Toward a State-of-the-Art of Fly-Rock Prediction Technology in Open-Pit Mines Using EANNs Model

Round 1

Reviewer 1 Report

This paper presents a technique to predict the fly rock in open-pit mining. The problem is well investigated in academia and the techniques in the proposed method are not expected to have a significant impact on improvement of the current techniques. The scientific innovation from this work and the contribution to industry is deemed marginal.

Author Response

Dear reviewer,

Thank you very much for your time and follow up. Please kindly see our explanations in this manuscript, as well as the revised manuscript.

Thanks.

Author Response File: Author Response.pdf

Reviewer 2 Report

The problem of forecasting the fly-rock is a difficult issue and requires the collection of real data on the parameters of blasting works and geological structure. Therefore, I would like to ask you to clarify the following:

How was the spread range of fly-rock measured for the tested series under real conditions. What was the exact range of variability of blasting works parameters. How were the blasting works parameters determined? Was the burden taken from surface measurements, or was it the smallest burden obtained as a result of 3D scanning and blasthole deviation testing? I propose to provide values of the parameters of blasting works for which the largest and smallest fly-rock were measured.

 

How were the data for training and testing of ANN models selected? What part was intended for training and which part for testing. Were attempts made to influence the amount of data intended for training on the quality of the model?

How does the introduction of EANN in reality (for a blasting engineer) improve the dispersion prediction, since the differences between the simplest ANN model and EANN are small in statistical indexes.

Which of the blasting works parameters most affects the training of the ANN model?

It seems to me that Figures 6-9 are unnecessary, one example Fig. 5 is enough, besides the descriptions under Figures 8 and 9 are the same, and the drawings show a different structure of ANN models.

Author Response

Dear reviewer,

Thank you very much for your time and valuable comments. We have addressed/corrected all your comments/suggestions in the revised manuscript. Please kindly see them in the revised version of the manuscript.

 

Thanks again!

Author Response File: Author Response.pdf

Reviewer 3 Report

This manuscript presents a state-of-the-art technology of fly-rock prediction based on ANN models and the associated robust combination EANNs model. The paper is easy to understand, and the results and analyses are reasonable and convincing. This an interesting paper and good contribution to the literature. Just few following corrections need to be addressed:

1. Line 81: notoriousànotable

2. Line 145: accuracyàAccuracy

3. I would suggest just keep Figure 5 enough from figures 5-9.

4. I would suggest add one more scenario to be compared: use the EANN 5-25-21-15-1 model (within three hidden layers) to train and test the original data.

Author Response

Dear reviewer,

Thank you very much for your time and review. We have addressed/corrected all your comments/suggestions properly in the revised version of the manuscript. Please kindly see them in the revised manuscript.

Thanks again!

Author Response File: Author Response.pdf