Research on Vehicle Congestion Group Identification for Evaluation of Traffic Flow Parameters

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper presents the results of research on the identification of different states of creating congestion groups and their relationship to road capacity, or the decrease in speed. The hypothesis was verified that when the capacity of the road is exceeded or almost reached, there is "always" a significant drop in the flow of traffic compared to when the capacity is not exceeded. 

The authors cite the most recent papers (especially published in 2022, 2023), the presented evaluation according to the identification and analysis of time gaps will be further used in the analysis of the traffic flow with the help of artificial intelligence. 

I think that this article could be accepted for publication after major revision. But there are three issues that need to be revised.

1- The "contribution"  is missing. In "Introduction" section, it only introduces what this paper does, but does not point out the significance of the research.

Contribution/innovation points are very important to an article.

2- A concern is that no formal significance analysis  of the results are done, to indicate whether the differences in performance are statistically significant or not.

For example; Friedman Aligned Rank Test, Wilcoxon Test, Quade Test, etc.

p-value can be calculated and compared with the significance level (p-value < 0.05).

3-Some pictures lack clarity.

For example Figure 2b, Figure 6a, Figure 13; The font in pictures should be "Times New Roman".

Author Response

Dear reviewer,

On behalf of the authors, thank you for the precise review of the article. Our answers, reactions are listed in the description below. I am attaching a complete version of the article with the comments of other reviewers.

This paper presents the results of research on the identification of different states of creating congestion groups and their relationship to road capacity, or the decrease in speed. The hypothesis was verified that when the capacity of the road is exceeded or almost reached, there is "always" a significant drop in the flow of traffic compared to when the capacity is not exceeded. 

The authors cite the most recent papers (especially published in 2022, 2023), the presented evaluation according to the identification and analysis of time gaps will be further used in the analysis of the traffic flow with the help of artificial intelligence. 

I think that this article could be accepted for publication after major revision. But there are three issues that need to be revised.

1- The "contribution"  is missing. In "Introduction" section, it only introduces what this paper does, but does not point out the significance of the research.

Contribution/innovation points are very important to an article.

Thank you for your comment. We would like to respond to your comment with the following text in the article:

Monitoring the flow of traffic in the territory contributes to the development of several disciplines. In addition to the basic traffic-engineering point of view, the parameterization of traffic conditions is beneficial, for example, in services (travel time), in the automotive industry (autonomous vehicles), and in other areas to make the supply and demand system more efficient. Research into the detection and categorization of congestion groups contributes to the creation of the structure of databases usable in the parameterization of traffic models or the development of AI solutions. The targeted intention is traffic management based on local conditions in the territory. The specific research was part of an extensive assignment of the Ministry of Transport of the Slovak Republic. We find applications in international projects that are focused on mobility and reducing emission impacts.

2- A concern is that no formal significance analysis  of the results are done, to indicate whether the differences in performance are statistically significant or not.

For example; Friedman Aligned Rank Test, Wilcoxon Test, Quade Test, etc.

p-value can be calculated and compared with the significance level (p-value < 0.05).

The statistical point of view is very important in the analysis. In the text, we focused on showing the formation of groups of vehicles during the measurement depending on different locations. We would like to add the following text to the statistical view in the article:

Changes in the representation of vehicles in groups were verified by Friedmen's test. The average values of the most represented group from two radars, separated by approx. 1 km on one section, in the given time intervals was used as a basic set. Based on data analysis using a t-test, we determined that the difference in the efficiency of the occurrence of a dominant group of vehicles between radars C2 and C3 is statistically significant (p < 0.05), indicating that there is a probability of less than 5% that this difference is only random. This means that the representation of the dominant group on the road section changes and is not just a result of chance (p=0.0027). This result is valid when identifying a group with a criterion of up to 2 and up to 5s. A higher p-value than 0.05 was calculated for the peak hour period for the 3s identification criterion (p=0.0833). At 5s, the dependence was still below 0.05.

3-Some pictures lack clarity.

For example Figure 2b, Figure 6a, Figure 13; The font in pictures should be "Times New Roman".

Thank you for your comments. They exported the images in higher quality and added a description somewhere (for example, Fig.12. and 13.)

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this paper, based on the statistics of road traffic flow characteristics, an algorithm has been used to optimize and improve traffic efficiency. The results of the article meet the requirements of sustainable. Modifications are recommended for acceptance.

1. The graphical quality of the article is too poor, e.g. Figure 12 is basically unreadable;

2. the conclusions of the article are too complicated, it is recommended to simplify them;

3. the references of the article are too wrong, e.g. literature 10 and 12 are duplicated.

Comments on the Quality of English Language

No comment.

Author Response

Dear reviewer,

On behalf of the authors, thank you for the precise review of the article. Our answers, reactions are listed in the description below. I am attaching a complete version of the article with the comments of other reviewers.

In this paper, based on the statistics of road traffic flow characteristics, an algorithm has been used to optimize and improve traffic efficiency. The results of the article meet the requirements of sustainable. Modifications are recommended for acceptance.

1. The graphical quality of the article is too poor, e.g. Figure 12 is basically unreadable;

Thank you for your comments. They exported the images in higher quality and added a description somewhere (for example, Fig.13)

2. the conclusions of the article are too complicated, it is recommended to simplify them;

We have modified some wording. For example, we replaced the second paragraph with the text:

The study confirmed the assumption of a dynamic process of creating vehicle groups. By detecting vehicles at the cross-section, it is possible to identify the stability of the traffic stream based on the formation of groups of vehicles. The described procedure makes it possible to detect and characterize the state of traffic based on the speed and number of dominant groups of vehicles. The highest differences were measured in places where the combination of no overtaking, short gradients, and layout to the formation of groups of vehicles. In some cases, speed differences on neighboring radars of up to 30 km/h were measured, and this was at the same intensities. Values of traffic density on sub-sections synergistically with speed. An increase in density of up to 26% was recorded on sections complicated by direction and slopes. In our analysis, we will continue to focus on precisely defining problem areas.

3. the references of the article are too wrong, e.g. literature 10 and 12 are duplicated.

Thank you for pointing out the errors in the literature. We checked the references once more and generated a new bibliography.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript Marek Drliciak, Michal Cingel, Jan Celko and Zuzana Panikova entitled ‘Research of vehicle congestion groups identification for evaluation of traffic flow parameters’ discussed the evaluation of traffic stream parameters in a road network by comparing volumes with maximum capacity, speed, and density. The findings is helpful for the development of traffic models for short time intervals based on actual spatial conditions. I recommend that this manuscript can be published in Sustainability after addressing the issues below:

1. A detailed explanation of data processing and evaluation methods is suggested to add to the manuscript.

2. The authors should carefully review the original text to ensure the removal of any errors, such as unremoved track changes or modification markers on line 208.

3. While mentioning the consideration of local conditions such as visibility and road surface, there's no detailed explanation of how these factors were comprehensively considered and their impact on results.

4. How are these parameters parameterized in microsimulation models to ensure the model's accuracy?

5. The vehicle groups are a key parameter. What are the primary factors influencing the formation of these groups, and how does the formation of groups impact the dynamics of traffic flow?

6. How about the applicability and generalizability of the model to other roads and regions?

Comments on the Quality of English Language

Double-check to avoid any mistakes.

Author Response

Dear reviewer,

On behalf of the authors, thank you for the precise review of the article. Our answers, reactions are listed in the description below. I am attaching a complete version of the article with the comments of other reviewers.

 

This manuscript Marek Drliciak, Michal Cingel, Jan Celko and Zuzana Panikova entitled ‘Research of vehicle congestion groups identification for evaluation of traffic flow parameters’ discussed the evaluation of traffic stream parameters in a road network by comparing volumes with maximum capacity, speed, and density. The findings is helpful for the development of traffic models for short time intervals based on actual spatial conditions. I recommend that this manuscript can be published in Sustainability after addressing the issues below:

1. A detailed explanation of data processing and evaluation methods is suggested to add to the manuscript.

Thank you for the constructive comment. We added the following paragraph to the collection and evaluation methodology:

Traffic data were collected on profiles that were located outside important intersections or sections with a temporary change of organization. The radars were always installed in April, May, June, September, October and November. After installing the measuring equipment, we conducted a short-term survey for the calibration setting of the vehicle type determination. As a rule, the measurement was processed within one week. The evaluation methodology was processed in successive steps. Saturated or most loaded time windows were selected from the course of the total load. Data were aggregated into 15 and 5-minute groups.

Subsequent identification of vehicle groups is given in the article.

2. The authors should carefully review the original text to ensure the removal of any errors, such as unremoved track changes or modification markers on line 208.

The removal of any errors in the text is essential. Thanks for the heads up. The current version of the article has been reviewed by several colleagues and errors have been removed.

3. While mentioning the consideration of local conditions such as visibility and road surface, there's no detailed explanation of how these factors were comprehensively considered and their impact on results.

This aspect deserves attention and we hope to incorporate your comment by adding the following text:

The aspect of horizontal and vertical geometry was taken into account especially when choosing the location of the profile survey and survey on the section. When calculating the capacity of the road, factors such as the curve angle, grade of climb, and the proportion of solid lines on the road are included in the calculation, according to Slovak regulations. The last figure characterizes the visibility. The data is freely available in the database of the Road network manager. We compared the data from the road administrator with the results of the works, where the degradation of the road was monitored from the point of view of skidding. The same approach was taken when evaluating the road surface. Surface properties are evaluated and geolocated in the administrator's database. From the database, we have selected sections that reach average life parameters. From the current evaluation, this effect appears to be significant, especially on mountain passes. The issue will be examined in more detail in the next part of the research.

4. How are these parameters parameterized in microsimulation models to ensure the model's accuracy?

Thank you for the excellent reminder. Our research is divided into several stages. In the first step, we focused on evaluation tools. Application in microscopic models is currently under development and will certainly be published. We would like to include your comment in the following text:

The long-term goal of the research is the parameterization of environmental influences into traffic flow models. In microscopic modeling, it is possible to set safety distances of vehicles, acceleration, deceleration, etc. The probability of the formation of groups on individual sections of the model is reflected in the individual types of roads. This step is possible using a script for speed control on sensors in the network.

 

 

5. The vehicle groups are a key parameter. What are the primary factors influencing the formation of these groups, and how does the formation of groups impact the dynamics of traffic flow?

We incorporated the answer into the text as follows:

The formation of vehicle groups depends on several factors. The presented results are mainly based on congested sections. Groups of vehicles will be created based on intensity, or for example acceleration noise. The study deals with different conditions on different profiles, which are parameterized according to the road manager's database. The impact on the traffic stream is significant, especially from the point of view of possible communication between vehicles. The early formation of groups with different numbers of vehicles causes instability in the traffic flow. By dynamic speed regulation, it is possible to increase the number of cars in a group, but paradoxically stabilize the traffic flow.

 

6. How about the applicability and generalizability of the model to other roads and regions?

We incorporated the answer into the text as follows:

The fragmentation of the territory and different horizontal and height design parameters affect the ride. The type of roads and regions are included especially for project parameters (narrower category, lower speed, or smaller values of horizontal and vertical geometry). By creating a complex database, it is possible to determine the clues for calculating the probability of the formation of vehicle groups. However, the results must be verified by in situ measurement.

 

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The main content of this article is about the study of particulate matter in the road tunnel environment in Slovakia. This paper describes the contribution of non-exhaust sources to particulate matter and proposes an assessment tool to identify the nature of traffic flow during peak hours. The study used the results of traffic surveys conducted on the road infrastructure of the Slovak Republic over the past five years as the data source. This paper also discusses the dynamics of traffic flow and the formation of vehicle groups, and analyzes the frequency variation of vehicle groups at different time intervals. In addition, the paper also discusses the specific characteristics of traffic flow and the changes under different weather conditions.

The advantages of this paper include: 1. Traffic congestion is an important problem in urban traffic management, and this study is of great significance for evaluating traffic flow and congestion. 2. Data from a traffic survey conducted on the road infrastructure of the Slovak Republic are used in this paper to increase the credibility and practicality of the study. 3. The paper makes a detailed analysis of the formation and influencing factors of traffic congestion groups, which provides a valuable reference for traffic planning and management.

The article could be improved in the following areas: 1. When describing the research method, the article could more clearly explain the specific steps of data collection and analysis, so that readers can better understand the research process. 2. When presenting the results, the paper can provide more specific data and charts to support the reliability and validity of the research conclusions.

Also, the literature review may be improved by citing more relevant papers. Just list several as follows.

AI-Empowered Speed Extraction via Port-Like Videos for Vehicular Trajectory Analysis

A hybrid visualization model for knowledge mapping: scientometrics, SAOM, and SAO

Overall, this paper makes valuable research on the evaluation of traffic flow parameters and the identification of traffic congestion groups, but there is still room for improvement in the method description, result presentation and discussion.

Comments on the Quality of English Language

Good

Author Response

Dear reviewer,

On behalf of the authors, thank you for the precise review of the article. Our answers, reactions are listed in the description below. I am attaching a complete version of the article with the comments of other reviewers.

The main content of this article is about the study of particulate matter in the road tunnel environment in Slovakia. This paper describes the contribution of non-exhaust sources to particulate matter and proposes an assessment tool to identify the nature of traffic flow during peak hours. The study used the results of traffic surveys conducted on the road infrastructure of the Slovak Republic over the past five years as the data source. This paper also discusses the dynamics of traffic flow and the formation of vehicle groups, and analyzes the frequency variation of vehicle groups at different time intervals. In addition, the paper also discusses the specific characteristics of traffic flow and the changes under different weather conditions.

The advantages of this paper include: 1. Traffic congestion is an important problem in urban traffic management, and this study is of great significance for evaluating traffic flow and congestion. 2. Data from a traffic survey conducted on the road infrastructure of the Slovak Republic are used in this paper to increase the credibility and practicality of the study. 3. The paper makes a detailed analysis of the formation and influencing factors of traffic congestion groups, which provides a valuable reference for traffic planning and management.

The article could be improved in the following areas: 1. When describing the research method, the article could more clearly explain the specific steps of data collection and analysis, so that readers can better understand the research process. 2. When presenting the results, the paper can provide more specific data and charts to support the reliability and validity of the research conclusions.

Thank you for the constructive comment. We added the following paragraph to the collection and evaluation methodology:

Traffic data were collected on profiles that were located outside important intersections or sections with a temporary change of organization. The radars were always installed in April, May, June, September, October and November. After installing the measuring equipment, we conducted a short-term survey for the calibration setting of the vehicle type determination. As a rule, the measurement was processed within one week. The evaluation methodology was processed in successive steps. Saturated or most loaded time windows were selected from the course of the total load. Data were aggregated into 15 and 5-minute groups.

Subsequent identification of vehicle groups is given in the article.

The statistical point of view is very important in the analysis. In the text, we focused on showing the formation of groups of vehicles during the measurement depending on different locations. We would like to add the following text to the statistical view in the article:

Changes in the representation of vehicles in groups were verified by Friedmen's test. The average values of the most represented group from two radars, separated by approx. 1 km on one section, in the given time intervals was used as a basic set. Based on data analysis using a t-test, we determined that the difference in the efficiency of the occurrence of a dominant group of vehicles between radars C2 and C3 is statistically significant (p < 0.05), indicating that there is a probability of less than 5% that this difference is only random. This means that the representation of the dominant group on the road section changes and is not just a result of chance (p=0.0027). This result is valid when identifying a group with a criterion of up to 2 and up to 5s. A higher p-value than 0.05 was calculated for the peak hour period for the 3s identification criterion (p=0.0833). At 5s, the dependence was still below 0.05.

Also, the literature review may be improved by citing more relevant papers. Just list several as follows.

AI-Empowered Speed Extraction via Port-Like Videos for Vehicular Trajectory Analysis

A hybrid visualization model for knowledge mapping: scientometrics, SAOM, and SAO

We have studied your proposed articles and incorporated them into the text. It is not a specific connection with our research. However, the focus points to other fields of application.

Overall, this paper makes valuable research on the evaluation of traffic flow parameters and the identification of traffic congestion groups, but there is still room for improvement in the method description, result presentation and discussion.

 

Author Response File: Author Response.pdf

Reviewer 5 Report

Comments and Suggestions for Authors

The manuscript proposes an approach to monitor and study traffic congestion. The introduction is concise and informative about the basics of traffic modeling. The manuscript is well-written and worthy of publication. I have a few minor comments that you may consider for revisions.

1. You may consider defining concepts such as vehicle density, intensity, and speed (lines 51-54) at their first mention rather than a bit later (lines 92-94). 

2. Lines 135-140: how did you select these sections? Were these sections representative enough to capture the key traffic behaviors?

3. Lines 144-46: what is a profile? Please define the term. Also, how long did you collect the data for, 1 day or multiple days? Any impact on the choice of days on the traffic behavior?

Comments on the Quality of English Language

NA

Author Response

Dear reviewer,

On behalf of the authors, thank you for the precise review of the article. Our answers, reactions are listed in the description below. I am attaching a complete version of the article with the comments of other reviewers.

The manuscript proposes an approach to monitor and study traffic congestion. The introduction is concise and informative about the basics of traffic modeling. The manuscript is well-written and worthy of publication. I have a few minor comments that you may consider for revisions.

1. You may consider defining concepts such as vehicle density, intensity, and speed (lines 51-54) at their first mention rather than a bit later (lines 92-94). 

Thank you for the reminder. We added the following text to the paper.

The traffic flow characteristics such as the traffic volumes and the density are estimated by averaging the vectors together in space and time . The density is an indication of the freedom of movement of the users. Density  is defined as the number of vehicles per unit length of the roadway. The density of vehicles is an essential traffic condition metric used in many traffic information systems. Traffic volume is defined as the rate at which vehicles travel through a particular point or highway segment. Speed is measured in units of distance per unit of time, typically per hour.

2. Lines 135-140: how did you select these sections? Were these sections representative enough to capture the key traffic behaviors?

Thank you for the reminder. We added the following text to the paper.

Traffic data were collected on profiles that were located outside important intersections or sections with a temporary change of organization. The radars were always installed in April, May, June, September, October and November. After installing the measuring equipment, we conducted a short-term survey for the calibration setting of the vehicle type determination. As a rule, the measurement was processed within one week. The evaluation methodology was processed in successive steps. Saturated or most loaded time windows were selected from the course of the total load. Data were aggregated into 15 and 5-minute groups.

3. Lines 144-46: what is a profile? Please define the term. Also, how long did you collect the data for, 1 day or multiple days? Any impact on the choice of days on the traffic behavior?

Thank you for the reminder. We added the following text to the paper.

The profile or cross-section indicates the place where the measuring device was installed. In the study, we focused on the characteristics of the traffic flow during working days. During the working week, the share of freight traffic is increased. During the weekend, the intensity of holiday traffic is particularly important.

The measurement description is processed in the reaction above.

 

 

 

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

No other concerns.