Revealing the Varying Impact of Urban Built Environment on Online Car-Hailing Travel in Spatio-Temporal Dimension: An Exploratory Analysis in Chengdu, China

Round  1

Reviewer 1 Report

 This paper aims to introduce a support for guiding and managing the development of online car-hailing, in order to integrate it into the transportation system and built environment. Geographically weighted regression (GWR) is used to check the spatial heterogeneity of the car-hailing travels and the most influential factors.

The paper is generally clear, well-written and well-structured. The proposed methodology is not new but the applied to an increasingly popular mode of urban mobility (online car-hailing). Tables and figures are clear, except figure 7, that I retain not very appropriate.

My major concern is about the literature review, because I think that it should be integrated and enriched. Specifically, the authors focused their review only on studies regarding built environment impact on travel behaviour but, in my opinion, they should furthermore talk about the importance of spatial analysis also in different field. As an example, some authors have recently introduced spatial analysis techniques for evaluating transit service quality at rail station (see “Spatial variation of the perceived transit service quality at rail stations”. Transportation Research Part A: Policy and Practice, 114, 2018, pp. 67-83).

In addition, the authors focus their review only on global regression models (page 3, row 70). Really, I would like to suggest more papers which used spatial association techniques and GWR analysis for analysing trip generation and their relationship with built environment, as the following:

“Spatial association techniques for analysing trip distribution in an urban area”. European Transport Research Review, 4(4), 2012, pp. 217-233.

“Factors impacting trip generation on Metro system in Madrid (Spain)”, Transportation Research Part D, Transport and Environment 67, 2019, pp.156-172.

“Exploring the Relationship among Urban System Characteristics and Trips Generation through a GWR Model”, International Journal of Innovations in Information Technology 1(2), 2013, pp. 51-61.

The suggested papers can be useful also for introducing other built environment attributes previously used in the literature, and for comparing the obtained results.

Author Response

Author Response File: Author Response.docx

Reviewer 2 Report

This paper explores interaction between built environment and ride hailing travel. The topic of this study is state-of-art and provides new insights, also they already pointed out critical limitations by themselves in conclusion. I have some comments.

1. Authors compared OLS and GWR models then concluded the GWR model shows better model fit. However, when I check estimation results in table 3 and 4, the variables used in each model is different. Is it fair comparison? How authors defend it? It should be much more emphasized in text.

2. 7p line 144, authors explain Wi as diagonal matrix in which the entries outside the main diagonal are all zero. However, when I checked equation 4, it seems the matrix has zero only for the  main diagonal. Please check the formulation and terminology.

3. How authors choose the date for “one-day data” ? Is there any specific reason?

4. In figure 5, 6, and 7, the index of X and Y-axis is not clear. What they mean? Please make it clear.

5. It would be better for readers to explain what is the “urban built environment”

Author Response

Author Response File: Author Response.docx

Reviewer 3 Report

 The topic is rather innovative and of interest. However, the manuscript seems to be more focused on its modelling than to its interpretation. There is no description of the built environment but just of general land use functions (which makes the manuscript title misleading) and for non-local readership, aside from the model description, it is very difficult to understand the phenomenon and its implications. Likewise, the discussion of results seems to be targeted to a native audience, who knows exactly places and names.

It is suggested a to revise the contents of the manuscript bearing in mind an international audience, who might be not familiar with the case study. The selection of the case study itself should be more elaborated, too, and its features in terms of built environment extensively described. 

Specific remarks

- r 5 reference not in the journal style, and repeatedly for more sources in the manuscript. Please amend

- From the list of POIs the study area seems with mixed land use, but a more detailed description on its built environment could help readers

- r 110 (above) which is the difference in thank between grid I and grid i In the formula?

- r 116 what are cold areas? 

- r 126 which are the 16 indicators? Describe them

- r 210 claimed inconsistency is maybe because of the comparison with non-Chinese reference (different land use and functional urban patterns). Probably referring to some local references could help dissipate inconsistency

Author Response

Author Response File: Author Response.docx

Round  2

Reviewer 2 Report

I am satisfying with current version.

Reviewer 3 Report

All the major issues have been addressed and the manuscript gained in clarity; the manuscript is now in shape for being published