Evaluating Prediction Models for Airport Passenger Throughput Using a Hybrid Method

Round 1

Reviewer 1 Report

1. Comments

An accurate predicting result of passenger throughput for an airport is one of the most important factors for its development decision on construction and expansion.The paper analyzed the airport passenger throughput (APT) of 203 airports in China, evaluating the performance of various methods for APT prediction and introducing a proposed evaluating method for five common prediction models. It studied the applicability of the five models on different airports with various developing conditions, which had some practical significance.

2. Questions

(1) The paper sets four criteria to assess the goodness of the model. Are the criteria reliable and comprehensive? Is there a certain theoretical basis?

(2) The basic versions of different models are employed as the components of the hybrid models.Whether such treatment is guaranteed to be effective? How to solve the possible mutually exclusive effects between different models?

(3) Table 5 shows the correlation coefficients between GDP and the other indicators, However, how should the value of 101% be interpreted?

3. Suggestions

The aim of the study is using a hybrid method to evaluate prediction models for airport passenger throughput, and it is suggested that the development of assessment criteria and the construction of hybrid model should be further clarified.

 

Author Response

Point 1: The paper sets four criteria to assess the goodness of the model. Are the criteria reliable and comprehensive? Is there a certain theoretical basis?

 

Response 1: The four evaluation criteria are with different levels of goodness. They are selected as parameters to assess the model suitability of different levels of goodness. The Mean Absolute Percentage Error (MAPE) in the criteria is one of the most common methods used to calculate the forecasting accuracy. The smaller the MAPE the better the forecast ^[1]. The four criteria are based on different value of MAPE to set different levels of goodness which can be used to show the model suitability of different levels of goodness. The aim and discussion are added in the revision.

 

Point 2: The basic versions of different models are employed as the components of the hybrid models. Whether such treatment is guaranteed to be effective? How to solve the possible mutually exclusive effects between different models?

 

Response 2: There are some misleading contents in the manuscript. The main focus of the paper is evaluating five models as a hybrid method. This combination method does not need to verify its effectiveness, but only needs to ensure that each subdivision model is effective. All misleading contents have been revised in the revision. The effectivity of all models and proposed five-model hybrid evaluating method are analyzed using the data of 203 airports in China.

 

Point 3: Table 5 shows the correlation coefficients between GDP and the other indicators, However, how should the value of 101% be interpreted?

 

Response 3: There was a mistake with the value of 101%. The latest data and interpretation have been revised in the revision.

 

Point 4: The aim of the study is using a hybrid method to evaluate prediction models for airport passenger throughput, and it is suggested that the development of assessment criteria and the construction of hybrid model should be further clarified.

 

Response 4: The assessment criteria and hybrid evaluation method are further clarified in the revision. Please see the revised vision.

 

Reference:

[1] (2000). MEAN ABSOLUTE PERCENTAGE ERROR (MAPE). In: Swamidass, P.M. (eds) Encyclopedia of Production and Manufacturing Management. Springer, Boston, MA . https://doi.org/10.1007/1-4020-0612-8_580

Reviewer 2 Report

This article explores 5 prediction models to study the applicability on different airports with various airport passenger throughput and developing conditions. A topic that is not of considerable interests to research community. Here some comments and suggestions:

1、Compared with previous research results, this study has no significant improvement.

2. The implementable solutions and suggestions should be put forward to make this study meaningful.

 

 

 

Author Response

Point 1: Compared with previous research results, this study has no significant improvement.

 

Response 1: The improvement and significance of the paper are further illustrated in the revision. An accurate predicting result of passenger throughput for an airport is one of the most important factors for its development decision on construction and expansion. The paper analyzed the airport passenger throughput (APT) of 203 airports in China, evaluating the performance of various methods for APT prediction and introducing a proposed evaluating method for five common prediction models. It studied the applicability of the five models on different airports with various developing conditions. Taking the five methods as a whole, a hybrid evaluating method is built to see if the passenger throughput of an airport can be predicted. When the hybrid evaluating method of the five common modelling models show high prediction accuracy, the best prediction result can be obtain using one of the appropriate models. However, other unique methods should be considered when the hybrid evaluating method shows bad prediction accuracy.

 

Point 2: The implementable solutions and suggestions should be put forward to make this study meaningful.

 

Response 2: The implementable solutions and suggestions are added in the revision. Please see the revised version.

Reviewer 3 Report

This article is interesting and well written. I have a couple of comments for the authors.

         1)     In this study, are there any information gaps that have not yet been addressed?

2)     The purpose of the study should be stated by the authors at the end of the "Introduction" session.

           3)     The discussion of the results is not enough details.

 

4)     The study's limitations should be stated more clearly by the authors.

 

 

Author Response

Point 1: In this study, are there any information gaps that have not yet been addressed?

 

Response 1: The prediction models and data used in the paper are all written in the manuscript. There are no information gaps. To make it clearer, the development of assessment criteria, the construction of hybrid evaluation methods and the analysis results are further clarified in the revision. This study mainly focuses on the applicability of prediction methods under natural growth scenarios. There is no unsolved information gap in this scenario. The gap is mainly due to the failure to evaluate the applicability of prediction methods in the presence of epidemic and other special events.

 

Point 2: The purpose of the study should be stated by the authors at the end of the "Introduction" session.

 

Response 2: The purpose is further illustrated in the “Introduction” as follow. “An accurate predicting result of passenger throughput for an airport is one of the most important factors for its development decision on construction and expansion. The paper analyzed the airport passenger throughput (APT) of 203 airports in China, evaluating the performance of various methods for APT prediction and introducing a proposed evaluating method for five common prediction models. It studied the applicability of the five models on different airports with various developing conditions.”

 

Point 3: The discussion of the results is not enough details.

 

Response 3: The discussion of the results is revised accordingly in the revision. Time series models, causal models, artificial intelligent models, market share methods and analogy-based methods are all commonly used methods for APT prediction. In the paper, a hybrid method of the aforementioned methods is developed and investigated. By conducting the evaluation on 5 kinds of models to predict the APT of 203 airports in China, it is found that for most prediction scenes, the models are applicative to short-term prediction of APT and the accuracy does not promote with the complexities of the models. 88% of the studied airports can be effectively predicted by using the evaluated prediction methods, and the MAPE is mostly within 10%. In addition, the constructed evaluation method can effectively predict the airport passenger throughput more than 10 million. The higher the airport passenger throughput level is, the more effective the segmentation prediction methods can be. The performance of the mentioned models in this paper is bad when there are insufficient historical data for modeling, or the APT of the airports changes abruptly owing to expansion, relocation or other kinds of external forces like earthquakes. When the airport enters the stable development period, the number of available prediction models in the hybrid method is significantly increased. The results show that there is no relationship between the prediction accuracy and the complexity of prediction models. Among the five used models, time series model, causal models and market share method usually have higher applicability than the other two models.

 

Point 4: The study's limitations should be stated more clearly by the authors.

 

Response 4: The study's limitations are that the prediction and evaluation methods in this paper are usually applicable for airports with high APT, on stable development period and sufficient historical data. based on stable environment. The predictions which consider unstable external environment under the influence of the COVID-2019 and other external forces like earthquakes or insufficient historical data should be further studied. The limitations are more clearly discussed in the conclusion part of the revision.

Round 2

Reviewer 2 Report

The authors have carefully revised this paper based on the reviewers' comments.

Author Response

The authors thank the reviewer for his/her valuable comments.