Civil Aviation Passenger Traffic Forecasting: Application and Comparative Study of the Seasonal Autoregressive Integrated Moving Average Model and Backpropagation Neural Network

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript analyzes the national civil aviation passenger transportation data from 2006 to 2019 by applying the SARIMA model, BP neural network model, and a combination of these two models. The feasibility and accuracy of the three models are verified by predicting passenger transportation volume in 2019. Based on the obtained optimal model, the civil aviation passenger transportation data from 2020 to 2023 is predicted, and the impact of the epidemic on civil aviation passenger transportation is compared with actual data. The final data results have certain significance for airlines to predict future passenger volume and formulate effective response strategies and measures for similar crises that may occur in the future. They also have certain reference value for research on predictive models or integrating multiple methods to improve the accuracy and stability of predictions. However, there are still issues worth considering.

1.The manuscript introduces the models used for prediction in the abstract section, but does not point out the innovative points and how they differ from the model predictions studied by other scholars.

2.The research listed in the introduction section of the manuscript should not only briefly describe its content, but also state its shortcomings or any references it may have for this article, as well as why the model used in the manuscript was chosen.

3.Is it inappropriate to state in the introduction section of the manuscript that this study evaluates the aviation passenger transportation industry's ability to respond to global crises? The conclusion in the manuscript only compares actual and predicted data of aviation passenger transportation affected by the epidemic, without describing this aspect.

4. The manuscript should briefly describe the Box Jenkins' model identification method in section 3.1.2 and explain how to obtain the results shown in Figure 6.

5.The manuscript points out in the conclusion that combining multiple prediction methods can improve the accuracy and stability of predictions. Do readers have any questions about whether the more combined prediction methods, the more accurate they will be? An explanation should be provided.

In summary, the author must sort out the logic of the article according to the writing standards of scientific and technological papers, check all details of the article, and if the author considers continuing to publish the paper in this journal, they should verify and resubmit it.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper is entitled “Civil aviation passenger traffic forecasting: an application and comparative study of SARIMA model and BP neural network”. In this case, the idea and results of the paper are interesting but the following comments can be utilized to improve this paper in future.

 

 

Abstract

1-     What is the meaning of BP? Authors must explain all abbreviation for the first time,

2-     When you mention evaluating the forecasting accuracy of the models, it would be beneficial to specify which metrics you used for comparison (e.g., mean absolute error, root mean square error, etc.).

3-     When discussing the comparison of the epidemic's impact on civil aviation passenger traffic with actual data, provide more details on how this comparison was conducted. Did you analyze specific aspects of the epidemic's impact (e.g., lockdowns, travel restrictions)?

4-     Conclude the abstract by briefly discussing the implications of your findings and any limitations of the study.

5-     Abstract need a simple introduction at the first part of that

6-     Abstract need to describe about existing problem which manuscript was arranged.

7-     What is the objectives of this research. It must be describe in the abstract

8-     In the following sentence, two models or three models?!

“By comparing the results with the actual data, this paper evaluates the forecasting accuracy of three models to determine the best forecasting method.”

 

Introduction

1-     While you mention the significance of comparing the BP neural network model and the SARIMA model, it would be helpful to explicitly state the specific gap or research question that your study addresses. What unanswered questions or limitations in existing literature motivated your research?

2-     Some sentences are long and complex, which can make them difficult to follow. Try breaking them down into shorter, more digestible sentences to improve readability and comprehension.

3-     Rather than simply listing previous studies, consider providing a brief context or summary of each one. This would help readers understand how each study contributes to the broader conversation and sets the stage for your research.

4-     Since the authors mixed introduction with literature review, they must describe more prior research which are related to the objective of this research

5-     This section needs a table to describe briefly for all utilized papers about their limitations and findings.

 

Research Methodology and Data

1-     Describe more about utilized data, type of data and procedure of preprocessing of data

 

Model Application and Analysis of Results

1-     “1.4. Comparison and Analysis of Results”: check the heading number

2-     For Comparison and Analysis of Results, which data utilized?! Is it same with modeling or different ones?

3-     Seems to be the procedure of evaluation of models (MSE, RMSE, R2, …..) and results were missed. Authors must add for their analysis.

*** Manuscript do not have suitable discussion related all objectives and findings. It must be added.

 

 

Final decision: This manuscript has interesting objectives however it is not suitable to publish now. It has major gap and problems for writing and organization. Therefore it needs "Reconsider after major revision"

Comments on the Quality of English Language

It needs some modification.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

In this paper, the combination of BP neural network model and SARlMA model is used to predict the passenger volume of civil aviation in the same period, and then the prediction accuracy of these three models is evaluated, and the best prediction method is determined, and the passenger volume of civil aviation from 2020 to 2023 is predicted by this method. Finally, the impact of the epidemic on civil aviation passenger transport is compared with the actual data, which is of practical significance for understanding and evaluating the impact of the epidemic on the aviation industry. However, at present, there are still the following problems in the abstract, literature review, result interpretation and innovation of the article, and it is recommended to modify:

Comment 1: Some of the research results are lacking in quantitative expression, so it is suggested to reorganize the language to show the innovation and important conclusions of the article and fully show the highlights of the paper.

Comment 2: The introduction is lacking in content, and the overall logic is a little confusing. The introduction of each model is only a single list, and the relationship between the development of each model is not pointed out. The impact of the epidemic on the development of global civil aviation is not deeply analyzed. The introduction and summary of the rest of the article can be deleted.

Comment 3:The literature review part in the introduction is not logical and lacks the mechanism analysis of the interaction between "the development of forecasting methods and models of civil aviation passenger volume" and "the impact of epidemic situation on global civil aviation". At the same time, the review of the existing research is not deep enough, which leads to the summary of the article innovation is not clear and accurate enough.

It is suggested that the author review the literature with reference to the mature logic of high-level papers at home and abroad; According to the key words such as "BP neural network" and "SARIMA", the author is advised to refer to and quote the following documents:

[1]  Survey data and human computation for improved flu tracking. Nat Commun 12, 194 (2021).

[2]  Trade-offs in land-use competition and sustainable land development in the North China Plain. Technological Forecasting and Social Change, 141, 36–46.

Comment 4: In the research method and data part, the specific ideas and selection criteria for the construction of the index system are not clearly stated, and it is suggested to supplement the specific meaning of the index and the data source and processing, and at the same time, avoid great duplication among the indicators. The research method is simple and the innovation is weak.

Comment 5:In the part of research results, the analysis and discussion of the results are not deep enough, which makes the significance of the research not profound. It is suggested to highlight the model test of the combination of the two models and reduce the description space of the two single models. In addition, the research only focuses on the comparison of the relative errors of the three models, but does not focus on the direct and fundamental reasons for the differences. If we can find out the driving factors behind them, this may make this paper more meaningful and innovative.

Comment 6: The conclusion part lacks the expression of policy suggestions, which is not practical, and the suggestions are further refined to make the research results more targeted and operable. In chapter 3.4, the title number is marked incorrectly.

Comments on the Quality of English Language

none

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addreessed all comments.

Comments on the Quality of English Language

It is suitable.

Author Response

谢谢！

Reviewer 3 Report

Comments and Suggestions for Authors

ACCEPT

Comments on the Quality of English Language

NONE

Author Response

Thank you!