Design and Analysis of an Effective Multi-Barriers Model Based on Non-Stationary Gaussian Random Fields

Round 1

Reviewer 1 Report

I have reviewed the manuscript entitled “Design and Analysis of An Effective Multi-Barriers Model Based on Non-Stationary Gaussian Random Fields” submitted to the Electronics journal. This manuscript is within the scope of the journal and the mathematical ingredients are quite well. Without a doubt, this study can be a proper candidate for publication. However, there is still room for improvement. Hence, to attract readers from both academia and industry, this manuscript needs a MAJOR REVISION. Consequently, the authors are encouraged to address ALL of the following comments for the potential publication:

1-      There exist some grammatical errors, typos and misleading sentences. Please double-check the text and address it. For example, “In (2), it was assumed “. It means “In Eq. (2)” or “In Fig (2)”? which one? What is “location?where”?

“In Table 1, Intercept is the intercept value, ?? is the standard deviation of the Gaussian random fields”

2-      The introduction part can be improved. I recommend adding explanation about random field and its application in both electronics and other engineering fields.

3-      In section 2.2.2, please add the corresponding reference.
“Lindgre et al.[?] also proposed that the solution of the SPDE can be approximated”

“In 2018, Bakka et al. [?] proposed the Barrier Model that divided”

4-      Some acronyms are not introduced for the first time. As an example, GRID and CCD are not defined.

5-      After introducing abbreviations, its recommended to use them! So, please follow a unified approach in this regard. One may note that some full forms are repeated after introducing the abbreviations.

6-      I recommend adding a flowchart to describe the whole methodology, to attract more readers from early career researchers and/or practitioners.

7-      Please add the numerical platform for the findings/simulations?

8-      Adding references from MDPI journals is appreciated. In accordance with comment #2, you can add some works in different fields which they applied the random field. For example, I recommend adding this paper from an MDPI journal which used random field theory with an engineering example: “https://doi.org/10.3390/w13030302”

9-      Its advisable to itemize the main contribution and novelties (optional)

10-  Please clarify the main limitations and scope of this study

Author Response

Point 1: There exist some grammatical errors, typos and misleading sentences. Please double-check the text and address it. For example, “In (2), it was assumed“. It means “In Eq. (2)” or “In Fig (2)”? which one? What is “location?where”?

“In Table 1, Intercept is the intercept value,  is the standard deviation of the Gaussian random fields”

Response 1: Thanks for your valuable advice. We have double-checked the manuscript. Some errors, e.g. the grammatical, typos and misleading sentences in the manuscript were modified and addressed. I will update the modified manuscript.

 

Point 2: The introduction part can be improved. I recommend adding explanation about random field and its application in both electronics and other engineering fields.

Response 2: Thanks for your valuable advice. We have improve the the introduction part and add the references to this part.

Point 3: In section 2.2.2, please add the corresponding reference.

“Lindgre et al.[?] also proposed that the solution of the SPDE can be approximated”

Response 3: Thanks for your valuable advice. We have add the corresponding reference in the modified manuscript. The the corresponding reference could improve the readability of articles, and the readers could understand the clear thinking/consideration of why the proposed approach can reach more convincing results.

Point 4: Some acronyms are not introduced for the first time. As an example, GRID and CCD are not defined.

Response 1: Thanks for your valuable advice. In the modified manuscript, we add the definition of some acronyms, e.g. GRID, CCD and LGCP etc.

Point 5: After introducing abbreviations, its recommended to use them! So, please follow a unified approach in this regard. One may note that some full forms are repeated after introducing the abbreviations.

Response 5: Thanks for your precious helpful suggetion. In the modified manuscript, firstly I introduce the abbreviations. Then the following paper writing is based on the unified abbreviations. Finally some full forms which are repeated are deleted. These modifications have been updated in the new manuscript.

 

Point 6: I recommend adding a flowchart to describe the whole methodology, to attract more readers from early career researchers and/or practitioners.

Response 6: Thanks for your precious helpful suggetion. The following figure is the flowchart of this paper.

 

Point 7: Please add the numerical platform for the findings/simulations?

Response 7: Thanks for your precious helpful suggetion. The simulations of the proposed method are implemented in python. All the experiments are carried out on a computer with a Intel(R) Core(TM) i5 processor running at 2.80GHz. We have add the introduction of platform into the updated manuscript.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments for author File: Comments.pdf

Author Response

Point 1: If I have understood the novelty of Multi-Barriers Model, the main advantage with respect to the previous specifications is the ability of this model to deal with spatial heterogeneity. Nevertheless, this term (spatial heterogeneity) is not mentioned in the paper. Furthermore, the natural comparison of the performance of Multi-Barriers Model should include models with spatial trends and, precisely, these type of models are not included neither in simulation nor in the empirical case.

Response 1: Thanks for your helpful suggetion. In the experiment there are not simulation results on spatial heterogeneity by these type of models. There are two reasons.

1) In fact, we did not have spatial heterogeneity data, especially in the criminal data.

2) The empirical results showed that Multi-Barriers Model is with good performance in the normal dataset. The Multi-Barriers Model must be with better performance in the spatial heterogeneity

Point 2: In the empirical case, the tables 1 and 2 show that Multi-Barrier Models reach a DIC greater than the other specifications, what is the advantage then of using Multi-Barrier Models in this cases?

Response 2: Thanks for your question. In the empirical results of table 1 and table 2, the value of DIC among the Nonstationary Gaussian Model is similar. There are two reasons as follow:

1) There is no data information in the obstacle area and the correlation of data between the obstacle area and the general research area is very low.

2) The Multi-Barrier Models could demonstrate the different types of obstacles by the modification of the thresholds of the left and right obstacles.

Point 3: There is no reference to the software used for the simulation and empirical case.

Response 3: Thanks for your helpful suggetion. The simulation and empirical case of the proposed method is implemented in python. All the experiments are carried out on a computer with a Intel(R) Core(TM) i5 processor running at 2.80GHz. We have add the introduction of platform into the section 5 of the updated manuscript.

Point 4: There are some typos in the text that should be corrected (e.g. lines 84, 90 among others)

Response 4: Thanks for your valuable advice. We have double-checked the manuscript. Some errors, e.g. the grammatical, typos and misleading sentences in the manuscript were modified and addressed. I will update the modified manuscript.

Author Response File: Author Response.docx

Reviewer 3 Report

This study proposed an extension of the Barrier Model, i.e., the Multi-Barriers Model, which could characterize an area of interest with different types of obstacles. Real data sets of burglaries in a certain area are used to compare the performance of the Stationary Gaussian Model, Barrier Model, and Multi-Barriers Model. The topic is very interesting; however, the manuscript needs revision. My specific comments are as under:

 

 

1.       Avoid lump-sum references in the introduction section.

2.       For Section 1, the authors should provide comments of the cited papers after introducing each relevant work. What readers require is, by convinced literature review, to understand the clear thinking/consideration of why the proposed approach can reach more convincing results. This is the very contribution from the authors.

3.       Please highlight the novelty of the current work point-wise. How is your work different from other published works? What contribution does it bring to the scholarly world?

4.       It is important to discuss the model estimation procedure.

5.       Based on the results in Tables 1 and 2, DIC is almost similar for all the models; hence, the proposal performance is the same compared to others.

 

6.       The conclusion should be shortened and precise.

Author Response

Point 1: Avoid lump-sum references in the introduction section.

Response 1: Thanks for your helpful suggetion. In the updated version, we have renewed references to avoid the lump-sum references in the introduction section.

Point 2: For Section 1, the authors should provide comments of the cited papers after introducing each relevant work. What readers require is, by convinced literature review, to understand the clear thinking/consideration of why the proposed approach can reach more convincing results. This is the very contribution from the authors.

Response 2: Thanks for your helpful suggetion. In Section 1 and Section 2 we have provided the comments of the cited papers. The development of the relevant work will be convenient for the readers to understand the clear thinking/consideration.

Point 3: Please highlight the novelty of the current work point-wise. How is your work different from other published works? What contribution does it bring to the scholarly world?

Response 3: Thanks for your helpful suggetion. The contributions of the this work are as follows:

• The Multi-Barriers Model is significantly closer to the real situations. Therefore, the model should be established according to the actual situation of the area of interest.
• The proposed Multi-Barriers Model is applicable for many different types of obstacles in the research area and such obstacles exert diverse obstructing effects on the correlation.

The contributions of the this paper was updated in the section 7.

Point 4: It is important to discuss the model estimation procedure.

Response 4: Thanks for your helpful suggetion. The model estimation procedure has been described in the section 5 and 6, which includes the following steps.

• Data: It includes the data simulation and the actual data construction.
• Parameter analysis: The parameters and DIC values of these models are calculated.
• Performance Analysis: In the proposed model we calculation the result values with different parameter values and make the conclution of different models.

Point 5: Based on the results in Tables 1 and 2, DIC is almost similar for all the models; hence, the proposal performance is the same compared to others.

Response 5: Thanks for your question. In the empirical results of table 1 and table 2, the value of DIC among the Nonstationary Gaussian Model is similar. There are two reasons as follow:

1) There is no data information in the obstacle area and the correlation of data between the obstacle area and the general research area is very low.

2) The Multi-Barrier Models could demonstrate the different types of obstacles by the modification of the thresholds of the left and right obstacles.

Point 6: The conclusion should be shortened and precise.

Response 6: Thanks for your valuable advice. In the updated manuscript, the conclusion has be shortened and precise.

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

In this round, I have reviewed the revised version of the manuscript entitled “Design and Analysis of An Effective Multi-Barriers Model Based on Non-Stationary Gaussian Random Fields” submitted to the electronics journal. The authors addressed my previous comments to some extent, while there is room for improvement. Please address the following minor issues:

1-      In figure 6, the axis label is written in the other languages. Please edit them.

2-      Please edit figure 2. It seems this figure is missing.

3-      Adding references from MDPI journals is appreciated. In accordance with comment #2, you can add some works in different fields which they applied the random field. For example, I recommend adding this paper from an MDPI journal which used random field theory with an engineering example: “https://doi.org/10.3390/w13030302”

 

Author Response

Point 1: In figure 6, the axis label is written in the other languages. Please edit them.

Response 1: Thanks for your valuable advice. The languages of the axis label have be modified.

Point 2: Please edit figure 2. It seems this figure is missing.

Response 2: Thanks for your careful inspection. In the modified manuscript the figure 2 is well edited.

Point 3: Adding references from MDPI journals is appreciated. In accordance with comment #2, you can add some works in different fields which they applied the random field. For example, I recommend adding this paper from an MDPI journal which used random field theory with an engineering example: “https://doi.org/10.3390/w13030302”

Response 3: Thanks for your valuable advice. We have add the corresponding reference in the modified manuscript.

Author Response File: Author Response.pdf