Safety Evaluation Method for Submarine Pipelines Based on a Radial Basis Neural Network
Round 1
Reviewer 1 Report
This paper proposes an evaluation index system for the risk factors of submarine pipeline defects using the Analytic Hierarchy Process (AHP); the idea is very interesting, but it requires some improvements, which are listed below:
- Related work and existing literature is merged into introduction section. It is suitable to have a separate "related work" section describing existing studies, focus area and problems identified from existing literature.
- Please give more references about this topic and improve the importance of this paper with comparing existing literature.
- Validation it is not clear; discuss the novelty of the work in comparison with:
1- "Pipeline defects risk assessment using machine learning and analytical hierarchy process." 2018 International Conference on Applied Smart Systems (ICASS). IEEE, 2018.
2- "A study on oil pipeline risk assessment technique based on fuzzy analytic hierarchy process." The Open Petroleum Engineering Journal 7.1 (2014).
3- Pipeline Risk Assessment Using Analytic Hierarchy Process (AHP). In International Conference on Offshore Mechanics and Arctic Engineering (Vol. 44366, pp. 1-11).
- Kindly improve the conclusion. It should address the key contributions and findings of the paper. It should summarize what the paper speaks about.
- Please give more future work directions at the end of the conclusion
The quality of writing of the paper must be improved, Typos must be fixed.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Reviewer Recommendation and Comments for manuscript sustainability-2415627-peer-review-v1.pdf with the title: “Safety Evaluation Method of Submarine Pipeline Based on Radial Basis Neural Network”, with follow authors Weidong Sun, Jialu Zhang, Yasir Mukhtar, Lili Zuo and Shaohua Dong.
This manuscript study Safety Evaluation Method of Submarine Pipeline using the Radial Basis Neural Network method.
The text is clearly written. The structure, content, and concept of the research work as well as the achievements, correspond to the new unpublished manuscript but some significant improvement are necessary. The manuscript is clearly written in order to understand scientific research but English language must be improved.
The abstract is clear and in the proper way present the scope of research, methods and achievement. Tables and Figures are properly presented but some improvement are mandatory. A lot of type mistake should be corrected. It is necessary to correct a significant number of typographical errors (indications, commas, dots,).
The introduction is short and there is no similar research in this area. It is necessary to cite similar studies that treat defect-related risk, material defects, work deviations, safety factors, or other.
Review of existing standard in this domain are relevant. In the paragraph 3, research of the safety factor are missing relevant data base of pipeline defects. It should be new paragraph.
The conclusion is correct and present main achievements of this manuscript. Future research directions are missing.
Some additional comments and necessary corrections that I find useful for improving the quality of the article are presented below:
Finding 1: (Line 24 and 28). It is not unique methods of literature resource (“.[1,2]” or “[3].”).
Finding 2: (Line 36). Literature resource is missing.
Finding 3: (Line 47). Incorrect text “results. keep. In”.
Finding 4: (Line 142). Incorrect text “[15],”. It should be “.”
Finding 5: (Line 172). Incorrect text “p 0 ”. It should be “p0 .”
Finding 6: (Line 201). It is necessary to prepare figure on proper way (reduce font). Add literature resource.
Finding 7: Line 206. Could you explain and add relevant literature: “After reviewing relevant literature and analyzing the causes of pipeline accidents both domestically and internationally”.
Finding 8: (Line 211). It is necessary to prepare figure on proper way (reduce font).
Finding 9: Figure 2, Table 3, some defects are not listed in the same way. It is necessary to unify the terms of defects or give numerical marks (Defect 1, …Defect 9) “Seawater Atmospheric corrosion or Atmospheric, or Seawater corrosion”.
Finding 10: (Line 224). It is necessary to clearly explain Risk factor importance specify in Table 3. Explain value 1 and other relevance relationships.
Finding 11: (Line 243-246). Formula 7.24 do not existing. Text mistake “y index, which is related 244 to the order n, and the corresponding value of RI can be found in Error! Reference source 245 not found..”
Finding 12: (Line 323). Paragraph 1.1. is not existing in manuscript.
Finding 13: (Line 375). Could you check do you have 9 or 12 independent variable???
Finding 14: (Figure 6 and Figure 7). It is necessary to unify all text and figures.
Due to this manuscript have a lot of type mistake, some details need explanation, some paragraph can be add or revised, database of pipeline defect are missing or some technical correction are necessary, I suggest major revision of this manuscript before publishing.
English language must be improved.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
1. As a research paper, there are many issues of improper writing in this article. Here are some suggestions:
(1) Abbreviations are not used consistently, such as Analytic Hierarchy Process (AHP) only need to be labeled the first time it appears, and the abbreviation can be used to describe it subsequently.
(2) The article extensively describes standards, basic concepts, sample information, and other research-related content. Although necessary, as a journal article, the focus of writing should be on method design, phenomenon description, and result analysis.
(3) Flowcharts and result diagrams in the article should be output using professional software (such as Microsoft Visio, Origin, etc.). Exporting as emf format is recommended. Figure 4 should be drawn after exporting the original data, avoiding screenshots directly from MATLAB.
(4) There is a high degree of overlap between the contents of Figure 3 and Figure 5.
(5) Figure 7 should be labeled with the titles of the horizontal and vertical axes.
(6) For content with the same meaning, the same vocabulary should be used, such as using "Radial Basis Neural Network (radial basis function neural network）" or directly using "RBF" to describe it.
(7) Lines 338-346 are recommended to be moved to Chapter 1.
(8) Line 375, x12?
2 The descriptions in Tables 2 and 3 are not clear enough, and the meaning of the fractions in Table 3 is easily confused. It is suggested to provide specific examples for clarification.
3. The article only provides a brief description of experimental phenomena, lacking in-depth analysis.
Minor editing of English language required.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 4 Report
You did not mention at all ANN in Introduction.
Please try to be more focused and direct. I think that Figure 3 present general explanation of this type of ANN, but it is not your ANN. The ANN from Figure 1 has two outputs, Y1 and Y2, while your findings suggest only one output as showed in Figure 5.
I am not sure how did you determined values of Risk Factor Importance in Table 3. Please explain.
Sections 2.1, 2.2, 2.3 describe technical standards, but Section 2.4 Finite element simulation evaluation method does not refer to any standard. Please explain.
Relevant paper about technical standards is: Brkić, D.; Praks, P. Proper Use of Technical Standards in Offshore Petroleum Industry. J. Mar. Sci. Eng. 2020, 8, 555. https://doi.org/10.3390/jmse8080555
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 5 Report
Dear Authors Thank you for submitting your work to Sustainability an MDPI journal. Editorial Comments: 1 E-mail of the corresponding author must be clear and not a number sequence. 2. The equation symbols must be explained properly. 3. All the acronyms must be analyzed the first time appeared in the document. 4. Figures 3 and 5 can be omitted. 5. The quality of the figures 1 and 2 must be improved. If they have already been published elsewhere then they must be added as a reference.
Technical comments. The figures 4, 6 and 7 need adequately explanations. Explain in detail which are the new outcomes of this study which have not been identified form other studies in the field. The pipelines are not only consist from metallic materials as declared in introduction. Composite materials are used also. This study must be applicable to this materials also.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
paper can be accepted for publication
quality of writing can be improved as well.
Author Response
Dear Reviewer 1,
We would like to express our sincere gratitude to you for your thoughtful comments and helpful suggestions.
Yours sincerely,
Shaohua Dong and all authors.
Reviewer 2 Report
Dear colleagues
Congratulations.
Author Response
Dear Reviewer 2,
We would like to express our sincere gratitude to you for your thoughtful comments and helpful suggestions.
Yours sincerely,
Shaohua Dong and all authors.
Reviewer 3 Report
The study proposed a method to evaluate the safety of subsea pipelines using RBF. The RBF model is developed for safety factor prediction by collecting multiple assessment indicators. However, The article does not fully demonstrate sufficient logic and coherence, and there are some problems in the article’s structure design. Therefore, I do not recommend this article for publication. The specific problems are as follows:
1. There is no detailed analysis of the structural optimization, parameter selection, and prediction results of RBF in the article. Instead, a lot of space is used to describe the composition process of input data. Although this is necessary content, such a structural arrangement makes the article appear like a product manual, which is clearly unreasonable. It is suggested that the background knowledge and data composition process, which are secondary content discussed in the article, be moved to the appendix.
2. Is there a specific selection criterion for Relative importance in Section 3.2 and for the entire process of scoring indicators? Is there an effect of data fluctuations caused by subjective choices on the prediction results?
3. From the description in Section 4.2, it is clear that the article simply constructs a regression problem and does not see the necessity of using RBFNN. A comparison with other methods is suggested.
4. Figure 3 is a screenshot of the MATLAB Neural Network Toolbox and it is not recommended to use such figures in research papers.
No
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 5 Report
Thank you for addressing the comments.
Author Response
Dear Reviewer 5,
We would like to express our sincere gratitude to you for your thoughtful comments and helpful suggestions.
Yours sincerely,
Shaohua Dong and all authors.
