The Importance of Weather Factors in the Resilience of Airport Flight Operations Based on Kolmogorov–Arnold Networks (KANs)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for submitting your manuscript, with an interesting and current topic, as the influence of weather changes (not necessarily caused by adverse phenomena) in the delays of flights. The Research is well-conducted and the manuscript follows the standards. My major concern regards the English language. Anyway, there are some issues listed below that you should solve before the acceptance of the work.

1. Figure 1. Please enlarge it (the labels are difficult to follow) and add the source in the caption (not only the reference number)

2. Figure 4 is also too small. In the caption indicate that the cyan dot corresponds to the airport. Add a scale bar to understand the magnitude of the flights.

3. Please explain in section 4 which are the usual weather causes of delay in the studied airport.

4. Enlarge Figure 7

5. Indicate in Figure 4 the airports presented in Table 3

6. The format (3D style) of Figures 8 and 9 difficults the understanding. I suggest changing to density plots. Besides, the axis text is difficult to read.

7. I am not able to understand the figure 10 because of its size. Anyway, it results very surprising the few differences between both curves. What is the forecasting time? (I was not able to see in any part of the text)

8. Again, the size text of the axis in Figures 11-13 is tiny, being not easily readable.

9. Which are the values associated with the wind and temperature changes? For example, do you think that a change of 5ºC has the same implications as 10ºC?

Best regards

 

 

 

 

Comments on the Quality of English Language

The manuscript English Language needs a clear improvement. The text is very confusing, and the Authors repeat the same sentences, making difficult the understanding of what they want to explain.

For example, the first four lines of the Abstract: "To analyze the impact of weather factors on the resilience of airport flight operations, particularly on flight operation, economic performance, and transportation capacity, Kolmogorov-Arnold Network (KAN) model was used to identify key weather features and establish the mapping relationship between weather factors and airport operational resilience." repeats the words "resilience" (2), "operational"/"operations" (3), "flight" (2), "airport" (2). This is one of the multiple examples.

I suggest the Authors contact an English native speaker to address all the limitations before resubmitting.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript describes the main weather factors and their relationship to airports' operational resilience with a focus on the Chinese  Xi'an Xianyang International Airport, via the KAN approach. There is not much to tell on this interesting manuscript, with clear methodology and results fully explained.

Just some recommendations to improve the manuscript attractiveness:

1) the problem description, when focusing on ground operations, seems not to dwell much on delays on taxiing operations; please, address this issue and consider the following paper as a prompt on this regard:  Di Mascio P, Corazza MV, Rosa N.R, Moretti L. Optimization of Aircraft Taxiing Strategies to Reduce the Impacts of Landing and Take-Off Cycle at Airports. Sustainability, 14(15), 2022, 9692.

2) equations 1 to 12: please, create a box or a subsection where it is possible to see a numerical application with data coming from the case study airport.

3) the above point 2 is related to the fact that the Authors do not comment the transferability of this approach to other case studies. Please, do that on the conclusions; also contemplate to comment on the methodology if applied to surface nodes (e.g. high speed rail stations) in case of similar weather conditions

4) this bring the focus on the conclusions, which appear like a proxy for a short summary. Clearly highlight, instead, the research limitations, transferability of results, work ahead, operational potential of the results. 

minor remarks:

- 181, explanation of eq. 1: probably something is missing in the sentence: "average delay of each flight in..." please clarify and introduce units of measurement wherever applicable.

- figures are too small; please enlarge them

- write equations' explanations with 1 item per line, so that it is clearer (for example, 250 - 256 explanation as a narrative is difficult to follow. 

- 335, eq. 13, is it incomplete? Please check equations with equation editors function

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

1. Title and Abstract

   - Title: The title is informative and reflects the content of the study well. It specifies the research focus on weather factors, airport flight resilience, and the use of Kolmogorov-Arnold Networks (KANs).

   - Abstract: The abstract provides a good overview of the research aims, methods, key findings, and implications. However, it could be enhanced by including specific details regarding the model’s performance metrics and any limitations identified during the research.

 

 2. Introduction

   - Strengths: The introduction provides a comprehensive background on the importance of airport resilience and how weather conditions impact flight operations. The authors also justify the choice of using KANs and Grey Relational Analysis (GRA) by emphasizing the limitations of traditional models.

   - Areas for Improvement: The introduction could benefit from a clearer statement of the research gap, perhaps by highlighting the shortcomings of existing resilience models in more detail. For example, mentioning why traditional neural networks may fail to capture the complexities of weather impacts would strengthen the justification for using KANs.

 

 3. Literature Review

   - Strengths: The authors cite relevant literature, including prior studies on resilience in air traffic networks, the influence of weather on airport operations, and the role of KPIs in measuring resilience. This demonstrates a thorough understanding of the topic.

   - Areas for Improvement: The literature review could be better organized, with clearer thematic sections. Additionally, some references appear outdated (e.g., studies from 2003 and 2012), and it would be beneficial to incorporate more recent research to reflect advancements in resilience modeling and machine learning applications in aviation.

 

 4. Methodology

   - Strengths: The methodological framework is sound. The choice of Kolmogorov-Arnold Networks for time-series forecasting is well-justified, and the authors provide a detailed explanation of how KANs differ from traditional neural networks. The Grey Relational Analysis (GRA) approach is appropriate for determining the importance of weather factors.

   - Areas for Improvement: While the methodology is technically robust, the explanation of certain technical details could be clearer for a broader audience. For example, the description of KANs' spline-based activation functions might be difficult for non-expert readers to follow. A simplified explanation or visual aids would help improve understanding.

   - The paper assumes that weather is the only variable affecting airport resilience, which is a significant simplification. The potential influence of other factors (e.g., air traffic control issues, mechanical failures) is not discussed, which could limit the generalizability of the findings.

 

 5. Data and Case Study

   - Strengths: The authors use real-world data from Xi'an Xianyang International Airport, making the study highly relevant. The period selected for analysis (February 23, 2023, to April 15, 2023) allows for a comprehensive assessment of weather variability and its impact on operations.

   - Areas for Improvement: The authors could include more details on the data quality and how missing or erroneous data points were handled. Additionally, the choice of time period could be expanded to include data from other seasons or airports to assess the generalizability of the results.

 

 6. Results

   - Strengths: The results are clearly presented with appropriate use of figures and tables to illustrate key findings. The analysis shows that weather factors like wind speed, wind direction, and temperature significantly influence flight operations, economic performance, and transportation capacity.

   - Areas for Improvement: The results could benefit from a deeper discussion on the limitations of the KAN model. For example, were there any cases where the model failed to predict disruptions accurately? Additionally, it is unclear how much improvement the KAN model offers compared to traditional neural networks. A comparison of prediction errors between KAN and more conventional models (such as Multi-Layer Perceptrons) would provide a stronger case for the superiority of the KAN approach.

 

 7. Discussion

   - Strengths: The discussion appropriately interprets the findings and connects them to the broader context of airport resilience. The implications for flight scheduling and airport management are clearly outlined, with practical recommendations for stakeholders.

   - Areas for Improvement: The discussion would benefit from a more critical reflection on the study’s limitations. For instance, the model only focuses on weather-related disruptions but does not account for the interconnected nature of flight networks, where disruptions in one airport can cause cascading effects. This could be a significant limitation when considering airport resilience holistically.

 

 8. Conclusion

   - Strengths: The conclusion effectively summarizes the study’s key findings and their practical implications. The focus on improving resilience through better weather forecasting and the use of advanced models is relevant and well-articulated.

   - Areas for Improvement: The conclusion could be strengthened by including more concrete recommendations for future research, such as expanding the model to include other variables (e.g., air traffic control or airport infrastructure issues) or applying the model to airports in different regions or countries.

 

 9. Quality of Writing and Presentation

   - Strengths: The paper is generally well-written, with a logical flow between sections. The figures and tables are helpful and enhance understanding of the results.

   - Areas for Improvement: Minor grammatical errors and awkward phrasing are present throughout the paper. For example, the phrase “flight distance... determined the weight of each route” is unclear in places. The writing could be made more concise and polished with a thorough editing pass.

 

 Comments and Suggestions for Authors:

 

1. Introduction and Literature Review: Consider restructuring the introduction to provide a clearer statement of the research gap. It would be helpful to include more recent studies in the literature review and organize the existing references thematically to enhance clarity.

 

2. Methodology: While the methodology is robust, some sections, particularly the technical explanation of KANs, could be simplified or supplemented with visual aids to improve accessibility for readers who may not be familiar with advanced neural network models.

 

3. Data Analysis: Expand the analysis to include data from multiple seasons to assess how weather patterns vary over time. Additionally, it would be useful to explain how missing or noisy data were handled.

 

4. Results and Model Comparison: The results could be enriched by providing a comparative analysis of the KAN model’s performance with traditional neural networks or other predictive models. This would make the advantages of KANs more concrete for readers.

 

5. Discussion of Limitations: A more in-depth discussion of the limitations of the model would be beneficial. While the focus on weather factors is important, other factors impacting airport resilience should be acknowledged.

 

6. Conclusion: Consider adding more specific recommendations for future research, such as applying the model to other airports or integrating additional variables that affect flight operations beyond weather.

 

7. Language and Clarity: The manuscript would benefit from minor editing to improve readability and eliminate grammatical errors. Sentences in some parts of the paper are overly complex and can be simplified.

Comments on the Quality of English Language

Overall, the quality of the English language in the manuscript is good, but there are several areas where improvements are needed to ensure clarity and professionalism. Below are some detailed comments and suggestions:

 

1. Sentence Structure:

   - Some sentences are overly long and complex, which can make them difficult to follow. For example, the sentence in the introduction: 

     > "With the continued growth of the aviation industry, airports have become essential hubs of economic development, especially in regions where air travel is the primary means of long-distance transportation."

     could be shortened or split into two sentences to improve readability.

   - Consider using shorter, more direct sentences to convey complex ideas more clearly.

 

2. Grammar and Tense Consistency:

   - There are a few minor grammatical issues, particularly with tense consistency. For instance, in some parts, present tense is used when past tense would be more appropriate. For example, 

     > "The study aims to address this gap by applying an advanced predictive model..."

     might work better as 

     > "This study aimed to address this gap by applying an advanced predictive model..."

   - Ensure that the tense usage is consistent throughout the text, particularly when describing past research and results.

 

3. Word Choice:

   - The use of technical terms is appropriate, but there are some instances where word choice could be refined to be more precise or professional. For example, the phrase:

     > "The weather changes severely affect flight operations..."

     could be improved to:

     > "Weather changes significantly impact flight operations..."

   - Avoid using informal or overly casual language. Terms like "a lot" or "big impact" should be replaced with more formal equivalents such as "significant" or "substantial."

 

4. Clarity and Flow:

   - The flow of ideas is generally logical, but some sections would benefit from clearer transitions between paragraphs. For example, the shift from discussing weather impacts on operations to introducing the Kolmogorov-Arnold Network (KAN) model could be made smoother with transitional phrases like "In light of these challenges" or "To address these complexities."

   - Some technical sections, particularly the explanation of the KAN model, could be simplified for better understanding without losing accuracy.

 

5. Punctuation:

   - Punctuation errors, particularly the use of commas and semi-colons, are scattered throughout the text. For example, commas are sometimes missing in compound sentences or are placed incorrectly. For instance:

     > "Temperature and wind speed changes are the key factors affecting flight operation and economic indicators."

     would be clearer with commas:

     > "Temperature and wind speed changes are the key factors, affecting both flight operation and economic indicators."

   - Ensure that commas are used to separate independent clauses or to set off non-essential information.

 

6. Use of Passive Voice:

   - The paper relies heavily on passive voice, which is not necessarily incorrect but can sometimes make the text feel less direct. Consider alternating between active and passive voice to improve readability and engagement. For example:

     > "The KAN model was used to analyze weather data."

     could be revised to:

     > "We used the KAN model to analyze weather data."

 

7. Spelling and Typos:

   - There are few spelling mistakes or typographical errors, but it's essential to conduct a thorough proofreading to ensure there are no small errors like “economy performance” (should be “economic performance”) or “degrade capacity” (should be “degraded capacity”).

 

8. Technical Jargon:

   - While the paper is directed at a technically knowledgeable audience, some explanations could be simplified or clarified. For example, the term “Grey Relational Analysis” is used without sufficient explanation in certain parts. While the term itself is accurate, additional context or a brief explanation would benefit readers who may not be familiar with the method.

 

 Suggested Edits:

1. Use shorter, clearer sentences where possible to enhance readability.

2. Ensure tense consistency throughout, especially in sections describing the research process.

3. Double-check punctuation, particularly the use of commas and semi-colons, to improve the logical flow of sentences.

4. Consider using more active voice to make the text feel more engaging.

5. Conduct a thorough spell-check to catch any minor typographical errors.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

 

English is still very poor and difficult to understand- You have changed some words, but it is very surprising to me that most of the changes have been to using "thus".

 

There are some improvements, but most of the previous questions are poorly solved. E.g. I asked to improve the text in the caption of Figure 4 and it has not been done.

 

Besides, what is the value of your research if you are indicating which are the elements that affect the operational use of the airport if you do not provide any significant threshold about these variables?

 

In the case of the weather causes, I referred to the weather conditions. Please, provide some description of the variables if not the weather regimes.

 

What is an epoch?

 

Again, what is the forecasting time? This is, with which lead time you can predict the changes in the interesting variables?

 

You need to provide more scientific information in your answers.

 

Best regards.

Comments on the Quality of English Language

I cannot distinguish relevant changes in the readability of the manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors provided partly met the revision requirements, missing to address the challenging issues raised. More specifically, no efforts seem to be placed in addressing requirement associated with:

comment 1 - it is important to stress the need of optmizing LTO operations in managing delays, not just reporting stats which are well-known or acknowledging the obvious relevance of taxiing operations. Please, elaborate this in relation to your research question and problem descrption, and in the conclusions. Start from the suggested reference and add more on this regards.

comment 2 - Table 2 is useful (provided to be bigger and more legible) but it is now what asked. The request was to create a box with real numbers as an example of calculations, so that the readership can appreciate the accuracy in the methodology

comment 3 - It is not just to replicate my suggestion about the transferability to other nodes, but comment/elaborate how this can be made. 

comment 4 - limitations: it is not just quantity and quality of data (to evidence which the numerical application was asked), but there is more to explain (throughput, recurrence of specific weather conditions, types of operations, typically cargo vs passengers, etc.).

The Authors are required to revise properly the manuscript, according to what has been asked. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf