Accelerated System-Level Seismic Risk Assessment of Bridge Transportation Networks through Artificial Neural Network-Based Surrogate Model
Round 1
Reviewer 1 Report
The paper is well written and use ANN to surrogate traditional transportation demand model.
My question is following:
1) What's the principle and method of TSTT prediction? Only the software name is not sufficient for people to feel the significance of this work.
2) How long does it take to simulate TSTT by traditional model?
3) Does traditional TSTT model contains statistical part which means itself is semi-statistical?
4) Is it fair to use synthetic results of traditional TSTT model to represent the real cases? Or better use the real data to train ANN and compare with traditional TSTT model?
5) Need to compare ANN model with other statistical models, e.g. Decision tree, non-parametric regression model to show this is not an easy task and should be solved by black box model like ANN.
Author Response
The authors are grateful to the reviewer for the valuable comments, which helped improve the manuscript significantly. The manuscript has been revised carefully according to the reviewer's comments. In the revised manuscript and response document, all the changes are highlighted in yellow, and the revisions are summarized in the response document.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The proposal is appealing and interesting, and the method deserves some consideration. Moreover, the paper is almost well written and well organized.
- The text, in general, reads well, but a grammatical revision could improve it further.
- The paper adequately puts the progress it reports in the context of previous work, representative referencing and first discussion.
- The authors could highlight better the new scientific contribution, for instance, analyzing several recent literature works.
Author Response
The authors are grateful to the reviewer for the valuable comments, which helped improve the manuscript significantly. The manuscript has been revised carefully according to the reviewer's comments. In the revised manuscript and response document, all the changes are highlighted in yellow, and the revisions are summarized in the response document.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper intends to investigate the system-level seismic risk of bridge transportation networks efficiently using ANN-based surrogate model. The authors provided sufficient literature and analysis to support their investigation. In addition, there are some inconsistencies between the discussions provided by the authors.
ANN is well known technique and using in many applications. But it require big dataset, which is difficult and time consuming for long term monitoring. Sometimes monitoring data could be interrupted due to different electrical and mechanical problem. How the authors deal with missing data was not clear? Because missing data could be big issue for validating the model. However, why such model was chosen to solve in the current study (What is the problem of other algorithms such as decision tree)? Also, each layer of ANN needs to be elaborated with proper equations.
Your recommendation and further research questions?
The discussion part needs some work out also.
Author Response
The authors are grateful to the reviewer for the valuable comments, which helped improve the manuscript significantly. The manuscript has been revised carefully according to the reviewer's comments. In the revised manuscript and response document, all the changes are highlighted in yellow, and the revisions are summarized in the response document.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
The revised version of this manuscript improve little bit. I accept it in current form.
