An Empirical Evaluation of a New Heuristic Method for Identifying Safety Improvement Sites on Rural Highways: An Oregon Case Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Title: The title is clear and accurately reflects the document's content. However, to make it more informative and engaging, you may consider adding specific details such as the location or country where the evaluation was conducted or the key findings or advantages of the proposed methodology. This will help readers understand the scope and relevance of the study at first glance. For example, "Empirical Evaluation of a New Heuristic Network Screening Methodology for Rural Highways in Oregon: Effective Identification of Sites with Potential Safety Improvements."

 

Introduction: The introduction section provides a clear overview of the importance of highway safety improvement programs, particularly in rural areas. It effectively highlights the disproportionate rate of vehicle crashes on rural roads and emphasizes the need for effective safety management programs.

One suggestion for improvement is to provide more background information on highway safety improvement programs (HSIPs). You could briefly explain what HSIPs are and how they are implemented, as this would give readers a better context for understanding the rest of the introduction.

Additionally, it would be helpful to define what "network screening" means in the context of highway safety improvement programs. This would ensure that readers clearly understand the term and its significance.

Consider including a sentence explicitly stating the study's research objectives or questions. This would help in providing clear directions for the rest of the paper.

Lastly, it would be beneficial to provide a brief overview of the proposed method for network screening. This would allow readers to understand the novelty and simplicity of the technique, as mentioned in the introduction, before diving into the details in later sections of the paper.

Overall, the introduction effectively highlights the importance of improving rural road safety and sets the stage for the rest of the paper. Addressing the above suggestions would further enhance the clarity and reader engagement of the introduction section.

 

Literature Review: The literature review section provides a good overview of the significant studies that have evaluated different network screening techniques. Here are a few suggestions for improvement:

Provide more specific details about the methodology used in each study. This will help the readers understand how the studies were conducted and the particular variables and data analyzed.

Include a summary of the limitations and drawbacks of each network screening technique studied. This will provide a more balanced view and help readers understand the potential weaknesses of each method.

Consider organizing the section by grouping studies that evaluated similar network screening techniques. This will help the readers compare and contrast different methods' effectiveness more easily.

Provide a brief explanation or definition of key terms or concepts mentioned in the section, such as Safety Performance Functions (SPFs) and Empirical Bayes adjustment. This will ensure that readers who may not be familiar with these terms can still understand the studies.

Consider including a concluding paragraph summarizing the findings and key takeaways from the reviewed studies. This will help the readers to understand the primary outcomes and contributions of the research in this field.

Overall, the literature review section provides a good foundation for understanding the existing studies on network screening techniques. By incorporating these suggested improvements, the area can become even more informative and insightful for the readers.

 

Study Motivation: The section you provided clearly and effectively explains the motivation for the research. The challenges faced by rural highways are described, and the need for a network screening method to address these challenges is highlighted. However, there are a few suggestions for improvement:

Clarify the research question: The section does not clearly state the research question or objective. Consider explicitly mentioning the research question that this study aims to answer.

Consider expanding on the challenges faced: Provide specific examples and details about rural highways' unique challenges. This will help the reader understand the context and significance of the proposed network screening method.

Justify the need for evaluation: Explain why evaluating the proposed network screening method is essential. Mention the potential consequences of not considering the technique and the benefits that an assessment can bring to the field.

Highlight the potential impact: Emphasize the benefits of an effective network screening method on safety management programs and sustainable mobility. This will help the reader understand the broader implications of the research.

Structure the paragraph more logically: Rearrange the sentences to create a more logical flow of ideas. For example, you can start by explaining the challenges faced by rural highways, then move on to the proposed network screening method, and finally discuss the need for evaluation and its potential impact.

Overall, the section provides a good overview of the motivation behind the research. By addressing the suggestions mentioned above, you can further enhance the clarity and impact of your explanation.

 

Overview of Proposed Screening Method: Overall, the proposed screening method seems to be based on established principles and guidelines, such as the Highway Safety Manual (HSM), a reliable source for conducting safety analyses. Using heuristic scoring schemes for intersections and roadway segments independently is a logical approach.

One suggestion for improvement would be to provide more details and explanations for the scoring criteria. For example, further clarification could benefit the criteria for determining the roadway and roadside characteristics score. Additionally, it would be helpful to include a brief description or definition of terms like "horizontal curve," "gradient," and "roadside fixed objects" to ensure a common understanding.

Regarding crash history, the scoring scheme appropriately assigns higher weights to fatal or severe injury crashes (N1) than other crashes (N2). However, including a threshold or guideline for determining when a crash history is considered significant enough to impact the score would be helpful.

The inclusion of traffic exposure through the use of multipliers based on traffic levels is a valuable addition to the screening method. However, providing more information on how these multipliers were derived using the HSM safety performance functions (SPFs) would be beneficial. Clarifying the process would increase transparency and allow for a better understanding and application of the method.

Lastly, while the research focuses on roadway segments, including information on the proposed scoring scheme for intersections would also be valuable. This would provide a more comprehensive understanding of the screening method and allow for a broader approach application.

Overall, the proposed screening method shows promise, but providing more details, definitions, and explanations would help enhance clarity and usability. Additionally, including information on scoring intersections would further enhance the overall impact of the method.

 

Study Area: Proofread for grammar and clarity: There are a few grammatical errors and awkward sentence constructions throughout the section. Proofreading the content and making necessary revisions can improve clarity and readability.

Overall, restructuring the section, providing more details, clarifying terminology, and proofreading for grammar and clarity can enhance the quality and effectiveness of the area.

 

Data Collection and Processing: Overall, the section provides a clear and detailed description of the data collection and processing methods used in the study. However, there are a few areas where improvements can be made:

Clarify the rationale for selecting state-owned rural two-lane roads as the target sample. Explain why these roads were chosen and how they relate to the research objectives.

Provide more information about the online geographic information system (GIS) data to identify state-owned rural two-lane roads. Describe the specific data sources and methodologies used to ensure accuracy and reliability.

Clearly define the variables used for segmentation and explain their relevance to the research objectives. Provide additional explanations or examples to help readers understand the segmentation process.

Consider adding visual aids, such as diagrams or illustrations, to help readers visualize the segmentation process described in Figure 2. This can enhance understanding and make the section more engaging.

Clarify the significance of using three different analysis periods (three years, five years, and ten years). Explain why these specific time frames were chosen and how they contribute to evaluating the proposed method.

When describing the evaluation of the proposed method using the Spearman rank correlation coefficient and root mean square error, provide more context or explanation of these statistical measures. This will help readers understand their significance in assessing the method's performance.

Consider providing more information or context on why the number and percentage of standard segments between the proposed method and observed crash data are essential in evaluating the method's performance. Explain why a higher number and percentage of common sites indicate better performance.

Discussing potential limitations or challenges in the data collection and processing methods would be helpful. This can provide transparency and help readers understand the potential impact on the reliability and validity of the results.

By addressing these suggestions, the section will be more informative and provide a clearer understanding of the data collection and processing methods used in the study.

 

Methodology: Overall, the methodology section clearly describes the approach used to evaluate the proposed network screening method. Here are a few suggestions for improvement:

Clarify the purpose of the evaluation: In the first paragraph, it would be helpful to explicitly state the objective of evaluating the proposed method. Are you aiming to determine the method's effectiveness, accuracy, or reliability? This will provide a more precise context for the subsequent analyses.

Provide more context for the Empirical Bayes method: While you mention that the proposed method was compared to the Empirical Bayes technique, it would be beneficial to provide a brief explanation of the Empirical Bayes method for readers who may not be familiar with it. Additionally, you could mention why the Empirical Bayes method is considered the gold standard and its specific merits or advantages.

Expand on the rationale for the study periods: You briefly mention that three-year and five-year analysis periods are consistent with current practice, but it would be beneficial to explain why these specific periods are chosen. Are there any particular reasons or considerations for these time frames? Additionally, for the ten-year analysis period, you mention that it alleviates the effect of randomness in crash data. Still, it would be helpful to explain or justify this assertion more.

Add more detail on the evaluation metrics: In the section discussing the evaluation of the proposed method, it would be helpful to provide more information about the specific analysis and evaluation metrics used. For example, you mention Spearman rank correlation coefficient and root mean square error, but explaining why these metrics were chosen and how they are calculated would be helpful.

Consider including a diagram or figure: To help readers visualize the evaluation process, consider a graph or figure illustrating the steps involved or the relationship between the different variables and metrics. This can enhance understanding and make the methodology section more engaging.

Proofread for sentence structure and clarity: A few sentences could be rephrased or clarified for better readability. For example, in Equation 1, consider rewording the equation explanation for better clarity.

Overall, the methodology section is well-structured and clearly explains the evaluation approach. By incorporating these suggestions, you can further enhance the clarity and comprehensibility of the methodology.

Study Results: After reviewing the results section, here are some suggested comments:

The results section provides a detailed analysis of the performance of the proposed screening method compared to the Empirical Bayes (EB) method. However, it would be helpful to briefly explain the Empirical Bayes method for readers who may not be familiar with it.

The scatterplots in Figure 1 and Figure 3 visually represent the relationship between the rankings from the proposed method and the EB method with crash history data. The figures clearly show a strong correlation between the rankings, with most observations clustering around the diagonal line. It would be beneficial to mention the statistical significance of these correlations.

The study evaluates the proposed method for different analysis periods, including three, five, and ten years. Interestingly, the consistency between the proposed method and crash history rankings increases with extended analysis periods. This finding could be discussed in more detail, including the implications for the effectiveness of the proposed method over time.

Table 2 presents the analysis periods' mean rank difference and root mean square error. It would be helpful to provide a brief interpretation of these metrics and their significance in assessing the performance of the proposed method.

The study also compares the performance of the proposed method for different road volumes. It is noted that the proposed method performs better for lower-volume roads, while the EB method is more effective for higher-volume roads. However, discussing the possible reasons behind this trend and its implications for road safety improvement strategies would be interesting.

It would be valuable to discuss the limitations of the proposed screening method and any potential constraints that may affect its generalizability and applicability in other regions or contexts.

 

Overall, the study's findings suggest that the proposed method effectively identifies sites with potential safety improvements, slightly outperforming the Empirical Bayes method. This conclusion could be reinforced by mentioning the practical implications and potential benefits of implementing the proposed strategy in real-world scenarios.

Comments for author File: Comments.pdf

Author Response

Reviewer 1 comments:

 

Title: The title is clear and accurately reflects the document's content. However, to make it more informative and engaging, you may consider adding specific details such as the location or country where the evaluation was conducted or the key findings or advantages of the proposed methodology. This will help readers understand the scope and relevance of the study at first glance. For example, "Empirical Evaluation of a New Heuristic Network Screening Methodology for Rural Highways in Oregon: Effective Identification of Sites with Potential Safety Improvements."

 

• The title has been changed to address the comment above.

 

Introduction: The introduction section provides a clear overview of the importance of highway safety improvement programs, particularly in rural areas. It effectively highlights the disproportionate rate of vehicle crashes on rural roads and emphasizes the need for effective safety management programs.

One suggestion for improvement is to provide more background information on highway safety improvement programs (HSIPs). You could briefly explain what HSIPs are and how they are implemented, as this would give readers a better context for understanding the rest of the introduction.

• A brief discussion of the HSIP and its implementation was added to the introduction section (lines 32-39).

Additionally, it would be helpful to define what "network screening" means in the context of highway safety improvement programs. This would ensure that readers clearly understand the term and its significance.

• The definition of network screening and its significance in highway safety improvement programs are briefly discussed in the introduction section of the revised manuscript (lines 40-50). This section also includes the challenges for existing network screening methods on rural highways.

Consider including a sentence explicitly stating the study's research objectives or questions. This would help in providing clear directions for the rest of the paper.

• An objective statement is added at the start of section 3 of the revised manuscript (lines 159-160).

Lastly, it would be beneficial to provide a brief overview of the proposed method for network screening. This would allow readers to understand the novelty and simplicity of the technique, as mentioned in the introduction, before diving into the details in later sections of the paper.

• Section 4 in the revised manuscript was expanded and more details were provided to explain the application of the proposed scoring scheme. A reference for the overview of the proposed method is added in the introduction (lines 67-68).  

 

 

 

Overall, the introduction effectively highlights the importance of improving rural road safety and sets the stage for the rest of the paper. Addressing the above suggestions would further enhance the clarity and reader engagement of the introduction section.

 

• The introduction section has been revised per the reviewer’s suggestion to increase clarity.

 

Literature Review: The literature review section provides a good overview of the significant studies that have evaluated different network screening techniques. Here are a few suggestions for improvement:

Provide more specific details about the methodology used in each study. This will help the readers understand how the studies were conducted and the particular variables and data analyzed. Include a summary of the limitations and drawbacks of each network screening technique studied. This will provide a more balanced view and help readers understand the potential weaknesses of each method.

• The literature review section was revised and more details added to address the reviewer’s comment above. Also, a summary section with limitations in the existing network screening methods is added at the end of section 2 of the revised manuscript (lines 147-156).   

Consider organizing the section by grouping studies that evaluated similar network screening techniques. This will help the readers compare and contrast different methods' effectiveness more easily.

• All the studies mentioned in the literature review section are in the same category. So, this section is left as it is with more details in every study along with its limitations and drawbacks.

Provide a brief explanation or definition of key terms or concepts mentioned in the section, such as Safety Performance Functions (SPFs) and Empirical Bayes adjustment. This will ensure that readers who may not be familiar with these terms can still understand the studies.

• The definition of SPF and EB methods have been added to the literature review section of the revised manuscript (lines 75-79).

Consider including a concluding paragraph summarizing the findings and key takeaways from the reviewed studies. This will help the readers to understand the primary outcomes and contributions of the research in this field.

• A paragraph summarizing the findings from the reviewed studies was added to the literature review section of the revised manuscript (lines 147-156).

Overall, the literature review section provides a good foundation for understanding the existing studies on network screening techniques. By incorporating these suggested improvements, the area can become even more informative and insightful for the readers.

 

Study Motivation: The section you provided clearly and effectively explains the motivation for the research. The challenges faced by rural highways are described, and the need for a network screening method to address these challenges is highlighted. However, there are a few suggestions for improvement:

 

Clarify the research question: The section does not clearly state the research question or objective. Consider explicitly mentioning the research question that this study aims to answer.

 

• An objective statement is added at the start of section 3 of the revised manuscript (lines 159-160).

Consider expanding on the challenges faced: Provide specific examples and details about rural highways' unique challenges. This will help the reader understand the context and significance of the proposed network screening method.

 

• Rural highway’s specific challenges were added to section 3 of the revised manuscript (lines 160-164).

 

Justify the need for evaluation: Explain why evaluating the proposed network screening method is essential. Mention the potential consequences of not considering the technique and the benefits that an assessment can bring to the field.

• A statement justifying the need for evaluation was added to section 3 of the revised manuscript (lines 167-168).  

Highlight the potential impact: Emphasize the benefits of an effective network screening method on safety management programs and sustainable mobility. This will help the reader understand the broader implications of the research.

• The benefits of an effective network screening method on safety programs and sustainable mobility are discussed in section 3 of the revised manuscript (lines 171-173).

Structure the paragraph more logically: Rearrange the sentences to create a more logical flow of ideas. For example, you can start by explaining the challenges faced by rural highways, then move on to the proposed network screening method, and finally discuss the need for evaluation and its potential impact.

• The flow of this section has been revised addressing the comment above. The current paragraph starts with the objective of the study, followed by challenges faced by rural highways and local agencies, the proposed screening method, and the need for evaluating the screening method along with its impact on sustainable mobility.

Overall, the section provides a good overview of the motivation behind the research. By addressing the suggestions mentioned above, you can further enhance the clarity and impact of your explanation.

 

 

Overview of Proposed Screening Method: Overall, the proposed screening method seems to be based on established principles and guidelines, such as the Highway Safety Manual (HSM), a reliable source for conducting safety analyses. Using heuristic scoring schemes for intersections and roadway segments independently is a logical approach.

One suggestion for improvement would be to provide more details and explanations for the scoring criteria. For example, further clarification could benefit the criteria for determining the roadway and roadside characteristics score. Additionally, it would be helpful to include a brief description or definition of terms like "horizontal curve," "gradient," and "roadside fixed objects" to ensure a common understanding.

• This section was expanded to include clarification of the scoring criteria and important definitions of roadway and roadside characteristics used in the scoring scheme (lines 184-191).

Regarding crash history, the scoring scheme appropriately assigns higher weights to fatal or severe injury crashes (N1) than other crashes (N2). However, including a threshold or guideline for determining when a crash history is considered significant enough to impact the score would be helpful.

• The logic behind the scores for observed crashes is explained in the revised manuscript (line 194-201).

The inclusion of traffic exposure through the use of multipliers based on traffic levels is a valuable addition to the screening method. However, providing more information on how these multipliers were derived using the HSM safety performance functions (SPFs) would be beneficial. Clarifying the process would increase transparency and allow for a better understanding and application of the method.

• An explanation of how the multipliers were derived is now added to section 4 of the revised manuscript (lines 204-207)

Lastly, while the research focuses on roadway segments, including information on the proposed scoring scheme for intersections would also be valuable. This would provide a more comprehensive understanding of the screening method and allow for a broader approach to application.

 

• As this study specifically investigated roadway segments, and to keep section 4 to a reasonable size, it was decided to focus on roadway segments only for its relevance to the current study. A statement to this effect was added to section 4 of the revised manuscript (lines 212-215).   

 

Overall, the proposed screening method shows promise, but providing more details, definitions, and explanations would help enhance clarity and usability. Additionally, including information on scoring intersections would further enhance the overall impact of the method.

 

Study Area: Proofread for grammar and clarity: There are a few grammatical errors and awkward sentence constructions throughout the section. Proofreading the content and making necessary revisions can improve clarity and readability.

Overall, restructuring the section, providing more details, clarifying terminology, and proofreading for grammar and clarity can enhance the quality and effectiveness of the area.

• The paragraph has been revised and proofread for more clarity.

 

Data Collection and Processing: Overall, the section provides a clear and detailed description of the data collection and processing methods used in the study. However, there are a few areas where improvements can be made:

Clarify the rationale for selecting state-owned rural two-lane roads as the target sample. Explain why these roads were chosen and how they relate to the research objectives.

• State-owned roads were selected for considerations related to data availability and accessibility. A statement to this effect was added in the revised manuscript (lines 236-237). Further, rural two-lane roadways were selected because the proposed method is for the network screening of the rural two-lane roads. This is explained in section 4, lines 175-176.

Provide more information about the online geographic information system (GIS) data to identify state-owned rural two-lane roads. Describe the specific data sources and methodologies used to ensure accuracy and reliability.

• Reference (data source) for the identification of the state-owned rural two-lane roads from the online GIS has been added in line 236.

Clearly define the variables used for segmentation and explain their relevance to the research objectives. Provide additional explanations or examples to help readers understand the segmentation process.

• The section was revised to address the comment above and an example for the clarification of the segmentation process was added (lines use of the segmentation is also added (lines 272-276).

Consider adding visual aids, such as diagrams or illustrations, to help readers visualize the segmentation process described in Figure 2. This can enhance understanding and make the section more engaging.

• Figure 2 has been revised and updated to ensure more clarity and better visualization.

Clarify the significance of using three different analysis periods (three years, five years, and ten years). Explain why these specific time frames were chosen and how they contribute to evaluating the proposed method.

• The manuscript clearly explains the rationale for using three different analysis periods (lines 306-314). Three and five years were used as these periods are common and typical in most safety studies. The ten-year period was used as it was believed to largely alleviate the effect of randomness in crash data, and to examine the consistency in study results.

When describing the evaluation of the proposed method using the Spearman rank correlation coefficient and root mean square error, provide more context or explanation of these statistical measures. This will help readers understand their significance in assessing the method's performance.

• The section was revised to address the reviewer’s comment above (lines 315-341).

Consider providing more information or context on why the number and percentage of standard segments between the proposed method and observed crash data are essential in evaluating the method's performance. Explain why a higher number and percentage of common sites indicate better performance.

• The reason behind the use of number and percentage of common sites between the proposed method and observed crash in evaluating the method’s performance is added in lines 345-347.

 

Discussing potential limitations or challenges in the data collection and processing methods would be helpful. This can provide transparency and help readers understand the potential impact on the reliability and validity of the results.

• Segments missing essential data were removed from the total sample to eliminate any potential effect on study results. A statement was added to the revised manuscript (lines 264-266).

By addressing these suggestions, the section will be more informative and provide a clearer understanding of the data collection and processing methods used in the study.

 

 

Methodology: Overall, the methodology section clearly describes the approach used to evaluate the proposed network screening method. Here are a few suggestions for improvement:

Clarify the purpose of the evaluation: In the first paragraph, it would be helpful to explicitly state the objective of evaluating the proposed method. Are you aiming to determine the method's effectiveness, accuracy, or reliability? This will provide a more precise context for the subsequent analyses.

• An objective statement was added to the start of section 7 (lines 290-291).

Provide more context for the Empirical Bayes method: While you mention that the proposed method was compared to the Empirical Bayes technique, it would be beneficial to provide a brief explanation of the Empirical Bayes method for readers who may not be familiar with it. Additionally, you could mention why the Empirical Bayes method is considered the gold standard and its specific merits or advantages.

• A brief discussion of EB method and its merits was added to the revised manuscript (lines 299-305).

Expand on the rationale for the study periods: You briefly mention that three-year and five-year analysis periods are consistent with current practice, but it would be beneficial to explain why these specific periods are chosen. Are there any particular reasons or considerations for these time frames? Additionally, for the ten-year analysis period, you mention that it alleviates the effect of randomness in crash data. Still, it would be helpful to explain or justify this assertion more.

• The manuscript clearly explains the rationale for using three different analysis periods (lines 306-314). Three and five years were used as these periods are common and typical in most safety studies. The ten-year period was used as it was believed to largely alleviate the effect of randomness in crash data, and to examine the consistency in study results.

Add more detail on the evaluation metrics: In the section discussing the evaluation of the proposed method, it would be helpful to provide more information about the specific analysis and evaluation metrics used. For example, you mention Spearman rank correlation coefficient and root mean square error, but explaining why these metrics were chosen and how they are calculated would be helpful.

• The reason behind the use of spearman rank correlation coefficient and root mean square error is discussed in the revised manuscript (lines 315-341). Also, the formula and process for the calculation of these metrics are added in that section.

Consider including a diagram or figure: To help readers visualize the evaluation process, consider a graph or figure illustrating the steps involved or the relationship between the different variables and metrics. This can enhance understanding and make the methodology section more engaging.

• The scatterplots in the study results section show the correlation between the two compared methods for network screening.

Proofread for sentence structure and clarity: A few sentences could be rephrased or clarified for better readability. For example, in Equation 1, consider rewording the equation explanation for better clarity.

• The equation explanation has been reworded for equations 1 and 2 in lines 326 and 338. A sentence is added (lines 372-373) to show what a perfect correlation looks like.

Overall, the methodology section is well-structured and clearly explains the evaluation approach. By incorporating these suggestions, you can further enhance the clarity and comprehensibility of the methodology.

Study Results: After reviewing the results section, here are some suggested comments:

 

The results section provides a detailed analysis of the performance of the proposed screening method compared to the Empirical Bayes (EB) method. However, it would be helpful to briefly explain the Empirical Bayes method for readers who may not be familiar with it.

• A brief explanation of the EB method is added in line 77-79 and 299-305.

The study evaluates the proposed method for different analysis periods, including three, five, and ten years. Interestingly, the consistency between the proposed method and crash history rankings increases with extended analysis periods. This finding could be discussed in more detail, including the implications for the effectiveness of the proposed method over time.

• With the increase in the analysis period, the randomness of the crash data is believed to be reduced. So, the increase in consistency for the proposed method with the increase in analysis period is both logical and expected. This explanation is added to the revised manuscript (lines 394-397).

Table 2 presents the analysis periods' mean rank difference and root mean square error. It would be helpful to provide a brief interpretation of these metrics and their significance in assessing the performance of the proposed method.

• The significance of the evaluation metrics is discussed in the methodology section (lines 315-341).

The study also compares the performance of the proposed method for different road volumes. It is noted that the proposed method performs better for lower-volume roads, while the EB method is more effective for higher-volume roads. However, discussing the possible reasons behind this trend and its implications for road safety improvement strategies would be interesting.

• The contribution of crash history in the EB expected number of crashes increases with traffic volume, and thus the correlation between crash history and the EB expected number of crashes. This explains why results for the low-volume roads and higher-volume roads are somewhat different. This explanation is provided in lines 569-574 of the revised manuscript.

It would be valuable to discuss the limitations of the proposed screening method and any potential constraints that may affect its generalizability and applicability in other regions or contexts.

• The limitations of the proposed network screening method and future research recommendations are added in the revised manuscript (lines 610-617).

Overall, the study's findings suggest that the proposed method effectively identifies sites with potential safety improvements, slightly outperforming the Empirical Bayes method. This conclusion could be reinforced by mentioning the practical implications and potential benefits of implementing the proposed strategy in real-world scenarios.

Reviewer 2 Report

Comments and Suggestions for Authors

Curve radius, grade and driveways greatly have a strong effect on crash density. It seems  they were non considered in the proposed metodology  For this a comparison with the ranking obtained including all the factors in table 1 would be very useful to understand the effect of the simplification adopted. Moreover the weight fir the factor considered were not explicitly reported.

Also, a direct comparison between the scoring obtained with the proposed methodology and the EB method would be useful.

Author Response

Reviewer 2 Comments:

Curve radius, grade and driveways greatly have a strong effect on crash density. It seems they were not considered in the proposed methodology. For this a comparison with the ranking obtained including all the factors in table 1 would be very useful to understand the effect of simplification adopted. Moreover, the weight for the factors considered were not explicitly reported.

• Curve radius, grade, and driveways are considered while calculating the final score for the subject roadway segment as shown in table 1. Similarly, weights for the factors considered in Table 1 were taken from the HSM CMFs for rural highways and are discussed in section 4: Overview of proposed Methodology.

Also, a direct comparison between the scoring obtained with the proposed method and the EB method would be useful.

• As the observed crash history is used as a reference in this study (typical in other studies as well), both the proposed method and the EB method were compared to the observed crash history and no direct comparison between the theoretical methods was provided.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I feel there is not novel concept and method in this work, also the evaluation for traffic safety is not well illustrated.

1. In equation (1), the "di = R(Xi) – R(Yi) is the difference between the two ranks of each observation" is unclear tome. What the meaning of the two ranks of each observation, please carefully specify this.

2. Although the proposed method and EB method have been investigated and compared using crash history rank as benchmark, it is not confident to conclude that the proposed method is effective in evaluating the safety level of the road network.

Comments on the Quality of English Language

N/A

Author Response

Reviewer 3 Comments:

1. In equation (1), the “di = R(Xi) – R(Yi) is the difference between the two ranks of each observation” is unclear to me. What the meaning of the two ranks of each observation is, please carefully specify this.

• di = R(Xi) – R(Yi) is the difference between the two ranks of segment i by two compared methods. This has been added in the explanation of the equation.

2. Although the proposed method and EB method have been investigated and compared using crash history as a benchmark, it is not confident to conclude that the proposed method is effective in evaluating the safety level of the road network.

• The observed crash history is usually used in safety studies as a reference in assessing improvement or deterioration of safety over time (e.g., the common before-after studies upon implementing safety treatments or other changes such as a change in speed limit). The EB method is included in the analysis to have a better understanding of the merits of the proposed method given that the EB method is well established in practice and has found increasing use by highway agencies and safety analysts.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Previous comments were due to the limited number of variables specified by the author for segmentation in the first version (they didn't include curve radius, grade etc) . Now, due to the word "etc",  it seems that more variable were considered. Anyway readers would benefit if author detailed all the variables adopted.

It's a pity that direct comparison between the proposed and EB methods is not provided

Author Response

1. Previous comments were due to the limited number of variables specified by the author for segmentation in the first version (they didn't include curve radius, grade etc.). Now, due to the word "etc.”, it seems that more variables were considered. Anyway, readers would benefit if author detailed all the variables adopted.

• For segmentation, “etc.” was removed and replaced by the remaining roadway element, i.e., lane type (line 276). Also, for the scoring schemes, all the roadway and roadside variables used are added with their definitions (line 186 and lines 194-196).

 

2. It's a pity that direct comparison between the proposed and EB methods is not provided.

• In safety analyses, it is usual to use observed crash data in validating theoretical models, and this study is no exception. However, given the widespread use of the EB method and the favorable performance of the method reported in the literature, the authors included the EB method in the comparisons using observed crash history as a reference. This would allow readers to compare the performance of the proposed and EB methods using the same benchmark, i.e., observed crash history. 

Reviewer 3 Report

Comments and Suggestions for Authors

1. Please specify the methodology to be evaluated in the abstract, providing more details than simply mentioning "a new proposed methodology."

2. In the literature review, please ensure the first mention of HSM (Highway Safety Manual) provides the full meaning.

3. In TABLE 1, Please provide specific meanings and units for N1, N2, and ADT to enhance clarity.

4. Note the inconsistency between the abstract and TABLE 1, where the abstract suggests the method's applicability without collision data, while TABLE 1 indicates a significant impact of crashes on the overall score. Clarify and align these points.

5. After presenting study results, ensure that all charts are correctly labeled for better understanding.

6. Throughout, please explicitly mention the use of available datasets for evaluation and validation. Clarify whether the evaluated method addresses the challenge of limited data acquisition. Clearly articulate the significance of the empirical evaluation and outline specific contributions made by the study.

Comments on the Quality of English Language

N/A

Author Response

1. Please specify the methodology to be evaluated in the abstract, providing more details than simply mentioning "a new proposed methodology."

• A statement with a brief description of the proposed methodology was added to the abstract (lines 13-16).

2. In the literature review, please ensure the first mention of HSM (Highway Safety Manual) provides the full meaning.

• The HSM was spelled out in the revised manuscript (line 74). Further, a definition of the HSM is added in the same paragraph (lines 76-78).

3. In TABLE 1, Please provide specific meanings and units for N1, N2, and ADT to enhance clarity.

• Table 1 was revised and the meaning of N1, N2, and ADT were made clear in the revised manuscript.

4. Note the inconsistency between the abstract and TABLE 1, where the abstract suggests the method's applicability without collision data, while TABLE 1 indicates a significant impact of crashes on the overall score. Clarify and align these points.

• Although not ideal, it is worth noting that this method can also be used if crash data is not available. In this scenario, the scores for the crash history will not be considered and the Relative Risk Compound Score (RRCS) would be based on roadway and roadside characteristics only. Table 1 was revised to clarify the application.     

5. After presenting study results, ensure that all charts are correctly labeled for better understanding.

• All the charts are revised to make sure labels are correct and free from typos.

6. Throughout, please explicitly mention the use of available datasets for evaluation and validation. Clarify whether the evaluated method addresses the challenge of limited data acquisition. Clearly articulate the significance of the empirical evaluation and outline specific contributions made by the study.

• The description of study data and segmentation is discussed in section 6. Also, the use of datasets for evaluation and validation is discussed in the methodology section (lines 296-305). The significance of the empirical evaluation is mentioned in the study motivation section (lines 169-175). The contributions made by the study are discussed in the abstract (lines 16-18), and in the introduction (lines 63-67).