Exploration and Prediction of the Elderly Travel Behavior Based on a Novel GR-GA-BP Hybrid Model

Round 1

Reviewer 1 Report

Based on the travel data of the elderly obtained from the field survey, from the perspective of poor information, a grey relational degree (GR) model was proposed to explore the internal relationship between individual characteristics and travel behavior of the elderly. The results showed that bus and subway station were the greatest influence on the times of weekly trips of the elderly, while the number of family population had little impact on the times of weekly trips and the maximum walking distance.

It is much potential for practical application and management. This is a carefully done study and the findings are of considerable interest. A few minor revisions are list below.

1.     The contributions of this manuscript with respect to the existing work is unclear.

2.     The horizontal and vertical coordinates of many graphs in the experimental results have no units.

3.     The references should be edited in the right form of the journal. Some references numbers are missing.

4.    The illumination of two figures in the FIGURE.2 should be added, and the quality of FIGURE.4 should be improved. and the data in the Table1-2 should be illustrated.The labels of Tables 2 and 3 should be switched.

5.     The paper should be reviewed carefully, in order to correct all the typing, grammar and English errors.such as Sentence 42-45 page 9 is difficult to understand and should be explained specifically.

Author Response

Dear :

Thank you very much for your valuable comments. Please refer to the attachment for relevant contents and replies.

Best Regards

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The presented article deals with the prediction of travel behavior for the elderly in Beijing, China using the GR-GA-BP model. The authors claim that it is the original GR-GA-BP model, but both in the Literature review and in the presentation of the model, they do not offer a clear explanation or proof that it is really the original model. In the literature review, I would expect a description of the existing state of the GR, GA and BP models and their use, and in the presentation of the model, a more precise description of what is new and different in the proposed model? Is this really a new model or the first use of such a model to predict the travel pattern of older people?

 

Below are a few more comments, the implementation of which could contribute to the quality of the article:

1. The abbreviation GR-GA-BP is used both in the title and in the abstract. As a rule, abbreviations must be described the first time they are used. The attentive reader should not look for the explanation of abbreviations throughout the article. The same applies to abbreviations that are expected to be widely known (eg MRE, MSE, AFC). Need to fix!

2. I believe that references are not cited and produced in accordance with the Instructions to Authors. Some regressions are not included in the reference list (eg Zhang et al., line 100). Need to fix!

3. GR-GA-BP similar models are not discussed in the Literature review. Need to add!

4. The authors claim in the text that the considered sample consisted of 3250 samples, while the figures (Fig.1, Fig.2 and Fig.3) show N=325. Which sample size is right? It is not clear from the article how the statistical relevance of the sample was discussed and proven?

5. Fig.1 to 3. are easily readable. With the goal of better readability, I propose a correction.

6. In Ch. 4, it is necessary to add information about what is new about this model or different from other models!

7. Equations must be entered in accordance with the Instructions for Authors!

8. Fig. 5 and 6 do not appear to be the work of the authors. If the authors of the article are not the authors of the mentioned pictures, it is necessary to add references or sources from where the images are used!

9. Does MSE=mse and MRE=mre apply? Records need to be corrected!

10. Which tools were used in the implementation of the proposed method? The variable "time" in Tables 4 and 5 probably refers to the computer time required for the calculation of each method. If this is the case, then information with which tool and on which computer the model was realized would be useful.

Author Response

Dear Prof.

Thank you very much for your valuable comments. Please refer to the attachment for relevant contents and replies.

Best regards

 

Author Response File: Author Response.pdf