, S.; Sołowczuk, A. Traffic Calming Measures and Their Slowing Effect on the Pedestrian Refuge Approach Sections. Sustainability, 2023, 15, 15265. [CrossRef]
  22. Gates, T. J.; Noyce, D. A.; Bill, A. R.; Van Ee, N. Recommended Walking Speeds for Pedestrian Clearance Timing Based on Pedestrian Characteristics, Transp. Res. Procedia, 2006, Paper No. 06-1826, TRB 2006 Annual Meeting. https://files.topslab.wisc.edu/publications/2006/noyce_2006_1826.pdf.
  23. Dommes, A.; Merlhiot, G.; Lobjois, R.; Dang, N.-T.; Vienne, F.; Boulo, J.; Oliver, A.-H.; Crétual, A.; Cavallo, V. Young and older adult pedestrians’ behavior when crossing a street in front of conventional and self-driving cars, Accid. Anal. Prev., 2021, Vol. 159, 106256. [CrossRef]
  24. Chai, C.; Shi, X.; Wong, Y. D.; Er, M. J.; Gwee, E. T. M. Fuzzy logic based observation and evaluation of pedestrians’ behavioral patterns by age and gender. TRANSPORT RES F-TRAF, 2016, 40, 104 118. [CrossRef]
  25. Ma, J.; Shen, Z.; Wang, N.; Xiao, X.; Zhang, J. Developmental differences in children’s adaptation to vehicle distance and speed in street-crossing decision-making, J Safety Res, 2024, Vol. 88, pp. 261-274. [CrossRef]
  26. Ferenchak, N. N. Pedestrian age and gender in relation to crossing behavior at midblock crossings in India. J. Traffic Transp. Eng, 2016, 3(4), 345-351. [CrossRef]
  27. Ištoka Otković, I. A Model to Predict Children’s Reaction Time at Signalized Intersections. Safety, 2020, 6, 22. [CrossRef]
  28. Gruden, C.; Ištoka Otković, I.; Šraml, M. Pedestrian safety at roundabouts: Their crossing and glance behavior in the interaction with vehicular traffic. Accid. Anal. Prev., 2021, 159 , 106290, 11. [CrossRef]
  29. Gruden, C.; Ištoka Otković, I.; Šraml, M. Safety Analysis of Young Pedestrian Behavior at Signalized Intersections: An Eye-Tracking Study. Sustainability 2021, 13, 4419. [CrossRef]
  30. Campisi, T.; Ištoka Otković, I.; Šurdonja, S.; Deluka-Tibljaš, A. Impact Of Social and Technological Distraction on Pedestrian Crossing Behaviour: A Case Study in Enna, Sicily. Transp. Res. Procedia, 2022, Vol.60, pp. 100-107. [CrossRef]
  31. Deluka-Tibljaš, A.; Ištoka Otković, I.; Campisi, T.; Šurdonja, S. Comparative Analyses of Parameters Influencing Children Pedestrian Behavior in Conflict Zones of Urban Intersections. Safety 2021, 7, 5. [CrossRef]
  32. Deluka-Tibljaš, A.; Šurdonja, S.; Ištoka Otković, I.; Campisi, T. Child-Pedestrian Traffic Safety at Crosswalks—Literature Review. Sustainability 2022, 14, 1142. [CrossRef]
  33. Ištoka Otković, I.; Deluka-Tibljaš, A.; Šurdonja, S. Non-signalized crosswalks design influence on children pedestrian behavior // XXVIIth World Road Congress - Prague 2023, 02.10.2023-06.10.2023.
  34. Gądek-Hawlena, T.; Michalski, K. Behavior and awareness of road traffic hazards among pedestrians in Poland under 18 and over 65 years of age. J. Mod. Sci, 2024, 56(2), 416-436. [CrossRef]
  35. Liu, Y. C.; Tung, Y. C. Risk analysis of pedestrians’ road-crossing decisions: Effects of age, time gap, time of day, and vehicle speed. Saf Sci, 2014, 63, 77-82. [CrossRef]
  36. Flötteröd, G.; Chen, Y.; Nagel, K. Behavioral calibration and analysis of a large-scale travel microsimulation. Netw. Spat. Econ. 2012, 12, 481–502. [CrossRef]
  37. Shahdah, U.; Saccomanno, F. Persaud, B. Application of traffic microsimulation for evaluating safety performance of urban signalized intersections, Transp. Res. Part C Emerg. Technol, 2015, Vol. 60, pp. 96-104. [CrossRef]
  38. Mahmud, S.M. S.; Ferreira, L.; Hoque, Md. S.; Tavassoli, A. Micro-simulation modelling for traffic safety: A review and potential application to heterogeneous traffic environment, IATSS Res, 2019, 43 (1), pp. 27-36. [CrossRef]
  39. Zečević, Đ. Analysis of the effects of reconstruction of the vehicle-pedestrian conflict zone according to traffic safety criteria, master thesis, Faculty of Civil Engineering and Architecture Osijek, Josip Juraj Strossmayer University of Osijek, 2022. https://urn.nsk.hr/urn:nbn:hr:133:470433.
  40. Rovetta A. Raiders of the Lost Correlation: A Guide on Using Pearson and Spearman Coefficients to Detect Hidden Correlations in Medical Sciences. Cureus, 2020, 12(11): e11794. [CrossRef]
  41. Schober, P.; Boer, C.; Schwarte. L.A. Correlation Coefficients: Appropriate Use and Interpretation. Anesth Analg. 2018, 126(5):1763-1768. PMID: 29481436. [CrossRef]
  42. Ištoka Otković, I.; Deluka-Tibljaš, A.; Šurdonja, S.; Campisi, T. Development of Models for Children—Pedestrian Crossing Speed at Signalized Crosswalks. Sustainability 2021, 13, 777. [CrossRef]
  43. Campisi, T., Šurdonja, S., Deluka Tibljaš, A., Ištoka Otković, I. Monitoring speed variation and pedestrian crossing distraction in Enna (Sicily) during different pandemic phases. Transp. Res. Procedia, 2023, Vol.69, pp. 647-654. [CrossRef]
  44. Wang, H.; Schwebel, D.C.; Tan, D.; Shi, L.; Miao, L. Gender differences in children's pedestrian behaviors: Developmental effects. J Safety Res. 2018, Dec;67:127-133. [CrossRef]
  45. Yao, B.; Chen, C.; Cao, Q.; Jin, L.; Zhang, M.; Zhu, H.; Yu, B. Short-Term Traffic Speed Prediction for an Urban Corridor. COMPUT-AIDED CIV INF, 2017, 32: 154-169.. [CrossRef]
  46. Miao, C.; Liu, H.; Zhu, G. G.; Chen, H. Connectivity-based optimization of vehicle route and speed for improved fuel economy, Transp Res Part C Emerg Technol, 2018, Vol. 91, pp. 353-368. [CrossRef]
Preprints 120592 g001
Figure 1. Diagram of the basic methodological steps. Reconstruction solutions (S1, S2, S3); existing solution model (MES); reconstruction solution models (MS1, MS2, MS3).
Figure 1. Diagram of the basic methodological steps. Reconstruction solutions (S1, S2, S3); existing solution model (MES); reconstruction solution models (MS1, MS2, MS3).
Preprints 120592 g001
Preprints 120592 g002
Figure 2. Observed pedestrian crosswalk [39].
Figure 2. Observed pedestrian crosswalk [39].
Preprints 120592 g002
Preprints 120592 g003
Figure 3. Traffic microsimulation model for existing solution (a) and proposed reconstruction solutions (b, c, d).
Figure 3. Traffic microsimulation model for existing solution (a) and proposed reconstruction solutions (b, c, d).
Preprints 120592 g003
Preprints 120592 g004
Figure 4. Cumulative speed diagram for an existing pedestrian crosswalk (normalized).
Figure 4. Cumulative speed diagram for an existing pedestrian crosswalk (normalized).
Preprints 120592 g004
Preprints 120592 g005
Figure 5. Cumulative speed diagram for a reconstructed pedestrian crosswalk (normalized).
Figure 5. Cumulative speed diagram for a reconstructed pedestrian crosswalk (normalized).
Preprints 120592 g005
Preprints 120592 g006
Figure 6. Average speeds for all traffic loads and vehicle encounter scenarios for an existing and reconstructed pedestrian crosswalk.
Figure 6. Average speeds for all traffic loads and vehicle encounter scenarios for an existing and reconstructed pedestrian crosswalk.
Preprints 120592 g006
Table 1. Measured and observed influencing variables.
Table 1. Measured and observed influencing variables.
Variables Data type Description Type of variable
1. Age group number Pedestrian are divided into age groups according to criteria 1= < 6y; 2 = 7 -10y; 3 = 11-14y; 4 = 15-18 y;
5 =19-24 y; 6 = 25-40 y; 7= 41-65 y; 8= >65y		 categorical
2. Gender 0/1 Gender of the pedestrian
girl →0, boys →1		 categorical
3. Supervision 0/1 Adult supervision for children
NO→0, YES→1		 categorical
4. Special need 0/1 A pedestrian with difficulties (e.g., motor or visual) NO→0, YES→1 categorical
5. Group-number number Movement of pedestrian in a group; number of pedestrians in the group; if the pedestrian moves individually, the number is 1 categorical
6. Mobile talk/listening music 0/1 Using a mobile phone without disturbing visual attention
NO→0, YES→1		 categorical
7. Mobile sms/Internet 0/1 Mobile phone use with visual distraction
NO→0, YES→1		 categorical
8. Crossing outside crosswalk 0/1 Crossing the road outside the pedestrian crosswalk
NO→0, YES→1		 categorical
9. Running 0/1 Crossing the road by running over
NO→0, YES→1		 categorical
10. Checking left 0/1 Checking the traffic situation before crossing the road (left side)
NO→0, YES→1		 categorical
11. Checking right 0/1 Checking the traffic situation before crossing the road (right side)
NO→0, YES→1		 categorical
12. Vehicle arrives left 0/1 The arrival of a vehicle towards a pedestrian from the left side of the pedestrian
NO→0, YES→1		 categorical
13. Vehicle arrives right 0/1 The arrival of a vehicle towards a pedestrian from the right side of the pedestrian
NO→0, YES→1		 categorical
14. Vehicle stopping left -1/0/1 The vehicle coming from the left has stopped in front of the pedestrian crosswalk
-1 the vehicle does not approach; 0 the vehicle did not stop; 1 vehicle stopped		 categorical
15. Vehicle stopping right -1/0/1 The vehicle coming from the right has stopped in front of the pedestrian crosswalk
-1 the vehicle does not approach; 0 the vehicle did not stop; 1 vehicle stopped		 categorical
16. Vehicle breaking left -1/0/1 The vehicle coming from the left slowed down/braked in front of the pedestrian crosswalk
-1 the vehicle does not approach; 0 the vehicle did not slow down/braked; 1 vehicle slow down/braked		 categorical
17. Vehicle breaking right -1/0/1 The vehicle coming from the right slowed down/braked in front of the pedestrian crosswalk
-1 the vehicle does not approach; 0 the vehicle did not slow down/braked; 1 vehicle slow down/braked		 categorical
18. Total number of children at crosswalk number The total number of children at the crosswalk at the time of observation, who may or may not be moving in a common group continuous
19. Total number of pedestrians at crosswalk number The total number of pedestrians at the crosswalk at the time of observation, together with child pedestrians continuous
20. Number of cyclists at crosswalk number Number of cyclists crossing the road using the observed pedestrian crosswalk continuous
21. V85 km/h 85th percentile speed of incoming traffic flow continuous
23. Vmax km/h The maximum recorded vehicle speed in the observed hour in the observed conflict zone continuous
24. Vexc % The percentage of vehicles that drive faster than the speed limit continuous
25. Vehicle traffic load veh/h Vehicle traffic load in the observed hour expressed through the number of vehicles per hour continuous
26. Pedestrian traffic load ped/h Pedestrian traffic load at the observed pedestrian crosswalk in the observed hour expressed in terms of the number of pedestrians per hour continuous
27. % of freight vehicles % Within the traffic structure, the percentage of freight vehicles in the observed hour continuous
28. % of buses % Within the traffic structure, the percentage of buses in the observed hour continuous
29. % of heavy goods vehicles % Within the traffic structure, the percentage of heavy goods vehicles in the observed hour continuous
30. % of bicycles and motorbikes % Within the framework of the traffic structure, the percentage of bicycles and motorcycles in the observed hour continuous
31. The length of the pedestrian crosswalk m The length of the observed pedestrian crosswalk continuous
32. The width of the pedestrian crosswalk m The width of the observed pedestrian crosswalk continuous
33. Pedestrian island 0/1 The existence of a pedestrian island at the observed pedestrian crosswalk
NO→0, YES→1		 categorical
34. Horizontal speed retarders 0/1 The existence of horizontal discontinuities as vehicle speed decelerators, shortly before or at the observed pedestrian crosswalk
NO→0, YES→1		 categorical
35. Vertical speed retarders 0/1 The existence of vertical discontinuities as vehicle speed decelerators, shortly before or at the observed pedestrian crosswalk
NO→0, YES→1		 categorical
36. Pedestrian crosswalk at the intersection 0/1 The observed pedestrian crosswalk is located at the intersection
NO→0, YES→1		 categorical
Table 2. Traffic load on approaches and pedestrian crosswalks.
Table 2. Traffic load on approaches and pedestrian crosswalks.
Drinska - main street Krbavska - side street
east-west west-east north-south south-north
Vehicles [veh/h] 195 164 38 63
Pedestrians [ped/h] 107 - 17 72
Table 3. Descriptive statistics of the measured speeds of the eastern approach.
Table 3. Descriptive statistics of the measured speeds of the eastern approach.
N Mean StDev Min Max
Measured speed [km/h] 195 51,50 11,09 21,00 76,00
Table 4. Comparison of traffic flow indicators.
Table 4. Comparison of traffic flow indicators.
Counted traffic Increase 100% Increase 150% Increase 200%
QLenmax VehDelay QLenmax VehDelay QLenmax VehDelay QLenmax VehDelay
Existing Inters. 10,38 1,59 28,28 3,85 35,95 4,56 38,5 11,97
solution 1 13,79 1,85 36,29 5,13 41,57 7,59 42,4 15,71
solution 2 22,63 2,06 38,44 4,31 39,84 11,17 43,62 20,41
Table 5. Comparison of approaching speeds (km/h) - the main road.
Table 5. Comparison of approaching speeds (km/h) - the main road.
Counted traffic Increase 100% Increase 150% Increase 200%
Existing Intersec. East 51,07 46,75 43,93 41,07
West 50,15 45,43 43,55 40,67
solution 1 East 38,95 35,15 32,65 29,80
West 27,88 28,56 30,52 21,20
solution 2 East 33,39 35,19 33,99 29,27
West 26,95 29,03 26,94 28,85
solution 3 East 38,5 36,1 33,66 30,49
West 28,87 29,8 28,08 28,93
Table 6. Basic Statistical Indicators of Pedestrian Behavior.
Table 6. Basic Statistical Indicators of Pedestrian Behavior.
N Mean StDev Min Max Median A-D p
Children - crossing time 89 4,53 0,67 2,96 6,78 4,53 0,701 0,065
Children - reaction time 89 0,76 0,91 0,00 3,94 0,43 5,75 0,000
Adults - crossing time 19 4,77 1,40 1,85 8,41 4,66 0,42 0,294
Adults - reaction time 19 1,24 1,80 0,00 7,61 0,87 1,84 0,000
Table 7. Comparative analysis of correlations using the Spearman Rho statistical test.
Table 7. Comparative analysis of correlations using the Spearman Rho statistical test.
Variables children and teenagers adult pedestrians
crossing reaction crossing reaction
SR p SR p SR p SR p
1. Age group 0,63 0,02 -0,51 0,04 0,48 0,03 0,63 0,02
2. Gender 0,05 0,63 0,02 0,89 -0,52 0,03 0,48 0,05
3. Supervision * * * * * * * *
4. Special need * * * * 0,49 0,04 0,35 0,09
5. Group-number -0,52 0,04 -0,17 0,12 0,47 0,04 -0,39 0,10
6. Mobile talk/listening music * * * * * * * *
7. Mobile sms/Internet 0,11 0,29 0,61 0,02 * * * *
8. Cross outside crossing -0,05 0,62 0,10 0,37 -0,39 0,11 0,05 0,83
9. Running -0,61 0,02 0,45 0,05 * * * *
10. Checking left 0,50 0,01 0,82 0,00 0,06 0,81 0,89 0,00
11. Checking right 0,48 0,01 0,78 0,00 -0,06 0,81 0,86 0,00
12. Veh arrives left -0,03 0,76 -0,01 0,90 0,02 0,90 -0,33 0,15
13. Veh arrives right -0,48 0,00 -0,01 0,90 0,07 0,75 -0,19 0,42
14. Veh stopping left 0,55 0,00 0,06 0,60 0,00 1,00 0,39 0,15
15. Veh stopping right 0,47 0,00 0,06 0,59 -0,01 0,90 0,11 0,65
16. Veh breaking left 0,01 0,93 0,04 0,72 0,00 1,00 0,34 0,15
17. Veh breaking right 0,45 0,00 0,03 0,77 0,08 0,75 0,19 0,42
18. Total number of children at ped crosswalk -0,33 0,01 -0,39 0,01 0,11 0,67 -0,63 0,04
19. Total number of pedestrians at ped crosswalk -0,88 0,00 -0,12 0,25 0,68 0,04 -0,59 0,04
20. Number of cyclists at ped crosswalk 0,09 0,38 0,07 0,55 * * * *
* All values in database are identical.
Table 8. Basic Speed Statistical Indicators – Existing Solution.
Table 8. Basic Speed Statistical Indicators – Existing Solution.
N Mean StDev Min Max Median Varianc A-D p
Counted traffic 559 51,07 7,53 12,5 59,87 53,42 56,63 44,7 0,000
Increase 50% 591 49,18 7,51 14,9 60,07 51,92 56,27 24,1 0,000
Increase 100% 599 46,75 8,38 12,4 58,30 49,07 70,14 18,5 0,000
Increase 150% 600 43,93 8,72 14,48 57,79 45,83 76,07 8,9 0,000
Increase 200% 600 41,07 9,31 13,4 57,29 42,22 86,61 4,7 0,000
Increase 250% 600 38,02 9,77 10,64 57,02 39,30 95,53 3,7 0,000
Table 9. Basic Speed Statistical Indicators – Reconstruction Solution.
Table 9. Basic Speed Statistical Indicators – Reconstruction Solution.
N Mean StDev Min Max Median Varianc A-D p
Counted traffic 562 38,95 6,00 7,9 59,08 40,85 36,00 46,7 0,000
Increase 50% 592 36,93 6,85 7,7 48,12 39,81 46,88 30,0 0,000
Increase 100% 600 35,15 6,94 7,8 48,87 36,77 48,19 17,2 0,000
Increase 150% 600 32,65 7,61 8,9 45,45 34,30 57,84 9,6 0,000
Increase 200% 600 29,80 7,87 7,6 43,72 30,43 61,99 3,0 0,000
Increase 250% 598 26,17 8,05 5,6 42,87 26,52 64,79 1,2 0,000
Table 10. Impact of traffic load on incoming vehicle speeds for the existing pedestrian crosswalk.
Table 10. Impact of traffic load on incoming vehicle speeds for the existing pedestrian crosswalk.
Counted/
Incr. 50%		 Counted/
Incr.100		 Count/
Incr.150		 Incr.50/
Incr.100		 Incr.100/
Incr.150		 Incr.150/
Incr.200		 Incr.200/
Incr.250
σ1 /σ2 1,00 0,90 0,86 0,90 0,96 0,94 0,95
V1/V2 1,01 0,81 0,74 0,80 0,92 0,88 0,91
Bonett -* -* -* -* -* 2,77 1,72
p 0,97 0,13 0,02 0,06 0,40 0,10 0,19
Levene 6,2 25,6 52,8 7,52 4,52 3,05 0,61
p 0,13 0,00 0,00 0,01 0,03 0,08 0,44
*In all cases where there is a difference in the number of data, the values of the Bonett test are not obtained, but the significance of the data is calculated.
Table 11. Impact of traffic load on incoming vehicle speeds for a reconstructed pedestr. crosswalk.
Table 11. Impact of traffic load on incoming vehicle speeds for a reconstructed pedestr. crosswalk.
Counted/
Incr. 50%		 Counted/
Incr.100		 Count/
Incr.150		 Incr.50/
Incr.100		 Incr.100/
Incr.150		 Incr.150/
Incr.200		 Incr.200/
Incr.250
σ1 /σ2 0,88 0,86 0,79 0,99 0,91 0,96 0,98
V1/V2 0,77 0,75 0,62 0,97 0,83 0,93 0,96
Bonett -* -* -* -* 3,76 0,76 -
p 0,07 0,03 0,00 0,81 0,05 0,38 0,55
Levene 18,92 39,7 73,10 2,27 6,37 0,90 0,79
p 0,00 0,00 0,00 0,13 0,01 0,34 0,37
*In all cases where there is a difference in the number of data, the values of the Bonett test are not obtained, but the significance of the data is calculated.
Table 12. Analysis of the impact of reconstruction on incoming vehicle speeds.
Table 12. Analysis of the impact of reconstruction on incoming vehicle speeds.
Counted Increase 50% Increase 100% Increase 150% Increase 200% Increase 250%
Exist Recon Exist Recon Exist Recon Exist Recon Exist Recon Exist. Recon
Vmean 51,07 38,95 49,18 36,93 46,75 35,15 43,93 32,65 41,07 29,80 38,02 26,17
σ 7,33 6,00 7,51 6,85 8,38 6,94 8,72 7,61 9,31 7,87 9,77 8,05
Varian 56,63 36,00 56,27 46,88 70,14 48,19 76,07 57,84 86,61 6,99 95,53 64,79
Bonett - - - 10,74 19,24 -
p 0,014 0,054 0,001 0,001 0,000 0,000
Levene 7,95 9,35 10,42 11,27 19,10 17,93
p 0,005 0,003 0,001 0,001 0,000 0,000
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

© 2024 MDPI (Basel, Switzerland) unless otherwise stated