Elsevier

Computers in Industry

Volume 113, December 2019, 103130
Computers in Industry

Review of digital twin applications in manufacturing

https://doi.org/10.1016/j.compind.2019.103130Get rights and content

Highlights

  • A deep literature review on DT applications in manufacturing is performed.

  • Rarely a DT environment offers a large set of services to the real system.

  • Almost never a DT share the elaborated analysis to the real counterpart.

  • A DT application is proposed in a Simulink environment to overcome these gaps.

  • The illustrated DT poses the basis for further improvements.

Abstract

In the Industry 4.0 era, the Digital Twin (DT), virtual copies of the system that are able to interact with the physical counterparts in a bi-directional way, seem to be promising enablers to replicate production systems in real time and analyse them. A DT should be capable to guarantee well-defined services to support various activities such as monitoring, maintenance, management, optimization and safety. Through an analysis of the current picture of manufacturing and a literature review about the already existing DT environment, this paper identifies what is still missing in the implemented DT to be compliant to their description in literature. Particular focuses of this paper are the degree of integration of the proposed DT with the control of the physical system, in particular with the Manufacturing Execution Systems (MES) when the production system is based on the Automation Pyramid, and the services offered from these environments, comparing them to the reference ones.

This paper proposes also a practical implementation of a DT in a MES equipped assembly laboratory line of the School of Management of the Politecnico di Milano. The application has been created to pose the basis to overcome the missing implementation aspects found in literature. In such a way, the developed DT paves the way for future research to close the loop between the MES and the DT taking into consideration the number of services that a DT could offer in a single environment.

Introduction

With the new paradigm of Industry 4.0, manufacturing evolves to include smart objects (Kusiak, 2017), basing on the concept of Cyber-Physical Systems (CPS), defined as “systems of collaborating computational entities which are in intensive connection with the surrounding physical world and its on-going processes, providing and using, at the same time, data-accessing and data-processing services available on the internet” (Monostori, 2015). These new autonomous systems are capable to elaborate and communicate data and to build a copy of real processes in a digital environment in real-time (Lee et al., 2015).

The main enabling technologies of the Industry 4.0 that characterise CPSs (such as Big Data and Cloud computing as support systems to read big sets of data from the field, store and analyse them (Cattaneo et al., 2018; Roblek et al., 2016) and Internet of Things (IoT) to remain connected and extract data) are also the basis for a new simulation approach, which leverages on the pervasive connectivity in production systems to offer a real-time synchronization with the field. This new simulation approach is generally referring to the elaboration of Digital Twins (DT) (Negri et al., 2017). Although much literature was produced on the topic, a general definition and an agreement over its features and scopes has not been reached yet. Considering its close link with production management issues, it becomes highly valuable to understand the features of the applications of DT and to compare them with what is suggested by literature. This paper aims at exploring the existent literature about the practical applications of DT, identifying the gaps between the theoretical features of the DT and the applications. Then, this paper proposes an application in a laboratory environment that poses the basis to overcome the found gaps.

To this aim, the next part of the introduction this paper is structured as follows: firstly, the automation pyramid structure, the most commonly used structure in industrial field, is illustrated in section 1.1. Then, in section 1.2 the innovations in this field are showed highlighting the role of the mentioned CPSs and the related Digital Twin (DT). Finally, in section 1.3 this paper will condense the main points for a reflection on the current scenario, needed to better illustrate the objectives of the paper.

Each industrial automated process is commonly based on the Automation Pyramid, a centralized structure composed of five layers, as in Fig. 1 (ISO/IEC, 2012). On the left of the figure lies the automation pyramid (https://visaya.solutions/video/old-new-automation-pyramid/), on the right the corresponding production management level (Govindaraju and Putra, 2016). The figure also shows the Manufacturing Execution System (MES) functions listed by the Manufacturing Execution System Association (MESA INTERNATIONAL, 1997).

  • (i)

    The three lower layers (0–1 - 2) are summarized as the control layer, including: electrical engineering layer, Programmable Logic Controllers (PLCs) / Distributed Control Systems (DCS) and Scada/HMI layers,

  • (ii)

    the middle layer (3) is the Manufacturing Execution System (MES) that guides the process, and

  • (iii)

    finally on top there is the Enterprise Resource Planning (ERP) layer (4) aimed at integrating organizational functions for better customer support and planning (Zuehlke, 2010).

In this structure, the MES guides the manufacturing process, monitoring all the steps of a product in the production system in a centralized way. Among the MESA-identified functions, some are directly linked to the production process (such as scheduling and quality control), while others (such as resource management and traceability) are best described as cross functions, not strictly related to the production process but supporting a higher point of view. Thus, all MES functions are not on the same level, as suggested by De Ugarte (De Ugarte et al., 2009).

Introducing the new technologies mentioned above, manufacturing industries future vision is often depicted in literature as based on CPSs (Rosen et al., 2015) or on Cyber Physical Production Systems (CPPS) (Weyer et al., 2015) and on the associated concept of the DT.

CPSs represent a promising future within the Industry 4.0 context and offer to enforce the flexibility of the automation pyramid-based manufacturing systems (Iarovyi et al., 2016), showed in section 1.1. CPSs, by definition, have both a cyber and physical nature. In the cyber part, they will be suitable to host computations, that cover data analysis and on-board simulations. The simulation may be:

- synchronized with the field (physical part of the CPS) but is limited to the boundaries of the CPS itself (e.g. a single workstation);

- a higher-level simulation, that is hosted outside the CPS system itself. This simulation may in this way replicate both single workstations and the whole production system behaviours, as data are collected from the single CPSs.

The latter type of simulation, when synchronized with the field, complies in all aspects with the definition of DT given by (Garetti et al., 2012) and reported in (Negri et al., 2017; Kritzinger et al., 2018): “The DT consists of a virtual representation of a production system that is able to run on different simulation disciplines that is characterized by the synchronization between the virtual and real system, thanks to sensed data and connected smart devices, mathematical models and real time data elaboration. The topical role within Industry 4.0 manufacturing systems is to exploit these features to forecast and optimize the behaviour of the production system at each life cycle phase in real time”. In this vision, the future of manufacturing is characterized by entities, CPSs or CPPSs, able to collect data directly from the field and to replicate the physical production system in the cyber world through various digital models that compose a DT; i.e. these models open the way to a real-time synchronized simulation of the physical equipment (Negri et al., 2017). The digital models also encompass a proper data modelling to allow for interoperability of various models and tools, to offer a common vocabulary and to provide to the system knowledge about the information the digital models elaborate: all of this is well supported by semantic models, such as ontologies (Garetti et al., 2015; Negri et al., 2016).

As said, the DT is related to the CPS and is seen as its digital counterpart. The concept of DT is still evolving and its definition has matured over the years as Negri reports in (Negri et al., 2017). The DT was initially employed by the NASA to replicate the life of the air vehicles. DT were used for health analyses i.e. crack propagation (Kraft, 2016), or for improving maintenance activity and planning. Then, it was also introduced to digitally mirroring the life of the physical entities and to support decision making through engineering and statistical analyses; in this wider perspective it started to be used also in other fields such as industrial engineering (Garetti et al., 2012). In its essence DT can be considered as an environment that can support different types of simulation or is considered as a simulation itself, which is synchronized with the field in a near to real-time fashion.

An interesting interpretation is given by Kritzinger, that proposes in (Kritzinger et al., 2018) a review of the DT literature depending on the interactions between the physical object and the relative virtual object. In this sense, a proper DT is the one where the virtual object exchange data flows with the physical one in both directions (physical-to-virtual and virtual-to-physical directions). This means that the virtual object can eventually act on the control system of the real one. This DT interpretation is depicted in Fig. 2 together with the other two categories mentioned by (Kritzinger et al., 2018). (i) The Digital Model does not entail any interactions between the physical and virtual objects. (ii) The Digital Shadow, where only the physical object sends data and updates the virtual one. It includes the Digital Model, the data acquisition protocols and the simulation software to simulate the digital model. In this case, the interaction between cyber models and physical objects is monodirectional. (iii) As stated by Kritzinger (Kritzinger et al., 2018), and illustrated in the context given by this paper in Fig. 2, the Digital Shadow can become a full Digital Twin only when the dotted line connecting data coming from the Digital Shadow (the virtual twin) with the real system is realized, therefore when the data flow is bidirectional from the real system to the DT and vice versa.

The DT is often used to offer specific analyses, related to the considered system and to its lifecycle, according to the services that a DT may offer, as listed by Tao et al (Tao et al., 2018). As we will recall later, not all services are needed for all systems, but having them at disposal in one DT environment may be useful for industrial decision makers that do not need to carry out different analyses in separate digital environments.

Considering the nine services listed by (Tao et al., 2018), they can be grouped in these categories:

  • Real-time state monitoring, used to update the virtual twin in real time;

  • Energy consumption analysis;

  • Product failure analysis and prediction, and product maintenance strategy, that have in common the analysis of the real-time state data and historical data to predict a fault and construct a maintenance strategy;

  • Intelligent optimization and update; this service is based on the analysis of the user’s operation habits and product behaviours data to improve the product and/or the production process;

  • Behaviour analysis and user operation guide, used to obtain the operations done from the users and/or giving them some user guidance to visualize the system updates with a user-friendly HMI (Human-Machine Interface);

  • Product virtual maintenance and product virtual operations; given a 3D environment or software, these services elaborate and show the operations or the maintenance strategy to the user.

The challenge in offering these services in a single environment leads to the fact that some of them need a 3D graphic interface and others only analyse data without requiring any graphics. Also noticeable is that some of them (e.g. “Intelligent optimization and update”) are based on this mutual exchange of information between the real and virtual object, the DT must offer specific services, that recalls again the concept introduced by Kritzinger (Kritzinger et al., 2018).

To sum up, the adoption of a DT must ensure that it is connected to the physical twin. In case the physical system it mirrors has the characteristics to be considered a CPS, this will make the DT implementation easier because its synchronization passes through a direct field connection hosted by the CPS itself. Also, the DT does not have a unique definition or a unique reference model yet, but Fig. 2 provides the reference concept of DT for the purpose of this work.

In practice, research on CPS and DT is still ongoing, and in most cases manufacturing systems at companies are equipped with traditional machinery which is hierarchically based on the automation pyramid: the production is planned in the long run by the company information systems (such as ERP), and the sequence of operations is controlled by the MES of the systems (referring to the levels of the automation pyramid of Fig. 1). In these cases, it is more difficult to introduce an implementation of DT in the systems and to integrate it with an existent MES-based equipment.

This paper wants to investigate the DT applications present in literature, to understand how the DT is used practically on real systems and what are the services offered. Particular attention will be given to the improvement that can be made to boost the usage of a DT inside the already existent systems, given the importance of the additional information and analysis provided by this tool.

The structure of this paper is the following: in section 2 the objective and approach of the work are better outlined; section 3 reports the literature review about the DT applications; in section 4 the creation of a DT in a laboratory environment and the related results are described; section 5 discusses the created DT environment; section 6 proposes hints for future work and final remarks.

Section snippets

Objective and approach

As already said, research about DT is ongoing as a promising topic for decision making in various fields, among which the manufacturing field is one of the most relevant (Macchi et al., 2018). The present paper aims at proposing an overall vision on the DT application for production systems. The main objective is to identify the discrepancies between literature and implementations of DT, starting from a literature review on the DT applications. The paper also aims at proposing a DT environment

Review of digital twin applications in literature

As mentioned, researchers and practitioners still do not agree on a unique definition of DT. A lot has been said about its role in the Industry 4.0 paradigm, however, only few practical examples are reported in scientific papers that do not show the same technological features and functionalities. It is therefore of paramount importance to investigate previous research works about DT application in production environments. Scopus was used as research literature database, considering the papers

Digital twin creation

This section illustrates the development of a DT that poses the basis to overcome the gaps found in the previous section. The proposed DT is built with the purpose of energy consumption monitoring, inside the Industry 4.0 Laboratory (I4.0 Lab) of the School of Management of Politecnico di Milano (Fumagalli et al., 2016).

Discussion

The proposed DT poses the basis to overcome the gaps from Section 3.4. The analysed DTs are limited in the number of services offered, never extended to the entire set of services of reference given by (Tao et al. (2018)), and they are usually not integrated with the existent control system, making the DTs closer to a “Digital Shadow” (Kritzinger et al., 2018) without reaching the full DT potentials mentioned in the “Introduction” section.

In the present state of research progress, the proposed

Conclusions and future work

In the most recent years, the DT in industry has received a relevant attention from both researchers and practitioners. Manufacturing is one of the most promising contexts where the DT may be successfully applied for concrete benefits in terms of maintenance and operations monitoring and optimization.

The article frames the research on DT as follows. Nowadays production systems are still “traditionally” controlled with the automation pyramid. However, the advent of CPS may lead to less

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This project has received funding from the European Union’s Horizon 2020 research and innovation programmes under grant agreements No 678556 (correspondent to the project shortly entitled “MAYA”, “MultidisciplinArY integrated simulAtion and forecasting tools, empowered by digital continuity and continuous real world synchronization, towards reduced time to production and optimization”) and No 680435 (correspondent to the project shortly entitled “PERFoRM”, “Production harmonizEd Reconfiguration

CHIARA CIMINO graduated in Automation and Control Engineering at the Politecnico di Milano in 2018. She started a PhD in Information Technologies in the department of Electronic, Information and Bioengineering at the Politecnico di Milano in collaboration with the Department of Management, Economics and Industrial Engineering carrying out activities in the Industry 4.0 laboratory. Her research is focused in dynamic modelling, simulation and control for optimized production management in the

References (85)

  • S. Haag et al.

    Digital twin – proof of concept

    Manuf. Lett.

    (2018)
  • L. Hu et al.

    Modeling of Cloud-based digital twins for smart manufacturing with MT connect

    Procedia Manuf., Elsevier B.V.

    (2018)
  • Z. Kemény et al.

    The MTA SZTAKI smart factory: platform for research and project-oriented skill development in Higher education

    Procedia CIRP

    (2016)
  • G.L. Knapp et al.
    (2017)
  • S. Konstantinov et al.

    The cyber-physical E-machine manufacturing system: virtual engineering for complete lifecycle support

    Procedia CIRP, The Author(S)

    (2017)
  • W. Kritzinger et al.

    Digital twin in manufacturing: A categorical literature review and classification

    IFAC-PapersOnLine.

    (2018)
  • J. Lee et al.

    A cyber-physical systems architecture for industry 4.0-based manufacturing systems

    Manuf. Lett.

    (2015)
  • E.A. Loeken et al.

    Design principles behind the construction of an Autonomous laboratory-scale drilling rig

    IFAC-PapersOnLine.

    (2018)
  • M. Macchi et al.

    Exploring the role of digital twin for asset lifecycle management

    IFAC-PapersOnLine.

    (2018)
  • E. Negri et al.

    Requirements and languages for the semantic representation of manufacturing systems

    Comput. Ind.

    (2016)
  • E. Negri et al.

    A review of the roles of digital twin in CPS-based production systems

    Procedia Manuf.

    (2017)
  • E. Negri et al.

    FMU-supported simulation for CPS digital twin

    Procedia Manuf.

    (2019)
  • J.O. Oyekan et al.

    The effectiveness of virtual environments in developing collaborative strategies between industrial robots and humans

    Robot. Comput. Integr. Manuf.

    (2019)
  • A. Padovano et al.

    A digital twin based service oriented application for a 4.0 knowledge navigation in the smart factory

    IFAC-PapersOnLine.

    (2018)
  • S. Rabah et al.

    Towards improving the future of manufacturing through digital twin and augmented reality technologies

    Procedia Manuf.

    (2018)
  • R. Rosen et al.

    About the importance of autonomy and digital twins for the future of manufacturing

    IFAC-PapersOnLine.

    (2015)
  • S. Sierla et al.

    Automatic assembly planning based on digital product descriptions

    Comput. Ind.

    (2018)
  • V. Toivonen et al.

    The FMS training center - A versatile learning environment for engineering education

    Procedia Manuf., Elsevier B.V.

    (2018)
  • J. Um et al.

    Plug-and-simulate within modular assembly line enabled by digital twins and the use of AutomationML

    IFAC-PapersOnLine.

    (2017)
  • P.D. Urbina Coronado et al.

    Part data integration in the shop floor digital twin: Mobile and cloud technologies to enable a manufacturing execution system

    J. Manuf. Syst.

    (2018)
  • X.V. Wang et al.

    Digital twin-based WEEE recycling, recovery and remanufacturing in the background of industry 4.0

    Int. J. Prod. Res.

    (2018)
  • S. Weyer et al.

    Towards industry 4.0 - standardization as the crucial challenge for highly modular, multi-vendor production systems

    IFAC-PapersOnLine.

    (2015)
  • P. Zheng et al.

    A systematic design approach for service innovation of smart product-service systems

    J. Clean. Prod..

    (2018)
  • S. Anand et al.

    Additive manufacturing simulation tools in education

    2018 World Eng. Educ. Forum - Glob. Eng. Deans Counc. WEEF-GEDC 2018

    (2019)
  • A. Ardanza et al.

    Sustainable and flexible industrial human machine interfaces to support adaptable applications in the industry 4.0 paradigm

    Int. J. Prod. Res.

    (2019)
  • D. Botkina et al.

    Digital twin of a cutting tool

    Procedia CIRP.

    (2018)
  • H. Brandtstaedter et al.

    Digital twins for large electric Drive trains

    Pet. Chem. Ind. Conf. Eur. Conf. Proceedings, PCIC Eur. 2018–June

    (2018)
  • A. da S. Barbosa et al.

    Virtual assistant to real time training on industrial environment

    Adv. Transdiscipl. Eng.

    (2018)
  • J. David et al.

    Leveraging digital twins for assisted learning of flexible manufacturing systems

    Proc. - IEEE 16th Int. Conf. Ind. Informatics, INDIN 2018. 589

    (2018)
  • A. De Carolis et al.

    Guiding manufacturing companies towards digitalization

    23rd ICE/IEEE Int. Technol. Manag. Conf.

    (2017)
  • A. De Carolis et al.

    A maturity model for assessing the digital readiness of manufacturing companies

    APMS 2017, Part I, IFIP AICT 513

    (2017)
  • B.S. De Ugarte et al.

    Manufacturing execution system - A literature review

    Prod. Plan. Control.

    (2009)
  • Cited by (479)

    View all citing articles on Scopus

    CHIARA CIMINO graduated in Automation and Control Engineering at the Politecnico di Milano in 2018. She started a PhD in Information Technologies in the department of Electronic, Information and Bioengineering at the Politecnico di Milano in collaboration with the Department of Management, Economics and Industrial Engineering carrying out activities in the Industry 4.0 laboratory. Her research is focused in dynamic modelling, simulation and control for optimized production management in the Industry 4.0 context.

    ELISA NEGRI received her PhD in 2016, with a thesis about the role of ontologies for smart manufacturing. She now works at the Department of Management, Economics and Industrial Engineering of the Politecnico di Milano in industrial, national and European projects on ontology-based engineering, industrial systems modeling, digital twin and Cyber-Physical Systems supporting advanced manufacturing solutions in the frame of Industry 4.0. During her studies and research work, she spent periods at prestigious international institutions, such as EPFL (Lausanne, Switzerland), FIR (Aachen, Germany), Tongji University (Shanghai, China) and Rutgers University (New Brunswick, USA).

    LUCA FUMAGALLI is Assistant Professor in the Department of Management, Economics and Industrial Engineering of Politecnico di Milano. After his Master of Science in Mechanical Engineering, he completed his PhD studies in 2010 about Innovation in Maintenance Management at the Department of Management, Economics and Industrial Engineering. His research interests are advanced production management, industrial services and maintenance management, with a specific concern on smart technological solutions. His research activity has been related also with European research funded projects. Luca Fumagalli is TeSeM observatory research responsible (www.tesem.net) and MEGMI Master vice-director (Executive Master on Industrial Maintenance Management) (http://www.mip.polimi.it/megmi).

    View full text