Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper proposed an OTCCPG based correlation diagnosis method in multivariate  process quality management. In general, the paper is well-written and organized. A comparison with the studies from the past literature would be improve the quality of the paper. 

One comment about the formatting:

- Please check the paper format in Page 4.

 

 

Comments on the Quality of English Language

English language looks good.

Author Response

Comments 1: This paper proposed an OTCCPG based correlation diagnosis method in multivariate  process quality management. In general, the paper is well-written and organized. A comparison with the studies from the past literature would be improve the quality of the paper.

One comment about the formatting:

Please check the paper format in Page 4.

Response 1:

Thank you for pointing this out. With reference to the journal template, the authors revised the irregular typography on page 4 and double-checked the typographical formatting of the full text to ensure that the layout of the revised manuscript complied with the journal requirements.

Please see the attachment for detailed revisions.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Review report. The manuscript is a good paper but has errors in formatting, organization, and some content errors that make it unacceptable. The referencing of figures, equations and tables, which are not well indicated, should be revised. There are equations that are not indicated in the content of the text, explained. There are figures and tables with formatting errors, not centered and located in the text, outside the content. There are missing basic references for the content of the research. Conclusions are scarce and do not reflect the research. It needs a wide and deep revision.

Author Response

Comments 1: Review report. The manuscript is a good paper but has errors in formatting, organization, and some content errors that make it unacceptable. The referencing of figures, equations and tables, which are not well indicated, should be revised. There are equations that are not indicated in the content of the text, explained. There are figures and tables with formatting errors, not centered and located in the text, outside the content. There are missing basic references for the content of the research. Conclusions are scarce and do not reflect the research. It needs a wide and deep revision.

Response 1:

Thank you for pointing this out. To address the formatting issues in the paper, with reference to the journal's template, the authors have revised the parts with irregular formatting and double-checked the full-text layout formatting to ensure that the revised version meets the journal's requirements. All the figures and tables in the paper are derived from the manufacturing site data collected by the authors, so there are no references marked; the equations in the theorem proving process are partly derived from the basic concepts and definitions of mathematics, and partly from the results of the authors' calculations, so there are also no references marked.

To address the issue of references related to the research content, the authors add a summary of papers based on intelligent diagnostic algorithms, including diagnostic methods based on artificial neural networks, Bayesian networks and support vector machine techniques. The efficiency of intelligent diagnostic algorithms can be significantly improved; however, the structure and parameters of the networks are generally only suitable for specific applications and their generalizability is limited. Therefore, the establishment of more general diagnostic algorithms is a key problem to be solved in the field of quality management and is the main work of this manuscript. 

In response to the problem of fewer conclusions, the authors supplemented the conclusion part. Compared with the existing diagnostic methods based on mathematical analysis, the proposed method has higher diagnostic efficiency and better accuracy of diagnostic results; compared with the intelligent diagnostic methods, the proposed method is based on mathematical analysis as the theoretical foundation, so it has the generality, and can be used as a general method for the correlation diagnosis in multivariate process quality (lines 478-500). In addition, the paper adds the prospect of the further research, by analyzing the change rules of manufacturing parameters in flexible manufacturing system, the diagnostic method based on large-batch manufacturing process will be applied to the flexible manufacturing systems with variable varieties and small batches (lines 501-514)

Please see the attachment for detailed revisions.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The article presents an approach to optimize the manufacturing process parameters developing an algorithm.

Elements needing more clarifications are as follows:

Compare the developed algorithm with other optimization solutions (such as experiments design) and explore why the presented methods are selected.

The description about ISO 9000 (page 4 line 154) is partially accurate. Moreover, there are no details about product quality and its' connection with manufacturing parameters. The article explores the process parameters and their correlation, but did not explain the coating process quality result (tangible metrics), What do we search to control ? How it works if this method is not applied ? Some tangible information is missing.

Some clarification about the 4 test number selection. Is there any connection with the 20 sample data number ?

Conclusion appears quite brief and future perspectives are missing.

Good continuation!

 

Author Response

Comments 1: Compare the developed algorithm with other optimization solutions (such as experiments design) and explore why the presented methods are selected.

Response 1:

Thank you for pointing this out. In the conclusion part of the revised manuscript, the authors compare and analyze the proposed method with the existing diagnostic methods. Compared with the diagnostic algorithms based on CCBD method, principal component analysis method and T² statistics orthogonal decomposition method, the proposed method in this manuscript has lower space complexity, higher diagnostic efficiency, and can avoid the redundancy of diagnostic information in the above methods, so as to provide more accurate diagnostic results for the manufacturing; compared with the intelligent diagnostic methods, the proposed method of is based on the theoretical foundation of mathematical analysis, and it has better generality (lines 478-500).

Comments 2: The description about ISO 9000 (page 4 line 154) is partially accurate. Moreover, there are no details about product quality and its' connection with manufacturing parameters. The article explores the process parameters and their correlation, but did not explain the coating process quality result (tangible metrics), What do we search to control ? How it works if this method is not applied ? Some tangible information is missing.

Response 2:

Thank you for pointing this out. The authors have added a brief description of the automotive purifier coating line in the case study section. by controlling 5 quality components to ensure that the coating coverage is uniform and the thickness is consistent to avoid defects such as too thin, too thick and leakage of the coating (lines 384-390).

Comments 3: Some clarification about the 4 test number selection. Is there any connection with the 20 sample data number?

Response 3:

Thank you for pointing this out. For a manufacturing process with unknown parameters, there are two stages to monitor the process:(1) Manufacturing parameter estimation. When the manufacturing process is in a steady state, enough sample data are collected to estimate the manufacturing parameters, which is demonstrated in lines 393-416 in the revised version. (2) Process monitoring. Using the estimated manufacturing parameters, the subsequent manufacturing process is monitored. The 4 sets of data in Table 2 were collected in the 2nd stage to monitor the state of the manufacturing process and diagnose the causes of the anomalies, and are not related to the of data in Table 1 (lines 419-425 in the revised version).

Comments 4: Conclusion appears quite brief and future perspectives are missing.

Thank you for pointing this out. The authors add to the conclusion section a comparative analysis of the proposed method with existing diagnostic methods from three perspectives: diagnostic efficiency, accuracy of diagnostic results, and generality of the proposed method, as well as an perspective for further research, i.e., to apply the diagnostic method applicable to large-batch manufacturing processes to small-batch, variable-variety flexible manufacturing by analyzing the regularity of the parameter changes in the flexible manufacturing process (lines 478-514 in the revised version).

Please see the attachment for detailed revisions.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

R2: The article has been improved

Good continuation!

 

Author Response

Comments 1: 

The article has been improved

Good continuation!

Response 1: Thank you very much for taking the time to review this manuscript. The authors rechecked the submitted manuscript and revised individual linguistic deficiencies.