A Survey on Data Mining for Data-Driven Industrial Assets Maintenance
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper is well-organized except the following concerns.
1) It is not well explained how robustness and running efficiency are affected by various factors.
2) It is not clear how robustness and efficiency changes across different data mining techniques and applications.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper provides a comprehensive review of data mining techniques for data-driven industrial assets maintenance. Some concerns and suggestions are as follows:
1. It would be better for the authors to improve the English presentation of this work. For example, some words are not necessarily capitalized. 'Data Mining' in the title, 'Michine Learning' and 'Deep Learning' in the abstract.
2. The 'Introduction' should be re-organized. Currently, readers may not easily get the logic behind the contents. It seems that too many graphs are simply stacked.
3. It is good to see that Section 2, Section 3 and Section 4 are well-organized. However, what are the core contributions of this work? Personally, it would be better for the authors to summarize 2-3 main contributions at the end of Introduction, just in front of the last paragraph of Section 1.
4. Though a survey paper, it is too long and redundant. On the one hand, a lengthy paper means that it contains a plenty of relevant content; on the other hand, it may lack of summary. Therefore, it would be better for the authors to condense the paper.
5. The title focuses on 'data-driven industrial assets maintenance'. However, asset maintenance was less mentioned in the last several sections. Please revise it.
6. In Figure 14, the concept of 'data mining' was divided into 'machine learning' and 'data mining based'. In my own opinion, it is not suitable for such a division.
Overall, I can not recommend it for publication in its current form, and a round of revision is needed.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsOverview:
This submission endeavors to survey data mining approaches employed for maintenance planning in production systems. Through this review, the authors present aspects of the relevant research, including themes, topics, and algorithms.
After reading the paper, I find it difficult to recommend it for publication in Technologies in its current form. I encourage the authors to improve the quality of the paper according to the comments below.
Comments:
-
A major concern is the length of the paper. In its current form, it reviews the relevant research and describes its aspects in detail, involving overlong sections, such as Abstract and Section 5. This considerably decreases the readability of the paper. The authors should consider shortening the paper.
-
In the Introduction, the statement “To the best of the authors’ knowledge, no existing survey comprehensively addresses all the topics covered in this research study” does not sufficiently support the novelty and contribution of the paper. In this regard, it addresses a well-studied field, which has already been reviewed by recent publications. Thus, the authors should emphasize the novelty and contribution of their work, illustrating the gaps and limitations of previous work.
-
In the same section, it is surprising that the authors failed to formulate research questions that define and convey the scope and the aim of the present paper.
-
In Section 2, the authors state that they compiled a list of 540 papers published within the years 1995-2023. Did the authors employ a systematic review approach (e.g., PRISMA) for selecting relevant papers? If not, it is sensible to assume that the conducted search yielded articles that are not even related to the investigated subject. Furthermore, their literature search was limited to research articles. Why did the authors exclude other document types, e.g., book chapters? What exclusion/inclusion criteria did they utilize? In my opinion, such papers are relevant to the addressed problem context.
-
In the review, the authors seemingly included preprint papers published in arxiv. This raises some concerns regarding the quality of the conducted review since these papers have not undergone formal peer review.
-
The statement “Articles from 2024 are not considered due to the specific topics defined for the search criteria” is confusing and questionable. That is, it is not clear why the authors imposed such a limitation since there are several relevant papers published in 2024. They should elaborate on that matter.
-
There are figures, including Figure 14, in the paper that are confusing and cover one page. They are overstuffed with redundant details, making it difficult to comprehend their meaning. It is advisable to refine or replace them with concise ones.
-
It is advisable to check again the cross-references throughout the paper since some of them are redundant. For example, on page 55, please cross-reference either the subsection or the figure in “As presented in Subsection 5.1, Figure 6”, not both. Furthermore, the authors should check the accuracy of their statements as well. For example, on page 12, please revise “Objective optimization, also known as optimization in Machine Learning” since it inaccurately states that objective optimization is only applied with machine learning methods.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have addressed all my concerns, and I can recommend the paper for publication.
Reviewer 3 Report
Comments and Suggestions for AuthorsThank you for addressing and responding adequately to my previous remarks and suggestions.