A Collision Risk Assessment Method for Aircraft on the Apron Based on Petri Nets

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

My general comments are as follows.

1. Interesting topic.

2. Generally, well written.

3. Please reformat headings throughout for consistent sequencing.

4. Please reformat paragraphs throughout for consistent spacing and indenting.

My specific comments are as follows.

5. Introduction. Review of the relevant literature belongs in a dedicated section.

6. Introduction. What is/are the research question/questions?

7. Introduction. What is the purported contribution of the study to the relevant literature?

8. Simulation Experiments. Highly recommend application of Space-Time Cubes in a GIS for a case study airport to extend present results and past results.

9. Conclusion. Completely inadequate. What is the contribution of the study to the literature? What are the strengths and the weaknesses of the study? What are the most fruitful directions for future research based on the results of the study?

Comments on the Quality of English Language

Please reformat accordingly.

Author Response

Comments 1: Interesting topic.

Response 1: Thank you for your affirmation！

Comments 2: Generally, well written.

Response 2: Thank you for your affirmation of the overall work of this paper, much appreciated!

Comments 3: Please reformat headings throughout for consistent sequencing.

Response 3: Agree. Through a thorough review of the paper, we have corrected the issues with the order of the titles. The following are the revisions made to the incorrect titles.

Comments 4: Please reformat paragraphs throughout for consistent spacing and indenting.

Response 4: Agree. After re-checking the article, we did find that the spacing of some paragraphs did not meet the requirements. All relevant paragraph issues have been addressed according to the template requirements. For instance, the spacing between titles and paragraphs has been adjusted accordingly as per the template guidelines.

Comments 5: Introduction. Review of the relevant literature belongs in a dedicated section.

Response 5: Thank you for your advice. The introduction of the paper does have some problems, and the literature review of the paper does not well describe the current research status in the research field. The paper has incorporated additional literature in the background section, making the analysis more comprehensive. For example, references [5], [6], and [7] are all research papers that utilize Petri nets to simulate the operation of complex systems. Additionally, the literature review has been moved to a separate section within the background of the study module for comprehensive analysis.

Comments 6: Introduction. What is/are the research question/questions?

Response 6: This paper primarily conducts a modeling analysis on the quantitative assessment of aircraft collision risks during apron operations, addressing the gap in current research on areas with potential collision risks. After quantifying the aircraft collision risks, the article applies relevant algorithms to classify and evaluate the data, facilitating better visualization. Additionally, the paper proposes a re-planning of aircraft routes based on the collision risk values, offering a new perspective to reduce collision risks in path planning. After revising the paper, the statements regarding the research questions can primarily be found in the first paragraph of the introduction and the final paragraph of the background section.

Comments 7: Introduction. What is the purported contribution of the study to the relevant literature?

Response 7: Through relevant revisions, the literature review has been presented as a separate chapter titled " Background of the study." This section organizes the related literature from various perspectives on the risk of aircraft collision. The main methods discussed include physical modeling of airborne aircraft operations, physical modeling of aircraft operations on the apron, and qualitative evaluation using indicators. Finally, the chapter summarizes the current state of research and highlights the significance of this study.

Comments 8: Simulation Experiments. Highly recommend application of Space-Time Cubes in a GIS for a case study airport to extend present results and past results.

Response 8: Thank you for your professional insights on the experimental analysis. We have carefully considered your suggestions. Given the characteristics of collision risk areas on the apron, using the entire apron as a base map for visualization may not yield the most effective results. Therefore, we have introduced a GIS image of the local airport, focusing on areas where collision risks are present, to provide a more targeted visualization. Additionally, we have selected several representative time points to visualize different areas, demonstrating how collision risks and risk areas change over time, as illustrated in the figures below.

Comments 9: Conclusion. Completely inadequate. What is the contribution of the study to the literature? What are the strengths and the weaknesses of the study? What are the most fruitful directions for future research based on the results of the study?

Response 9: Thank you to the reviewer for the thorough review. We acknowledge that the conclusion of the paper needed improvement. During the revision process, we added a discussion section before the conclusion. In the discussion, we addressed both the strengths and limitations of the paper, as well as future research directions to address the gaps. Subsequently, we provided a more concise and comprehensive summary in the conclusion section. The following is the revised conclusion section:

6. Conclusion

   (1)A method for assessing aircraft collision risks on aprons based on Petri nets is proposed. By using predictive calculations of aircraft operational data, the method accurately quantifies the collision risks between aircraft and identifies high-risk collision areas on the apron. The paper presents the results showing different collision probabilities corresponding to various risk zones, addressing the current gap in research on collision risk areas. The study employs relevant neural network training methods to classify and evaluate collision risks, with visual representation through color coding. This helps apron managers to precisely monitor current collision risks during operations. Furthermore, the paper optimizes aircraft taxi routes based on collision risks in specific areas, thereby reducing risks and ensuring operational safety. This approach enhances the practical significance of collision risk management in operations, allowing the risk values to be effectively applied in real-world scenarios.    

(2) The research results indicate that areas with higher taxiway occupancy rates are more likely to become collision risk zones. On peak days, the collision risk scenarios are more complex due to the increased workload, leading to higher risk levels and making it more challenging to reduce collision risks through path planning. Therefore, during peak day operations, airports should focus on areas with higher taxiway occupancy rates and implement rational taxi path planning to ensure the safe operation of the apron.

4. Response to Comments on the Quality of English Language

The quality of English interpretation has been improved as required by reviewers

 

Reviewer 2 Report

Comments and Suggestions for Authors

The main question addressed in the study is: How can aircraft collision risk on airport aprons be assessed and mitigated, particularly during peak operational periods? The study focuses on developing a method to quantify and manage collision risks on aprons, identify high-risk areas, and explore peak traffic's impact on collision risk. It also looks into how path re-planning can reduce these risks, especially in areas with high taxiway occupancy. The paper's subject is interesting and in line with the aims and scope of the Journal. The research gap addressed in this study is the lack of precise classification and targeted risk mitigation strategies for aircraft collisions on airport aprons, an area underexplored in applied sciences, particularly using Petri nets for operational risk assessment. The paper is well-structured, although it lacks a proper literature review and discussion. I suggest a revision of the paper according to the following comments.

1.     The authors should highlight the novelty and main contributions of the study in the Abstract.

2.     The literature review is scarce and mostly outdated. There are only several references from the previous three or four years. Also, the literature doesn't cover all the main aspects of the study. A literature review on Petri nets and their possible applications, in general, and for the problem under consideration in this study, is missing. Some of the potential references that the authors could cite are:

-      Brozovic, V., Kezic, D., Bosnjak, R., & Krile, S. (2023). Implementation of International Regulations for Preventing Collisions at Sea Using Coloured Petri Nets. Journal of marine science and engineering, 11(7), 1322.

-      Tang, X., & Ji, X. (2023). Research on early runway incursion warning based on Petri net. Simulation, 99(5), 503-514.

-      Żuchowska, D., & Stelmach, A. (2023). Modeling a negotiation process between aircraft using Petri nets. Zeszyty Naukowe. Transport-Politechnika Śląska, 121.

Additionally, a literature review on some other methodologies that could be used for solving this kind of problem, is missing. Some potential papers to be considered are:

-      Korkmaz, H., Filazoglu, E., & Ates, S. S. (2023). Enhancing airport apron safety through intelligent transportation systems: Proposed FEDA model. Safety Science, 164, 106184.

-      Zhu, H., Xu, Z., Zhang, B., & Zhang, W. (2023). Collision Avoidance Study for Towbarless Aircraft Taxiing Systems on the Airport Apron Considering the Measuring Uncertainty (No. 2023-01-0782). SAE Technical Paper.

3.     Most of what is currently Introduction should be moved to another section under the title “Background of the study” (or Literature review) since it mostly deals with the background of the problem and literature review. Within this section, the authors should highlight the research gaps that this paper is trying to cover. The authors should leave in the introduction only a short overview of the problem and highlight the purpose and the aim of the study, as well as the main results, conclusions, and scientific contributions. Also, a paragraph shortly describing the rest of the paper should be added at the end of the Introduction section.

4.     The connection between the methodology and the numerical experiments is not well established. The authors should quote the equations presented within the methodology section in the application section to allow the reader to understand how the results are obtained.

5.     The paper does not have a discussion. The authors did not discuss how the results can be interpreted from the perspective of previous studies. Discussion should clearly and concisely explain the significance of the obtained results to demonstrate the actual contribution of the article to this field of research when compared with the existing and studied literature.

6.     The authors did not provide any managerial (practical) or theoretical implications of the paper. Who can use the results of this study and for what?

7.     The authors did not provide any limitations of the study.

8.     The authors didn’t provide any future research directions. There should be at least 3-5 solid future research directions interesting to most of the Journal readers.

9.     Some technical issues should be addressed:

a)     There should be at least a couple of sentences between the headings of different levels (e.g. between section 3 and sub-section 3.1).

b)    Sub-sections 3.1.1.-3.1.5 are incorectly numbered (they should be 3.2.1.-3.2.5). The conclusion should be numbered as section 4.

c)     References in the reference list are not formatted according to the Instructions for Authors (e.g. journal names are not abbreviated).

d)    Some references are not complete. They are missing important information such as volume, issue, or page numbers.

e)     All figures and tables present in the paper must be mentioned somewhere in the main text. This is not the case with most figures and tables in the paper.

f)     All references from the reference list must be quoted somewhere in the main text and vice versa. For example, reference [19] is not quoted anywhere in the main text.

g)    Abbreviations/acronyms should be defined the first time they appear in the paper. For example, the abbreviation “SVM” is not defined. Check the rest of the paper.

Comments on the Quality of English Language

English is acceptable. Only minor syntax and style errors were identified.

Author Response

Comments 1: The authors should highlight the novelty and main contributions of the study in the Abstract.

Response 1: Thank you for your thorough review. The abstract of the paper did not sufficiently highlight the novelty and primary contributions of the research, and we have made adjustments to address this. The following is the most recently updated abstract section:

Abstract: The airport apron is a high-risk area for aircraft collisions due to its heavy operational load and high aircraft density. Currently, existing quantitative models for apron collision risk provide limited consideration and classification of risk areas. In response, this paper proposes a Petri net-based method for assessing aircraft collision risk. The method predicts the probability of aircraft reaching different areas at different times based on operational data, enabling the calculation of collision risks within the Petri net framework. This approach highlights areas with potential collision risks and provides a classification evaluation. Subsequently, aircraft path re-planning is carried out to reduce collision risks. The model simplifies the complex operations of the apron system, making the calculation process clearer. The results show that, during the mid-phase of aircraft taxiing, there is a significant deviation between the actual and ideal positions of aircraft. Areas with high taxiway occupancy are more prone to collision risks. On peak days, due to relatively high flight volumes, the frequency of collision risks is 14% higher than on regular days, with an average risk increase of 23.3%, and the risks are more concentrated. Therefore, reducing collision risks through path planning becomes more challenging. It is recommended to focus attention on areas with high taxiway occupancy during peak periods and carefully plan routes to ensure apron safety.

Comments 2: The literature review is scarce and mostly outdated. There are only several references from the previous three or four years. Also, the literature doesn't cover all the main aspects of the study. A literature review on Petri nets and their possible applications, in general, and for the problem under consideration in this study, is missing. 

Response 2: Thank you for your thorough review of the paper. We have referred to the relevant literature you provided and incorporated it into the background section to enrich the paper's literature content. The following are partial screenshots from the cited literature section:

Comments 3: Most of what is currently Introduction should be moved to another section under the title “Background of the study” (or Literature review) since it mostly deals with the background of the problem and literature review. Within this section, the authors should highlight the research gaps that this paper is trying to cover. The authors should leave in the introduction only a short overview of the problem and highlight the purpose and the aim of the study, as well as the main results, conclusions, and scientific contributions. Also, a paragraph shortly describing the rest of the paper should be added at the end of the Introduction section.

Response 3: Agree, thank you for your professional advice. Based on your suggestions, we made the corresponding revisions. In the introduction, we presented the research problem, methods, results, and main contributions, and added a paragraph at the end of the introduction to outline the remaining sections.

We then transferred most of the content to the background section to describe the context of the problem.

Comments 4: The connection between the methodology and the numerical experiments is not well established. The authors should quote the equations presented within the methodology section in the application section to allow the reader to understand how the results are obtained.

Response 4: Agree, thank you for your feedback. We have incorporated references to relevant formulas before presenting the data analysis in the paper. These formulas briefly explain the data calculation process, allowing readers to clearly understand the specific formulas applied.The following is the revised content, which incorporates references to the previous formulas:

Comments 5: The paper does not have a discussion. The authors did not discuss how the results can be interpreted from the perspective of previous studies. Discussion should clearly and concisely explain the significance of the obtained results to demonstrate the actual contribution of the article to this field of research when compared with the existing and studied literature.

Response 5: Thank you for your careful feedback. You are right, the paper indeed lacked this section. We have now added a discussion section, where we explore the significance of the results and analyze the contributions of the paper. Additionally, we have included the practical implications of the study, its limitations, and directions for future research in the discussion. The following is the newly added discussion section:

 

The content regarding the contributions of the paper is as follows:

  Most previous studies on collision risk have focused on numerical evaluations, lacking discussion on areas with imminent collision risks. This paper applies Petri nets to discretize airport operations, presenting the collision risks across different areas and using color mapping for intuitive visualization. This method helps apron controllers better manage the safety of apron operations by allowing them to more effectively identify aircraft with collision risks and their corresponding potential areas. Additionally, the paper offers an approach for re-planning aircraft taxi routes with the goal of reducing collision risks. This concept can serve as a reference for future studies on path planning and provide valuable insights for apron personnel when directing aircraft taxiing.

Comments 6: The authors did not provide any managerial (practical) or theoretical implications of the paper. Who can use the results of this study and for what?

Response 6: Agree, this issue has already been addressed in the discussion section. Thank you for your advice. We have added an analysis of the theoretical and practical significance of the paper in the discussion section. The relevant sentences have already been presented in the discussion section and will not be redundantly reiterated.

Comments 7: The authors did not provide any limitations of the study.

Response 7: Sorry , we did not initially analyze the limitations of the study. Therefore, we have addressed the limitations in the discussion section, where we have analyzed certain issues and influencing factors that were not considered in the research. It mainly includes the impact of human factors on operations, the interaction between specialized vehicles and aircraft, and other objective factors influencing the need for aircraft path planning.

Comments 8: The authors didn’t provide any future research directions. There should be at least 3-5 solid future research directions interesting to most of the Journal readers.

Response 8: Thank you for your suggestion. After analyzing the three limitations of this study in the discussion section, we propose that future research could focus on overcoming these limitations: incorporating human factors into the model, including the collision risk between special vehicles and aircraft in the calculations, and integrating airport scheduling and airline demands as constraints in the path planning process.

Comments 9: Some technical issues should be addressed.

Response 9: Thank you for your careful review. The relevant issues have been corrected in the paper.

(a) There should be at least a couple of sentences between the headings of different levels (e.g. between section 3 and sub-section 3.1).

Response a: Narrative sentences have already been added between the different levels of headings:

(b) Sub-sections 3.1.1.-3.1.5 are incorectly numbered (they should be 3.2.1.-3.2.5). The conclusion should be numbered as section 4.

Response b: The issue has already been addressed and revised.

(c)  References in the reference list are not formatted according to the Instructions for Authors (e.g. journal names are not abbreviated).

Response c: The issue has already been addressed and revised.

(d) Some references are not complete. They are missing important information such as volume, issue, or page numbers.

Response d: The missing information in the references has been supplemented.

(e) All figures and tables present in the paper must be mentioned somewhere in the main text. This is not the case with most figures and tables in the paper.

Response e: Based on the review of the entire text, references to the charts and tables have been included in the paper.

Example:

(f) All references from the reference list must be quoted somewhere in the main text and vice versa. For example, reference [19] is not quoted anywhere in the main text.

Response f: Thank you for your thorough review. We have completed the cross-checking of all the references.

(g) Abbreviations/acronyms should be defined the first time they appear in the paper. For example, the abbreviation “SVM” is not defined. Check the rest of the paper.

Response g: Sorry for the inconvenience caused by our oversight. We have promptly corrected the related issues.

4. Response to Comments on the Quality of English Language

The quality of English interpretation has been improved as required by reviewers

 

Reviewer 3 Report

Comments and Suggestions for Authors

Comments:

1.          Figure 10 clearly shows that SVM accuracy is below 80%, yet the authors state SVM accuracy is 95% in page 11, line 326. Please clarify this inconsistency.

2.          Figure 11: The language must be in English. Translate all non-English words for clarity.

3.          The quantities represented in the horizontal axis in Figures 12 and 13 are not clear. Please specify what is the parameter and what is the unit.

4.          Table 4: What is the meaning of place? How was this measured?

5.          Lines 317-318: The classification risk categories are based on road traffic collisions? How can they be applicable to aircrafts. I would expect aircrafts to have high risk levels at lower risks compared to vehicles, because planes have less maneuverability compared to road vehicles. Please explain this inconsistency.

6.          It is not clear how data from Tables 1 and 2 were used to predict aircraft arrival probabilities mentioned in Figure 5. Based on the abstract, it seems that the authors are trying to predict trajectories. That would usually mean collecting actual trajectory data for planes and using the trajectory data to predict trajectories at a future point in time. Please clarify this issue.

7.          Figure 16: How was the collision risk calculated after path replanning? Do the data (after path replanning) come from real world data or do they come from simulated data?

8.          Figure 10: Given the nature of the problem, I would expect Graph Neural Networks and Quantum Neural Networks to perform well. Clarify why those results were not included.

9.          There are various methods to calculate collision risk including using surrogate safety measures such as time to collision, post encroachment time, deceleration required to stop, etc. Please justify why your chosen method is superior to other more common methods of collision risk calculation.

10.        Air traffic control operates in real-time, where conditions change rapidly due to aircraft movements, weather, and other factors. Petri nets, traditionally used for static analysis, may struggle to keep up with the real-time requirements needed for collision risk assessment and immediate decision-making. How do the authors address this challenge.

11.        Accurate modeling relies on precise data about aircraft positions, speeds, trajectories, and control inputs. However, real-world data often contains inaccuracies or uncertainties due to sensor limitations, delays in data transmission, or incomplete information, which can affect the reliability of the model. How did the authors check for data quality?

Comments on the Quality of English Language

English:

1.          Lines 246-248: should be broken into 2 sentences. There are many more instances of this throughout the manuscript. Please ensure correct grammar usage.

2.          Figure 14 and 15: should be “updated path”, not “update path”

Author Response

Comments 1: Figure 10 clearly shows that SVM accuracy is below 80%, yet the authors state SVM accuracy is 95% in page 11, line 326. Please clarify this inconsistency.

Response 1: Thank you for your careful feedback. The numerical analysis and presentation in that section of the paper had issues, which have now been corrected accordingly.

Comments 2: Figure 11: The language must be in English. Translate all non-English words for clarity.

Response 2: Agree. The issue has already been corrected.

Comments 3: The quantities represented in the horizontal axis in Figures 12 and 13 are not clear. Please specify what is the parameter and what is the unit.

Response 3: The horizontal axes of Figures 12 and 13 represent the collision risk classification levels and taxiway occupancy rates, respectively, and these parameters do not have precise units. After our discussion, we believe it may be appropriate to keep this part as it is.

Comments 4: Table 4: What is the meaning of place? How was this measured?

Response 4: The "Place" in the table 4 represents the Place in the apron Petri net, which is a component of the apron operation Petri net. The specific visual explanation can be seen in the diagram below:

The establishment process was introduced earlier in the paper, and it follows the rules for constructing a Petri net, enabling the simulation of the aircraft taxiing process.

Comments 5: Lines 317-318: The classification risk categories are based on road traffic collisions? How can they be applicable to aircrafts. I would expect aircrafts to have high risk levels at lower risks compared to vehicles, because planes have less maneuverability compared to road vehicles. Please explain this inconsistency.

Response 5: Currently, there is a lack of precise definitions for aircraft collision risk classification methods in this field. Since there are certain similarities between the operation of road vehicles and the taxiing of aircraft on aprons—both being confined to a fixed path—this study references the classification standards for collision risks used in road-related research. Furthermore, as you correctly noted, aircraft have lower maneuverability, limited rearward visibility for pilots, and generally operate at lower speeds on aprons compared to vehicles on roads. Therefore, this study also incorporates relevant aviation incident standards [30] and the analysis of aircraft collision parameters from reference [2] to establish this classification standard. Although there are similarities between aircraft operation and vehicle operation on roads, the differences are more significant, leading to different classification outcomes. Additionally, this classification method proved to be the most effective during subsequent classification training.

Comments 6: It is not clear how data from Tables 1 and 2 were used to predict aircraft arrival probabilities mentioned in Figure 5. Based on the abstract, it seems that the authors are trying to predict trajectories. That would usually mean collecting actual trajectory data for planes and using the trajectory data to predict trajectories at a future point in time. Please clarify this issue.

Response 6: Table 1 presents the operational data values set for the aircraft in the simulation experiment, which were determined based on multiple studies. These data set the parameters for the size and operating speed of the aircraft in the simulation experiments. Of course, all of these parameters are scaled proportionally.

   

Table 2 contains extracted operational data from Guangzhou Baiyun Airport during a specific period. The simulation experiment sets the flight operation schedule according to the data in Table 2, which includes flight numbers, operating times, types of takeoff and landing, and the docked gallery bridge.

  By configuring the simulation based on the data in Table 2, the aircraft takeoff and landing processes during a specific time period at the airport can be simulated, thereby modeling the entire apron operation. In this paper, simulation experiments were conducted based on flight plans and corresponding aircraft data. During the simulation process, model calculations were incorporated to determine the probability of each aircraft arriving at different locations at various times. Based on these calculations, the collision risk values between aircraft were computed, and in subsequent data analysis, the collision risks were classified, evaluated, and further processed.

Comments 7: Figure 16: How was the collision risk calculated after path replanning? Do the data (after path replanning) come from real world data or do they come from simulated data?

Response 7: This paper analyzes the causes of collision risks during aircraft taxiing and finds that most of these risks arise from conflicts along the taxiing routes. Therefore, the Dijkstra path planning algorithm is employed with the goal of reducing collision risks, allowing for the re-planning of aircraft taxiing paths. After path planning, each aircraft still has its own taxiing route, and the collision risk is recalculated based on the updated path using the model, which is essentially a repetition of the previous experimental operations. Although the taxiing data used comes from simulation experiments, all operational parameters are based on real-world data.

Comments 8: Figure 10: Given the nature of the problem, I would expect Graph Neural Networks and Quantum Neural Networks to perform well. Clarify why those results were not included.

Response 8: Since XGBoost is more efficient in handling small-scale structured data, whereas neural networks typically perform better with large-scale, unstructured data such as images or speech, XGBoost was chosen. The collision risk values calculated in this study are not extensive, and XGBoost offers faster training speed, yielding better performance during the classification phase of the research. In 2018, Huang et al. also analyzed this phenomenon in their article titled "Application of Machine Learning Methods in Stock Index Futures Prediction: A Comparative Analysis Based on BP Neural Network, SVM, and XGBoost." They concluded that XGBoost is more efficient and performs better when processing small-scale data.

Comments 9: There are various methods to calculate collision risk including using surrogate safety measures such as time to collision, post encroachment time, deceleration required to stop, etc. Please justify why your chosen method is superior to other more common methods of collision risk calculation.

Response 9: Traditional point-to-point collision physical models, while effective in quantifying collision risks between two objects, involve relatively complex computational modeling.  These models require intricate coupling when assessing collision risks among multiple aircraft, limiting their overall applicability.  In comparison, the model proposed in this study is developed from a macro-level perspective of apron operations.  It not only effectively addresses collision risks between individual entities but also facilitates the calculation of collision risks among multiple aircraft with greater ease.  Additionally, the model is capable of accurately identifying areas of potential collision risk and computing the corresponding risk values for each area—an advantage that other risk quantification models typically lack.

Comments 10: Air traffic control operates in real-time, where conditions change rapidly due to aircraft movements, weather, and other factors. Petri nets, traditionally used for static analysis, may struggle to keep up with the real-time requirements needed for collision risk assessment and immediate decision-making. How do the authors address this challenge.

Response 10: In response to the reviewer's concern regarding real-time changes at the airport, the apron operation Petri net is a tool designed to simulate aircraft movements on the apron, capable of meeting the need for real-time operational simulations. In 2023, Tang et al., in their study titled "Research on Early Runway Incursion Warning Based on Petri Net," also utilized Petri nets to simulate aircraft operations on the apron. While their model effectively simulates real-time aircraft movements, it lacks an analysis of collision risks between aircraft. Therefore, this study adopts the Petri net approach to simulate aircraft operations on the apron, addressing this gap by incorporating collision risk analysis. In the event of unforeseen factors such as adverse weather conditions, adjustments can be made by closing or modifying certain places in the Petri net to restrict token entry. Additionally, the operational parameters of aircraft can be adjusted based on relevant data for different weather conditions, allowing the simulation to reflect the operational state under various weather scenarios.

Comments 11: Accurate modeling relies on precise data about aircraft positions, speeds, trajectories, and control inputs. However, real-world data often contains inaccuracies or uncertainties due to sensor limitations, delays in data transmission, or incomplete information, which can affect the reliability of the model. How did the authors check for data quality?

Response 11: The real-time data is sourced from the airport's control system, where multi-sensor fusion is used to comprehensively calibrate the operational data of aircraft. This means that the data is not solely dependent on ground-based sensors but can also be obtained through satellite positioning of aircraft to gather relevant operational information. Additionally, the operational data at airports are often interrelated. We can evaluate the reasonableness of a specific metric by coupling decisions based on multiple data points. Furthermore, the metrics themselves typically exhibit differential relationships and should fall within a reasonable range of fluctuations. By employing this method, we can assess the validity of the metric values. Moreover, during data collection, we can verify the consistency of the data with historical records to ensure that the data meets the predefined standards at the point of collection. Furthermore, we can adopt data processing techniques used by large data management companies, such as data cleaning, which involves removing duplicate entries, correcting erroneous information, and filling in missing values. These steps help ensure the completeness and accuracy of the data.

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

None.

Comments on the Quality of English Language

Quality of writing is acceptable.

Author Response

Comments 1: None.

Response 1: Thank you for your affirmation! We also appreciate your previous professional advice,thank you!！

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have invested a substantial effort to address all issues identified in the previous review round thus significantly improving the quality of their paper. Therefore I suggest an acceptance of the paper in its present form.

Author Response

Comments 1: The authors have invested a substantial effort to address all issues identified in the previous review round thus significantly improving the quality of their paper. Therefore I suggest an acceptance of the paper in its present form.

Response 1: Thank you, we appreciate your feedback and the previous suggestions for revisions on my paper!

Reviewer 3 Report

Comments and Suggestions for Authors

The following comment needs to be addressed.

 

11.  Provide regional collision risk level color map (like in Figure 11) for the conditions of normal day and peak day,, and place them beside the corresponding figures of Figure 12 and 13. This will help in comparison.

Author Response

Comments 1: Provide regional collision risk level color map (like in Figure 11) for the conditions of normal day and peak day,, and place them beside the corresponding figures of Figure 12 and 13. This will help in comparison.

Response 1: Agree! Thank you for your professional advice. We have made the corresponding revisions based on your suggestions and have attached the relevant images in the appropriate sections, as shown below: