Design and Simulation Debugging of Automobile Connecting Rod Production Line Based on the Digital Twin

Round 1

Reviewer 1 Report

Check that the abstract provides an accurate synopsis of the paper. It is very vague in its present form.

The figures are of poor quality, not appropriately labeled, and not visible.

A few key steps in implementing a digital twin for this purpose are listed here and need to be incorporated. The flow of the paper can be:

·       Data collection: The first step in creating a digital twin is to gather data on the automobile connecting rod production line. This data can come from various sources, including sensors, CAD models, and historical production data. The more available data, the more accurate the digital twin will be. This is a major missing in the paper.

·       Modeling: Once the data has been collected, a production line model can be created. This model should include all relevant components and processes, such as the machining centers, inspection stations, and assembly areas. The model should be detailed enough to simulate the production line accurately but not so complex that it becomes unwieldy.

·       Validation of the Model: After creating the model, it should be validated against real-world data to ensure it accurately represents the production line. This can involve running simulations and comparing the results to actual production data. This is also missing.

·       Control system design: Once the model has been validated, the control logic can be added to simulate the behavior of the production line under different conditions. This logic can include things like machine tool settings, production rates, and quality control measures. This should be clearer in terms of PLC.

·       Simulation and Debugging: With the model and control logic in place, the digital twin can simulate the production line and identify potential problems. By simulating different scenarios, engineers can identify the root cause of issues and test out potential solutions without interrupting production. There should be a demonstration of at least one case study.

·       Model Deployment: Once the digital twin has been thoroughly tested and refined, it can be deployed to help optimize the production line. This is also missing in the paper.

·       Model Refinement: As issues are identified and addressed, the digital twin can be refined to better reflect the production line's behavior. This can involve tweaking control logic, adjusting machine settings, or adding new components to the model. This could be considered future work.

The manuscript could be further improved by including some demonstrations and case studies to illustrate the application of digital twin technology in the automobile industry. Another option can be to provide more detailed information on the data sources and sensors used to collect data for the digital twin.

The scope of MEMS-based sensors integrated with an open source controller, such as industrial versions of Arduino, Raspberry pi, etc., for measurement of real-time signals has widened a lot. This has significantly reduced the cost, and the data acquisition systems development has become open, flexible, and cheap. You may refer to the articles where MEMS-based sensors are integrated with an open-source controller to collect real-time data for decision-making.

Novel Machine Health Monitoring System, Tyre Pressure Supervision of Two Wheeler Using Machine Learning, Overview of contemporary systems driven by open-design movement

This would open new opportunities for authors and readers as well. The authors could also provide more information on validating the model against real-world data, an essential aspect of the process. This would help readers better appreciate the technology's practical aspects and potential benefits.

Author Response

Dear Professor

I'm glad to receive your amendments, and your suggestions have helped me greatly.

You not only clearly point out the missing part of my article, but also give a new chapter assignment for the full text. This part of the content I really learned a lot of knowledge when I revised it, and I feel that I have gained a lot. Thank you again!

There are many things that need to be modified in this article. I have modified them one by one according to your suggestions. The details are as follows. The italic content is the modification suggestion given by you. The following non-italic is my corresponding modified reply content.

Check that the abstract provides an accurate synopsis of the paper. It is very vague in its present form.

---- I rewrite the abstract part. Now its content is mainly a summary of the key steps you give.

The figures are of poor quality, not appropriately labeled, and not visible.

---- I've replaced all the pictures with high-quality ones and labeled the key ones for easy reading.

A few key steps in implementing a digital twin for this purpose are listed here and need to be incorporated. The flow of the paper can be:

---- I have modified the structure of the article according to your suggestions.

Data collection: The first step in creating a digital twin is to gather data on the automobile connecting rod production line. This data can come from various sources, including sensors, CAD models, and historical production data. The more available data, the more accurate the digital twin will be. This is a major missing in the paper.

---- The main data source of this paper is the historical production data in the factory processing. I have supplemented the three views of the connecting rod and the production process schedule.

Modeling: Once the data has been collected, a production line model can be created. This model should include all relevant components and processes, such as the machining centers, inspection stations, and assembly areas. The model should be detailed enough to simulate the production line accurately but not so complex that it becomes unwieldy.

---- The automobile connecting rod production line model is shown in Fig.4.This automobile connecting rod production line model is mainly divided into five areas : processing area, assembly area, cleaning area, inspection area and inventory area. The details have been marked on the figure.

Validation of the Model: After creating the model, it should be validated against real-world data to ensure it accurately represents the production line. This can involve running simulations and comparing the results to actual production data. This is also missing.

----Because the simulation steps are set according to table 3, and the table 3 data is from the historical data of the factory processing, the generated simulation sequence is the actual processing process, so the operation of the simulation must be completely consistent with the actual processing process. At the same time, the simulation process is given by using the change of 8 pictures. The simulation process is shown in Fig.9.

Control system design: Once the model has been validated, the control logic can be added to simulate the behavior of the production line under different conditions. This logic can include things like machine tool settings, production rates, and quality control measures. This should be clearer in terms of PLC.

---- mainly including the main program, reset program, sequence control system flow program, human-computer interaction and so on.

Simulation and Debugging: With the model and control logic in place, the digital twin can simulate the production line and identify potential problems. By simulating different scenarios, engineers can identify the root cause of issues and test out potential solutions without interrupting production. There should be a demonstration of at least one case study.

---- through the debugging of the connecting rod in the process of transportation in the process of sudden slip case is analyzed. The causes of connecting rod slipping in debugging are analyzed carefully, and the shortcomings of planning in actual processing are found out, and the corresponding solutions are given.

Model Deployment: Once the digital twin has been thoroughly tested and refined, it can be deployed to help optimize the production line. This is also missing in the paper.

---- Through the case of 2.5.4 section, it can be clearly seen that when the automobile connecting rod production line model is deployed, it can help optimize the production line during the debugging process. In addition, in the debugging process found another concern, it is the problem of workpiece accumulation. Then the causes are analyzed in detail, and the corresponding solutions are given considering various factors.

 Model Refinement: As issues are identified and addressed, the digital twin can be refined to better reflect the production line's behavior. This can involve tweaking control logic, adjusting machine settings, or adding new components to the model. This could be considered future work.

----In this part, I discussed the limitations of my research and the direction of future work.

The manuscript could be further improved by including some demonstrations and case studies to illustrate the application of digital twin technology in the automobile industry. Another option can be to provide more detailed information on the data sources and sensors used to collect data for the digital twin.

----A new part of the content is added, and the changes of eight pictures are used to show the whole process of simulation operation. A case of connecting rod slipping from the row crane during the debugging process is added. The causes of their virtual simulation are carefully analyzed and the planning deficiencies in the actual processing process are found, and the specific solutions are given. It also discusses the case that the workpiece to be processed accumulates on the production line after the production line is simulated for a certain period of time, analyzes the causes of this situation, and gives the corresponding solutions according to the actual situation. In addition, the three views of the connecting rod and the production process schedule are added.

The scope of MEMS-based sensors integrated with an open source controller, such as industrial versions of Arduino, Raspberry pi, etc., for measurement of real-time signals has widened a lot. This has significantly reduced the cost, and the data acquisition systems development has become open, flexible, and cheap. You may refer to the articles where MEMS-based sensors are integrated with an open-source controller to collect real-time data for decision-making.

Novel Machine Health Monitoring System, Tyre Pressure Supervision of Two Wheeler Using Machine Learning, Overview of contemporary systems driven by open-design movement

This would open new opportunities for authors and readers as well. The authors could also provide more information on validating the model against real-world data, an essential aspect of the process. This would help readers better appreciate the technology's practical aspects and potential benefits.

----I have read the three articles you listed. In the process, I have learned a lot. Thank you for your suggestions. Because of your specific suggestions, I have really learned a lot, even more than the knowledge content itself. After learning through the key steps that you listed, I have a better grasp of the structural level of digital twins than before.

Finally, thank you for taking the time out of your busy schedule to read this letter. 

I wish you all the best!

Best regards

Jiayan

Author Response File: Author Response.pdf

Reviewer 2 Report

The Digital Twin concept is herein used in the automobile connecting rod production line, and the virtual model is used to verify the feasibility, but also in the optimization of the production line. The work appears very interesting and the concept is worthy of investigation. However, some doubts arise. The authors are invited to answer the following questions.

1.    Some improvements are necessary for the figures and tables. Indeed, they are too small to appreciate. Moreover, a centralization of them could help the readability.
2.    The Digital Twin concept is strictly related to the Industrial Internet of Things (IIoT). Some drawbacks can arise. For example, DT need to be driven by massive data distributed devices, some issue concerning the privacy, competition, and security, the real-time interactions between the DT and the entity object (see for example DOI: 10.1109/TII.2022.3183465). These concepts are absent or little addressed in the paper.
3.    A contributions list at the end of the introduction could help to better understand the novelties apported by this work.
4.    The authors affirm that the proposed approach allows to reduce the development cycle and to improve the debugging speed. These advantages have been investigated against some specific method known from the current literature? And according to which metrics have been evaluated?

Author Response

Dear Professor

I'm glad to receive your amendments, and your suggestions have helped me greatly.

This part of the content I really learned a lot of knowledge when I revised it, and I feel that I have gained a lot. Thank you again!

There are some things that need to be modified in this article. I have modified them one by one according to your suggestions. The details are as follows. The italic content is the modification suggestion given by you . The following non-italic is my corresponding modified reply content.

The Digital Twin concept is herein used in the automobile connecting rod production line, and the virtual model is used to verify the feasibility, but also in the optimization of the production line. The work appears very interesting and the concept is worthy of investigation. However, some doubts arise. The authors are invited to answer the following questions. 

1. Some improvements are necessary for the figures and tables. Indeed, they are too small to appreciate. Moreover, a centralization of them could help the readability.

---- I've replaced all the pictures with high-quality ones and labeled the key ones for easy reading.

2. The Digital Twin concept is strictly related to the Industrial Internet of Things (IIoT). Some drawbacks can arise. For example, DT need to be driven by massive data distributed devices, some issue concerning the privacy, competition, and security, the real-time interactions between the DT and the entity object (see for example DOI: 10.1109/TII.2022.3183465). These concepts are absent or little addressed in the paper.

---- I carefully read the article that you listed, I learned a lot in this article. I have cited this article as a reference and discussed it. 

3. A contributions list at the end of the introduction could help to better understand the novelties apported by this work.

---- I have added a list of contributions to the end of the introduction.

4. The authors affirm that the proposed approach allows to reduce the development cycle and to improve the debugging speed. These advantages have been investigated against some specific method known from the current literature? And according to which metrics have been evaluated?

----A new part of the content is added, and the changes of eight pictures are used to show the whole process of simulation operation. A case of connecting rod slipping from the row crane during the debugging process is added. The causes of their virtual simulation are carefully analyzed and the planning deficiencies in the actual processing process are found, and the specific solutions are given. It also discusses the case that the workpiece to be processed accumulates on the production line after the production line is simulated for a certain period of time, analyzes the causes of this situation, and gives the corresponding solutions according to the actual situation. In addition, the three views of the connecting rod and the production process schedule are added.

Finally, by comparing the traditional product debugging mode with the virtual simulation de-bugging mode of the automobile connecting rod production line based on digital twin, it is concluded that the virtual simulation debugging of the automobile connecting rod production line based on digital twin will greatly reduce the actual debugging time of the production line, thus speeding up the research and development progress and improving the industrial competitiveness.

The details are shown in 3. Discussion and Figure 23.

Finally, thank you for taking the time out of your busy schedule to read this letter.

I wish you all the best!

Best regards

Jiayan

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed all my comments. 

Reviewer 2 Report

Authors have implemented all the suggestions and addressed all questions. I don't have any further comments. The paper can be accepted.