The Chinese Aviation Network: An Empirical Temporal Analysis on Its Structural Properties and Robustness

Response to Reviewer  Comments

1. Summary

Dear Editor,

Thank you very much for taking the time to review the manuscript. We are uploading 

(a) our point-by-point response to the comments (below) (response to reviewers),

(b) a revised article which has highlighted the modifications in blue to make it easier to identify the changes we made.

 Once again, we thank you for your careful review and helpful suggestions. Your feedback has positively influenced our research. We have made every effort to address the issues you raised, and we believe that the revised manuscript has significantly improved in terms of quality and content.

Best regards, 

Ruoshi Yang, Wei Sun, Meilong Le and Hongyan Zhang

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1:

The solution proposed of a sequence of static graph driven by time events is well justified, however, excessive redundant information will be stored, which can constraint the applicability (and the interest) of the proposed solution for future research on temporal and structural analysis.

Response 1:

We appreciate your attention to the potential redundant information issues in the solution and the applicability and interest in future research on time and structure analysis. Although using time event driven static graph sequences may lead to redundant information storage, we believe that this method has multiple reasonable reasons and advantages that exceed this limit. By using a series of static graphs to capture the temporal characteristics of data, our solution can comprehensively analyze the structure and temporal characteristics of the data. In addition, in the final chapter of this article, we added a discussion section and plan to explore more advanced temporal network models to find a balance between computational cost and applicability.

Comments 2:

The introduction should be extended with a better overview of the use of Graph Theory in the temporal analysis of airspace. Thus, it is recommented to check papers https://doi.org/10.3390/aerospace8120364 and https://doi.org/10.3390/aerospace9100613.

Reference to above papers will better support sentences such as "...Integration of the time dimension into the network model results in more complex interrelationships and presents novel challenges to static network theory... "

Response 2:  

We argee with this comment. In order to address this recommendation, we have thoroughly examined relevant papers in the field to enhance our understanding and provide a more comprehensive overview. Therefore, we have added an overview of Graph Theory in the temporary analysis of airspace in the Introduction section (revised version Page2 Line 65- Line 74). 

Comments 3:

Besides "degree, betweenness, and clustering coefficient." it is suggested to consider also the "Nearest Neighbor Degree" to improve the Attack analysis since more robustness could be justified.

Response 3:  

We appreciate your insight and agree that considering "Nearest Neighbor Degree" could enhance the robustness of our study. As a result, we have incorporated additional concepts related to the nearest neighbor (revised version Page 6 Formula 8) and conducted corresponding experiments. These include analysis of nearest neighbor degree centrality (revised version Page 15 Line 403-Line 410 and Page 16 Figure 9) and the study of robustness based on nearest neighbor attacks (revised version Page 15 Line 428-Line 437, Page 18 Figure 12, and Page 19 Figure 13).

The experimental results show that there is a correlation between the centrality of nearest neighbors and regional hub airports. The nearest airport with a greater proximity is usually the hub airport in the region. Further analysis reveals that most of these airports are provincial capital airports and airports in economically developed regions. In the robustness analysis experiment, we found that the failure of regional hub airports will seriously affect the operational efficiency of the aviation network, but will not lead to large-scale collapse of the aviation network.

Comments 4:

Authors should validate if the data provided of 4,961,074 data entries in page 7 corresponds to expected annual flights in the aerospace area under study.

Response 4:  

We appreciate your concern regarding the accuracy and reliability of the data presented in our work. Our data is provided by VariFlight Company, an aviation data and technology company headquartered in China. As a leading aviation data service provider in China, VariFlight obtains aviation data through various channels, including airlines, aviation control departments, aviation business systems, and other reliable data providers. They establish cooperative relationships with all parties to ensure the reliability and accuracy of data sources.

After you proposed this suggestion, we also conducted cross validation on the data. We obtained all flight takeoff and landing data within China (excluding Hong Kong, Macao, and Taiwan) in 2018 from the OAG database and found that the data volume was not as large as the data provided by VariFlight. Due to the fact that VariFlight data is a domestic aviation data service provider in China and has corresponding cooperation with the Civil Aviation Administration of China, we are more inclined to believe in the data provided by VariFlight.

Therefore, in revised version Page 7 Line 238, we incorporated the authoritative data provided by VariFlight Company to ensure the accuracy of our data.

Comments 5:

Page 7 "...it became evident that such missing values were typically associated with distinct situations like go-arounds.." A go-around can be issued by the ATC due to RW pronlems or by the landing pilot due to abnormal condiitions in the cockpit. Please justify why a go-around can generate missing values?

Response 5:  

We appreciate the opportunity to clarify the association between "go-around" and missing values. However, before addressing your question, we would like to clarify any potential misunderstanding caused by our original statement about missing values.

The missing values mentioned in the original text refer to the missing data in actual takeoff and landing time, rather than data missing specifically related to the go-around process itself. After comparing the data, we observed that certain missing values in the actual takeoff and landing time were often associated with go-arounds, diversions, or other exceptional circumstances, which could result in data discontinuity.

Regarding whether the go-around process itself can generate missing values, we fully agree with your point. A go-around can be initiated by air traffic control (ATC) due to runway issues or by the landing pilot based on abnormal conditions in the cockpit. However, we do not believe that the go-around process itself would directly result in missing data.

Therefore, we have made modifications to the original sentence, and the modified sentence is “Upon comparison with actual operational data, it became evident that the presence of missing values was often observed in conjunction with distinct situations, such as diversion or midair returns.” in revised version Page 8 Line 260-Line 263. Through such modifications, we emphasize the observation of missing values in the dataset and clearly indicate that these missing values may be related to different situations such as diversion.

Minors Comments 1:

Page 2 "...Selecting a window that has small results in numerous invalid.." improve.

Response Minors Comments 1:  

We appreciate your input and recognize that the sentence can be improved for clarity and readability. We have made the following modifications:

A small time window setting results in numerous invalid empty network structures, whereas using the entire observation window as the time window renders the temporal network static, leading to the loss of temporal information. （Page 2 Line 60-62）

Minors Comments 2:

Page 6: "...The data consist of ten valid attributes " but in table 1 there are 11 attributes (it seems "Departure delay in seconds" is an attribute) . Please check.

Response Minors Comments 2:  

Thank you for pointing out the inconsistency in the number of attributes mentioned in the text and the actual count in Table 1. We apologize for the oversight. We have made the following modifications:

• Revised version Line 241: ten→ eleven

– This bug has been fixed. 

Minors Comments 3:

Page 6: It is meaningless to replicate table 1 in table 2 removing file "delay informatioN". Improve

Response Minors Comments 3:  

Thank you for pointing out the issue of replicating Table 1 in Table 2 without meaningful changes. Therefore, we have deleted Table 2 and made the following modifications to the original text:

• Original version Line 224 - Line 228: Delete this paragraph

- Delete “The redundancy of some redundant variables, such as aircraft and airport affiliation information, which were not related to our research objectives, could not be used in our research and could potentially affect data processing speed. To address this issue, we removed the unnecessary variables and added delay attributes. The remaining relevant attributes are presented in Table 2.”

• Revised version Line 241 - Line 244: Add explanation for redundant data operations

– The redundancy of some redundant variables which were not related to our research objectives, could not be used in our research and could potentially affect data processing speed. To address this issue, we removed the delay information. 

Minors Comments 4:

Page 7 : "...The redundancy of some redundant variables.:" improve

Response Minors Comments 4:  

• Original version Line 224 - Line 228: Delete this paragraph

- Delete “The redundancy of some redundant variables, such as aircraft and airport affiliation information, which were not related to our research objectives, could not be used in our research and could potentially affect data processing speed. To address this issue, we removed the unnecessary variables and added delay attributes. The remaining relevant attributes are presented in Table 2.”

• Revised version Line 241 - Line 244: Add explanation for redundant data operations

– The redundancy of some redundant variables which were not related to our research objectives, could not be used in our research and could potentially affect data processing speed. To address this issue, we removed the delay information.

Minors Comments 5:

Page 8: Robustness measurements is one of the main outcomes of the research. I suggest to improve the title since "Temporal and Structural properties" is too wide and do not reflecte the content of the paper. Note for example that paper https://doi.org/10.3390/aerospace10100847 also provides a spatio-temporal analysis of the Airspace considering the structures, but with a drastic different scope.

Response Minors Comments 5:  

Thank you for your valuable comment regarding the title of our paper. Considering your suggestion, we propose revising the title : "Chinese Aviation Network: an Empirical Temporal Analysis on Its Structural Properties and Robustness". We believe this revised title better captures the essence of our research and emphasizes the specific aspect of robustness measurements.

4. Response to Comments on the Quality of English Language

Response :   

Page 2 Line 26 : Chi Li-Ping -> Li-Ping

Page 2 Line 71 : In order to better represent -> To better represent

Page 3 Line 94 : which interact -> interacting

Page 4 Line 145: to be used to -> can be used to

Page 5 Line 157: Noted that -> Note that

Page 5 Line 187: the function to remove airport nodes and routes randomly -> the function that removes airport nodes and routes randomly

Page 7 Line 242: Line which were not related to our research->, that were not related to our research

Page 7 Line 250: Delte “Generally speaking, the validity of the data was closely related to the research question. ”

Page 12 Line 327: Moreover -> Besides

Page 14 Line 379-384：This tendency originated from a higher rate of increase in flight routes relative to the number of airports between 5:00 and 7:00 and a lower rate of increase in flight routes relative to the number of airports between 7:00 and 9:00. ->

This phenomenon can be attributed to a temporal asymmetry in the rate of growth between routes and airports within the network during distinct time intervals. Specifically, during the period from 5:00 to 7:00, the rate at which routes become part of the network surpasses the rate at which airports are incorporated. In contrast, between 7:00 and 9:00, the rate at which routes integrate into the network lags behind the rate at which airports join.

Page 14 Line 388: reveal -> revealed

Response to Reviewer  Comments

1. Summary

Dear Editor,

Thank you very much for taking the time to review the manuscript. We are uploading 

(a) our point-by-point response to the comments (below) (response to reviewers),

(b) a revised article which has highlighted the modifications in blue to make it easier to identify the changes we made.

 Once again, we thank you for your careful review and helpful suggestions. Your feedback has positively influenced our research. We have made every effort to address the issues you raised, and we believe that the revised manuscript has significantly improved in terms of quality and content.

Best regards, 

Ruoshi Yang, Wei Sun, Meilong Le and Hongyan Zhang

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1:

It is recommended to present more main results in the abstract, maybe even in a numerical expression (for example, you write "an obvious morning departure peak and an evening arrival peak existed at the airline hub" - maybe there is an opportunity to specify the time when it is the most intense, etc. ).

Response 1:

We appreciate your valuable feedback regarding the presentation of main results in the abstract of our paper. We agree that providing more specific and numerical expressions in the abstract will enhance the clarity and comprehensiveness of our research findings.

Thank you for pointing this out.

(1) through the spatial-temporal aviation network, we found that there are morning departure peak (7:00-8:00) and an evening arrival peak at the airline hub (20:00 to 21:00). (Revised version Page1, Line 9 - Line 10).

(3)the robustness of the CATN was found to be unaffected by time windows, but it displayed poor resilience against deliberate attacks, particularly when subjected to betweenness and closeness attacks, which target the network's shortest paths.(evised version Page1, Line 13 - Line 14)

Comments 2:

In the 50th line you write "... reachability . However" - there must be a dot not in the middle. In line 312, the same note about the dot between sentences.

Response 2:

Thank you for your attention to detail and for pointing out the punctuation errors in our manuscript. We sincerely apologize for the oversight and appreciate your careful review. We will make necessary corrections to ensure that the punctuation format of the paper is correct.

• revised version Line 52: ”...reachability . However...”→”...reachability. However...”

• revised version Line 335: ”...locations .The...”→”...locations.The...”

– This bug has been fixed. 

Comments 3:

At the beginning of Chapter 2, you need an introduction, explaining what you are going to do in that chapter and why.

Response 3:

We agree with this comment. Therefore, At the beginning of Chapter 2 we explained what we are going to do in that chapter and why.

This section focuses on the techniques and approaches used to analyze and understand, we employ complex networks as a modeling tool to capture the dynamic behavior of the CATN  and investigate its properties. Besides, we utilized robustness measurements to quantify the network's structure and its ability to withstand potential disruptions. (Revised version Line 84 - Line 87)

Comments 4:

If possible, it is recommended not to end the subsection with a picture and to have a logical transition to the next subsection.

Response 4:

We have carefully considered your comments and made the necessary revisions to address these concerns. To ensure a better flow and logical transition between subsections, we have reorganized the structure of the manuscript. We have moved the picture from the end of the subsection and placed it within the relevant content to provide better context and support for the text. This change allows for a smoother transition to the next subsection, maintaining a logical and coherent structure throughout the paper.

Comments 5:

To clarify the title of 2.2.4. That general note to subsection 3.1

Response 5:

We appreciate your suggestion to clarify the title of subsection 2.2.4 and provide a general note referencing subsection 3.1. 

In Subsection 2.2.4, we did not explain the strength in the text, so in Revised version Line 169, we added the following explanation: The strength of a node is usually represented by the sum of the routes or flights.

In section 3.1, we made the following modifications:

• Title of subsection 3.1: Data -> Flight Data

Comments 6:

In Figure 3, the units of measurement for "Time" are unclear

Response 6:

We apologize for the oversight and understand the importance of providing clear. In Figure 3, we add “(Hours)” in x axis.

Comments 7:

It is recommended to change the name of the 3rd chapter to Results and Discussions. In addition, this chapter misses the authors' critical insights about the obtained results.

Response 7:

We greatly appreciate your suggestion to change the name of the third chapter to "Results and Discussions." We understand that you believe there is a lack of critical insights about the obtained results in the third chapter "Empirical analysis". If this is the case, we fully agree with your recommendation and will make the necessary modifications accordingly.

• Add Summary and critical insights about temporal attributes of nodes and edges in Line323-Line330

- Overall, the temporal variation of airline routes and flights in China highlights the challenges in the operational capacity of CATN, particularly in handling peak traffic. These limitations lead to flight schedule delays, and airports resort to scheduling "red-eye flights" to accommodate all daily flights, consequently increasing the risk of aviation operations. Moreover, evaluations of various representative airports reveal that they continue to encounter these constraints, although their operational capacities vary. Therefore, when developing flight optimization plans, it is crucial to optimize them based on the distinct operational capacities of different airports.

• Add Summary and critical insights about degree and clustering coefficient in Line385-393

- The findings of the study reveal the dynamic nature of the network through the examination of the degree and clustering coefficient. The analysis of the temporal and spatial variations in node degree illustrates the dynamic nature of airport operations and the fluctuating distribution of air traffic throughout the day. The fluctuation of the clustering coefficient reflects the changing patterns of connectivity and the growth of the network scale. This provides important data support for airport management and operational optimization, as well as information for the decision-making process of expanding routes and allocating airport resources to ensure effective connectivity and overall transportation performance.

Comments 8:

The conclusions recommend specifying which indicators "Besides, we tried to enhance the modeling ability of temporal networks for dynamic systems and proposed additional indicators.......". Limitations of the research you conducted must also be named.

Response 8:

We apologize for the confusion regarding the additional indicators mentioned in the conclusions section. It was a result of our poor wording choice. Our intention was to provide a discussion and outlook for future research. We deeply regret any misunderstanding caused by our oversight.

To address this issue, we will make the necessary revisions to clearly articulate our discussion and outlook for future research in the conclusions section. Additionally, we modify the heading of the conclusions section to "Conclusions and Discussion" to emphasize the discussion of the limitations of our study. In the discussion section, we will explicitly state that this paper represents a preliminary study and highlight our future research directions.

The specific changes are as follows:

• Title of section 5: Conclusion -> Conclusion and Discussion

• Original version delete Line 420-423, add discussion in revised version Line 472-482:

- As a preliminary study, we verifies the feasibility of temporal network theory in CATN research and discusses the law of robustness variation in CATN. In our future research, we intend to conduct an in-depth exploration from two perspectives. First, we will employ the theory of temporal networks to investigate the intricate dynamics of networks in CATN. This investigation aims to offer practical solutions concerning the optimization and prevention of future flight schedules. It involves addressing cascading failures triggered by nodes or links within the network, as well as the dispersion and propagation of delays. Second, we will engage in an exploration of more advanced techniques for temporal network modeling, aiming to accurately capture the dynamic structure of CATN. For instance, event-based temporal networks or dynamic graph neural networks can be utilized for this purpose.

Response to Reviewer  Comments

1. Summary

Thank you very much for taking the time to review the manuscript. We are uploading

(a) our point-by-point response to the comments (below)

(b) a revised article

(c) a revised article which has highlighted the modifications in blue to make it easier to identify the changes we made

Once again, we thank you for your careful review and helpful suggestions. Your feedback has positively influenced our research. We have made every effort to address the issues you raised, and we believe that the revised manuscript has significantly improved in terms of quality and content.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1:

It is recommended not to use the word "we" in the article, but sentences in which it is formulated in the general form, so the text will sound more scientific.

Response 1:

We appreciate your attention to the potential redundant information issues in the solution and the applicability and interest in future research on time and structure analysis.

• revised version Line 6: ”We...”→”This study...”

• revised version Line 9: ”...,we found that...”→”...,the authors found that...”

• revised version Line 78: ”We analyzed...”→”The researchers analyzed...”

• revised version Line 81: ”...,we utilize the theory of temporal networks to conduct an empirical analysis of the CATN.”→”...,the theory of temporal networks was utilized to conduct an empirical analysis of the CATN.”

• revised version Line 96: ”We...”→”This study...”

• revised version Line 112: ”...we can create a temporal network...”→”...a temporal network can be established by utilizing...”

• revised version Line 117: ”...we can generate temporal snapshot sequences...”→”... one can generate temporal snapshot sequences...”

• revised version Line 126: ”We introduced a novel technique...”→”A novel technique has been introduced...”

• revised version Line 190: ”We used the Rand function...”→”The Rand function was utilized...”

• revised version Line 192: ”We measured the network’s performance...”→”The network's performance was measured...”

• revised version Line 217: ”We used network efficiency and the maximum connected...”→”Network efficiency and the maximum connected sub-graph proportion were used...”

• revised version Line 248: ”...,we...”→”...,the researchers...”

• revised version Line 256: ”...,we...”→”...,this study...”

• revised version Line 268: ”...,we...”→”...,the researchers...”

• revised version Line 272: ”...,we obtaineded total of 4,915,350 valid data records...”→”..., a total of 4,915,350 valid data records were obtained...”

• revised version Line 284: ”...,we...”→”...,the authors...”

• revised version Line 299: ”We weighted the network with the number of flight...”→”The network was weighed based on the number of flights...”

• revised version Line 308: ”...,we...”→”...,the authors...”

• revised version Line 311: ”We...”→”The researchers...”

• revised version Line 317: ”We...”→”This study...”

• revised version Line 321: ”...,we...”→”...the authors...”

• revised version Line 337: ”We illustrate the spatiotemporal characteristics of the air transportation network...”→”...the spatiotemporal characteristics of the air transportation network were illustrated...”

• revised version Line 360: ”...,we...”→”...the authors...”

• revised version Line 371: ”..., we ignored these small connected sub-graghs...”→”...the small connected sub-graphs were ignored...”

• revised version Line 399: ”We...”→”This study...”

• revised version Line 400: ”To examine the centrality and robustness of the aviation network during these specific time intervals, we combined all flight information coming from these four time periods in 2018.”→”In order to examine the centrality and robustness of the aviation network during these specific time intervals, all flight information from these four time periods in 2018 was combined.”

• revised version Line 423: ”We...”→”This study...”

• revised version Line 457: ”...,we found that...”→”...it was observed...”

• revised version Line 462: ”...,we...”→”...,the authors...”

• revised version Line 464: ”...,we adjusted the attack parameters to target multiple noeds...”→”...,the attack parameters were adjusted to target multiple nodes simultaneously.”

• revised version Line 465: ”...,we...”→”...,the authors...”

• revised version Line 474: ”We employed complex network and temporal network theories to explore the temporal and topological characteristics of CATN..”→”Complex network and temporal network theories was employed to explore the temporal and topological characteristics of CATN..”

• revised version Line 483: ”Furthermore, we explored the robustness of CATN at different times against random and deliberate attacks using the proportion of maximum connected sub-graph nodes and network efficiency as robustness measures.”→”Furthermore, this study explored the robustness of CATN at different times against random and deliberate attacks using the proportion of maximum connected sub-graph nodes and network efficiency as robustness measures..”

• revised version Line 490 - Line 492: ”As a preliminary study, we verifies the feasibility of temporal network theory in CATN research and discusses the law of robustness variation in CATN.”→”As a preliminary study, the feasibility of temporal network theory in CATN research was verified, and the law of robustness variation in CATN was discussed.”

• revised version Line 492 - Line495: ”In our future research, we intend to conduct an in-depth exploration from two perspectives. First, we will employ the theory of temporal networks to investigate the intricate dynamics of networks in CATN.”→”In future research, an in-depth exploration will be conducted from two perspectives. Firstly, the theory of temporal networks will be employed to investigate the intricate dynamics of networks in CATN, aiming to provide practical solutions for optimizing flight schedules.”

• revised version Line 497 - Line 500: ”Second, we will engage in an exploration of more advanced techniques for temporal network modeling, aiming to accurately capture the dynamic structure of CATN.”→”Second, an exploration of more advanced techniques for temporal network modeling will be undertaken, aiming to accurately capture the dynamic structure of CATN..”