Virtual Engineering and Commissioning to Support the Lifecycle of a Manufacturing Assembly System

Round 1

Reviewer 1 Report

This paper proposes a method based on digital twin technology to extend the life cycle of a Manufacturing Assembly System. I have the following comments:

1. Please explain the mechanism of the digital twin technology used in this paper.

2. Is the adopted twin technology technique data-based? Is the used data requried to be drived online?

3. Please explain the relationship between the digital twin and cyber-physical systems

4. In my opinion, the digital twin technology which charaterizes the physical laws or integrated "knowledge" may be more useful and meaningful.

Author Response

The authors thank the reviewer for the valuable comments. They are addressed below.

This paper proposes a method based on digital twin technology to extend the life cycle of a Manufacturing Assembly System. I have the following comments:

1. Please explain the mechanism of the digital twin technology used in this paper.

The Digital Twin presented in this work has the features below, which are adapted from the Industrial Internet Consortium publication titled: “Digital Twin Architectures and Standards”, available here:

https://www.iiconsortium.org/news-pdf/joi-articles/2019-November-JoI-Digital-Twin-Architecture-and-Standards.pdf

The text below was added in the methodology section

Digital Twin capabilities

• Document management: this includes the drawings and instructions throughout the lifecycle.
• Model: a suitable digital representation that mirrors the properties and behaviour.
• Simulation: a representation of the physical device in a simulation environment.
• Data Model: a Standardised data model for connectivity, analytics, and/or visualisation.
• Visualisation: a graphical representation of the object.
• Model synchronisation: the alignment of the model with real-world parameters.
• Connected analytics: the algorithms and computational outcomes.

To achieve this, the next section describes the utilised apparatus.

The text below was also added in the results section

• Document management: all documents, drawings, CAD models, components, MODAPTS tables, etc. are stored in the cloud database and can be accessed by the vueOne application during each stages of the DT development. e.g. at the validation stage, the MODAPTS documents can be generated to compare the virtual model to the physical and find the cause of the differences. The maintenance documents, recordings can be accessed online by fault tracker option.
• Model: the virtual model of the machine mimics the physical machine, its behaviour, features, and functionalities.
• 3D representation: the physical machine devices properties are represented and mapped to the 3D digital representations.
• Simulation: behaviour of the devices of the physical machines is represented in a simulation environment. The necessary studies were made: such as kinematics behaviour and properties, physics (where necessary – e.g. gravity conveyor), manual, semi-auto and auto processes.
• Data model: the data model is standardized according to IEC, IEEE, and OPC standards. Connection is executed via OPC UA, TCP IP and PROFINET.
• Visualisation: there is a dynamic graphical representation of the physical machine with the connection to a PLC real-time.
• Model synchronisation: The model is connected online to the physical machine’s PLC, where it has access to all the parameters, data and input-outputs of the actuators and sensors. The virtual twin of the machine can mimic the physical machine online via OPC UA interface connected to the machine’s PLC and behave according to the physical machine’s signals.
• Connected analytics: the physical machine properties are being measured, recorded and analysed by applications, such as the Fault Tracker application, KPI analytics, etc. This is done for future predictive maintenance, operation planning and control.

The connected analytics feature is still in development. The fault tracker can track, store and represent all the faults, can predict the fault or maintenance needed according to the usage of the component, its historical and real-time data. The data is stored in a SQL database and accessed online by the vueOne applications.

2. Is the adopted twin technology technique data-based? Is the used data required to be derived online?

Yes, the DT technology described in the article is Data-based. The data is derived online, it is also stored in the database and can be accessed online any time. This argument was made clear in the added text mentioned above.

3. Please explain the relationship between the digital twin and cyber-physical systems

The authors clarified in the paper that DT is a part of the CPS, which agrees with the literature as the one below, which is referenced in the text.

“Digital Twin-driven smart manufacturing: Connotation, reference model, applications and research issues”

https://www.sciencedirect.com/science/article/pii/S0736584519302480?via%3Dihub

The text below is quoted from this article:

“A CPS is characterized by a physical asset and its Digital Twin. In contrast, a Digital Twin is limited to the digital model, not the twinning physical asset, though a Digital Twin cannot live without its twining asset in the physical space. In other words, Digital Twin represents the prerequisite for the development of a CPS”.

The text below is added:

The digital twin is regarded as a prerequisite of the CPS, where DT is the digital aspect of the CPS

4. In my opinion, the digital twin technology which characterizes the physical laws or integrated "knowledge" may be more useful and meaningful.

Thanks for highlighting this. Indeed, it is more helpful to include the physical laws and knowledge. In this work, the physical laws in terms of kinematics are included, as the target of this work was to model the manufacturing processes during production. The DT monitors and controls the pressure force, peening force etc., but is not knowledge in the form or advanced algorithms at the moment.

Please see attached the original submission with changes highlighted, and the new submission (clean copy).

Author Response File: Author Response.pdf

Reviewer 2 Report

Congratulations authors, great paper.

Author Response

Thank you!

Please see attached the original submission with changes highlighted, and the new submission (clean copy).

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript's clarity, logical order of contents, and originality are problematic. Authors must fix those before publication. Some of the hints to improve the manuscript are as follows:

Figures 1,…,5. All figures are from other authors. The manuscript has 12 figures. This means that half of the figures are not original figures of the authors. Therefore, the reviewer recommends removing at least four figures from Figures 1,…,5. Most arguments are redundant, i.e., they can be rewritten concisely.

In Figures 6 and 7, the authors introduced conceptual task allocation diagrams for operating a manufacturing system with some computerized manufacturing enablers. Are these conceptual task allocation sequences user-generated? If yes, then what is the contribution of the digital twin? First, is digital twin must be able to generate the task allocation sequences. Otherwise, it has no meaning. It is worth mentioning that there are three types of digital twins as far as smart manufacturing is concerned: object twin, process twin, and phenomenon twin. A twin, whatever may be its type, consists of five modules: input module, modeling module, simulation module, validation module, and output module. In the case of the authors, the process twin is the relevant one. The process twin here creates task allocation sequences to operate the manufacturing system. Thus, the contents starting from Figure 6 up to Figure 11 need revision showing the physical work and cyber world integrated by some OPC UA client-server as midware. In doing so, the author needs to reorder the figures (6 to 11).

The authors cited references mainly from conference articles (including procedia). The reviewer recommends citing journal articles instead.

The writing style needs to be revisited.

"The process of the rotor assembly is presented at the Figure 7b. The station is designed for the rotor assembly from the hub and two lamination rings. The sequence summary is as follows: operator loads hub and first lamination ring from the trolley to the machine, the machine presses the lamination ring onto the hub, then the operator loads second ring to be pressed. When both lamination rings are pressed, peening automatic operation starts to fix laminations and the hub."

The reviewer rewrites, as follows.

The rotor assembly station is shown in Figure 7b. As seen in Figure 7b, the station assembles the rotor using the following sequence of operations:

1. The operator loads the hub and first lamination ring from the trolley to the machine.
2. The machine presses the lamination ring onto the hub.
3. The operator loads the second ring to be pressed.
4. After pressing both lamination rings, an automatic peening operation starts to fix laminations and the hub.

Author Response

The authors thank the reviewer for the comments, they are addressed as follows:

The manuscript's clarity, logical order of contents, and originality are problematic. Authors must fix those before publication. Some of the hints to improve the manuscript are as follows:

Figures 1,…,5. All figures are from other authors. The manuscript has 12 figures. This means that half of the figures are not original figures of the authors. Therefore, the reviewer recommends removing at least four figures from Figures 1,…,5. Most arguments are redundant, i.e., they can be rewritten concisely.

Figures 1,2 are removed in the modified copy.

Figures 3,4,5 are based on [13] which are the authors’ previous work.

The rest are coming from the case study.

The language of the manuscript was revised: some phrases are removed.

In Figures 6 and 7, the authors introduced conceptual task allocation diagrams for operating a manufacturing system with some computerized manufacturing enablers. Are these conceptual task allocation sequences user-generated? If yes, then what is the contribution of the digital twin?

The authors propose a process twin, the initial sequence tasks, and the main sequence are user-generated. The application generates specific tasks such as MODAPTS, detailed machine sequences and generates the PLC code via vueOne toolset.

First, is digital twin must be able to generate the task allocation sequences. Otherwise, it has no meaning. It is worth mentioning that there are three types of digital twins as far as smart manufacturing is concerned: object twin, process twin, and phenomenon twin. A twin, whatever may be its type, consists of five modules: input module, modeling module, simulation module, validation module, and output module. In the case of the authors, the process twin is the relevant one. The process twin here creates task allocation sequences to operate the manufacturing system.

In the modified version, the authors refer to the features of the targeted digital twin, which is adapted from the Industrial Internet Consortium (IIC). The evaluation is done according to this internationally recognised body.

Thus, the contents starting from Figure 6 up to Figure 11 need revision showing the physical work and cyber world integrated by some OPC UA client-server as midware. In doing so, the author needs to reorder the figures (6 to 11).

 A considerable amount of details (in text) is provided to demonstrate the conducted work. Another Figure showing “showing the physical work and cyber world integrated by some OPC UA client-server as midware. “ is added to the manuscript, with an explanatory text.

The authors cited references mainly from conference articles (including procedia). The reviewer recommends citing journal articles instead.

It is understandable that the reviewer recommends citing journal articles. Some of the Procedia articles received citations more than a lot of journal papers. Below are some of the references’ citations based on the Scopus database:

The following conference and proceeding papers are deleted:

• Next Generation Manufacturing Systems and the Virtual Enterprise [9]
• A novel framework for virtual recommissioning in reconfigurable manufacturing systems [21]
• Software-in-the-loop technique: an approach to support reconfiguration of manufacturing system [25]
• Automatic PLC Code Generation Based on Virtual Engineering Model. [26]
• Automatic model generation for virtual commissioning based on plant engineering data. [34]
• Keeping the digital twin up-to-date Process monitoring to identify changes in a plant [36]
• A concept in synchronization of virtual production system with real factory based on anchor-point method [39]
• Mechatronic concept of heavy machine tools [42]
• Virtual commissioning of a robotic cell: an educational case study [43]

The writing style needs to be revisited.

Some spotted language mistakes were corrected

"The process of the rotor assembly is presented at the Figure 7b. The station is designed for the rotor assembly from the hub and two lamination rings. The sequence summary is as follows: operator loads hub and first lamination ring from the trolley to the machine, the machine presses the lamination ring onto the hub, then the operator loads second ring to be pressed. When both lamination rings are pressed, peening automatic operation starts to fix laminations and the hub."

The reviewer rewrites, as follows.

The rotor assembly station is shown in Figure 7b. As seen in Figure 7b, the station assembles the rotor using the following sequence of operations:

1. The operator loads the hub and first lamination ring from the trolley to the machine.
2. The machine presses the lamination ring onto the hub.
3. The operator loads the second ring to be pressed.
4. After pressing both lamination rings, an automatic peening operation starts to fix laminations and the hub.

Thanks for rewriting this, it is added to the paper.

Please see attached the original submission with changes highlighted, and the new submission (clean copy).

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed my previous comments. I have no further comments.

Reviewer 3 Report

authors made some efforts to improve. it is good. the assembly system design is completely human-intelligence driven. we just need to program and integrate existing facility. 

however, please consider revisting the English. Still expressions are hard to follow.

some minor corrections are as follows:

Please delete SolidWorks mark (figure 5). We do not promote any vender in scholarly article. Figure 3. Please remove color from the black-colored blocks. Simple white background is easy for readers to follow.

Figure 13. please reduce the font size (make it as large as the text).