Review Reports - Robotic Welding Technologies for Intersecting and Irregular Pipes and Pipe Joints Toward Automated Production Line Integration: A Review

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Robotic Welding Technologies for Intersecting and Irregular 2 Pipes and Pipe Joints toward Automated Production Line Inte- 3 gration: A Review

Review Comments

The manuscript presents a systematic literature review on robotic welding for complex pipe geometries, employing a four-step methodology that integrates AI-based screening. The topic is timely and addresses a significant challenge in industrial automation—specifically the handling of intersecting and irregular pipe joints.

Nevertheless, the current presentation suffers from a lack of critical technical analysis. The manuscript focuses predominantly on listing existing studies (what has been done), whereas the discussion on comparing the efficacy, precision, and limitations of different sensing and path-planning algorithms remains insufficient. The contribution would benefit significantly from a deeper comparative analysis rather than a simple enumeration of references.

In addition, several "careless" errors, including logical contradictions in counting categories, incomplete references, and citation of outdated "under review" manuscripts, were observed. These issues significantly undermine the rigorousness expected of an academic review.

Major Revision is recommended.

Logical Contradiction in Categorization

Logical Contradiction in Categorization (Page 5, Line 173): The manuscript states: "The authors of this review proposed the following five categorization of these 84 papers...". However, immediately following this statement and in Table 1, only four categories are listed: (1) General pipe welding, (2) Intersecting pipes, (3) Boilers..., and (4) Control and modelling.

This is a glaring logical error. The authors must verify whether a category was accidentally omitted or if the count "five" is a typo. Such a mathematical discrepancy in the core methodology section is highly unacceptable.

Incomplete and Inappropriate References

Ref [43] (Page 12, Line 436): The citation is listed as "Wu, H.W... (No further journal info provided)". Submitting a manuscript with incomplete placeholder text in the reference list is a sign of negligence.

Ref [22] and [92]: These references are cited as "In Review 2024". Given that the current context implies a timeline beyond 2024 (as indicated by the dates in the header "2026"), citing manuscripts that were "in review" two years ago is inappropriate. If these papers were rejected or never published, they should be removed. If they were published, the citations must be updated.

Recommendation: Clarify the fusion mechanism with a detailed diagram and explicitly state the dataset size (training/validation/test splits).

Insufficient Comparative Analysis

Although the manuscript categorizes various technologies (e.g., visual servoing vs. laser scanning), no quantitative comparison with these baselines is provided.

To substantiate the practical value of this review, the authors should include a comparative table highlighting the pros/cons, typical accuracy ranges, and processing speeds of different sensing methods discussed in Section 3.2 and 3.4. Currently, the text is too descriptive and not sufficiently analytical.

Critical Errors and Redundancies in Tables and Figures

Table 2 contains repeated blocks of content, indicating an apparent copy–paste or editing error. Such redundancy is highly problematic in a review article, as it undermines confidence in the accuracy of the classification and the editorial rigor of the manuscript.

In addition, some figures lack clear definitions of their statistical scope. For example, the publication trend analysis does not clearly specify whether it is based on the final 84 articles or a larger dataset, nor does it clearly justify the selected time intervals.

All tables and figures should be carefully checked, corrected, and accompanied by unambiguous captions and definitions.

Limitations and Future Work

The current conclusion provides an outlook on future work but fails to explicitly address the limitations of the review itself.

It is recommended to add a discussion on the potential bias introduced by the specific keywords chosen or the AI screening tool. Additionally, the "Future Research" section should explicitly discuss the integration of Digital Twins and VR/AR in welding training and monitoring, which are critical trends in Industry 4.0 not fully explored in the text.

Minor Comments

1.Several grammatical and stylistic issues are present throughout the manuscript (e.g., incorrect plural forms, awkward phrasing, and inconsistent terminology such as “modeling” vs. “modelling”). A thorough language revision by a fluent English speaker is recommended.

2.Minor typographical and formatting issues (line breaks, hyphenation artifacts, and spacing inconsistencies) should be corrected in the final version.

3.Terminology should be standardized across sections to avoid confusion, particularly in the naming of categories and subcategories.

Overall Assessment

The manuscript addresses a relevant and valuable topic and demonstrates a considerable literature survey effort. However, due to insufficient methodological rigor, limited critical analysis, and notable editorial errors, the manuscript does not yet meet the standards expected for publication in Applied Sciences.

If the authors comprehensively address the issues outlined above—particularly those related to review reproducibility, system-level integration analysis, and table/reference accuracy—the manuscript could become a meaningful and publishable contribution.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

2.Minor typographical and formatting issues (line breaks, hyphenation artifacts, and spacing inconsistencies) should be corrected in the final version.

3.Terminology should be standardized across sections to avoid confusion, particularly in the naming of categories and subcategories.

Author Response

Please refer to the attached PDF file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This article conducts a literature review on robot pipe welding, employing a four-step approach: data collection, data cleaning, artificial intelligence-based screening, and manual screening. The analysis indicates that digital, automated, and adaptive robot solutions are showing a clear trend, and the influence of artificial intelligence and multi-sensor data collection is increasing. This review has certain industrial guidance value, but there are also some deficiencies that need to be improved.
1. From the system, it can be seen that the text duplication rate of the manuscript has reached 44%, far exceeding the acceptable standard of the journal. Therefore, the author is requested to reduce this duplication rate index and suggest it to be below 25%.
2. The introduction of the research background in this article is too extensive, accounting for almost half of the text. It needs to be carefully streamlined; moreover, the results of the research have not been mentioned at all, and there is not even a quantifiable data. Therefore, this abstract needs to be rewritten.
3. There are many reviews on welding robots. Please elaborate on where this article differs from the existing literature, what gaps it fills, and what innovative work it has done.
4. In the last paragraph of the introduction, the author should indicate the structure, content of the review, and propose the novelty of this article. Please add it.
5. In Figure 1, the text in the box is not clear. Additionally, what are the meanings of the left and right sides of this figure? Please provide a detailed explanation and rewrite it.
6. Figures 3 and Tables 1-5 only show the number of documents, but do not indicate the advantages and disadvantages of these documents, especially important application scenarios. It is recommended to make improvements.
7. For review papers, some actual welding images and related data drawings should be provided as support. However, the article does not have any of these. It is hoped to make revisions to improve it.
8. With the development of society, digital, automated, and adaptive robot solutions are definitely the future development trend. It is hoped that the author can extract some new content from the review.
9. In the conclusion section, it is recommended that the author condense and simplify the content based on specific results and point them out, to improve readability.

Author Response

Please refer to the attached PDF file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Robotic welding, and robotic production cells in general, are a significant practical and scientific issue. The paper therefore aligns with current trends in industry and science. The paper is interesting, but it has some areas that require clarification:

The number of publications referred to by the authors is acceptable, but the description seems to be a bit limited for a review paper. In fact, 9 pages seems short for a review paper. Therefore, I encourage the authors to expand on the issues discussed. The next 6 pages are a bibliography.
Lines 43-47 - The authors outlined the health risks associated with welding. While these are real, this paragraph should be expanded upon to describe the solutions used in practice to mitigate the negative health impacts of welding products. Robotics isn't the only solution.
Lines 64-67 - These are very obvious claims about the use of sensors, machine vision, and AI in robotic cells. This statement alone adds absolutely nothing to the reader's understanding. Please elaborate. What are the current trends and latest developments in this field?
Line 68 - please provide specific applications in the automotive industry.
In the introduction section, it is worth adding a paragraph about Industry 4.0 and Industry 5.0 from the point of view of process robotization. Please refer to: https://doi.org/10.3390/app15062976 and https://doi.org/10.3390/su16072837 In these articles, the authors stated that, in addition to the traditional enterprise drive toward robotization of production processes, the latest trends also include efforts to increase the conditions of collaboration between robots and humans, as well as to achieve sustainable development, including energy efficiency. Among other things, the authors analyzed various robot movement parameters to minimize energy consumption.
I have the impression that the authors, in the main part of the paper, primarily provided classifications of groups of topics most frequently discussed in the literature. This is a good introduction to the paper. But, forgive me, it's just an introduction. The paper definitely needs to be expanded upon. And now, let's discuss at least the most interesting and current papers. This work is described in terms of entries, not in detail.
Line 302 - Future studies should truly explore new research. The authors suggest focusing on Industry 4.0, but Industry 5.0 has been in the works for several years. Therefore, these new research findings appear to be at least five years old.

In summary, the paper has potential and thematic value. However, it requires significant content expansion. It is too general.

Author Response

Please refer to the attached PDF file.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Reviewer’s Second-Round Comments

The revised manuscript addresses a relevant and potentially valuable topic, and I appreciate the authors’ efforts to revise the paper. In particular, the revised version appears to make some attempt to strengthen the methodological explanation and to add broader discussion on future research directions. For example, the methodology section now includes a brief statement on the limitations of the review process, including possible bias introduced by database selection, keyword choice, AI instructions, and manual screening. This is a positive step.

In addition, the revised manuscript introduces a more structured comparative element in the discussion of the literature, including dimensions such as technical focus, validation level, type of robot control, and integration level. This partially responds to my previous concern that the first version was overly descriptive and insufficiently analytical.

However, despite these efforts, I do not believe that the manuscript is yet suitable for publication in its current form. Several of the major concerns raised in the first review remain unresolved, and the revised version contains numerous new editorial and formatting problems that substantially undermine the scientific rigor and readability of the paper. Therefore, I must recommend Major Revision again.

Major Concerns

The manuscript still does not fully resolve the categorization inconsistency.
In the first-round review, I explicitly pointed out the logical contradiction regarding the number of categories. In the revised manuscript, this issue does not appear to have been cleanly corrected. More importantly, the revision seems to contain overlapping old and new text, which makes it difficult to determine the final intended categorization structure with confidence. The methodology and classification narrative remain textually inconsistent.
The comparative analysis is improved, but still not sufficiently technical.
Although the authors have added comparative dimensions and some tabulated synthesis, the revision still falls short of the deeper technical comparison requested in the first review. The manuscript would benefit from a dedicated comparison of sensing and path-planning methods in terms of practical strengths and weaknesses, typical accuracy, computational burden or processing speed, robustness to geometry variation, and industrial applicability. At present, the new comparison remains closer to a classification framework than a true technical benchmark or critical evaluation.
Serious editorial contamination is present throughout the revised PDF.
This is currently the most critical problem. The revised manuscript contains visible editing artifacts such as repeated blocks of text, formatting marks, duplicated methodological passages, and remnants of prior draft content. For example, in the methodology section, old and revised versions of the same description are merged together rather than properly replaced. The PDF also displays repeated “Formatted:” markers, which should never appear in a clean revised manuscript. These problems make the manuscript appear not as a finalized revision, but as an improperly exported tracked-changes or internally corrupted version.
Reference formatting remains unacceptable and, in fact, appears more problematic than before.
One of the previous concerns was incomplete or inappropriate references. In the current revision, the reference list appears to suffer from numbering corruption and formatting overlap, with entries such as “3.4.”, “4.5.”, “5.6.” and similar numbering anomalies. This raises serious concerns about the reliability of the citation system in the manuscript. Even if some individual references may have been updated, the present reference section is not in a publishable state.
The revised manuscript still lacks adequate editorial control over tables and figure-related presentation.
My earlier review noted redundant content in tables and insufficiently defined statistical scope in figures. The revised manuscript still does not convincingly resolve these issues. Because the document contains duplicated text and formatting corruption, it is difficult to verify whether all tables and figures have been systematically corrected. The paper requires a complete audit of every table, figure, caption, cross-reference, and in-text callout before it can be evaluated fairly on scientific merit.
Add some recent literature into the introduction and also in the results and discussion section to strengthen your discussion. You can add many more but few are given for your guidance:

AHP-based welding position decision and optimization for angular distortion and weld collapse control in T-joint multipass GMAW, https://doi.org/10.1016/j.jmapro.2024.05.014

Autonomous Decision Making of Welding Position during Multipass GMAW with T-joints: A Bayesian Network Approach, 10.1109/TIE.2021.3076710

Additional Comments

Language quality and consistency still require substantial improvement.
Beyond the formatting problems, the manuscript still contains awkward phrasing, inconsistent terminology, and grammatical issues. These were already noted in the first round and remain unresolved at a level expected for an international review article.
The current version should be thoroughly recompiled as a clean manuscript before any further evaluation.
At this stage, the authors should not submit another response based only on localized edits. Instead, they should prepare a fully clean version of the manuscript with all tracked changes, hidden formatting artifacts, duplicated text fragments, placeholder remnants, and corrupted reference numbering removed. Only after that can the scientific revisions be assessed properly.

Recommendation to the Authors

Before resubmission, I strongly recommend that the authors do the following:

Prepare a clean, fully regenerated manuscript file with no track-changes residue, formatting marks, or duplicated paragraphs.

Recheck the entire categorization framework and ensure that the number of categories is logically and consistently stated in the text and tables.

Add a true technical comparative table for the main sensing and path-planning methods, including at least typical advantages, limitations, accuracy level, computational demand, and application scenario.

Rebuild the reference list and in-text citations from scratch if necessary, to eliminate numbering corruption and ensure that all references are complete, current, and appropriate.

Carefully verify all tables, figures, captions, and cross-references for redundancy, numbering errors, and scope definition.

Perform a comprehensive English language revision by a fluent academic writer or professional editing service.

Final Recommendation

The topic remains worthwhile, and the manuscript has the potential to become a useful review. However, the present revised version still does not meet the editorial and scholarly standards required for publication. I therefore recommend Major Revision.

Comments for author File: Comments.pdf

Author Response

Please refer to the attached PDF file.

As requested, the manuscript is attached in a clean version, without track changes.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

After the author's careful revision, the quality of the manuscript has significantly improved and it can be considered for publication.

Author Response

Please refer to the attached PDF file.

Author Response File: Author Response.pdf