Building a Large-Scale Micro-Simulation Transport Scenario Using Big Data

Round 1

Reviewer 1 Report

The authors propose a method to estimate traffic flow at the road level considering different transport modes, including active modes. The method could be useful for transport planners and policy-makers and the case study is well detailed. I have the following comments:

• The paper seems a bit imbalanced in the sense that the Introduction is very long, whereas the interpretation of the results and the explanation of the model are quite brief. For instance, I think that the last paragraph of Section 3 (starting “It is interesting to note how, by setting…”) should be expanded, as it provides useful insights about the method proposed.
• In the same vein, the analysis of measured flows vs simulated flows could be enriched. For instance, it could be of value to analyze the error depending on road types or to assess the presence of spatial autocorrelation in model residuals. At least the first of these two model quality indicators should be incorporated into the paper, in my opinion.
• The authors use the term “big data” from the very beginning of the paper. Although the term is widely used, maybe it would be better that the authors clarify what they are referring to by “big data” in this case. It may seem obvious, but I still think that a comment is needed (maybe by comparing current “big data” data sources against traditional data sources).

Other minor issues:

• The authors use a structured abstract, which is not common in IJGI.
• English is good, in my opinion, but a revision of the text is still needed. For instance, the verb “allow” needs a pronoun in case the infinitive is used, or a gerund in the absence of a pronoun. Besides, I have found other mistakes/typos such as “mayor federal and provincial roads”, or strange sentences such as “with an alpha1 is fixed at zero”.
• In Lines 478 and 485, the subscripts of the alphas are not rightly typed.
• In Line 569 capital letters should not be used (“where fij…”).

Author Response

Please see attached file with answers

Author Response File: Author Response.docx

Reviewer 2 Report

This paper introduces a comprehensive agent-based traffic microsimulation model for the city of Bologna based on SUMO, accounting for multiple transport modes. The authors claimed that the simulation results have been validated against real-world traffic data, indicating the effectiveness of building a large-scale activity-based model for transportation planning and management.

My major concerns are:

• The simulation is primarily based on a simple mode choice model, which is computationally efficient. The validation tests are not sufficient to support the conclusion since no other state-of-the-art models were involved and compared. You many implement a few popular micro-simulation models and compare their performance with your model;
• The consistency between simulated and observed traffic is not significant (0.61 in R2). Could you examine the performance in more details, such as evaluating which parts of the city have inferior performance and explain why？
• The readability of the plan choice model (section 3) can be improved. The current form can be re-structured: a workflow figure may help. The intellectual contribution is unclear. Please highlight your contribution in this part.

Author Response

Please see attached file with answers.

Author Response File: Author Response.docx

Reviewer 3 Report

Major

1. You may need to re-structure the manuscript. The introduction is usually structured as follows: problem statement, the research gaps, why it is now good time to address the research gaps, and research questions (or aims), and general structure of manuscript. In addition, research questions were not clearly stated.

• In addition, Literature review section is usually an independent section, which is not a sub-section of introduction.
• Also, method and result sections need to be separated. It may hinder the readers to understand your manuscript.  

2. To make this manuscript logically sound, you may need to properly cite the previous studies. For example, in your model, you considered about the vehicle ownerships. For this, you can explain the Pattern-Oriented Modeling (POM) paradigm (Grimm et al., 2005). Following POM, the patterns of modeling elements should be similar to those from observations.

Grimm, V., Revilla, E., Berger, U., Jeltsch, F., Mooij, W. M., Railsback, S. F., ... & DeAngelis, D. L. (2005). Pattern-oriented modeling of agent-based complex systems: lessons from ecology. science, 310(5750), 987-991.

3. In agent-based modeling community, it is required to share the ODD (Overview, Design concept and Description) to promote the reproducibility (Grimm et al., 2010). Otherwise, people may not replicate your model. In other words, it limits impacts of your study.

Grimm, V., Berger, U., DeAngelis, D. L., Polhill, J. G., Giske, J., & Railsback, S. F. (2010). The ODD protocol: a review and first update. Ecological modelling, 221(23), 2760-2768.

4. It is a good idea to leverage three indicators for validation. According to Kang et al., (2019), simulated outputs need to be compared to observation at multiple space-time scales. Then, it can help to avoid “equifinality” in agent-based modeling.

Kang, J. Y., & Aldstadt, J. (2019). Using multiple scale spatio-temporal patterns for validating spatially explicit agent-based models. International Journal of Geographical Information Science, 33(1), 193-213.

5. You may need to bring proper reference to support your future direction and your general research idea for making your manuscript reasonable.

Minor

1. Abstract style. IJGI does not have subsection in Introduction, but you have background, method, results, and conclusion.
2. Please double-check the words.

• e.g., km2 should be km²

3. Sec needs to be Section.

Author Response

Please see attached file with answers.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have widely improved the paper and correctly answered to my comments. I would only suggest them to improve the quality of the Figures.

Reviewer 2 Report

My commetns have been addressed.