Prediction of the Ultimate Tensile Strength (UTS) of Asymmetric Friction Stir Welding Using Ensemble Machine Learning Methods

Round 1

Reviewer 1 Report

Journal: Processes (ISSN 2227-9717)

Manuscript ID: processes-2171284

 

The authors presented an article on “Prediction of the Ultimate Tensile Strength (UTS) of Asymmetric Friction Stir Welding using Ensemble Machine Learning Methods”. The authors clearly demonstrate the difference from previous studies using all input parameters. This aspect supports the originality of the study. Similarity rate is 20%. I think the article is well organized and suitable for the "Processes" journal. But the article will be ready for publication after a minor revision. Comments are listed below.

 

1.      On page 15, line 402, "Table 9.1" should be corrected to "Table 10". Subsequent erroneous table numbers should also be corrected.

2.      Units are not written in most of the tables. It should be added.

3.      The discussion section is too short. It should be discussed further with current studies from the literature.

4.      The article contains numerous typographic and language errors. It should be corrected.

5.      The article should be rearranged by taking into account the journal writing rules and citation rules.

6.      The paper is well-organized, yet there is a reference problem. First, your reference list contains no article from the “Processes” journal. If your work is convenient for this journal's context, then there are many references from this journal. Secondly, cited sources should be primary ones. Namely, the indexed area shows the power of a paper and directly your paper's reliability. Please make regulations in this direction.

 

*** Authors must consider them properly before submitting the revised manuscript. A point-by-point reply is required when the revised files are submitted.

Comments for author File: Comments.docx

Author Response

Thank you very much for your suggestion. We have studied your comment carefully and perform the following improving of our article.

1. On page 15, line 402, "Table 9.1" should be corrected to "Table 10". Subsequent erroneous table numbers should also be corrected.

Answer : Table 9.1 was change to Table 10 as shown on page 16, line 405.

2. Units are not written in most of the tables. It should be added.

Answer : The authors have added the units in Tables 1,3,4,10 and 13 as shown in lines 167,205,228,405 and 430 respectively.

3. The discussion section is too short. It should be discussed further with current studies from the literature

Answer : We have added the discussion from lines 454 to 483 in order to compare our study's findings with those of other methodologies found in the literature.

4. The article contains numerous typographic and language errors. It should be corrected.

Answer : The manuscript was submitted to MDPI for checking language errors.

 

 

5. The article should be rearranged by taking into account the journal writing rules and citation rules.

Answer : The manuscript has been arranged into the format following the journal writing rules. The manuscript was also submitted to MDPI for checking format.

6. The paper is well-organized, yet there is a reference problem. First, your reference list contains no article from the “Processes” journal. If your work is convenient for this journal's context, then there are many references from this journal. Secondly, cited sources should be primary ones. Namely, the indexed area shows the power of a paper and directly your paper's reliability. Please make regulations in this direction.

Answer : The following references from processes have been included to the manuscript, and the primary source of the cited sources has been verified across the entire paper.

[28] Desai, P.S.; Granja, V.; Higgs, C.F., III. Lifetime Prediction Using a Tribology-Aware, Deep Learning-Based Digital Twin of Ball Bearing-Like Tribosystems in Oil and Gas. Processes 2021, 9, 922. https://doi.org/10.3390/pr9060922

[29] Schmoeller, M.; Weiss, T.; Goetz, K.; Stadter, C.; Bernauer, C.; Zaeh, M.F. Inline Weld Depth Evaluation and Control Based on OCT Keyhole Depth Measurement and Fuzzy Control. Processes 2022, 10, 1422. https://doi.org/10.3390/pr10071422

 

[30] Shaheen, B.W.; Németh, I. Integration of Maintenance Management System Functions with Industry 4.0 Technologies and Features—A Review. Processes 2022, 10, 2173. https://doi.org/10.3390/pr10112173

 

[33] Chainarong, S.; Srichok, T.; Pitakaso, R.; Sirirak, W.; Khonjun, S.; Akararungruangku, R. Variable Neighborhood Strategy Adaptive Search for Optimal Parameters of SSM-ADC 12 Aluminum Friction Stir Welding. Processes 2021, 9, 1805. https://doi.org/10.3390/pr9101805

 

[35] Srichok, T.; Pitakaso, R.; Sethanan, K.; Sirirak, W.; Kwangmuang, P. Combined Response Surface Method and Modified Differential Evolution for Parameter Optimization of Friction Stir Welding. Processes 2020, 8, 1080. https://doi.org/10.3390/pr8091080

 

[42] Yin, L.; Du, X.; Ma, C.; Gu, H. Virtual Screening of Drug Proteins Based on the Prediction Classification Model of Imbalanced Data Mining. Processes 2022, 10, 1420. https://doi.org/10.3390/pr10071420

Author Response File: Author Response.docx

Reviewer 2 Report

The topic is interesting, but the study has some technical issues that need to be addressed. It might be accepted by editors only if required amendments can be done by authors in the final submission.

1.  The authors of the paper need to make some changes to the introduction section so that it more effectively conveys the originality of their work. It is not entirely clear what novel aspects this approach incorporates. Which of the authors' previous works did they draw inspiration from when writing this one? What's the latest? Please use references that are more relevant and recent, with an emphasis on works that are comparable.

2.  Explain What research gaps is your work intended to fill? What limitations in previous work does it address? Make it in bullet points. What contributions to knowledge does it make to fill the previous gaps?

3.    Table 1: In many cases, a checkmark (Ö) is used to denote a value; in the remaining cases, either a minus sign (-) or an x (x) must be used.

4.    Equations 2 and 3 are confusing. Double-check everything and re-add it using the proper expressions and sequence.

5.    Figure 2 and 3 captions need to revise.

6.    To ensure the validity of the data, the authors should replace Figure 3 with a new picture of the actual experimental setup in the lab.

7.    Table 6: explain how 100 random vectors are generated. How do they relate to equations 5 and 6?

8.    Line 338: The authors stated that the differential evolution algorithm (DE) can solve such a problem. Authors must include complete DE algorithm steps to demonstrate how the algorithm can be used to solve the specified problems.  

9.    The conclusion of the paper needs to be rewritten to include a summary of the most crucial points made throughout the work. In addition, authors should avoid citing previously published references in the conclusion section.

10. The future recommendations need to be rewritten in bullet point format, with an emphasis on using the most recent technologies and various machine learning methodologies.

11.  Some Latest related references are required.

 

Author Response

The topic is interesting, but the study has some technical issues that need to be addressed. It might be accepted by editors only if required amendments can be done by authors in the final submission.

 

1. The authors of the paper need to make some changes to the introduction section so that it more effectively conveys the originality of their work. It is not entirely clear what novel aspects this approach incorporates. Which of the authors' previous works did they draw inspiration from when writing this one? What's the latest? Please use references that are more relevant and recent, with an emphasis on works that are comparable.

Answer : Thank you very much for the valuable comment. The introduction has been reformulated and re-write as shown in line “ 47 to 50” and line “ 66-101”.  The added lines have been used to covey the originality of this research and also make more clear the research contribution (line 113 to 118).

2. Explain What research gaps is your work intended to fill? What limitations in previous work does it address? Make it in bullet points. What contributions to knowledge does it make to fill the previous gaps?

Answer : As demonstrated in Line 66-101, the research gap is described in three bullet points. Each of the prior literatures has been reviewed, along with the identification of research gaps and the reason for further study.

3. Table 1: In many cases, a checkmark (Ö) is used to denote a value; in the remaining cases, either a minus sign (-) or an x (x) must be used.

Answer : Table 1 has been modified by replacing the black cell with a minus sign (-) in case of there is no value reported.

4. Equations 2 and 3 are confusing. Double-check everything and re-add it using the proper expressions and sequence.

Answer : The section containing equation 2 and 3 was rewritten by replacing the previous equations with the simplified forms. Equation explanations are rewritten accordingly as shown in line 253 to line 266

5. Figure 2 and 3 captions need to revise.

Answer : The captions of Figure 2 and 3 have been corrected as demonstrated in line 195 and 221.

6. To ensure the validity of the data, the authors should replace Figure 3 with a new picture of the actual experimental setup in the lab.

Answer : Thank you very much for the comment. The author has replaced the new picture in Figure 3. The new picture was taken from an actual experiment in the laboratory as shown in line 220

7. Table 6: explain how 100 random vectors are generated. How do they relate to equations 5 and 6?

Answer : I greatly appreciate your suggestion. We have reorganized and included more information regarding the origin of the 100 random vectors and their relationship to equation (5). (6). These equations have also been rewritten to make them easier to understand for the reader. The variables connection equation (Eq.7) has been proposed, so Equations (5), (6), and (7-10) are connected. The information mentioned above is shown in line “ 333 to 367”.

8. Line 338: The authors stated that the differential evolution algorithm (DE) can solve such a problem. Authors must include complete DE algorithm steps to demonstrate how the algorithm can be used to solve the specified problems. 

Answer : We reorganized and rewritten DE with a step-by-step explanation, as indicated by lines "343 to 367." These lines have all addressed the specifics of how the DE can be applied to solve the presented problem (Optimal weigh finding)

9. The conclusion of the paper needs to be rewritten to include a summary of the most crucial points made throughout the work. In addition, authors should avoid citing previously published references in the conclusion section.

Answer : The lines "499 to 510" have been added to the conclusion to describe how to build the experiment and the model, which is a crucial component of the designed model used to forecast the UTS.

10. The future recommendations need to be rewritten in bullet point format, with an emphasis on using the most recent technologies and various machine learning methodologies.

Answer : We rewrote and added additional future research methods to lines "543-554," and four bullet points were inserted between these lines.

11. Some Latest related references are required.

Answer : In the manuscript, references 14,15,30,36,39,40,41,42,74 and 75 have been included. The publication dates of these articles are 2022 and 2023.

[14] B. Anandan, M. Manikandan,Machine learning approach for predicting the peak temperature of dissimilar AA7050-AA2014A friction stir welding butt joint using various regression models,Materials Letters,Volume 325,2022,132879.

[15] Wei Guan, Yanhua Zhao, Yongchang Liu, Su Kang, Dongpo Wang, Lei Cui,Force data-driven machine learning for defects in friction stir welding,Scripta Materialia,Volume 217,2022,114765.

[30] Shaheen, B.W.; Németh, I. Integration of Maintenance Management System Functions with Industry 4.0 Technologies and Features—A Review. Processes 2022, 10, 2173.

[36] Schmoeller, M.; Weiss, T.; Goetz, K.; Stadter, C.; Bernauer, C.; Zaeh, M.F. Inline Weld Depth Evaluation and Control Based on OCT Keyhole Depth Measurement and Fuzzy Control. Processes 2022, 10, 1422.

[39] B. Anandan, M. Manikandan,Machine learning approach with various regression models for predicting the ultimate tensile strength of the friction stir welded AA 2050-T8 joints by the K-Fold cross-validation method,Materials Today Communications,Volume 34,2023,105286,ISSN 2352-4928.

[40] Prasitpuriprecha, C.; Pitakaso, R.; Gonwirat, S.; Enkvetchakul, P.; Preeprem, T.; Jantama, S.S.; Kaewta, C.; Weerayuth, N.; Srichok, T.; Khonjun, S.; Nanthasamroeng, N. Embedded AMIS-Deep Learning with Dialog-Based Object Query System for Multi-Class Tuberculosis Drug Response Classification. Diagnostics 2022, 12, 2980.

[41] Prasitpuriprecha, C.; Jantama, S.S.; Preeprem, T.; Pitakaso, R.; Srichok, T.; Khonjun, S.; Weerayuth, N.; Gonwirat, S.; Enkvetchakul, P.; Kaewta, C.; Nanthasamroeng, N. Drug-Resistant Tuberculosis Treatment Recommendation, and Multi-Class Tuberculosis Detection and Classification Using Ensemble Deep Learning-Based System. Pharmaceuticals 2023, 16, 13.

[42] Yin, L.; Du, X.; Ma, C.; Gu, H. Virtual Screening of Drug Proteins Based on the Prediction Classification Model of Imbalanced Data Mining. Processes 2022, 10, 1420.

[74] Gonwirat, S. and Surinta, O. (2022). DeblurGAN-CNN: Effective Image Denoising and Recognition for Noisy Handwritten Characters. IEEE Access, 10, 90133-90148. doi: 10.1109/ACCESS.2022.3201560.

[75] Noppitak, S. and Surinta, O. (2022). dropCyclic: Snapshot Ensemble Convolutional Neural Network Based on a New Learning Rate Schedule for Land Use Classification. IEEE Access, 10, 60725-60737 .

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The authors have addressed the most of of suggestions in the most recent version of the manuscript.

Again, the entire document must be proofread thoroughly, and all errors must be eliminated. There are still some grammatical errors.

Good luck,.

Author Response

Reviewer 2

1. Comments and Suggestions for Authors The authors have addressed the most of of suggestions in the most recent version of the manuscript.
2. Again, the entire document must be proofread thoroughly, and all errors must be eliminated. There are still some grammatical errors.

Answer : The manuscript was submitted to MDPI for checking language errors.

Author Response File: Author Response.pdf