Next Article in Journal
Reducing Computational Time in Pixel-Based Path Planning for GMA-DED by Using Multi-Armed Bandit Reinforcement Learning Algorithm
Previous Article in Journal
Adaptive Aberration Correction for Laser Processes Improvement
 
 
Article
Peer-Review Record

Manufacturing Process Optimization Using Open Data and Different Analysis Methods

J. Manuf. Mater. Process. 2025, 9(4), 106; https://doi.org/10.3390/jmmp9040106
by Md Tahiduzzaman 1, Angkush Kumar Ghosh 2,* and Sharifu Ura 2,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
J. Manuf. Mater. Process. 2025, 9(4), 106; https://doi.org/10.3390/jmmp9040106
Submission received: 19 February 2025 / Revised: 21 March 2025 / Accepted: 24 March 2025 / Published: 25 March 2025

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I thoroughly enjoyed reading this article up to Section 3. However, in that section, it is unclear how ANOVA or PD methods are applied to process optimization. Typically, as discussed in the review section, optimization involves system models (e.g., regressions) followed by metaheuristic or other techniques to explore new parameter combinations, such as ant colony optimization, bee algorithms, or genetic algorithm optimizers.

1) I was expecting Section 3 to cover a different approach, more in line with what was presented in the literature review. Instead of detailing the mathematics behind well-known statistical methods, I would suggest that the authors enhance the discussion, as they did for the Taguchi method.

2) Additionally, if the open data was generated using a JSON file, please share it with the community via a GitHub repository and reference it in the paper. This is essential for ensuring the article meets publication requirements.

3) The discussion in Section 5 appears trivial because the methods used are quite basic. It is well understood that reduced contact time with the workpiece leads to increased tool life, though it may degrade surface quality. At this point, surface quality should be considered for a more comprehensive and nuanced optimization approach. It would be beneficial to remove some of the simpler methods from Section 3 and instead present at least a linear regression model combined with a metaheuristic approach. This would allow for a more meaningful exploration of how different parameter combinations could improve the process.

4) Issues like the one discussed in 770-790 may be related to the number of samples and repetitions. Without sufficient repetition and data, such problems can arise. Please specify how many data points are included in the selected dataset before proceeding with Point 3 of this review-report, as there may not be enough to conduct this study reliably.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

 

Dear Authors,

Please see the comments:

 

Abstract

Please give the results numerically. "The discoveries suggest" in line 20 is too general a wording.

 

Introduction

Please explain in detail why only the cuttig process was optimized.

Figure 1: The Machining Phenomena part of the figure is not clearly visible, please correct it.

It would be worthwhile to justify why these methods were chosen. (lines 78, 79). Please delete 82-85. lines, this is not necessary.

 

Methods

Please create a figure about Research Methodology plan to visualise it

 

Results

Cells in tables are too large. These are not pretty! Please correct the tables.

 

Appendix A, B:

This chapter is valuable but not complete. It is not an important part of the manuscript. It can be left out.

 

Summary

The manuscript gives the impression that the authors are not convinced that their results can really be used.  Very good research concept, but since the method was only applied to the cutting process, the results are questionable. The user is not convinced to perform his own optimization as shown.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The research on open data processing and application holds significant theoretical and practical importance. The following are the identified shortcomings and recommendations for improvement:

1. While the single objective optimization problem focused in this paper has been solved by mature algorithms, the paper fails to adequately highlight the value of open data and the advanced nature of the proposed method. It is recommended to conduct a more in-depth analysis and exploration of the value of open data by showing the complexity of the problem.

2. The literature review lacks clarity. To enhance readability, it is suggested that a table summarizing the research contents and methods of other scholars cited in the literature review be included.

3. In the study of WM1-TM2, the ANOVA results indicate that feed speed f has no significant effect on tool wear Tw, whereas the SNR results suggest optimal performance at f = 0.15mm. Although the author attempts to explain the discrepancy between these two sets of results using the probability distribution (PD) method, this approach does not fully resolve the issue. It is recommended to integrate variance analysis, signal-to-noise ratio (SNR), and PD into a unified framework to provide a more comprehensive and clear evaluation and analysis of the results.

4. The combination of PD with traditional statistical methods is a notable highlight of the paper; however, its advantages over single methods, such as handling uncertainty, are not clearly defined. It is recommended to quantitatively compare the performance of different methods on open data, including metrics such as calculation time and error rate.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thanks for addressing all my comments. I have a few minor issues that still need to be addressed:

  1. Please add a Limitations section discussing the challenges related to surface quality in the optimisation process and the issue of missing data. Additionally, you can further elaborate on the responses provided in the reply file to ensure a more comprehensive discussion in the paper.

  2. Furthermore, please include a Future Work section outlining potential alternative methods that could be explored, such as more complex Artificial Neural Networks (ANNs) for modelling and Genetic Algorithms (GA) or Particle Swarm Optimisation (PSO) for optimisation. You may refer to paper [1] to gain a clearer understanding of what I mean. It would also be valuable to discuss future perspectives on open datasets in this context.

  3. Please double check the provided data link

[1] https://doi.org/10.1007/s40684-025-00705-4

Author Response

Please refer to the attached file.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Thank You for the correction.

Author Response

We gratefully acknowledge the approval of the reviewer.

Reviewer 3 Report

Comments and Suggestions for Authors

In response to the comments, the author provided detailed explanations and made targeted supplements and modifications, thereby enhancing the quality of the paper. However, several issues remain: First, regarding the literature review, the newly added table (A1, A2) lacks sufficient generality, and the analysis of literature on OD is not sufficiently in-depth. Consequently, the distinctions between this study and existing literature, as well as the innovative contributions of this paper, are not clearly articulated. Second, Figures 15, 16, and 17 lack clarity, which impairs the readability of the document.

Author Response

Please refer to the attached file.

Author Response File: Author Response.pdf

Back to TopTop