![The rationale of digital twins-based system controls commissioning [70].](https://www.researchgate.net/profile/Jiewu-Leng/publication/351862149/figure/fig5/AS:1039248492871684@1624787557534/The-rationale-of-digital-twins-based-system-controls-commissioning-70_Q320.jpg)
Content uploaded by Jiewu Leng
Author content
All content in this area was uploaded by Jiewu Leng on Jun 27, 2021
Content may be subject to copyright.
Journal of Manufacturing Systems 60 (2021) 119–137
Available online 25 May 2021
0278-6125/© 2021 Published by Elsevier Ltd on behalf of The Society of Manufacturing Engineers.
Digital twins-based smart manufacturing system design in Industry 4.0:
A review
Jiewu Leng
a
,
b
,
c
, Dewen Wang
a
, Weiming Shen
b
, Xinyu Li
b
, Qiang Liu
a
,
*, Xin Chen
a
a
Guangdong Provincial Key Laboratory of Computer Integrated Manufacturing System, State Key Laboratory of Precision Electronic Manufacturing Technology and
Equipment, Guangdong University of Technology, Guangzhou, 510006, China
b
State Key Laboratory of Digital Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan, 430074, China
c
Department of Information Systems, City University of Hong Kong, Hong Kong, 999077, China
ARTICLE INFO
Keywords:
Digital twins
Manufacturing system design
Cyber-physical systems
Smart manufacturing
Function-structure-behavior-control-
intelligence-performance
ABSTRACT
A smart manufacturing system (SMS) is a multi-eld physical system with complex couplings among various
components. Usually, designers in various elds can only design subsystems of an SMS based on the limited
cognition of dynamics. Conducting SMS designs concurrently and developing a unied model to effectively
imitate every interaction and behavior of manufacturing processes are challenging. As an emerging technology,
digital twins can achieve semi-physical simulations to reduce the vast time and cost of physical commissioning/
reconguration by the early detection of design errors/aws of the SMS. However, the development of the digital
twins concept in the SMS design remains vague. An innovative Function-Structure-Behavior-Control-Intelligence-
Performance (FSBCIP) framework is proposed to review how digital twins technologies are integrated into and
promote the SMS design based on a literature search in the Web of Science database. The denitions, frame-
works, major design steps, new blueprint models, key enabling technologies, design cases, and research di-
rections of digital twins-based SMS design are presented in this survey. It is expected that this survey will shed
new light on urgent industrial concerns in developing new SMSs in the Industry 4.0 era.
1. Introduction
The general birth process of a manufacturing system can be
described as follows: a designer generates an idea in his mind, then
draws it in drawings (forms a model) and tries to make some physical
prototypes for testing, and nally assembles the system. Table 1 pro-
vides an overview of the manufacturing system design in different stages
of industry development from Industry 1.0–4.0.
In Industry 1.0 and 2.0, the information in the R&D stage exists in the
drawings; and in the physical prototype stage, the information uses the
physical system as the carrier. The conventional manufacturing system
design method requires physical prototypes to be assembled to check
and verify whether the virtual space model is accurately matched,
including structure, ergonomics, and performance indexes. However,
using physical prototypes to verify the design is very costly. To avoid
unnecessary risks on physical prototypes, simulations of the perfor-
mance and manufacturability of a model in the virtual space are critical
[1].
In Industry 3.0, many Computer-Aided Design (CAD) software
providers (e.g., PTC, Siemens, ANSYS, Dassault) proposed the concept of
virtual prototype, digital prototype, active prototype, and others. A
digital prototype is a kind of rehearsal in which the information model
replaces physics. Among various digital prototype concepts, the Digital
Mock-Up [2] concept is widely used, which emphasizes complex map-
pings and contextual relationships among 3D model simulations. The
existence of digital prototypes greatly reduces the failure of physical
prototypes (sometimes they simply directly replace real prototypes).
Based on a virtual simulation of the physical manufacturing process, the
SMS performance can be evaluated early to guide cutting down the
reconguration costs/losses in establishing physical SMS prototypes (e.
g., Plant Simulation). Using virtual reality could greatly enhance the
convenience and efciency of SMS design (e.g., FlexSim). Due to the
consistency of information transmission in digital prototypes, the dif-
culty of the manufacturing system design is greatly reduced [3].
In Industry 4.0, smart manufacturing has become a development
direction of the world’s manufacturing industry. New national advanced
manufacturing strategies around the world resulted in the increasing
need of designing new smart manufacturing systems. A Smart
* Corresponding author.
E-mail address: liuqiang@gdut.edu.cn (Q. Liu).
Contents lists available at ScienceDirect
Journal of Manufacturing Systems
journal homepage: www.elsevier.com/locate/jmansys
https://doi.org/10.1016/j.jmsy.2021.05.011
Received 20 March 2021; Received in revised form 16 May 2021; Accepted 17 May 2021
Journal of Manufacturing Systems 60 (2021) 119–137
120
Manufacturing System (SMS) is a multi-eld physical system composed
of intelligent machines, materials, products, and complex couplings
among various elements. In the digital design process, an SMS can be
broken down into digital models of various granularities in digital space,
while the physical products and manufacturing processes exist in
another physical space. In the SMS design process, high-delity cyber
models mapping the real worlds of SMSs are critical to fullling the gap
between its design domain and operation domain [5].
Digital twins (DT) technology is a solution. Nowadays, the digital
twins trend is gaining momentum because of rapidly evolving simula-
tion and modeling capabilities, better interoperability and IoT sensors,
and more availability of tools and computing infrastructure (www2.
deloitte.com/). Recent MarketsandMarkets research suggests that the
global digital twin market size was valued at USD 3.1 billion in 2020 and
is projected to reach USD 48.2 billion by 2026 (www.marketsandma
rkets.com). A digital twin could realize the feedback of the digital
model of cyberspace to the real physical system. In this way, it is possible
to ensure the coordination of the digital and physical spaces within the
scope of the entire life cycle. Zhang et al. [5] rstly applied the digital
twins-based approach in the manufacturing system design sector. After a
review of existing SMS development models, Mahmoud and Grace [4]
suggested a digital twins-based simulation approach for SMS congu-
ration. Because of the supremacy of digital continuity, realistic
modeling, and interaction between disciplines against conventional
simulation, the creation of a digital twin in the SMS designing phase will
be highly valuable during all the lifecycle of the SMS.
However, the application of the digital twins concept in the SMS
design remains vague [6]. Therefore, it is necessary to explore and sort
out the research on the rational integration of digital twins technology
into the SMS design. This article attempts to show how digital twins
technology is integrated into the SMS design to truly promote the
development of smart manufacturing. A literature search on digital
twins-based SMS design is conducted in the Web of Science database.
The collected literature is further rened by following three phases. The
rst phase is to screen related papers via keyword retrieving. Two
combinations of keywords, namely, {“digital twin” and “manufacturing
system design”} and {“digital twin” and “manufacturing system plan-
ning”}, are used for retrieving papers. The retrieving yields 202 papers
and 54 papers, respectively (up to 31 December 2020). After the dele-
tion of duplicated papers, this retrieving yield 220 papers for further
screening. The second phase is a theoretical screening process to identify
high-quality studies related to the concept, key enabling techniques,
models, systems, frameworks, and case studies on SMS design. Research
not related to the digital twins or SMS design domain and short papers
less than four pages are excluded. In this phase, 159 papers are nally
included to discover the key issues, new solutions, challenges, and
promising directions in the research of the digital twins technology in
the SMS design process. Additionally, 33 additional papers are referred
to make this survey more comprehensive. Finally, this survey includes
192 references, and the statistics of collected literature are shown in
Fig. 1.
The denitions, frameworks, major design steps, new blueprint
models, key enabling technologies, design cases, and research directions
of digital twins-based SMS design in industry 4.0 will be presented in
this survey. The organization of this survey is outlined in Fig. 2.
2. A framework for reviewing the digital twins-based SMSD
2.1. Denition of smart manufacturing system design
Manufacturing System Design (MSD) conventionally includes the
modeling, analyzing, and optimizing of system layout, production ca-
pacity, production ow, material handling system, manufacturing
methods, system exibility, and operation strategies [7]. Decisions on
the selection and conguration of resources should be optimized in the
design process. The goal of the MSD is to nd the design solution that
yields the optimal performance measures. In terms of the modeling of
the couplings among the system elements, the mathematical
programming-based quantitative approaches cannot model and analyze
the stochastic disturbances (e.g., machine tool breakdowns, resource
operation conicts) that greatly affect the system performance. On the
other hand, simulation models, due to the ability to account for the
time-variant dynamic behaviors, have been utilized to optimize the
system performance.
However, MSD is facing new challenges under the smart
manufacturing blueprint. Smart Manufacturing System Design (SMSD)
refers to the design of an SMS. SMS coordinates various manufacturing
elements (e.g., machine tools, material, human, and environment) and
holistically optimizes the operations based on a unied environment of
Cyber-Physical Systems (CPS). CPS for manufacturing enables optimi-
zation of product development and total control of production system
through real-time exchange of all information required for production
based on Internet-of-Things (IoT) [8]. SMSD is a complicated process
composed of modeling, analyzing, mining, and learning data from many
sources in SMS. Besides the uncertainties of the input demands and the
process disturbances, the complex interactions, couplings, conicts
among the design variables and goals make the SMSD a highly iterative
and cost/time-consuming task. Since the major difference between an
SMS and a conventional manufacturing system is industrial intelligence,
the SMSD is more challenging in terms of the design of industrial arti-
cial intelligence models compared to the conventional MSD.
2.2. The difference between digital twins method and simulation methods
2.2.1. Denition of digital twins
The digital twins concept is a development of modeling and simu-
lation technology (Fig. 3). Traditional simulation methods are of limited
capabilities in evaluating system performance [7]. Digital twins tech-
nology represents the breakthrough of limitations on the modeling and
engineering analysis capabilities of simulation by integrating the IoT
technology [9].
The virtual SMS is described by its digital models. Digital models
Table 1
Manufacturing system design in different stages of industry development.
Stages Industry 1.0 Industry 2.0 Industry 3.0 Industry 4.0
Period 1760s~1850s 1860s~1960s 1970s~2010s 2010s~Now
Mfg. Paradigm Mass Production Mass Production +Mass Customization +Mass Individualization
Mfg. Goal Scale +Speed +Knowledge +Intelligence
Mfg. System Type Mechanical Production Line Automated Mfg. Syst. Computer-Aided Mfg. Syst. Smart Mfg. Syst.
Design Elements Machinery +Automation +Digitalization +Complex Algorithms
Design Goal Tolerance +Interchangeability +Flexibility/Recongurability +Sustainability/Intelligence
Design Methods 2D Drawings 2D Drawings 3D Models & Simulations Digital Twins-based
Typical Blueprint – – Digital Prototype/Mock-Up Lifecycle Digital Twins
Major Challenges Huge Commissioning Cost Huge Commissioning Cost Complex Assemblies +Multi-dimension Couplings
Ref. – – – [4]
Notes: “+” is followed with additional items in the column compared to the former column; “-” denotes vacancy/unknown (Similarly hereinafter).
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
121
could be categorized into the digital replica, digital shadow, and digital
twin [10]. A digital replica emphasizes the automatic projection of
system constructions. The digital shadow emphasizes mathematical
modeling to describe the physical/chemical attributes of a system. A
digital twin is an integrated multi-physics and multi-scale simulation of
a product/system that can model the mechanical, electrical, software,
and other discipline-specic properties across its lifecycle. A digital twin
is supposed to be able to optimize the physical product/system based on
the updated real-time data synchronized from sensors [11]. Other
similar digital twins concept includes digital surrogates [12] and intel-
ligent beings [13]. Valckenaers [13] argued that real-world smart sys-
tems consist of digital twins and intelligent agents. The digital twins
mirror the real systems without imposing articial limitations, while
intelligent agents convert NP-hard problems into
low-polynomial-complexity computations. Digital twins technology is
being developed to optimize manufacturing, aviation, healthcare, and
medicine elds [14–16]. It is recognized as one of the technological
pillars for achieving smart manufacturing and Industry 4.0 [17–20].
2.2.2. The advantages of digital twins-based SMSD approach
Traditionally, simulation technology is restricted to a standard tool
for supporting designers to solve specic engineering problems.
Communication-By-Simulation is the core concept of Model-Based Sys-
tems Engineering. Extending the simulation from the design phase to
subsequent lifecycle activities is the trend. As shown in Fig. 4, the
existing MSD methods lack an effective iterative optimization of
"execution-analysis-adjustment" in the design process. The design has
not been veried so it is difcult to guarantee the decision quality.
Although simulation models are useful for supporting the design visu-
alization, how to build a unied model across multi-discipline of SMS is
challenging. The solution is to build a digital twin that reects and ob-
serves the real-world SMS behavior and use it in all design procedures
Fig. 1. Statistics of collected literature.
Fig. 2. The organization of this survey.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
122
including system deployment and commissioning [21], which is named
as Digital twins-based Smart Manufacturing System Design (DT-SMSD)
in this survey. Digital twins are established based on combining a
multi-physics simulation model with the real-time data of a pro-
duct/system [22]. Digital twin models could be timely modied and
validated to avoid bottlenecks that happen in the SMS simulation and
development processes. The design process of an SMS implies the con-
cepting, forming, and ne-tuning of corresponding digital twins. The use
of digital twins in three stages is different, for example:
•Concepting stage of digital twins (Function model design and
Structure model design): Using multi-disciplinary simulations to
quickly verify the conceptual scheme. The goal in this stage is not the
pursuit of simulation accuracy but a fast simulation.
•Forming stage of digital twins (Behavior model design and Control
model design): Key technical design parameters are determined by
simulations. Although the purpose of digital twins is to replace
physical commissioning, necessary semi-physical system commis-
sioning in practice needs to be retained, especially as required by
some industry specications. The semi-physical system commis-
sioning is the focus of this stage.
•Fine-tuning stage of digital twins (Intelligence model design and
Performance model design): The digital twins are used to help build
the prototype SMS and plan the system test scheme. The digital twins
could reduce trials and errors to accurately locate the points that
need to be intelligentized. The optimal performance could be ob-
tained in a lean way. In another word, the digital twin approach
could achieve the SMS validation purpose with fewer test times and
improve the design efciency. The key point to the ne-tuning of
corresponding digital twins lies in the excellent recongurability and
exibility of SMS to support alternative design/prototyping solutions
[23].
The most direct value of implementing digital twins is to replace the
costly pure-physical commissioning and test. However, the ultimate goal
of implementing digital twins is not simply validation but design inno-
vation. The established digital twin in the SMS design phase can also be
further utilized for monitoring, optimizing activities, diagnostics, and
Fig. 3. The evolution of simulation technology from the design view.
Fig. 4. The transition from conventional MSD to digital twins-based SMSD approach.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
123
prognostics in the following-up lifecycle phases of SMS [24].
2.3. Function-structure-behavior-control-intelligence-performance
framework
The SMS is conceptually considered as an integration of complex
interconnected functional models. The essence of the SMSD is to
establish entities that meet individual demands, such as layout con-
straints, production capacity, used equipment integration, and
manufacturing efciency. In the SMSD process, the personalized re-
quirements from the customer and product domains are transferred to
the SMS conguration domain and corresponding execution domain,
including the customized industrial articial intelligence models in SMS.
This survey proposes a Function-Structure-Behavior-Control-
Intelligence-Performance (FSBCIP) framework to review the major
steps of DT-SMSD, as shown in Fig. 5.
•Function model. It is a structured description of the activities for
manufacturing the product and the relationships among the activ-
ities. This model highlights the inuence of manufacturer demand in
the conceptual design on product manufacturability, assembly,
maintainability, and safety. What kind of functions the SMS has is
determined by the product features it faces and its development re-
quirements. This model usually includes process planning, equip-
ment selection, and mechanical design.
•Structure model. It is a structured description of the fusion,
connection, and assembly relations among the mechanical structures
that realize manufacturing functions, principles, and behaviors. The
interrelation of the mechanical structure is the foundation for the
transferring and transformation of the material, energy, information,
and motion behavior of the manufacturing system. This model usu-
ally includes topology denition, layout planning, xture/carrier
designing, and buffer/cache designing in SMS.
•Behavior model. It is a structured description of the mechanical
motion transmission, the motion form transformation, and their
mutual relations. The material, energy, and information connection
that existed among the SMS are mainly reected in the transfer
relationship of the mechanical motion behavior. This model usually
includes equipment motion, WIP motion, material delivery/
handling, and process engagement in SMS. The essential elements of
the motion behavior mainly include force, displacement, velocity,
acceleration, jerk, angular displacement, angular velocity, angular
acceleration, angular jerk, snap, and higher derivatives.
•Control model. This model is built to deal with the structure, oper-
ation, or calculus of a process by statistical or engineering methods to
control the output of the process. It usually includes electrical &
pneumatic system designing, control & sensor networking, equip-
ment controls designing, and multi-process coordination.
•Intelligence model. This model is supposed to able to describe,
develop, and validate the learning ability, optimization ability, and
autonomy ability of an SMS from the control design perspective. It
includes a systematic method and a variety of reasoning rules, ma-
chine learning algorithms, and computational optimization algo-
rithms for achieving sustainable performance in SMS.
•Performance model. The differences in design decisions will result in
the uctuation of manufacturing execution efciency. This model
usually includes the evaluation and optimization of the system per-
formance such as efciency, exibility, recongurability, robustness,
adaptivity, extensibility, and resilience.
The input of SMSD includes personalized requirements and param-
eters. The design process includes the design of the above six models.
The output of the SMSD is the whole cyber models, motion scripts,
control schemes, execution systems, and intelligent algorithms, which
form the nal optimized/ne-tuned digital twins. The proposed FSBCIP
framework is a development of the conventional function-behavior-
structure model [25]. Conventional MSD research usually focused on
the functional, structural, and behavior dimensions. The design scope
review in this paper is extended to the control, intelligence, and per-
formance dimensions, which are critical to efciently realize the
comprehensive design goals of SMSD in the context of Industry 4.0.
Abstracting and making full use of similarities in the FSBCIP framework
at a higher level could effectively enhance the accuracy and efciency of
the customized design of similar SMSs.
3. Digital twins-based SMSD
This section will review the research progress and issues in the
critical steps of SMSD that could be enhanced by the digital twins
technologies from the proposed FSBCIP view, including 6 dimensions
and 16 sub-dimensions.
3.1. Digital twins-based function design of SMS
3.1.1. Digital twins-based process planning
Manufacturing process planning is a process performed by engineers
to analyze manufacturability, cost, and efciency in the early phase of
MSD based on the manufacturing feature recognition [26] from product
design. It includes manufacturing operation selection and operation
sequencing. Few studies have been conducted on how to deal with the
Fig. 5. An FSBCIP framework for smart manufacturing system design.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
124
random disturbances and complex loadings in the process plan process
[27]. Concerning current simulation-based process planning ap-
proaches, the obtaining of adequate process information and searching
for optimal decisions needs 74 % of the planning time [28]. Process
planning is the rst step in SMSD that could be enhanced by digital twins
technology.
The digital twin for manufacturing processes improves the optimi-
zation speed and accuracy of planning. Lu et al. [29] established a
bidirectional cyber-physical mapping for planning and optimization of
re-manufacturing processes based on the discrete event simulation and
IoT technology. Liu et al. [30] proposed a digital twins-based process
evaluation approach on the basis of the real-time mapping between the
physical machining process and the design model. Further, to express
the evolutionary characteristics of product processing, Liu et al. [31]
proposed a digital twin process model based on the
knowledge-evolution machining features. Sierla et al. [32] derived a
digital twin from a digital product description to concurrently conduct
the assembly design and planning, and automatically coordinate the
manufacturing resources in one manufacturing unit. The resulting ver-
satile manufacturing system is supposed to be able to cope with a large
variety of products at minimal reconguration cost. The application of
digital twins technologies in process planning is not yet a common
practice. More advantages, as well as decits, are supposed to be
discovered in using digital twins in the manufacturing process planning,
as well as the rapid mining and learning of the collected data.
3.1.2. Digital twins-based equipment mechanical design
A mechanical design of equipment should be conducted based on a
deep understanding of its machining process. It is necessary to simulate
the physical machining processes of operations to design high-
performance equipment. For instance, in the design of laser processing
equipment, a digital twin model of the interaction process between laser
and materials should be established to have a deeper understanding of
the manufacturing process of laser materials through simulations. Haber
et al. [33] presented a digital twins-based approach for optimizing the
controller parameters of the backlash peak amplitude/time and the
hysteresis amplitude of an ultra-precision motion stage system to guar-
antee an appropriate dynamic response in the presence of backlash and
friction in a cost-effective manner. Actually, the digital twinning of the
equipment implies the digital twinning of the physical process of the
equipment operation. The semi-physical simulation of the process could
be carried out to reect all activities/status of the physical equipment.
3.2. Digital twins-based structure design of SMS
3.2.1. Digital twins-based layout planning and topology optimization
The layout planning (e.g., U-shaped, O-shaped, L-shaped, 1-shaped,
and mixed type) of SMS is largely decided by the geometric con-
straints of given site space. On the other hand, the logistics intension is a
critical index to evaluate the layout planning solution. Based on the
digital twins technologies, the physical manufacturing process could be
mirrored and validated with its digital model in cyberspace promptly,
and thus could support the automatic retrieving of layout information
and facilitate planning an optimized layout in the digital twin of phys-
ical SMS.
According to the process route and the priority relationship between
processes, the system topology and the connection relationship between
equipment could be determined. The topology design of SMS is the basic
idea of using numerical computation and optimization algorithms for
seeking to achieve the optimal topology structure and satisfy various
constraints in a given design space. Multi-disciplinary design optimiza-
tion is a well-known design direction aimed at solving the coupling and
tradeoff problems of multiple disciplines in the design of large-scale
complex engineering systems, including SMSs. It fully explores and
utilizes the synergistic mechanism of interaction in engineering systems,
considers the interaction between various disciplines, and optimizes the
design of complex engineering systems from the perspective of the
whole SMS. The digital twins model applied in the SMSD process should
naturally have the ability of optimization. Guo et al. [34,35] introduced
a digital twins-based Graduation Intelligent Manufacturing System that
is capable of real-time task allocation and execution for xed-position
assembly islands. Generally, the digital twins-based coordination
among production processes ensures that each equipment is allocated in
the right position and utilized to the right activities with enhanced
visibility.
3.2.2. Digital twins-based tooling and buffer designing
More than 20 % of the complex products’ manufacturing time is the
auxiliary preparation time, which is thereby the key point to improve
manufacturing efciency [36]. The tooling and buffer designing is a
critical step affecting the auxiliary preparation time of product
manufacturing. Through the takt time analysis and productivity
balancing analysis, the type and quantity of machines and buffer ca-
pacity in each process could be determined. However, the stochastic
logistics, massive disturbances, and various routing patterns in SMS
make tooling and buffer designing a very challenging task.
The digital twins technology is a potential solution to tackle the
difculty of tooling/buffer designing and bridge the gap between
designing and controlling. For instance, Liu et al. [36] proposed a digital
twins-based approach for dynamic clamping and positioning of exible
tooling systems. The geometry data of the manufacturing parts are
collected in real-time to drive optimizing of the clamping forces and
positioning. Wang et al. [37] proposed a digital twins-based CPS for the
design and control of roller conveyor lines, which is critical for
improving the sorting speed and delivery speed of workpieces in SMS
logistics.
3.3. Digital twins-based behavior design of SMS
3.3.1. Digital twins-based equipment behavior design
Equipment behavior design refers to dening the actions (including
high-level action patterns) of equipment (process equipment and inter-
mediate equipment) and the action sequence of multi-process coordi-
nation according to process route and process requirements. It should
avoid unnecessary collisions and deadlocks. Although building virtual
equipment is useful for simulating equipment’s capabilities safely and
cost-effectively, it is difcult to exactly simulate the physical equipment.
Establishing the digital twin of equipment could benet the behavior
design, as well as the further diagnosis and prediction of failures/faults
in the operation of physical equipment. Vatankhah et al. [38] proposed a
digital twins-driven motion planning approach by integrating with
agent-based rapid optimization algorithms in a robotic manufacturing
cell. Cai et al. [39] utilized sensory data to model the actual status of the
machine tools to develop their digital twins for enhancing accountability
and manufacturing capabilities. Further, Cai et al. [40] presented an
augmented reality-based simulation approach for robotic toolpath
planning based on communicating the layout information between a
recongurable additive manufacturing system and its’ digital twin.
The intelligent machine tool (IMT) is a vision of digital twins-driven
digitalized, highly-efcient, and autonomous machine tools in smart
manufacturing blueprint. Tong et al. [41] presented a digital
twins-based real-time machining data application service for IMT. Data
transmission and storage are completed using the MTConnect protocol
and components. The digital twin of IMT is equipped with the optimi-
zation functions related to the machining trajectory, machining status,
energy consumption, machine tool dynamics, contour error estimation,
and cutting-tool compensation. Also, as a key component of machine
tools, the digital twin of the cutting tool is critical to the optimization of
machining solutions [42].
3.3.2. Digital twins-based ergonomics design
Due to the limited articial intelligence level of software models and
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
125
the limited exibility/recongurability of hardware in SMS, dynamic
demands and contextual randomness still require humans (workers) to
participate in the highly exible tasks. The ergonomics design in the
interactions between workers and machines is an indispensable
dimension to evaluate and optimize the system efciency, exibility,
and recongurability in the SMSD. Also, the advent of collaborative
robots makes the use of digital twins that involve humans grow in the
future. The discrete event simulation methods widely used during the
design stage cannot mirror the continuous movement behavior of
workers in SMSs. It is critical to enhancing the ergonomics simulation
model with the continuous movement, working environment, and per-
formance routines of workers in SMSs [43].
A form of hybrid automation is taking benet from the synergistic
effect of human-robot collaboration. The need for adaptability and the
dynamics of human presence are keeping the full potential of human-
robot collaborative systems difcult to achieve [44]. When used in the
assembly, the requirement of exibility, adaptability, and safety makes
the design and redesign of human-robot collaborative systems a complex
and prone to error process [45]. Based on the data collected from
wearable sensors on workers, virtual reality-based digital twins tech-
nology is the key to designing, monitoring, and optimizing the ergo-
nomics performance of manual work activities, as well as guiding the
improvement of the working conditions in human-machine interactions.
For instance, Graessler and Poehler [46] presented a digital twin
approach to enable workers on the shop oor to take part in computa-
tional decision-making. Latif and Starly [47] proposed a data-driven
simulation algorithm to model the complex and manual
manufacturing process in a generic-reusable way, from which the pro-
duction managers can make more informed early decisions. Greco et al.
[48] proposed a methodological human digital twin framework that
enables the monitoring and optimizing of the ergonomics performances
of workers. The actions and personal preferences of workers should be
captured in the digital twin [49]. Malik and Bilberg [50] presented a
digital twin approach to help the designing and controlling of
human-machine interactions/cooperation’s in manual assembly work.
After the SMSD phase, the digital twin of ergonomics should be
continuously updated in the following operating activities of the SMS
from a comprehensive cyber-physical-social-connected systematic view
[51]. The realization of digital twins for human-robot collaborative
tasks is facing challenges in the lack of high-delity simulation models at
various detailed levels of human endeavors, as well as the seamless
integration of different modules.
3.3.3. Digital twins-based WIP/material handling
Unsynchronized production can easily cause problems such as the
backlog of intermediate warehouses, unsmooth production, and long
production cycles. Synchronized production helps to improve overall
efciency and reduce waste. It requires the planning of the material
handling, production logistics path, movement pattern, suspension, and
cache mode of the WIP (i.e., work-in-process) according to the action
and behavior mode of the equipment. Material handling refers to the
unloading, distributing, and delivery of raw materials to the production
unit and warehouse associated with the entire manufacturing process.
Robots are common tools for material delivery/handling in SMSs due to
their high exibility. Using simulation/emulation is an effective
approach of optimization designing of WIP and material handling of
manufacturing system [52]. An adaptable material handling solution is
supposed to be able to deal with the dynamic-change product
manufacturing requirements under the mass individualization para-
digm. The optimization design of WIP and material handling cannot be
separated from the consideration of three aspects: 1) the impact of the
manufacturing process on the logistics; 2) the requirement of
manufacturing capacity on the logistics facilities; 3) the impact of pro-
duction takt time on the material ow.
These dynamic factors and impacts could be more efciently evalu-
ated by digital twins technologies. Digital twins technology enables the
system to achieve excellent logistics traceability, controllability, and
efciency via a design-for-manufacturing manner. For instance, Wang
et al. [53] proposed a proactive digital twins-based material handling
method for CPS-enabled shop-oor to reect the production status of
physical SMS. Banyai et al. [54] introduced a concept of matrix pro-
duction as a CPS concentrating on material handling and described an
extended and real-time optimization model of supply routing and
assignment. Yerra and Pilla [55] created a digital twin for each physical
part in smart material ow control in automotive composites body shop.
Instead of using the century-old assembly line philosophy in the
manufacturing layouts, they proposed an assembly grid layout to
improve space utilization, inspired by a colony of ants working to
accomplish a common goal. It has the potential to improve exibility to
product variabilities. Nevertheless, more realistic virtual models of WIP
need to be developed to further fulll the gap between design and
manufacturing and to map the real and cyber worlds.
3.4. Digital twins-based control design of SMS
3.4.1. Digital twins-based process optimization and equipment control
designing
Virtual equipment is usually built to simulate the machining toolpath
of various workpieces/parts, and thereby emulate each machining
process to guide the generation of controls. If the digital twin of physical
equipment is established, designers could not only conrm its func-
tionality to greatly decrease the design cycle but also conveniently make
innovations in manufacturing processes. For instance, Zhu et al. [56]
developed a functional simulation system to establish digital twins of the
physical machining tools. It is capable of generating the robot manipu-
lator path to produce a given workpiece, which could be straightly
controlled by the CAD tool employed in the product design phase. Cao
et al. [57] proposed a STEP-NC standard-based digital twin system for a
computer-aided manufacturing system. Zhao et al. [58] proposed a
digital twin-driven cyber-physical system for autonomously controlling
of micro punching system. Gohari et al. [59] developed the digital twin
inspection system to optimizing the inspection controls of geometrically
complex surfaces of workpieces. The closed-loop digital twins could
avoid the randomness involved in the inspection process based on the
high-volume contextual data collected from distributed sensors in
workpieces.
From the workpiece view, incorporating the geometry assurance
model into digital twins is critical for realizing the process optimization
and controls designing, thereby guarantying the desired geometry of the
product. For instance, Tabar et al. [60] proposed a digital twin approach
emphasizing the geometry assurance to secure the geometry of the in-
dividual sheet metal assemblies based on a spot-welding sequence
optimization method. Based on an overview of employing the digital
twins approach in geometrical variations management in the context of
industry 4.0 [61], Schleich et al. [62] presented a Skin Model Shapes
concept-oriented reference model serving as the digital twin of the
physical product to simultaneously optimize its design and
manufacturing. Cerrone et al. [63] established a nite element model of
the as-manufactured specimen to resolve the crack-path ambiguity, and
thus predict the crack path. The modeling of as-manufactured properties
is essential to establish the digital twins of workpieces.
Equipment calibration and subsequent controller-based compensa-
tion are critical steps in the initial controls design to improve the
volumetric performances in the ultra-precision SMS domain. Digital
twins-based technology is a potential solution. For instance, Montavon
et al. [64] adopted a digital view regarding machine tool calibration and
dened the layered semantics in the CPS of SMS, in which the controller
of equipment additionally acts as the edge computing device to coor-
dinate with upper management systems in the manufacturing process.
Although some performance checking models have been proposed for
predicting the volumetric performance degradation and thereby avoid-
ing undesirable breakdown, the insights in the equipment calibration
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
126
are still limited due to the fragmentation of sensory data.
3.4.2. Digital twins-based system controls commissioning
The design and runtime validation of controls are critical for product
quality management but challenging since SMSs in Industry 4.0 vision
are highly complex and heterogeneous. The system controls commis-
sioning is a complex task requiring modeling of low-level and high-level
control functions in different software environments and connecting
them through specic interfaces, which is a critical step to verify and
validate the basis design models.
The commissioning of an SMS mainly includes three approaches:
virtual commissioning, physical commissioning, and semi-physical
commissioning. The physical commissioning needs to wait for the
completion of physical equipment assembly and control logic to com-
plete the physical debugging work, resulting in waiting-time waste.
There will inevitably be design errors, logic errors, coding errors, and
other problems, resulting in irreversible consequences such as me-
chanical equipment misoperations. There exist many virtual simulation
commissioning software tools, such as Siemens Tecnomatix, Rockwell
Emulate3D, VisualComponents, and Simumatik3D. The results obtained
from virtual commissioning suffer from low condence. Simply writing
the model behavior logic on the simulation platform is unable to debug
and verify whether the real controller can drive the physical equipment
correctly and reasonably [65]. After the design of the actual
manufacturing system is adjusted, the behavior logic prepared by the
simulation model also needs to be changed accordingly, which cannot
meet the iterative optimization design process of the physical
equipment.
Therefore, it is essential to carry out the semi-physical (hardware-in-
the-loop) commissioning. The application of hardware-in-the-loop
commissioning is to create reasonable and effective virtual equipment,
which requires the combination of control technology and digital twins
technology. The digital twin of the physical equipment is taken as the
debugging object to run the control logic in the virtual environment, to
optimize and verify the rationality and feasibility of the logic. A digital
twin can truly reect the changes of the physical equipment, and
continuously accumulate relevant knowledge, optimize and analyze the
current situation under specic controls, and thus support the design
decision.
In the development process of an SMS, many sets of equipment may
be supplied by different manufacturers, and it requires a concurrent
control design of multiple virtual machines. Conventionally, the control
program/code written by engineers relying on their personal experience
is neither conducive to others’ understanding of the logical action ow
nor conducive to the rapid response of the iterative optimization of the
control debugging process. Automated control code generation is critical
and, in the long run, makes it easier to modify and upgrade programs
and software. For instance, Julius et al. [66] proposed a method to
automatically convert the Grafcet specication language into the ST text
code of Programmable Logic Controller (PLC) while preserving the hi-
erarchical structure described by the processing logic. A methodology of
automatic generation, formal verication, and implementation of safety
PLC program can save the time of equipment design/verication, and
thus avoid the duplication of tasks and design errors [67].
Developing a digital twins-based commissioning system that could
automatically generate the control programs (e.g., PLC codes) is of great
signicance to further shorten the preparation time of integrated system
commissioning and continuously iterative design optimization. For
instance, Orive et al. [68] presented a digital twins-based design
approach to conveniently emulate controlling reactions to disturbances
and misoperations during system commissioning. Toivonen et al. [69]
introduced a digital twins-based versatile learning environment for
supporting engineers to design and program controls in cyberspace
before commissioning physical SMS. Digital twins could cut down the
reconguration cost since it enables the timely discovery of design
errors/aws in the functional, structural, and behavioral dimensions of
SMS during semi-physical commissioning. Leng et al. [70] proposed
digital twins-based remote semi-physical commissioning of ow-type
smart manufacturing systems. As shown in Fig. 6, the I/O addresses
and signals of distributed machines (hardware PLC) are congured and
mapped to the software PLC of digital equipment models in the digital
twin system via Industrial Internet protocols. When the designers start
the simulation of the whole material ow, workers put the needed raw
material in the machines according to the instructions. Except for the
simulation of whole material ow, many situations of simulation (e.g.,
machine kinetics inside equipment, and control compatibility) may be
performed via a no-load running test manner [70].
A prerequisite is a current virtual plant model of the structured,
mechatronic resource components of the real plant during the different
design phases. Talkhestani et al. [71,72] presented a cross-domain
mechatronic data structure-based deviations detection method of the
virtual manufacturing cells in the automotive industry in the context of
RAMI 4.0. Kang et al. [73] proposed a digital twin framework named
DigTwinOps for runtime verication of CPS. The DigTwinOps is char-
acterized by an execution engine that drives the cyber model to map
states of physical manufacturing. Therefore, a designer could validate
the controls before deploying them to the physical SMS.
Generally, the greatest difference between the simulation-based
design method and the digital twins-based design method lies in the
difference between conventional simulation-based virtual commis-
sioning and digital twins-based semi-physical commissioning. However,
it is still challenging to evaluate the environmental and human factors of
the equipment control in a digital twin model.
3.5. Digital twins-based intelligence model design of SMS
Intelligence is an important manifestation of the digital twins model.
It is a common vision that an SMS could realize self-perception, self-
decision, self-execution, autonomy, self-learning, and self-improvement
in its production process, including the optimal control of parameters in
the production process. Its optimization not only includes classic and
modern optimization control technology but also integrates big data and
articial intelligence technology. Without digital twin’s accurate
modeling of the actual system, the intelligence of an SMS cannot be
analyzed and reasoned. The industrial articial intelligence that exists in
SMS could be categorized into data-driven machine learning models and
computational optimization models.
3.5.1. Digital twins-based machine learning model design of SMS
The SMS is characterized by its complex, dynamic, and chaotic be-
haviors [74]. Machine learning saw fast developments in terms of not
only promising results but also usability to satisfy the demand for
high-quality customized products efciently. The intelligent control
models based on machine learning (e.g., naive Bayes, decision tree,
random forest, support vector machine, dimensionality reduction algo-
rithms, and neural network) exist in the different levels from
manufacturing equipment, production line, workshop, and factory to
the cross-factory level [75].
The selection and design of an appropriate machining learning sys-
tem to handle dynamic events and enable intelligent process control is
an essential issue in SMSD. The machine learning model design of SMS is
to develop and validate a variety of machine learning algorithms to
achieve sustainable performance. For instance, to account for the
multimodal and time-varying properties of system dynamics in SMSs,
Sun et al. [76] proposed a hybrid machine learning modeling approach
based on an extended/comprehensive state-space descriptive system, in
which the system control parameters are learned for responding
different running conditions. The machine learning systems can act as an
intelligent execution engine in a digital twin of the physical process.
However, if problems in the industry could not be properly modeled, this
technology will not yield true value but misleading.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
127
3.5.2. Digital twins-based computational optimization model design of SMS
Computational optimization in SMSs includes a variety of algorithms
in control and management optimization, such as large-scale optimiza-
tion, complexity theory, network optimization, and robustness/sensi-
tivity analysis. The computation cost and the numerical accuracy of
computational optimization algorithms are inuenced by the initiali-
zation of hyper-parameters, which are sensitive to application situa-
tions. However, the current mass individualization paradigm implies
signicant changes in SMS. In SMSD practice, designers are searching
for robust computational optimization algorithms to decrease the
sensitivity to the changes/disturbances as much as possible [77].
Therefore, the main goal of the computational optimization model
design of SMS is to nd a robust design that could handle a large variety
of system changes and disturbances. Leng et al. [78] proposed a digital
twins-driven warehousing optimization model for a large-scale auto-
mated high-rise warehouse under a mass individualization paradigm.
The digital twins technology provides a highly condent runtime envi-
ronment to simulate the complex system for searching robust compu-
tational optimization models.
The hybrid autonomy models that integrate data-driven machine
learning models and computational optimization models are also com-
mon practices. Real-time data-based digital twin systems could support
alternative articial intelligence, computational optimization, and
hybrid autonomy models to nd better production patterns/rules [79].
Generally, the abstraction of the optimization problems and their
coupling relationships contained in manufacturing systems is a critical
design step.
3.6. Digital twins-based performance design of SMS
Various blueprints of SMS have envisioned it to be highly exible,
intelligent, adaptive, and autonomous [80]. These SMS performance
goals should be evaluated and optimized in the design phase to deal with
more unpredictable changes and demands [81]. Scholars are making
efforts to improve system performance (e.g., quality, exibility, and
recongurability) based on digital twins technology.
3.6.1. Quality-oriented SMSD based on digital twins
Product quality is an essential metric to evaluate the performance of
a newly design SMS. On one hand, from the product perspective, the
mathematical modeling of tolerance specications is critical for
enabling overall tolerance analysis, which could be enhanced by the
digital twins technology. Luo et al. [82] proposed a parametric space
envelope-based tolerance modeling approach that can facilitate
Fig. 6. The rationale of digital twins-based system controls commissioning [70].
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
128
different computer-aided tools to satisfy individualized tolerancing re-
quirements. On the other hand, from the perspective of the
manufacturing system, the full-eld displacement perceiving of key
parts is desirable in the ultra-precision manufacturing domain. Zidek
et al. [83] discussed how to create a digital twin for an experimental
assembly system based on a belt conveyor system and an automatized
line for quality production check. Liang et al. [84] presented an online
full-eld displacement perceiving model by incorporating the matrix
completion theory with multi-points displacement tracking technology.
Experiments on the assembly of large aircraft showed that the maximum
displacement error could satisfy the ultra-precision manufacturing de-
mands efciently. However, these two studies did not address how to
use the digital twins technology to simulate and optimize the product
manufacturing quality in the early design and conguration stages of an
SMS.
3.6.2. Flexibility-oriented SMSD based on digital twins
In the mass customization/individualization/personalization para-
digm, the changeover of the SMS is extremely frequent, which calls for
fast system design/reconguration technologies as well as high exi-
bility of SMS. The SMS should be able to ensure the scalability and
compatibility of production to meet the needs of continuous product
development and services for users.
Many studies have been conducted on MSD that embody production
exibility for different product orders. In the blueprint of Industry 4.0,
each element in SMS (e.g., machine tool, workpiece, robot) is supposed
to be described with the Asset Administrative Shell technique. It is a
knowledge structure to model the functionality and interactions of each
element in SMS, which could be recognized as the key techniques of the
digital twins model. Seif et al. [85,86] created AASs to seamlessly
integrate assets into an SMS and presented a Model Factory system to
enhance the connectivity and the agility in a Mini Factory context.
Gallego et al. [87] proposed a holistic capacity management model for
manufacturing organizations to change or manage their capacity to
react to all potential demand scenarios. Generally, an effective
exibility-oriented SMSD solution should evaluate the upgrading of
products and the combination of different generations and architectures
of products [88].
3.6.3. Recongurability-oriented SMSD based on digital twins
The conventional methods to SMSD are usually limited in
manufacturing one single type/generation/family of products and suffer
from tremendous cost for redesigning/reconguration when new type/
generation/family of products/models are introduced. As the upgrading
and renewing of a product become faster, this “new-product-then-new-
system-design” logic is less ineffective. An optimized design for a given
product may be inexecutable for the updated generation of the product
since the recongurability and adaptivity of the design are unclear in
terms of the indeterministic and unpredictable product changes, which
makes it challenging to identify a design that is adaptive for different
type/generation/family of products/models.
Digital twins technology provides a solution for recongurability-
oriented SMSD. For instance, Leng et al. [89] proposed a digital
twins-driven fast reconguration method for SMSs based on the high
recongurability of the open-architecture machine tool. Jeon and Suh
[90] presented a smart factory architecture based on big data and digital
twins technology. Zhang et al. [91] presented 5D digital twin
model-based service function block strategies to increase the reusability
of functions/algorithms in an automatic reconguration of
robotics-based SMSs. Still, there is a need to integrate the computational
optimization techniques into the digital twins technology to improve the
design effectiveness and efciency so that recongurability can be
automatically and intelligently optimized [92].
3.6.4. Sustainability-oriented SMSD based on digital twins
With increasing pressures to reduce energy consumption,
manufacturers are faced with the challenge of increasing the sustain-
ability of the SMS [93]. Optimal machining parameter design/decision
is imperative for energy conservation, emission reduction, and cost
saving of SMS [94]. The SMS should have the ability to continuously
coordinate the relationship between multiple components according to
the characteristics of its internal structure, to ensure that the internal
coordination can always be maintained during the continuous devel-
opment process. However, little attention has been paid to formally
dene and optimize energy efciency and carbon emission reduction in
the design phase of SMS [95].
In Industry 4.0, the digital twins technology is a powerful tool to
understand and optimize the interaction between all relevant material,
energy, and resource ows in an SMS [96]. For instance, Barni et al. [97]
proposed a digital twins-based sustainability evaluation framework that
could simulate achievable results for different process plans. Wang et al.
[98] presented a digital twins-based system for supporting the design
and recovery decisions in the waste electrical and electronic equipment
recovery. Generally, the digital twins play a critical role in providing a
data-rich runtime for realizing the sustainability-oriented SMSD.
3.6.5. Security/Safety-oriented SMSD based on digital twins
Process safety is a serious issue for SMS control. The digital twin can
predict system failures [99], which could be used to optimize the process
safety and information ow of SMSs in the design phase. For instance,
Becue et al. [100] presented a digital twins-based Cyber-Factory blue-
print that could realize the iterative design of the safety function of
cobots to reinforce the ability on quality monitoring of avionics elec-
tronics for withstanding manipulation attacks. Kloibhofer et al. [101]
proposed a digital twins-based smart factory blueprint for operating
with secured product conguration and secured data communication.
Lee et al. [102] applied a system thinking methodology for imple-
menting the digital twins to improve the safety of manufacturing sys-
tems. A standardized language/ontology is critical for enabling the
reasoning to associate modules and processes in SMS. From the control
security perspective, fault tolerance is a critical capability of controllers
to adjust system behavior to cope with faults without manual inter-
vention. Designing active control strategies could improve the capability
of fault tolerance of SMS, which could also be called robustness of the
system. Rodriguez et al. [103] used digital twins technology to design
adaptive control strategies to increase the robustness of controllers.
In fact, it is difcult for many manufacturers to nd a real SMS that
can fully realize all the above goals. And in the actual SMS, due to the
diversity and contradiction of various goals, the improvement of some
performances will lead to the decline and deterioration of other per-
formance goals. For instance, exibility and productivity are usually
contradictory. Therefore, how to reasonably congure and coordinate
various goals to jointly realize the overall high performance of the SMS
is a complicated issue. Understanding the couplings among different
design objectives is a premise for solving this issue [104]. A unied
environment that could be utilized to decouple and coordinate different
design objectives is favorable to increase design efciency [105].
4. Digital twins-based SMSD models
Few existing models can fully meet the design requirements of SMSs,
while two models of digital twins-based SMSD are discussed as the po-
tential solution.
4.1. Quad-play CMCO model
A similarity transferring mechanism exists among the function
model, structure model, behavior model, control model, intelligence
model, and performance model of SMS. Using this similarity mechanism
could effectively improve design efciency. Liu et al. [106] proposed a
digital twins-based quad-play CMCO (Conguration design-Motion
planning-Control development-Optimization decoupling) architecture
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
129
for supporting the rapidly customized designing of ow-type SMSD
(Fig. 7). Conguration Model was dened as the topological layout and
resource conguration of the ow-type SMS. Motion Model was dened
as the motion form of equipment and the logistics form of WIP in the
ow-type SMS. Control Model was dened as the data acquisition and
processing model and control system structure of the ow-type SMS. The
optimization model was dened as the coupled optimization in the
execution of the ow-type SMS.
In CMCO, the Conguration Design dimension includes requirements
analysis, process plan, typology denition, layout plan, and the cong-
uration of equipment and buffer. The Motion Planning dimension in-
cludes dening the motion of the machine tool, the material handling,
and the buffering/caching. The Control Development dimension includes
designing the bi-directional sensory and controlling mechanism. The
Optimization Decoupling dimension includes solving the coupled process
optimization problems and exporting the execution engine to drive the
SMS operation. The dynamics of an SMS during its operation undoubt-
edly determine its performance. Usually, designers can only consider the
SMS in the design and development stage based on the limited cognition
of the dynamics, but actually, they use the static intellectual entity
thinking to investigate the SMS. Concurrently conducting the SMSD
tasks and developing a unied model to effectively imitated every
interaction and behavior of the manufacturing process are challenging.
To cope with this challenge, the CMCO model includes four typical
iterative design paradigms (Fig. 8) that could be implemented based on
practical design goals. The CMCO model is a bi-level architecture, in
which the “FW” and “FB” are two symbols to separate the physical
design level (including Conguration design and motion planning) and
logic design level (including Control development and optimization
decoupling). The CMCO model is an iterative "execution-analysis-
adjustment" optimization for balancing the static conguration scheme
and the dynamic execution engine. Therefore, the CMCO architecture is
a promising approach for effectively improving design efciency in DT-
SMSD.
4.2. Digital twin shop-oor model
As a basic constituent part of SMS, the design and optimization of the
shop oor are important. Qi and Tao [107] proposed the concept of
digital twin shop-oor and discussed the connotation, conguration
mechanism, operation mechanism, and key technologies of digital twin
shop-oor (Fig. 9). It mainly includes the design and conguration of a
physical shop-oor, a virtual shop-oor, and a shop-oor service system.
The physical shop oor refers to the physical entities (e.g., machine
tools) in the shop oor. The physical shop oor receives manufacturing
instructions and performs manufacturing tasks. To be able to understand
the status and consequence of manufacturing operations, including the
machines and production plans required to perform specic operations,
it is necessary to design and congure the embedded system for auto-
matic identication and data perception. The design of physical
shop-oor and virtual shop-oor refers to establishing high-precision
digital simulation models and interactions of various shop-oor
elements.
Digital twins could help realize the correlation and dynamic
adjustment between manufacturing planning and implementation, as
well as promote fault prediction, diagnosis, and maintenance in a digital
way [107]. To establish a digital twin workshop, it is necessary to pro-
vide modeling tools (e.g., Plant Simulation and Demo3D) and initialize
the digital twin model of physical shop-oor using a reference model
with static properties, motion scripts, control schemes, and communi-
cation interfaces. On this basis, the real-time operating performance of
the workshop could be simulated and predicted, including process
planning, manufacturing operations, and abnormal/error. After estab-
lishing the digital model, an interoperability channel between cyber-
space and physical space could be built from a data-driven view. The
design of the shop-oor service system includes establishing a data
channel for cyber-physical interoperability. The development of various
data service applications (such as the deep neural network) makes
manufacturing data valuable for operation decision support in the
shop-oor service system.
5. Key enabling technologies
The integration of various key enabling technologies such as the
Industrial Internet of Things, cloud computing, articial intelligence
algorithms makes the digital twin system more reliable and efcient for
SMSD. This section will review these key enabling technologies shown in
Fig. 10.
5.1. Industrial internet of things
The digital twin of a physical process allows simulations over the
process in a reliable environment [108]. The Industrial Internet of
Things (IIoT), as a network composed of massive various types of sensor
and actuator nodes to realize the dynamic perceiving of the physical
system, is a carrier for feeding the digital twin with high-quality in-situ
data for high-delity virtual modeling and simulation computing [109].
For instance, Chhetri et al. [110] proposed an IIoT-based model to
establish digital twins by employing indirect side-channels (to percept
the energy consumptions/emissions) of the SMS, which can precisely
predict the product quality. Li et al. [111] proposed a communication
architecture for connecting heterogeneous industrial automation sub-
systems in an Industry 4.0 SMS by integrating the Open Platform
Communications Unied Architecture (OPC UA) and Time-Sensitive
Networking technologies. Liu et al. [112] presented a cyber-physical
machine tools platform with standard and interoperable interfaces by
integrating the OPC UA and MTConnect technologies. Generally,
concurrently conducting the design and semi-physical commissioning
allows designers to use the IIoT data from the prototype SMS to enable
the digital twin system to validate the designs in advance in the early
development phase, and then carry out corresponding redesign or
adjustment decisions if inefciencies and errors are identied [113].
5.2. Multi-domain physical-chemical modeling
The SMS is a multi-eld physical (mechanical-electric-hydro-ther-
mal-magnetic-control, etc.) integrated system. The traditional way of the
Fig. 7. The architecture of the quad-play CMCO design model [106].
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
130
manufacturing system design is that designers in various elds design
their relatively independent subsystems separately and then integrate
them. Although there are overall considerations, discussions, and ne-
gotiations in the design process, it is still difcult to grasp the complex
coupling of various parts of the system. The models of various profes-
sional disciplines have been widely used in various aspects of SMS
design, but the models lack unied coding and can not be shared.
Modeling work is still in the "chimney" mode of information trans-
mission, forming model information silos without a suitable system
engineering workow. Based on the Unied Modeling Language, Model-
Based Systems Engineering (MBSE) is a common approach for
describing SMSs. The MBSE becomes the hub of the models of various
Fig. 8. Four patterns in the iterative design logic of the quad-play CMCO model [106].
Fig. 9. Conceptual model of digital twin shop-oor [107].
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
131
professional disciplines. For instance, a software provider developed a
corresponding support SysML (System Modeling Language) tool
modeling language, which could be integrated with existing professional
analysis software such as Finite Element Analysis and CAD. Designers
from all parties carry out requirements analysis, system design, simu-
lation, and other work around the system model, to facilitate the
collaborative work of the SMSD team. Therefore, a new generation of
digital design methods based on multi-domain modeling and multi-
disciplinary simulation knowledge is needed.
The digital twin model of an SMS may contain models reecting
physical dynamics/characteristics in different dimensions, such as uid,
structural, thermodynamic, stress, fatigue damage, and material state (e.
g., stiffness and strength). How to correlate these dimensions is critical to
the establishment of digital twins. Models and data are the cores of the
digital twins. The digital twin model should reect the inner operating
law of the physical system. The profound understanding of the nature of
the laws, phenomena, and processes occurring in the SMSs would allow
one to develop and apply relevant multi-domain physical-chemical
models. For example, the construction of a digital twin model of laser
processing equipment should go deep into the mechanism of the inter-
action between laser and material.
If the digital twin model has the simulation ability including the
power/spectrum/force/heat distribution and coupling relationships, it
can guarantee the output design solution with better processing per-
formance. Lin et al. [114] proposed a cyber-physical SMS simulation
architecture based on the integration of aggregate software controllers
in cyberspace and distributed hardware controllers in physical space.
Zhang et al. [115] proposed an information modeling approach for CPPS
based on the standardized format AutomationML. Ghosh et al. [116]
adopted hidden Markov models for the construction and abstraction of
dynamics in digital twins of the manufacturing systems. Malykhina et al.
[117] proposed an information-measuring system model to extract
knowledge from massive manufacturing information in digital twin
models. Zheng et al. [118] proposed a three-level Ele-
ment-Behavior-Rule modeling method for SMS. Kim et al. [119] pro-
posed a three-stage modeling environment with an automatic module
synthesis engine for supporting user-level customizing of an SMS.
Despite the remarkable achievements, further improvements of the
mathematical tools are associated with the need to improve the ef-
ciency of multi-threaded calculations in solving multi-parametric design
optimization problems of SMSs [120].
5.3. Real-time synchronization and discrete event simulation
The model simulation is the core technology of creating/running a
digital twin and ensuring an effective closed-loop between the digital
twin and corresponding physical entity. Simulation is a technology that
simulates the physical world by transforming a model containing
deterministic laws and mechanisms into the software. As long as the
model is correct and fed with complete context information and envi-
ronmental data, it can accurately reect the characteristics and pa-
rameters of the physical world. The real-time cyber-physical
synchronization is a major difference between the digital twins method
and conventional simulation methods. Network latency is a key metric
to evaluate the synchronization capability of a digital twin system. For
instance, Szabo et al. [121] suggested a 5 G ultra-reliable low-latency
communication solution for establishing an accurate digital twin of a
robot cell. Leng et al. [122] utilized a software-comprehensible PLC
model to connect the cyber models, physical PLC, and other systems.
Discrete Event Simulation (DES) is useful in the process ow
designing and planning of SMSs. Dos et al. [123] analyzed the feasibility
of the DES to support building the digital twins of non-automated pro-
cesses. Silva et al. [124] presented a Petri Net-based DES method to
enable digital twins of SMSs. Sierla et al. [32] presented an automatic
assembly planning model using UML and a 3D simulation environment
employing the Automation Markup Language. To get rid of the issue of
poor-quality data from heterogeneous sources, Mieth et al. [125]
introduced a real-time indoor localization system to provide reliable
manufacturing data into the digital twin. To obtain a deep cognition of
material properties of pharmaceutical powders in the manufacturing
stage, Bhalode and Ierapetritou [126] incorporated material calibration
into a DES-based feeder simulation model to achieve a particle-level
understanding of the impacts of material properties on feeder perfor-
mance. These digital twin methods provided deep micro/meso/macro
insights into the system. Still, automatic simulation model generation is
critical for material ow simulation in the discrete-type SMS [127].
5.4. Visualization & virtual/augmented reality
SMSD is a complicated and prone-to-error human-computer inter-
action process. Traditional 3D simulation systems usually cannot sup-
port the designers to experience the nal SMS as an end-user in an
immersive manner. Some intelligent systems with virtual reality (VR),
augmented reality (AR) or mixed reality (MR) have been proposed to
support the engineers to design and operate. The boundary between
Fig. 10. An overview of key enabling technologies in DT-SMSD.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
132
"virtual" and "real" becomes blurred in such a human-machine system
supplemented by VR/AR/MR technologies [128]. Scholars are making
efforts to improve design efciency by enhancing the digital twin system
with the VR/AR/MR technologies. For instance, Han et al. [129]
developed a framework to extract and visualize 3D models that could be
used in the AR/VR of an industrial plant in real time. Fera et al. [130]
combined wearable sensors and simulation tools for analyzing
manufacturing performances of SMS at varying demands. Malik et al.
[45] developed a unied VR-enhanced human-robot simulation frame-
work to estimate the human-robot cycle times and thereby helping de-
signers to make process plans, optimize system layouts, and program
controls. Perez et al. [131] presented an immersive VR-enhanced pro-
cess automation design methodology to create a digital twin of SMS for
guiding its physical implementation. Generally, introducing immersive
VR/AR/MR-enhanced simulation could provide safe cyberspace for
commissioning and validating thus easing the SMSD.
5.5. Big data analytics and industrial articial intelligence
The direct goal of SMSD is to satisfy the individualized requirements,
such as the site space constraint, manufacturing capacity, product
quality, and production efciency. The individualization requirements
from the product domain are translated to the SMS domain step-by-step.
There exists a similarity in translating the six dimensions of the FSBCIP
model of SMS, which can be utilized to improve the design efciency.
The development of comprehensive knowledge models in the digital
twin of manufacturing processes/operations is critical to make use of
these design similarities and make accurate predictions on the overall
process dynamics for each design parameter/decision [132].
The conventional knowledge-based systems could be categorized
into case-based reasoning systems and rule-based reasoning systems. A
better way to learn in the digital intelligence age is to effectively and
efciently discover hidden knowledge and patterns in large-scale case
data to enhance design efciency [133]. There are so many underlying
interrelated factors and correlations in a complex SMS that it is often
difcult for domain experts to realize, which could be, however,
captured by big data analytics and deep learning methods. In short, the
digital twin model with learning ability can make the SMSD more and
more intelligent. The growing availability of data coming from historical
design cases and contextual operations enables data analytics towards
the creation of added value [134]. Ringsquandl et al. [135] introduced a
representation learning model for synchronizing semantics from existing
manufacturing knowledge graphs and time-dependent operational data.
Huerkamp et al. [136] presented a cyber-physical production
system-based framework that integrates simulation and machine
learning techniques to form a smart digital twin. A
Finite-Element-Method surrogate model was established for predicting
the interface bond strength quality independence of the process settings.
Cavalcante et al. [137] discussed how and when machine learning and
simulation can be combined to create digital twins for supporting the
design of risk mitigation strategies in supply disruption management
models.
However, considering the randomness in the manufacturing process,
the proper design of digital twins remains difcult to enable its
robustness and adaptability [138]. For instance, even small deviations of
the temperature settings may result in great defects in the composite
structure [136]. Deep learning (e.g., deep reinforcement learning), as
one of the most promising articial intelligence technologies [139],
could be used to stimulate and expand human intelligence in the SMSD.
Today’s digital twin models are mostly limited to the understanding and
simple reasoning of design decisions of SMS. In the future, incorporating
more high-level industrial articial intelligence algorithms into the
digital twin models could be expected to realize the ability to auto-
matedly create and generate new design solutions.
5.6. Industrial blockchains and smart contracts
Since the SMSD involves multi-eld (e.g., mechanical, electric,
hydro, thermal, magnetic, and control) knowledge and expertise, it is
necessary to establish a reliable and interconnected runtime that enable
designers from all parties to collaboratively carry out analysis, design,
and simulation work around one unied model. The existing digital twin
systems are mostly centralized and suffer from the shortcomings of low
data auditability and traceability. The data security in this inter-
connected runtime is a critical issue under cyber-attack [140]. The
traditional cybersecurity methods need to evolve to address the attacks
that threaten the collaboration network [141].
Blockchains provide a new solution for secure collaborative SMSD.
Blockchains use cryptographic technology and distributed consensus
protocols to ensure the security of network transmission and access, and
thereby to enhance the cyber-credit among various participants [142].
Much research on blockchains-empowered smart manufacturing appli-
cations in Industry 4.0 had been conducted [143,144], including
blockchains-secured data sharing in collaborative DT-SMSD. Hasan et al.
[24] proposed a blockchain-based digital twin solution to ensure
auditability, accessibility, and traceability of design data. They used
smart contracts to manage and trace data transactions in digital twins.
Chirico et al. [145] exploited the assume-guarantee reasoning method
by using smart contracts to model a virtual production line for efciently
and accurately simulating the manufacturing operations. Leng et al.
[122] proposed an iterative bi-level hybrid intelligence framework
called ManuChain to avoid the inconsistency between planning and
execution in individualized SMS. Spellini et al. [146] argued to use a
smart contract-based design mechanism to cope with the increased
automation of SMSD. However, introducing blockchain technologies
into SMSD has been greatly hindered by practical issues such as system
scalability and exibility [147], which should be further addressed in
the future.
5.7. Cloud computing and web services for distributed digital twins
Besides blockchains, another kind of key enabling technology to
realize the collaborative SMSD is the cloud computing and web services
for supporting the analysis, design, and simulation work in a distributed
digital twin environment. In the context of collaborative SMSD, both the
designers and computing resources are geographically distributed
[148]. The cloud computing and web service systems allow to cooper-
atively share designing activities across multiple similar units located in
remote areas [149]. The demand for high-performance computing for
simulation work in SMSD is another driver to introducing cloud
computing technology, which could provide low-cost computing
resource sharing, dynamic allocation, and exible expansion [150].
Integrating web services with cloud computing enables designers to
conduct the SMSD work remotely and distributed.
6. Digital twins-based SMSD cases
Under the increasingly diverse product demands and the evolution of
technologies, the manufacturing paradigm has evolved from mass pro-
duction, mass customization to mass individualization/personalization.
The corresponding manufacturing systems are being transformed from
dedicated production lines, exible/recongurable manufacturing sys-
tems, to SMSs. Table 2 provides an overview of manufacturing system
design cases for different manufacturing paradigms. The exibility,
autonomy, and sustainability of SMS are critical for enabling the low-
cost mass-individualized production. Open architecture is recognized
as a potential solution of SMS architecting, which is characterized by a
standardized platform and swappable individualized modules [89].
Designers can exibly redesign and recongure the SMS following
process planning by swapping individualized modules onto the platform
[151]. As the open architecture SMSD paradigm is enriched with the
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
133
wild innovation of crowds in the community, the scale, and variety of
module databases will grow signicantly. The open architecture in
SMSD is supposed to go into a deeper open-source level ultimately
instead of simply build unied interfaces. Simultaneously, the intellec-
tual property protection consensus will be updated accordingly to make
it more suitable for the distributed design and open innovations.
The SMS will suffer from unexpected situations with different forms
and data [154]. The design elements, design goals, and major challenges
of different levels (e.g., machine tool/center, manufacturing units, pro-
duction line, workshop, factory) of SMSD differ from each other. Table 3
provides an overview of different levels of SMSD. The SMSD at the
production line level has been intensively researched. For instance,
Gericke et al. [155] proposed a digital twins-based optimization model
for decreasing the bottleneck/delay in a water-bottling production line.
Liu et al. [156] presented a digital twins-driven approach for fast indi-
vidualized designing of the furniture production line. Further, Liu et al.
[106] utilized the designing of a hollow glass production line as a case to
demonstrate the formerly-mentioned CMCO design model. Actually, the
proposed fundamental SMSD solution for this hollow glass production
line mainly includes the design of a computational decoupling model for
a coupled optimization of "grouping-blanking-loading-planning" that
affects the production performance. The types of the workshop in
Table 3 could be further categorized into job shop [29,123,127], ow
shop [5,156], and open shop. For instance, Leng et al. [151] presented a
digital twins-driven manufacturing cyber-physical system for a
board-type product manufacturing ow-shop. Despite the remarkable
achievements, the existing DT-SMSD model is still insufcient to meet
the requirements of precisely and completely interconnecting and fusing
the virtual model/data and physical SMS [157]. An important step to-
wards the success of DT-SMSD is the establishment of practical standards
and reference architectures [158].
7. Research directions
This section suggests four directions for future research on digital
twins-based SMSD.
7.1. Integrating the SMSD with product development
From the product development perspective, SMSD could be catego-
rized into the SMSD for a given product and the SMSD for a new product.
SMSD and new product development are highly coupled [170], which,
however, are conventionally sequentially conducted without coordina-
tion [171]. The concept of digital twins promises high potentials for
integrating product development and SMSD [172]. Integrating the SMS
function, structure, behavior, and control modeling with product design
could help to achieve a balance between product cost and performance
efciently and effectively from a holistic product lifecycle view [173].
This direction has received wide attention. Kuliaev et al. [174] pre-
sented a digital twins-based modeling & assembly planning framework
for product-centric design. Eddy et al. [175] introduced an optimal
composition of simulation models in the product development for more
precisely predicting the manufacturing system performance based on
the digital twins model, which could thus cut down errors in SMS
commissioning at the product development phase. Ma et al. [164] pro-
posed a digital twins-driven production life-cycle management system
framework to support a CPS of manufacturing workshop, including the
design and manufacturing stage. Loizou et al. [152] proposed a platform
that supports the consumers and the manufacturers to co-design prod-
ucts based on the integration of view-based modeling, visualization, and
monitoring technologies, and thereby creating digital twin models of
SMS towards more fact-based manufacturing decisions. Schuetzer et al.
[176] proposed a smart digital twin model to integrate the product twin
and the twin of its development process.
Actually, product development involves supply chain integration. A
product goes through a group of manufacturers involved in a series of
coordinated manufacturing activities before nished. Although the
digital twins technology offers great superiorities in this research di-
rection, however, many manufacturers develop their digital twins in
their own business processes. Existing models for the implementation of
digital twins are limited in isolated physics domains inside one enter-
prise. Manufacturers need continually adapt their supply chain to
changing product designs. Therefore, the development of digital twins
for automating businesses towards supply chain integration is critical
but in absence. Currently, the lack of real-time data available and
responsive designing/planning systems make the adaptation of digital
twins difcult. A unifying of digital twin models in integrating smart
product development and supply chain integration with SMSD is
promising in future product R&D since it will allow us to holistically
optimize the lifecycle activities of the product [177].
7.2. Standards for digital twins-based SMSD
Implementing digital twins in small and medium-size manufacturers
is time-consuming since the desired standards are unclear [178].
Moreover, there are often large differences between actual physical
manufacturing and the cyber system. Standards should be developed to
answer the question of how to realize a good DT-SMSD and to dene the
Table 2
Manufacturing system design cases for different manufacturing paradigms.
Paradigms Mass
Production
Mass Customization Mass
Individualization
Typical Mfg.
Systems
Dedicated
Production
Line
Flexible/
Recongurable
Manufacturing System
Smart
Manufacturing
System
Design Goal Efciency and
Quality, etc.
+Flexibility and
Recongurability, etc.
+Sustainability,
etc.
Mfg. System
Architecture
Unied
Architecture
Modular Architecture +Open
Architecture
IP Protection Closed Source Closed Source +Open Source
Major
Challenges
System
Reliability, etc.
+Module Reusbility,
etc.
+Cost Effectivity,
etc.
Typical
Industries
Mobile Phone,
Chemicals
Furniture, Automobile Apparel, Printed
Circuit Board
Ref. – [119,152] [80,89,122,151,
153]
Table 3
Different levels of smart manufacturing system design.
Levels Machine Tool/Center Mfg. Units Production Line Workshop Factory Distributed System
Design
Elements
Parts/Components,
Controls, etc.
+Robots, AGVs, etc. +Layout, Typology,
Performance, etc.
+Material
Handling, etc.
+Load Balancing,
Intelligence, etc.
+Coordination
Mechanism, etc.
Design Goal Product Quality +Autonomy +Recongurability +Productivity +Flexibility +Sustainability
Typical
Blueprint
Open Architecture
Machine tools, etc.
Autonomous Mfg.
Units, etc.
Recongurable Production
Line, etc.
Smart workshop,
etc.
Smart Factory, etc. e.g., Decentralized
Autonomous Mfg.
Major
Challenges
Manufacturing Accuracy,
etc.
Deadlock
Prevention, etc.
Bottleneck Prevention, etc. Adaptive Layout,
etc.
Compatibility with New
Products, etc.
Consensus Efciency, etc.
Ref. [39,41,64,75,112,159,
160]
[56,154,155,161,
162]
[5,36,130,145,163] [151,156,164] [80,86,87,90,100,132,
157,165–168]
[149,169]
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
134
evaluation criteria. Such standards could be used to calibrate the dif-
ference/gap between the DT-SMSD results and the nal operating per-
formance of real SMS. Even if the difference/gap is large, the results can
be considered reasonable as long as the gap is constant and can be
calculated and rened by a certain data or method. This "precision" is
determined by standards instead of software, calculation methods, or
processing methods. With the help of DT-SMSD standards, the calibra-
tion deviation of design results compared to the nal operating perfor-
mance of real SMS could be identied and revised. The ultimate goal of
developing DT-SMSD standards is to make sure each design result can be
reproduced and traced, instead of pursuing the design results to look
more like the real SMS. In another word, if we cannot keep the consis-
tency of design results, this DT-SMSD method is unreliable.
Developing DT-SMSD standards is a process to establish the specic
provision of technical guidelines and enforcement documents, including
technology principles, premise assumptions, processing methods, soft-
ware tools, model denitions, boundary conditions, design steps, vari-
able controls, result evaluations, and result calibrations in DT-SMSD.
The following evaluation dimensions may be includes in the developing
DT-SMSD standards: 1) digital thread breadth, 2) data synchronization
frequency, 3) articial intelligence type/level, 4) simulation capabil-
ities, 5) reference model richness, and 6) user interface [179]. Compared
with the use of digital twin-based CAD software, DT-SMSD standards are
much more complex and more difcult to form, which requires not only
a lot of test verication of the designing process but also a lot of error
calibration work.
7.3. Data schema and reference architecture of digital twins-based SMSD
Critical to the creation and evaluation of a DT-SMSD solution is a
digital thread that allows the continuous collection and linking of a wide
range of relevant data and analytical models throughout the lifecycle of
a physical SMS. However, the lack of a universal data schema for syn-
chronizing, coordinating, and integrating cross-domain digital twin
models of SMS results in a massive investment. It is critical to capture
scheme variances in the specic domains of electrics, mechanics, ergo-
nomics, and software and build a consistent interface standard and data
model of the digital twins [165]. This direction has also received
attention. Park et al. [180] designed a reference activity model-based
data schema for supporting the development of a cloud-based digital
twin manufacturing system. Kumar et al. [181] presented a federated
multimodal data platform integrating modern semantic and big data
technologies.
Although signicant strides have been made to the digital twin ar-
chitecture development [153,182–186], there is still a need for
comprehensive reference architectures that can help guide the imple-
mentation of DT-SMSD. The methodological, technological, operative,
and business aspects of developing, operating, and evaluating DT-SMSD
solutions should be dened in the reference architecture. The reference
architecture answers the question of where and by what means to use
digital twins. How to dene the design task in which digital twins can
exert the greatest value and ensure correct decision guidance when
necessary is the mission of establishing reference architecture. The
DT-SMSD reference architecture is supposed to address requirements of
reusability, interoperability, interchangeability, maintainability, exten-
sibility, and autonomy across the entire digital twin lifecycle. In addi-
tion, considering that the maturity of enterprise digital twin system is
gradually evolving, the reference architecture should not only stipulate
"what digital twins implementation work should be done when the
SMSD is done with fully-mature digital twin system", but also stipulate
"what design functions should be tailored at different maturity levels of
the digital twin system".
7.4. Servitization of digital twins-based SMSD
Servitization of SMSD is a trend in industries, which adopts service
business models to offer not only a single design solution but also ser-
vices extended to the operation phase to satisfy individualized adjust-
ment needs. This servitization trend has driven an IT-driven business
paradigm. Similar to the concept of service-oriented manufacturing and
smart product-service system [187,188], the servitization of DT-SMSD
aims to provide on-demand design capabilities to gain individualized
satisfaction with less environmental impact by utilizing data-driven
digital twins as the media and tool. Servitization of DT-SMSD is a
customer-oriented trend that achieves the continuous conversion of
manufacturers’ individualized requirements into the elements, features,
and parameters of SMS. Detailed knowledge of SMSD must be provided
[189]. Considering the dynamic behavior and interactions in SMS [190],
the trend is towards mining decision-support knowledge and patterns
from large-scale historical case data to enhance SMSD services [191]
and context-aware value co-creation [192].
8. Concluding remarks
This article surveys how the digital twins technologies are integrated
into and promote the SMS design based on a literature search in the Web
of Science database. Based on the denitions of SMSD and the advantage
of DT-SMSD, a Function-Structure-Behavior-Control-Intelligence-
Performance (FSBCIP) framework is proposed to review the major
steps of DT-SMSD. Major design steps in SMSD that could be enhanced
by the digital twins technology are reviewed. New blueprint models
including the CMCO design architecture and digital twin shop-oor are
discussed, yet there still is an urgent need for further DT-SMSD meth-
odology development. Key enabling technologies such as IIoT, multi-
domain physical-chemical modeling, virtual reality, data analytics, in-
dustrial articial intelligence, blockchains, and cloud computing for
supporting the DT-SMSD are analyzed. Design cases in different
manufacturing paradigms and different manufacturing system levels are
presented. Based on the insights obtained from the analyses and dis-
cussions, we then suggest four future directions of research in DT-SMSD.
It is expected that this survey will shed new light on urgent industrial
concerns in developing new SMSs in the Industry 4.0 era.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
This work was supported by the National Key R&D Program of China
under Grant No. 2018AAA0101704 and 2019YFB1706200; the National
Natural Science Foundation of China under Grant No. 52075107 and
U20A6004; the Science and Technology Planning Project of Guangdong
Province of China under Grant No. 2019A1515011815,
2019B090916002, and 2019A050503010.
References
[1] Fowler JW, Rose O. Grand challenges in modeling and simulation of complex
manufacturing systems. Simulation 2004;80(9):469–76.
[2] Rooks B. A shorter product development time with digital mock-up. Assem
Autom 1998;18(1):34–8.
[3] Wu D, et al. Cloud-based design and manufacturing: a new paradigm in digital
manufacturing and design innovation. Comput Des 2015;59:1–14.
[4] Mahmoud MA, Grace J. A generic evaluation framework of smart manufacturing
systems. Procedia Comput Sci 2019;161:1292–9.
[5] Zhang H, et al. A digital twin-based approach for designing and multi-objective
optimization of hollow glass production line. IEEE Access 2017;5:26901–11.
[6] Wang K, Lee T, Hsu Y. Revolution on digital twin technology-a patent research
approach. Int J Adv Manuf Technol 2020;107(11–12):4687–704.
[7] Nujoom R, Mohammed A, Wang Q. A sustainable manufacturing system design: a
fuzzy multi-objective optimization model. Environ Sci Pollut Res - Int 2018;25
(25):24535–47.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
135
[8] Choi S, et al. Cyber-physical systems: a case study of development for
manufacturing industry. Int J Comput Appl Technol 2017;55(4):289–97.
[9] Frontoni E, et al. In: DePaolis LT, Bourdot P, DePaolis LT, Bourdot P, editors.
Cyber physical systems for industry 4.0: towards real time virtual reality in smart
manufacturing, in lecture notes in computer science. CHAM: Springer
International Publishing Ag: Cham; 2018. p. 422–34.
[10] Zakoldaev DA, et al. In: Sapozhkov SB, Sapozhkov SB, editors. Digital forms of
describing Industry 4.0 objects, in IOP Conference series-materials science and
engineering. Bristol: Iop Publishing Ltd; 2019.
[11] Negri E, Fumagalli L, Macchi M. A review of the roles of Digital Twin in CPS-
based production systems. Procedia Manuf 2017;11:939–48.
[12] Shao GD, Kibira D. Digital manufacturing: requirements and challenges for
implementing digital surrogates. In: Winter Simulation Conference Proceedings;
2018. p. 1226–37.
[13] Valckenaers P. Perspective on holonic manufacturing systems: PROSA becomes
ARTI. Comput Ind 2020;120(103226).
[14] Barricelli BR, Casiraghi E, Fogli D. A survey on digital twin: denitions,
characteristics, applications, and design implications. IEEE Access 2019;7:
167653–71.
[15] Zobel-Roos S, et al. Accelerating biologics manufacturing by modeling or: is
approval under the qbd and pat approaches demanded by authorities acceptable
without a digital-twin? Processes 2019;7(942).
[16] Chen Y, et al. Digital twins in pharmaceutical and biopharmaceutical
manufacturing: a literature review. Processes 2020;8(10889).
[17] Tao F, et al. Digital twin in industry: state-of-the-art. IEEE Trans Ind Inform 2019;
15(4):2405–15.
[18] Lu Y, et al. Digital twin-driven smart manufacturing: connotation, reference
model, applications and research issues. Robot Comput Integr Manuf 2020;61
(101837).
[19] Yan J, Li B. Research hotspots and tendency of intelligent manufacturing. Chin Sci
Bull-Chin 2020;65(8):684–94.
[20] Cimino C, Negri E, Fumagalli L. Review of digital twin applications in
manufacturing. Comput Ind 2019;113(103130).
[21] Qi Q, Tao F. Digital twin and big data towards smart manufacturing and Industry
4.0: 360 degree comparison. IEEE Access 2018;6:3585–93.
[22] Tao F, et al. Digital twins and cyber-physical systems toward smart
manufacturing and Industry 4.0: correlation and comparison. Engineering 2019;5
(4):653–61.
[23] Konstantinov S, et al. The cyber-physical E-machine manufacturing system:
virtual engineering for complete lifecycle support. Procedia CIRP 2017;63:
119–24.
[24] Hasan HR, et al. A blockchain-based approach for the creation of digital twins.
IEEE Access 2020;8:34113–26.
[25] Qian L, Gero JS, et al. Function-behavior-structure paths and their role in
analogy-based design. AI EDAM 1996;10(4):289–312.
[26] Zhang Y, et al. Intelligent feature recognition for STEP-NC-compliant
manufacturing based on articial bee colony algorithm and back propagation
neural network. J Manuf Syst 2021. https://doi.org/10.1016/j.
jmsy.2021.01.018).
[27] Zhao P, et al. The modeling and using strategy for the digital twin in process
planning. IEEE Access 2020;8:41229–45.
[28] Uhlemann T, Lehmann C, Steinhilper R. The digital twin: realizing the cyber-
physical production system for industry 4.0. Procedia CIRP 2017;61:335–40.
[29] Lu Y, et al. An IoT-enabled simulation approach for process planning and
analysis: a case from engine re-manufacturing industry. Int J Comput Integr
Manuf 2019;32(4–5SI):413–29.
[30] Liu J, et al. Dynamic evaluation method of machining process planning based on
digital twin. IEEE Access 2019;7:19312–23.
[31] Liu J, et al. Dynamic design method of digital twin process model driven by
knowledge-evolution machining features. Int J Prod Res 2021:1–19. https://doi.
org/10.1080/00207543.2021.1887531.
[32] Sierla S, et al. Automatic assembly planning based on digital product descriptions.
Comput Ind 2018;97:34–46.
[33] Haber Guerra R, et al. Digital twin-based optimization for ultraprecision motion
systems with backlash and friction. IEEE Access 2019;7:93462–72.
[34] Guo D, et al. Digital twin-enabled graduation intelligent manufacturing system
for xed-position assembly islands. Robot Comput Integr Manuf 2020;63
(101917).
[35] Guo D, et al. Graduation Intelligent Manufacturing System (GiMS): an Industry
4.0 paradigm for production and operations management. Ind Manag Data Syst
2020.
[36] Liu J, et al. A digital twin-based approach for dynamic clamping and positioning
of the exible tooling system. Procedia CIRP 2019;80:746–9.
[37] Wang P, et al. Digital twin-driven system for roller conveyor line: design and
control. J Ambient Intell Humaniz Comput 2020.
[38] Vatankhah Barenji A, et al. A digital twin-driven approach towards smart
manufacturing: reduced energy consumption for a robotic cellular. Int J Comput
Integr Manuf 2020.
[39] Cai Y, et al. Sensor data and information fusion to construct digital-twins virtual
machine tools for cyber-physical manufacturing. Procedia Manuf 2017;10:
1031–42.
[40] Cai Y, Wang Y, Burnett M. Using augmented reality to build digital twin for
recongurable additive manufacturing system. J Manuf Syst 2020;56:598–604.
[41] Tong X, et al. Real-time machining data application and service based on IMT
digital twin. J Intell Manuf 2020;31(5):1113–32.
[42] Botkina D, et al. Digital twin of a cutting tool. Procedia CIRP 2018;72:215–8.
[43] BAINES TS, KAY JM. Human performance modelling as an aid in the process of
manufacturing system design: a pilot study. Int J Prod Res 2002;40(10):2321–34.
[44] Malik AA, Brem A. Digital twins for collaborative robots: a case study in human-
robot interaction. Robot Comput Integr Manuf 2021;68:102092.
[45] Malik AA, Masood T, Bilberg A. Virtual reality in manufacturing: immersive and
collaborative articial-reality in design of human-robot workspace. Int J Comput
Integr Manuf 2020;33(1):22–37.
[46] Graessler I, Poehler A. Integration of a digital twin as human representation in a
scheduling procedure of a cyber-physical production system. IEEE; 2017.
[47] Latif H, Starly B. A simulation algorithm of a digital twin for manual assemble
process. Procedia Manuf 2020;48:932–9.
[48] Greco A, et al. Digital twin for monitoring ergonomics during manufacturing
production. Appl Sci 2020;10(775821).
[49] Graessler I, Poehler A. Intelligent control of an assembly station by integration of
a digital twin for employees into the decentralized control system. Procedia
Manuf 2018;24:185–9.
[50] Malik AA, Bilberg A. Digital twins of human robot collaboration in a production
setting. Procedia Manuf 2018;17:278–85.
[51] Frazzon EM, et al. Manufacturing networks in the era of digital production and
operations: a socio-cyber-physical perspective. Annu Rev Control 2020;49:
288–94.
[52] Buˇ
ckov´
a M, et al. Designing of logistics systems with using of computer
simulation and emulation. Transp Res Procedia 2019;40:978–85.
[53] Wang W, Zhang Y, Zhong RY. A proactive material handling method for CPS
enabled shop-oor. Robot Comput Integr Manuf 2020;61(101849).
[54] Banyai A, et al. Smart cyber-physical manufacturing: extended and real-time
optimization of logistics resources in matrix production. Appl Sci 2019;9(12877).
[55] Yerra VA, Pilla S. IIoT-enabled production system for composite intensive vehicle
manufacturing. SAE Int J Engines 2017;10(2):209–14.
[56] Zhu CY, Pires JN, Azar A. A novel multi-brand robotic software interface for
industrial additive manufacturing cells. Ind Robot 2020;47(4):581–92.
[57] Cao X, Zhao G, Xiao W. Digital Twin-oriented real-time cutting simulation for
intelligent computer numerical control machining. Proceed Instit Mech Eng Part
B 2020. 0954405420937869.
[58] Zhao R, et al. Digital twin-driven cyber-physical system for autonomously
controlling of Micro Punching system. IEEE Access 2019;7(1):9459–69.
[59] Gohari H, Berry C, Barari A. A digital twin for integrated inspection system in
digital manufacturing. IFAC Papersonline 2019;52(10):182–7.
[60] Tabar RS, et al. Efcient spot welding sequence optimization in a geometry
assurance digital twin. J Mech Des 2020;142:10200110.
[61] Schleich B, et al. Geometrical variations management 4.0: towards next
generation geometry assurance. Procedia CIRP 2018;75:3–10.
[62] Schleich B, et al. Shaping the digital twin for design and production engineering.
CIRP Ann Manuf Technol 2017;66(1):141–4.
[63] Cerrone A, et al. On the effects of modeling as-manufactured geometry: toward
digital twin. Int J Aerosp Eng 2014;(439278). 2014.
[64] Montavon B, et al. A digital perspective on machine tool calibration. Int J Autom
Technol 2020;14(3SI):360–8.
[65] Berger T, et al. Arezzo-exible manufacturing system: a generic exible
manufacturing system shop oor emulator approach for high-level control virtual
commissioning. Concurr Eng Res Appl 2015;23(4):333–42.
[66] Julius R, et al. Transformation of GRAFCET to PLC code including hierarchical
structures. Control Eng Pract 2017;64:173–94.
[67] Niang M, et al. A methodology for automatic generation, formal verication and
implementation of safe PLC programs for power supply equipment of the electric
lines of railway control systems. Comput Ind 2020;123:103328.
[68] Orive D, et al. Fault injection in Digital Twin as a means to test the response to
process faults at virtual commissioning. In: IEEE International Conference on
Emerging Technologies and Factory Automation-ETFA; 2019. p. 1230–4.
[69] Toivonen V, et al. The FMS Training Center - a versatile learning environment for
engineering education. Procedia Manuf 2018;23:135–40.
[70] Leng J, et al. Digital twins-based remote semi-physical commissioning of ow-
type smart manufacturing systems. J Clean Prod 2021;306:127278.
[71] Talkhestani BA, Schloegl W, Weyrich M. Synchronization of digital models -
Application of an anchor-point method for production cells. ATP Edition 2017;
(7–8):62–9.
[72] Talkhestani BA, et al. Consistency check to synchronize the Digital Twin of
manufacturing automation based on anchor points. Procedia CIRP 2018;72:
159–64.
[73] Kang S, Chun I, Kim H. Design and implementation of runtime verication
framework for cyber-physical production systems. J Eng 2019;(2875236). 2019.
[74] Leng J, et al. Contextual self-organizing of mass individualization process under
social manufacturing paradigm: a cyber-physical-social system approach. Enterp
Inf Syst 2020;14(8):1124–49.
[75] Wojcicki J, Bianchi G. A Smart Spindle Component concept as a standalone
measurement system for Industry 4.0 Machine Tools. New York, USA: IEEE; 2020.
p. 278–82.
[76] Sun B, et al. A comprehensive hybrid rst principles/machine learning modeling
framework for complex industrial processes. J Process Control 2020;86:30–43.
[77] Orosz T, et al. Robust design optimization and emerging technologies for
electrical machines: challenges and open problems. Appl Sci 2020;10(665319).
[78] Leng J, et al. Digital twin-driven joint optimisation of packing and storage
assignment in large-scale automated high-rise warehouse product-service system.
Int J Comput Integr Manuf 2019:1–18.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
136
[79] Lugaresi G, Travaglini D, Matta A. In: A LEGO (R) Manufacturing System as
Demonstrator for a Real-Time Simulation Proof of Concept, in Winter Simulation
Conference Proceedings; 2019. p. 2025–36.
[80] Park KT, et al. Design and implementation of a digital twin application for a
connected micro smart factory. Int J Comput Integr Manuf 2019;32(6):596–614.
[81] Gu P, Rao HA, Tseng MM. Systematic design of manufacturing systems based on
axiomatic design approach. CIRP Ann-Manuf Technol 2001;50(1):299–304.
[82] Luo C, et al. A framework for tolerance modeling based on parametric space
envelope. J Manuf Sci Eng-Trans ASME 2020;142(0610076).
[83] Zidek K, et al. Digital twin of experimental smart manufacturing assembly system
for industry 4.0 concept. Sustainability 2020;12(36589).
[84] Liang B, et al. A displacement eld perception method for component digital twin
in aircraft assembly. Sensors 2020;20(516118).
[85] Seif A, Toro C, Akhtar H. Implementing Industry 4.0 asset administrative shells in
mini factories. Procedia Comput Sci 2019;159:495–504.
[86] Toro C, Seif A, Akhtar H. Modeling and connecting asset administrative shells for
mini factories. Cybern Syst 2020;51(2SI):232–45.
[87] Gallego-Garcia S, Reschke J, Garcia-Garcia M. Design and simulation of a
capacity management model using a digital twin approach based on the viable
system model: case study of an automotive plant. Appl Sci 2019;9(556724).
[88] Ko J, Hu SJ. Manufacturing system design considering stochastic product
evolution and task recurrence. J Manuf Sci Eng-Trans ASME 2009;131(0510125).
[89] Leng J, et al. Digital twin-driven rapid reconguration of the automated
manufacturing system via an open architecture model. Robot Comput Integr
Manuf 2020;63(101895).
[90] Jeon B, Suh S. Design considerations and architecture for cooperative smart
factory: MAPE/BD approach. Procedia Manuf 2018;26:1094–106.
[91] Zhang C, et al. Digital twin-enabled recongurable modeling for smart
manufacturing systems. Int J Comput Integr Manuf 2019. https://doi.org/
10.1080/0951192X.2019.1699256.
[92] Kurniadi KA, Lee S, Ryu K. Digital twin approach for solving reconguration
planning problems in RMS. IFIP advances in information and communication
technology. Cham: Springer International Publishing; 2018. p. 327–34.
[93] Mawson VJ, Hughes BR. The development of modelling tools to improve energy
efciency in manufacturing processes and systems. J Manuf Syst 2019;51:95–105.
[94] Xiao Q, et al. Meta-reinforcement learning of machining parameters for energy-
efcient process control of exible turning operations. IEEE Trans Autom Sci Eng
2021;18(1):5–18.
[95] Supekar SD, et al. A framework for quantifying energy and productivity benets
of smart manufacturing technologies. Procedia CIRP 2019;80:699–704.
[96] Ma S, et al. Data-driven sustainable intelligent manufacturing based on demand
response for energy-intensive industries. J Clean Prod 2020;274(123155).
[97] Barni A, et al. Exploiting the digital twin in the assessment and optimization of
sustainability performances. New York, USA: IEEE; 2018.
[98] Wang XV, Wang L. Digital twin-based WEEE recycling, recovery and
remanufacturing in the background of Industry 4.0. Int J Prod Res 2019;57(12SI):
3892–902.
[99] He B, Bai K. Digital twin-based sustainable intelligent manufacturing: a review.
Adv Manuf 2020.
[100] Becue A, et al. CyberFactory#1-Securing the Industry 4.0 with cyber-ranges and
digital twins. New York, USA: IEEE; 2018.
[101] Kloibhofer R, Kristen E, Jakˇ
si´
c S. Safety and security in a smart production
environment. Lecture notes in computer science. Cham: Springer International
Publishing; 2018. p. 190–201.
[102] Lee J, Cameron I, Hassall M. Improving process safety: what roles for
digitalization and Industry 4.0. Process Saf Environ Prot 2019;132:325–39.
[103] Rodriguez-Guerra J, et al. On fault-tolerant control systems: a novel
recongurable and adaptive solution for industrial machines. IEEE Access 2020;8:
39322–35.
[104] Cochran DS, et al. A decomposition approach for manufacturing system design.
J Manuf Syst 2001;20(6):371–89.
[105] Hibino H, et al. The development of an object-oriented simulation system basedon
the thought process of the manufacturing system design. Int J Prod Econ 1999;
(60–61):343–51.
[106] Liu Q, et al. Digital twin-based designing of the conguration, motion, control,
and optimization model of a ow-type smart manufacturing system. J Manuf Syst
2021;58:52–64.
[107] Tao F, Zhang M. Digital twin shop-oor: a new shop-oor paradigm towards
smart manufacturing. IEEE Access 2017;5:20418–27.
[108] Souza V, et al. A digital twin architecture based on the industrial internet of
things technologies. In: IEEE International Symposium on Consumer Electronics;
2019.
[109] Cheng J, et al. DT-II:digital twin enhanced Industrial Internet reference
framework towards smart manufacturing. Robot Comput Integr Manuf 2020;62
(101881).
[110] Chhetri SR, et al. In: Ramachandran GS, Ortiz J, Ramachandran GS, Ortiz J,
editors. QUILT: quality inference from living digital twins in IoT-enabled
manufacturing systems. New York, USA: Assoc Computing Machinery; 2019.
p. 237–48.
[111] Li Y, et al. Practical implementation of an OPC UA TSN communication
architecture for a manufacturing system. IEEE Access 2020;8:200100–11.
[112] Liu C, et al. A cyber-physical machine tools platform using OPC UA and
MTConnect. J Manuf Syst 2019;51:61–74.
[113] Zhang K, et al. Digital twin-based opti-state control method for a synchronized
production check toroperation system. Robot Comput Integr Manuf 2020:63.
[114] Lin WD, et al. Integrated cyber physical simulation modelling environment for
manufacturing 4.0. In: International Conference on Industrial Engineering and
Engineering Management IEEM; 2018. p. 1861–5.
[115] Zhang H, Yan Q, Wen Z. Information modeling for cyber-physical production
system based on digital twin and AutomationML. Int J Adv Manuf Technol 2020;
107(3–4):1927–45.
[116] Ghosh AK, Ullah AMMS, Kubo A. Hidden Markov model-based digital twin
construction for futuristic manufacturing systems. Artif Intell Eng Des Anal Manuf
2019;33(3):317–31.
[117] Malykhina GF, Vasilyev AN, Tarkhov DA. Information measurement systems in
the digital society. J Phys Conf Ser 2019;1379(1):12037.
[118] Zheng Y, et al. Digital twin for geometric feature online inspection system of car
body-in-white. Int J Comput Integr Manuf 2020. https://doi.org/10.1080/
0951192X.2020.1736637.
[119] Kim BS, Jin Y, Nam S. An integrative user-level customized modeling and
simulation environment for smart manufacturing. IEEE Access 2019;7:
186637–45.
[120] Golubev VO, et al. Systems and aids of mathematical modeling of the alumina
renery methods: problems and solutions. Non-Ferrous Met 2019;(1):40–7.
[121] Szabo G, et al. Digital twin: network provisioning of mission critical
communication in cyber physical production systems. New York, USA: IEEE;
2019. p. 37–43.
[122] Leng J, et al. ManuChain: combining permissioned blockchain with a holistic
optimization model as Bi-Level intelligence for smart manufacturing. IEEE Trans
Syst Man Cybern-Syst 2020;50(1):182–92.
[123] Dos Santos CH, et al. Use of simulation in the industry 4.0 context: creation of a
Digital Twin to optimise decision making on non-automated process. J Simul
2020.
[124] Silva JM, Javales R, Silva JR. A new requirements engineering approach for
manufacturing based on petri nets. IFAC Papersonline 2019;52(10):97–102.
[125] Mieth C, Meyer A, Henke M. Framework for the usage of data from real-time
indoor localization systems to derive inputs for manufacturing simulation.
Procedia CIRP 2019;81:868–73.
[126] Bhalode P, Ierapetritou M. Discrete element modeling for continuous powder
feeding operation: calibration and system analysis. Int J Pharm 2020;585
(119427).
[127] Reinhardt H, Weber M, Putz M. A survey on automatic model generation for
material ow simulation in discrete manufacturing. Procedia CIRP 2019;81:
121–6.
[128] Simoes B, et al. Cross reality to enhance worker cognition in industrial assembly
operations. Int J Adv Manuf Technol 2019;105(9SI):3965–78.
[129] Han Y, et al. 3D CAD data extraction and conversion for application of
augmented/virtual reality to the construction of ships and offshore structures. Int
J Comput Integr Manuf 2019;32(7):658–68.
[130] Fera M, et al. Towards digital twin implementation for assessing production line
performance and balancing. Sensors 2020;20(971).
[131] Perez L, et al. Digital twin and virtual reality based methodology for multi-robot
manufacturing cell commissioning. Appl Sci 2020;10(363310).
[132] Taylor C, et al. Dening production and nancial data streams required for a
factory digital twin to optimise the deployment of labour. Communications in
computer and information science. Singapore: Springer Singapore; 2018. p. 3–12.
[133] Moyne J, Iskandar J. Big data analytics for smart manufacturing: case studies in
semiconductor manufacturing. Processes 2017;5(393).
[134] Gao RX, et al. Big data analytics for smart factories of the future. CIRP Ann-Manuf
Technol 2020;69(2):668–92.
[135] Ringsquandl M, et al. Knowledge fusion of manufacturing operations data using
representation learning. In: Lodding H, Lodding H, editors. IFIP advances in
information and communication technology. Cham: Springer International
Publishing; 2017. p. 302–10.
[136] Huerkamp A, et al. Combining simulation and machine learning as digital twin for
the manufacturing of overmolded thermoplastic composites. J Manuf Mater
Process 2020;4(923).
[137] Cavalcante IM, et al. A supervised machine learning approach to data-driven
simulation of resilient supplier selection in digital manufacturing. Int J Inf
Manage 2019;49:86–97.
[138] Wang J, et al. Digital Twin for rotating machinery fault diagnosis in smart
manufacturing. Int J Prod Res 2019;57(12SI):3920–34.
[139] Leng J, et al. A loosely-coupled deep reinforcement learning approach for order
acceptance decision of mass-individualized printed circuit board manufacturing
in Industry 4.0. J Clean Prod 2021;280(2):124405.
[140] Danilczyk W, Sun YL, He H. In: ANGEL: An Intelligent Digital Twin Framework
for Microgrid Security, in North American Power Symposium; 2019.
[141] Gupta N, et al. Additive manufacturing cyber-physical system: supply chain
cybersecurity and risks. IEEE Access 2020;8:47322–33.
[142] Leng J, et al. Makerchain: a blockchain with chemical signature for self-
organizing process in social manufacturing. J Clean Prod 2019;234:767–78.
[143] Leng J, et al. Blockchain-secured smart manufacturing in Industry 4.0: a survey.
IEEE Trans Syst Man Cybern Syst 2021;51(1):237–52.
[144] Leng J, et al. Blockchain-empowered sustainable manufacturing and product
lifecycle management in industry 4.0: a survey. Renew Sustain Energy Rev 2020;
132:110112.
[145] Chirico R, et al. A contract-based methodology for production lines validation. In:
IEEE International Conference on Industrial Informatics; 2019. p. 695–8.
[146] Spellini S, et al. Languages and formalisms to enable EDA techniques in the
context of industry 4.0.. New York, USA: IEEE; 2019.
J. Leng et al.
Journal of Manufacturing Systems 60 (2021) 119–137
137
[147] Leng J, et al. Blockchain security: a survey of techniques and research directions.
IEEE Trans Serv Comput 2020:1.
[148] Leng J, Jiang P. Evaluation across and within collaborative manufacturing
networks: a comparison of manufacturers’ interactions and attributes. Int J Prod
Res 2018;56(15):5131–46.
[149] Avventuroso G, Silvestri M, Pedrazzoli P. A networked production system to
implement virtual enterprise and product lifecycle information loops. IFAC
Papersonline 2017;50(1):7964–9.
[150] Cohen Y, et al. Design and management of digital manufacturing and assembly
systems in the Industry 4.0 era. Int J Adv Manuf Technol 2019;105(9SI):3565–77.
[151] Leng J, et al. Digital twin-driven manufacturing cyber-physical system for parallel
controlling of smart workshop. J Ambient Intell Humaniz Comput 2019;10(3SI):
1155–66.
[152] Loizou S, et al. In: Filipe J, editor. A smart product Co-design and monitoring
framework via gamication and complex event processing; 2019. p. 237–44.
Setubal, Portugal.
[153] Park KT, et al. Digital twin-based cyber physical production system architectural
framework for personalized production. Int J Adv Manuf Technol 2020;106(5–6):
1787–810.
[154] He YC, Zhang NS, Wang AM. Digital twin process and simulation operation
control technology for intelligent manufacturing unit. In: IOP Conference Series-
Materials Science and Engineering; 2020.
[155] Gericke GA, et al. Design of digital twins for optimization of a water bottling
plant. IEEE industrial electronics society. New York: IEEE; 2019. p. 5204–10.
[156] Liu Q, et al. Digital twin-driven rapid individualised designing of automated ow-
shop manufacturing system. Int J Prod Res 2019;57(12SI):3903–19.
[157] Chen G, et al. The framework design of smart factory in discrete manufacturing
industry based on cyber-physical system. Int J Comput Integr Manuf 2020;33(1):
79–101.
[158] Redelinghuys AJH, Basson AH, Kruger K. A six-layer architecture for the digital
twin: a manufacturing case study implementation. J Intell Manuf 2020;31(6):
1383–402.
[159] Jeon B, et al. The architecture development of Industry 4.0 compliant smart
machine tool system (SMTS). J Intell Manuf 2020;31(8):1837–59.
[160] Wang K, Lee Y, Angelica S. Digital twin design for real-time monitoring - a case
study of die cutting machine. Int J Prod Res 2020. https://doi.org/10.1080/
00207543.2020.1817999.
[161] Wang B, et al. Intelligent welding system technologies: state-of-the-art review and
perspectives. J Manuf Syst 2020;56:373–91.
[162] Zhang X, Zhu W. Application framework of digital twin-driven product smart
manufacturing system: a case study of aeroengine blade manufacturing. Int J Adv
Robot Syst 2019;16. 17298814198806635.
[163] Lin WD, Low MYH. Concept and implementation of a cyber-physical digital twin
for a SMT line. In: International Conference on Industrial Engineering and
Engineering Management; 2019. p. 1455–9.
[164] Ma J, et al. A digital twin-driven production management system for production
workshop. Int J Adv Manuf Technol 2020;110(5–6):1385–97.
[165] Talkhestani BA, et al. A concept in synchronization of virtual production system
with real factory based on anchor-point method. Procedia CIRP 2018;67:13–7.
[166] Nafors D, et al. Application of a hybrid digital twin concept for factory layout
planning. Smart Sustain Manuf Syst 2020;4(2SI):231–44.
[167] Cheng Y, et al. Cyber-physical integration for moving digital factories forward
towards smart manufacturing: a survey. Int J Adv Manuf Technol 2018;97(1–4):
1209–21.
[168] Longo F, Nicoletti L, Padovano A. Ubiquitous knowledge empowers the Smart
Factory: the impacts of a Service-oriented Digital Twin on enterprises’
performance. Annu Rev Control 2019;47:221–36.
[169] Schimanski CP, et al. Pushing digital automation of congure-to-order services in
small and medium enterprises of the construction equipment industry: a design
science research. Appl Sci 2019;9(8):3780.
[170] Tomiyama T, et al. Development capabilities for smart products. CIRP Ann-Manuf
Technol 2019;68(2):727–50.
[171] Steimer C, Aurich JC. Analysis of information interdependencies between product
development and manufacturing system planning in early design phases. Procedia
CIRP 2016;50:460–5.
[172] Schleich B, et al. Shifting value stream patterns along the product lifecycle with
digital twins. Procedia CIRP 2019;86:3–11.
[173] Promyoo R, Alai S, El-Mounayri H. Innovative digital manufacturing curriculum
for industry 4.0. Procedia Manuf 2019;34:1043–50.
[174] Kuliaev V, et al. Towards product centric manufacturing: from digital twins to
product assembly. In: IEEE International Conference on Industrial Informatics;
2019. p. 164–71.
[175] Eddy D, et al. Framework for design from manufacturing data mapping. New
York, USA: ASME; 2020.
[176] Schuetzer K, et al. Contribution to the development of a Digital Twin based on
product lifecycle to support the manufacturing process. Procedia CIRP 2019;84:
82–7.
[177] Wagner R, et al. Challenges and potentials of digital twins and Industry 4.0 in
product design and production for high performance products. Procedia CIRP
2019;84:88–93.
[178] Kurfess TR, et al. A review of modern communication technologies for digital
manufacturing processes in industry 4.0. J Manuf Sci Eng-Trans ASME 2020;142:
11081511.
[179] Bevilacqua M, et al. Digital twin reference model development to prevent
operators’ risk in process plants. Sustainability 2020;12(10883).
[180] Park Y, Woo J, Choi S. A cloud-based digital twin manufacturing system based on
an interoperable data Schema for smart manufacturing. Int J Comput Integr
Manuf 2020.
[181] Kumar VS, et al. A federated, multimodal digital thread platform for enabling
digital twins. Nav Eng J 2020;132(1):47–56.
[182] Ahuett-Garza H, Urbina Coronado PD. A reference model for evolving digital
twins and its application to cases in the manufacturing oor. Smart Sustain Manuf
Syst 2019;3(2):1–13.
[183] Catarci T, et al. In: Bertino E, editor. A conceptual architecture and model for
smart manufacturing relying on service-based digital twins. New York, USA: IEEE;
2019. p. 229–36.
[184] Damjanovic-Behrendt V, Behrendt W. An open source approach to the design and
implementation of Digital Twins for Smart Manufacturing. Int J Comput Integr
Manuf 2019;32(4–5SI):366–84.
[185] Moyne J, et al. A requirements driven digital twin framework: specication and
opportunities. IEEE Access 2020;8:107781–801.
[186] Stark R, Fresemann C, Lindow K. Development and operation of Digital Twins for
technical systems and services. CIRP Ann-Manuf Technol 2019;68(1):129–32.
[187] Zhang G, et al. An integrated framework for active discovery and optimal
allocation of smart manufacturing services. J Clean Prod 2020;273(123144).
[188] Zheng P, et al. A systematic design approach for service innovation of smart
product-service systems. J Clean Prod 2018;201:657–67.
[189] Olivotti D, et al. Creating the foundation for digital twins in the manufacturing
industry: an integrated installed base management system. Inf Syst E-Business
Manag 2019;17(1):89–116.
[190] Li Y, Li L. Enhancing the optimization of the selection of a product service system
scheme: a digital twin-driven framework. Strojniski Vestnik-J Mech Eng 2020;66
(9):534–43.
[191] Leng J, et al. Combining granular computing technique with deep learning for
service planning under social manufacturing contexts. Knowl Based Syst 2018;
143:295–306. https://doi.org/10.1016/j.knosys.2017.07.023.
[192] Cong J, et al. A holistic relook at engineering design methodologies for smart
product-service systems development. J Clean Prod 2020;272(122737).
J. Leng et al.
