Evolution and Future of Manufacturing Systems

JID: CIRP
ARTICLE IN PRESS [m191;July 20, 2021;18:02]
CIRP Annals - Manufacturing Technology 00 (2021) 1 24
Contents lists available at ScienceDirect
CIRP Annals - Manufacturing Technology

journal homepage: https://www.editorialmanager.com/CIRP/default.aspx
Evolution and future of manufacturing systems

Hoda ElMaraghy (1)a,b,*, Laszlo Monostori (1)c,d, Guenther Schuh (1)e,
Waguih ElMaraghy (1)a,b
a
Intelligent Manufacturing Systems Centre (IMSC), University of Windsor, Ontario, Canada
b
Department of Mechanical, Automotive and Materials Engineering (MAME), University of Windsor, Ontario, Canada
c €tvo
Centre of Excellence in Production Informatics and Control, Institute for Computer Science and Control, Eo nd Research Network, Budapest, Hungary
€s Lora
d
Department of Manufacturing Science and Technology, Budapest University of Technology and Economics, Budapest, Hungary
e
Laboratory for Machine Tools and Production Engineering (WZL), RWTH Aachen University, Aachen, Germany
A R T I C L E I N F O A B S T R A C T
Article history: The evolution of manufacturing systems, influenced by changes along four axes - products, technology, busi-
Available online xxx ness strategies and production paradigms - is presented. Adoption of human-centric decision making in
meshed collaboration with intelligent systems is examined. Implications and preparedness for the shift
Keywords: towards more responsive, intelligent adaptive systems are reviewed. Research and industrial use cases are
Manufacturing systems
presented. A vision for the new future Adaptive Cognitive Manufacturing System (ACMS) paradigm and its
Sustainable development
Cognitive adaptability
characteristics, drivers and enablers are articulated highlighting the digital and cognitive transformations.
Perspectives and insights are offered for future research, education, and work to realize the evolution of
manufacturing systems.
© 2021 CIRP. Published by Elsevier Ltd. All rights reserved.
1. Introduction
Manufacturing systems continue to evolve in design, configura-

tion, operation, and control in an eco-system characterized by new
drivers, more advanced enablers and disruptive technologies and
business models. Socio-technical developments and business strate-
gies will shape their future.
1.1. Challenges and motivation
This section sets the stage with brief discussion of: importance
of manufacturing, various challenges (technical, economic, social,
strategic, business), drivers (cost, quality, variety, efficiency, value, Fig. 1. Manufacturing systems evolution trends and scope.
sustainability), transformative innovations, complexity, respon-
siveness, knowledge-based and data-intensive manufacturing,
digitalization, connectivity and communication, demographic 1.2. Scope and objectives
changes, human capital development and future work. Tracking
and analysing effects of industrial revolutions on changes in Manufacturing systems changes over many decades are driven by
manufacturing systems and enabling axes of evolution is a good advances in production and other technologies, introduction of new
predictor of what is to come, and what industry and experts are materials and complex products requiring new processing techni-
saying about needed developments. The paper scope is illustrated ques, organizational strategies seeking to minimize cost, increase
in Fig. 1. quality and reliability, maximize profit and concerns about societal
and sustainability goals as well as humans’ interaction with systems
elements and the future of work.
The focus of this keynote paper is about the evolution and future
of manufacturing systems for discrete parts/products production as
well as the characteristics, enablers, and drivers of manufacturing
systems paradigms.
* Corresponding author at: Intelligent Manufacturing Systems Centre (IMSC), Uni- Manufacturing systems encompass both the physical and logical
versity of Windsor, Ontario, Canada. aspects of production. Their physical configuration consists of
E-mail address: imscadmin@uwindsor.ca (H. ElMaraghy).
https://doi.org/10.1016/j.cirp.2021.05.008
0007-8506/© 2021 CIRP. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: H. ElMaraghy et al., Evolution and future of manufacturing systems, CIRP Annals - Manufacturing Technology (2021),
JID: CIRP
2 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24
machines, workstations, robots, and other equipment arranged in

various layouts and integrated or connected physically by material
handling equipment and logically via computer control and software
applications. People are an integral part of manufacturing systems
and play an important role in their design, planning, operation, and
control. Computers have played a significant role in the operation
and control of manufacturing systems and their modules including
software applications for computer-aided design and computer-
aided manufacturing (CAD/CAM), computerized numerical control
(CNC), product life cycle management (PLM), process planning, pro-
duction planning and scheduling, inspection, quality control, mainte- Fig. 2. Keywords for manufacturing systems literature survey.
nance, inventory, and supply chain management. Hence,
manufacturing systems are found physically on shop floors with logi-
cal support systems within and beyond the factory.
The objectives of this keynote paper are manifold including: a)
reviewing the evolution of manufacturing systems to date by survey-
ing the most relevant and important literature and identifying the
major milestones along their evolution path and their defining fea-
tures and characteristics, b) reviewing recent advances, trends and dis-
ruptors, relevant to the four axes of manufacturing systems evolution,
and discussing their expected/anticipated effects on future
manufacturing systems, c) reviewing the evolution of manufacturing
systems, their future, and related concepts and paradigms, d) review-
ing and assessing the readiness and state-of-implementation of the
“new technologies and business strategies” in manufacturing systems,
obstacles and challenges, and discussing illustrating use cases in indus-
try and academia, e) articulating a vision for a new adaptive cognitive
Fig. 3. Publications sources for manufacturing systems literature.
manufacturing systems paradigm, its drivers, characteristics, and ena-
blers, work force requirements, and societal expectations and respon-
sibilities, and f) offering future perspectives and establishing a map for Fig. 3 shows the sources of these publications contributed to 10
research needs and challenges towards achieving smarter, more adap- high impact journals and conferences.
tive, more sustainable and human-centred manufacturing systems.
1.5. Keynote paper organization
1.3. Terminology
The new trends and anticipated changes in the manufacturing
“Production” and “manufacturing” have been used interchange- landscape, strategies, and manufacturing systems paradigms provided
ably over the years in the literature. While manufacturing and pro- the motivation for this keynote paper and shaped its scope. The fol-
duction have similar fundamentals, they are distinct with many key lowing outlines the structure of the paper. Section 1 paints a picture of
differences. This keynote paper focuses on manufacturing systems, the eco-system in which manufacturing systems exist, change direc-
which are broader and encompass production systems. The differ- tions and expectations, and outlines the keynote paper motivation,
ence between “manufacturing” and “manufacturing systems” is objectives, and scope. Section 2 includes a survey of relevant literature,
emphasized particularly with regard to the paradigms names that definitions, and innovations, introduces four axes of evolution of
have been introduced, mostly by CIRP, and generally agreed to in the manufacturing systems and illustrates their co-evolution. It presents
literature to-date, up to the ongoing fourth industrial revolution. In major manufacturing systems paradigms, nature-inspired products/
addition, Cyber-Physical Production Systems (CPPS), Industrie 4.0 or manufacturing systems co-evolution, biologicalisation in manufactur-
Industry 4.0, and Smart Manufacturing Systems (SMS) share many ing systems, and manufacturing systems life cycle and sustainability.
similar features and characteristics. They are used interchangeably in Section 3 reviews the advances in cyber-physical systems (CPS) and
the literature and in this keynote paper. Cognitive manufacturing lev- smart manufacturing systems including smart manufacturing drivers
erages Industry 4.0 technologies, including big data analytics and and enablers such as internet of production (IoP) or industrial internet
artificial intelligence and generates actionable insights and interac- of things (IIoT), big-data analytics, communication, connectivity, smart
tions between humans and machines. This is discussed in greater data-driven manufacturing systems design and control, and bio-
detail later in this keynote paper. Most of the used terminology is inspired methodologies for manufacturing systems design and opera-
consistent with, and often explicitly defined in, the CIRP Dictionary tion and their co-development In the context of the on-going smart
and the CIRP Encyclopedia [97] . manufacturing (Industry 4.0) evolution, Section 4 focuses on smart
systems industrial implementations and applications of digital
1.4. Publications trends and bibliometric analysis manufacturing systems in car assembly, auto-parts tooling and com-
munication, and illustrates the use of IoT, IoP and digital shadows. It
Extensive literature survey was conducted from a variety of per- discusses industry readiness and Industry 4.0 maturity models and
spectives and sources. Various search approaches and keywords indices. It includes joint research-industry use cases of many aspects
were used, with a focus on manufacturing systems documents pub- of smart manufacturing systems examples of human-machine interac-
lished to date. It was noted that the frequency of publication related tion and learning. Section 5 introduces a vision for the new adaptive
to earlier manufacturing paradigms, such as flexible manufacturing cognitive manufacturing systems (ACMS) paradigm, its characteristics,
systems (FMS), reconfigurable manufacturing systems (RMS), and drivers, and enablers and explains why it represents a paradigm shift.
changeable manufacturing systems (CMS) had peaked in 2010, whilst It discusses bio-intelligent manufacturing and human-centric adaptive
papers relating to smart and cyber-physical manufacturing systems manufacturing. The multiple facets of systems adaptation are pre-
are on a steep rise. Fig. 2 illustrates the results of a Scopus search for sented in a new classification including cognitive adaptation which is
the selected keywords in documents published annually since the at the core of the new paradigm. The ability of current manufacturing
year 2010. As illustrated, the number of publications relating to adap- systems to respond and adapt under the impact of extreme disruption,
tive manufacturing systems and smart manufacturing systems, dur- such as those caused by the COVID-19 pandemic, are analysed.
ing this period, is steadily increasing. Human-centric cognitive manufacturing systems and evolution of
JID: CIRP
H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 3
cognitive digital twins (CDT) are discussed. Finally, Section 6 presents directly improved manufacturing systems design, and rationalization
insights, reflections, and conclusions on how manufacturing systems and standardization of processes. Modularity is the degree to which a
are being re-imagined. It offers an outlook and perspectives on current complex system (product, process plan, manufacturing system, etc.)
challenges and future research. can be divided into sub-units (modules) which can be reconfigured
as needed [170]. Product modularity is enabled by design clustering
2. Evolution of manufacturing systems and enablers and granularity methods, and reflected in the used manufacturing
technologies, processes, machines clusters, and manufacturing cells,
Manufacturing systems have evolved over many decades driven and is mirrored by organizational units and clusters of suppliers.
by advances in production technology, machine tools, information Modularity reduces complexity and cost of design, manufacture, and
technology, materials, and products, as well as the evolution of orga- repair, increases reliability, improves maintainability, and prolongs
nizational strategies seeking to minimize cost, increase quality and products life by facilitating selective updating of modules.
reliability, maximize productivity and profits, and promote sustain- Manufacturing systems are products that also require use of design
ability. The proliferation of products variety and desire to increase methodologies, collaboration and complexity management
competitiveness through differentiation had a significant impact on [49,50,159,161], and quality prediction [51].
the evolution of manufacturing systems and motivated the develop- The products platforms design concept, where clustering is used
ment of several manufacturing systems paradigms. to form a core of common components or modules that can later be
customized to generate product variants belonging to the family,
2.1. Four axes of evolution were developed as well as methods to optimize the design of the
product platform. It is an important enabler of product mass customi-
Earlier evolution of manufacturing systems, from dedicated to zation by designing manufacturing systems where products differen-
flexible and reconfigurable, was motivated by the need to manage tiation is delayed, allowing manufacturing the product platform in
changes in production volumes and products variety. The next waves large quantities, i.e. mass production with push business model, then
of manufacturing systems evolution towards smart, cognitive and individual product variants are produced in smaller volumes per vari-
more adaptable systems are influenced by disruptive advances along ant, following a pull business model [70]. The effects of increased
four axes depicted in Fig. 4, which shows: a) products evolution, b) variety on products design, manufacturing systems design and strate-
technological evolution, c) business strategies evolution and d) pro- gies, industrial enterprises and supply chains, as well as management
duction/manufacturing evolution manifested in the industrial revolu- strategies on all levels were discussed extensively [47].
tions to date which collectively gave rise to the development of new Design methods for products reconfiguration, such as open archi-
manufacturing systems paradigms. The four axes of evolutions are tecture products (OAP) [67,92] were introduced for added adaptation
highlighted in the coming sub-sections. by allowing modules to be added/removed/swapped to change prod-
uct features and functionality. Research followed to identify the opti-
mal design of OAP and assess their assembly and disassembly
complexity [210]. Open architecture products coupled with modular-
ity, scalability and standard interfaces between product modules
enabled, and increased the efficiency of corresponding reconfigurable
manufacturing systems. These approaches apply to products made by
reconfigurable machine tools, reconfigurable robots, universal/recon-
figurable end-effectors, reconfigurable molds and fixtures, and uni-
versal tooling used in FMS and RMS. The discussed product design
methodologies have evolved along with flexible, reconfigurable sys-
tems and the hybrid additive/subtractive technologies to satisfy the
increasing need for products customization and personalization.
The use of new materials in consumer products continued to
increase partly because of customer’s demand for products with spe-
cific characteristics and performance. Light weight materials (alumi-
num, polymers, and composites) were introduced to auto-
manufacturing to reduce vehicles weight and fuel consumption,
which triggered related manufacturing technologies, processes, and
machines. Environmental concerns motivated the design and devel-
opment of new car engines and power trains using alternate fuels as
Fig. 4. Four axes of manufacturing and systems evolution.
well as electric mobility led to major disruption in the automotive
powertrain production, which significantly impacted their
2.1.1. Products evolution manufacturing systems design, configuration and operation [80]. The
In the early days of mass production, products had simple shapes manufacturing systems of mobility vehicles have witnessed signifi-
and features due to limitations of technology, used materials and cant disruption in design, configuration, size, location, and produc-
manufacturing processes. Products with intricate shapes, complex tion volume not only due to these technological advances but also
features, composite materials, and smart functionalities continued to the changing business strategies such as ride sharing and integrated
emerge giving rise to associated design, modelling, machines, and product-service models. This is not a unique example of the inter-
manufacturing and services innovations and technologies to meet twined nature of the four axes of manufacturing systems evolution.
the new challenges. More examples can be seen in many large and small consumer prod-
Advances in computer science and technology along with intro- ucts in various sectors. The innovation helix best represents that
duction of several design theories and methodologies have directly intertwined nature of products, technology, and business innova-
contributed to the design of products and their manufacturing sys- tions; and manufacturing systems evolution which mark the compet-
tems. Parts/products coding, classification and group technology lead itiveness frontier. Examples include smart materials capable of
to the efficient formation of product families and machine cells, and responding to external stimuli with shape change, self-actuation,
improved the efficiency of numerical control (NC)/CNC programming, self-sensing, self-diagnosing and self-healing behaviours [20]. Advan-
fixtures and tooling design and process planning [17]. Modular prod- ces in 3D printing of products influenced products design. Additive
uct design lends itself to flexible, reconfigurable and changeable manufacturing use in printing organs and tissues, and responsive
manufacturing systems and led to many methodologies for design materials for 4D printing [62] led to the development of special prod-
for ease of manufacture (DFM) and ease of assembly (DFA) which uct design methods and additive manufacturing machines. In the
JID: CIRP
textile industry, smart materials for wearables that can adapt to tem- controlled manufacturing systems to be part of connected and data
perature and moisture to improve comfort are developed. All these intensive CPSs, Industry 4.0 and digital manufacturing systems.
applications call for commensurate manufacturing systems capabili- Artificial Intelligence (AI) plays an important role in smart
ties. machines and intelligent manufacturing systems by enabling the
The demand for smart products which are intelligent, connected application of essential features of natural intelligence such as sens-
and highly responsive is increasing [188]. Smart products contain ing, perception, learning, reasoning and decision making in areas
cyber-physical mechatronic components which have distinct differ- such as operating and controlling manufacturing systems, process
entiating features and capabilities including sensing with integrated planning and production planning, and in planning robot-human col-
or imbedded sensors, connectedness, and communication via inter- laboration and deep learning-based human motion trajectory recog-
net (IoT), networking with other smart products, interaction with nition [39,142,200]. Throughout the life of products and
users, processing data and intelligence. Smart products are complex manufacturing systems, large amounts of data are collected. The role
multi-domain/multi-disciplinary in nature. Manufacturing their indi- of data analytics in supporting smart manufacturing systems,
vidual components represents challenges for embedding sensors, machines, products, and related technologies and business strategies
assembly and testing and increases complexity as they have highly has been discussed [61,101].
integrated hardware and software [114]. Cyber-physical systems (CPS) which integrates cyber components
such as embedded sensors with the physical resources in the produc-
2.1.2. Technological evolution tion system are important enablers for implementing smart systems
Many technological advances have emerged in the manufacturing [119]. Achieving realtime data acquisition, processing and decision
field. Important developments and disruptive trends, which influence making, development of computational dynamic systems theory for
manufacturing systems evolution are discussed in this section. modelling and analysis, standardization of communication protocols,
Automation and digitalization. Production systems have been and data security are fundamental for the success of CPS applications.
evolving from standalone machinery with proprietary controllers to Cloud manufacturing refers to decentralized and networked
flexible and reconfigurable manufacturing systems with added mod- manufacturing resources that can be accessed by manufacturers as
ularity, mobility, and more open and smarter control architecture. needed enabled by cloud computing, IoT and service-oriented archi-
The automation technologies developed since the 1st and 2nd indus- tecture. Cloud manufacturing differs from computer integrated
trial revolutions, such as computer-controlled programmable manufacturing (CIM) and CPS in that the connection between the
machines (NC, CNC) made it possible to move from hard-wired logic physical and cyber domains proceeds through services [150]. Optimal
controls, punched cards and hard automation machines and produc- allocation and scheduling of physical and computational resources is
tion lines to digital programmable machines/ workstations and pro- required for efficient use of cloud manufacturing. Cloud manufactur-
grammable, reconfigurable and smart logic control. Advances in ing models and protocols for cloud-based usage are not suitable for
communication control layers and systems from mobile application all manufacturing. Cost benefit analysis and validation are needed to
protocols (MAP) to supervisory control and data acquisition (SCADA) assess their performance, effectiveness and technical and financial
made it possible to control large and complex manufacturing systems feasibility [124].
in hierarchical and distributed manner with two-way communica- Digital transformation leverages information technology to disrupt
tion and feedback. These are the backbone technologies the evolution traditional industry models and business practices to deliver excep-
of which paralleled that of manufacturing systems. Versatile multi- tional customer and business value and create sustainable competi-
axis, multi-tasking flexible machines along with advances in laser tri- tive advantage. Artificial intelligence and machine learning, big data,
angulation and global positioning systems (GPS), bar code readers predictive analytics, and business process automation are important
and proximity sensors used in automated and self-guided vehicles, rapidly emerging digital transformation enablers which are being
made alternate part-machine assignment possible and enabled alter- incorporated into the organization and modernization strategies of
nate routing flexibility. manufacturing systems such as self-awareness, self-learning, self-
Sensors, RFIDs (Radio-frequency identification), IoT and IIoT, and 4G healing, and cognitive adaptation characteristics which are discussed
and 5G internet communication protocols are essential elements of in Sections 4 and 5.
digitalization and evolution of advanced manufacturing systems.
Sensors collect digital and analogue data from manufacturing sys- 2.1.3. Business models evolution
tems and smart products for use in monitoring equipment condi- Business models followed by companies for creating value and
tions, operations execution, and feedback. Large number of sensors profit have changed over time to satisfy consumers demands and
with different sizes and functionalities are required in any advanced requirements. This section overviews and classifies the evolution of
manufacturing systems implementation. Hence, both capacity and classical and emerging business models as they affect manufacturing
speed of wireless networks and communication infrastructure as systems. In craft manufacturing, a product was designed and made
well as techniques for sensor data integration and fusion to utilize for one customer following a pull model. In mass production, prod-
and interpret the measured data are crucial to the success of IIoT and ucts were designed and made for customers a priori and offered in
IoP [162] in digital manufacturing systems implementation. large quantities following a push business model. To satisfy more cus-
Additive manufacturing (AM) is an example of disruptive tomer requirements, yet keep manufacturing cost manageable, flexi-
manufacturing technologies as parts are manufactured by adding ble, reconfigurable, and changeable manufacturing and mass
material in successive layers instead of removing it. It can produce customization were introduced. Customers configure their products
intricate shapes of many different materials with simple setup and based on pre-designed and grouped features (packages) offered by
minor post processing. This technology evolved from rapid prototyp- the manufacturer. Therefore, the customer is satisfied by ordering
ing of products using plastics in the 1980s to applications beyond the chosen product configuration, which is not unique, while the
prototyping such as producing automotive parts, aerospace products manufacturer is able to group similar orders to increase efficiency
and medical instruments where end products can be manufactured and reduce cost. Mass customization follows a hybrid push-pull busi-
in economical quantities [19,185]. Hybrid additive-subtractive ness model where the product family platform is mass produced
manufacturing supports production of new products with better (push model) then customized according to customers configurations
functionality more cost effectively. A macro-process planning meth- (pull model). Personalized manufacturing drives customization fur-
odology for optimally selecting the type and sequence of hybrid addi- ther where customers become closely involved in the design of many
tive-subtractive processes was developed [46,125]. Multi-tasking features of the product. The followed business model clearly drives
machine tools for both additive and subtractive processes are now the design, configuration and control of manufacturing systems and
available. These advances in AM facilitate the design of systems for the utilized manufacturing processes and technologies to achieve the
personalized manufacturing [92]. Advanced digital AM machines desired outcome. Personalized products cost more than mass cus-
with imbedded sensors are easily integrated with computer- tomized ones. Bespoke personalized products would cost even more,
JID: CIRP
where the production system is akin to craft manufacturing, albeit human involvement. Upon receiving an order, single or few products
using more advanced machines and technologies. are made to specification and tailored to the customer requirements.
Value creation within industrial enterprises has been gradually The work is carried out manually and/or using standalone versatile
shifting from manufacturing products to providing services or a com- machines with varying degrees of automation and sophistication.
bination of both [178]. This type of manufacturing still exists to satisfy the need for special
The production era - push model: Mass production follows the push customized products.
model of Make-to-Forecast and the material resource planning (MRP) Several “variety-oriented” manufacturing systems have evolved
principle of Make-to-Stock, where products are produced in large over time influenced by changes in products, production technology
quantities based on volume forecasts regardless of customer orders and processes, production volume and varying degrees of automa-
[56] and production planning and control flows from upper levels to tion, intelligence and adaptation as depicted in Fig. 5.
the shop floor adding value through manufacturing processes.
The marketing era - pull model: Manufacturing companies adopted
Make-To-Order strategy in which parts are ordered from suppliers
based on actual customer demand. In this strategy, firms used economic
ordering quantity principle, work-in-process, Kanban and base stock
strategies for inventory management and control. A hybrid approach
utilizes the benefits of push and pull business models where a portion
of the manufacturing system is dedicated to producing common com-
ponents shared among a product family, i.e. product platform, using a
push model and Make-to-Stock strategies. Differentiating product com-
ponents are manufactured based on customer orders, following the pull
business model / Make-to-Order strategies, hence, delaying the differen-
tiation point in manufacturing and maximizing efficiency and profit.
The relationship era - JIT model: In the seventies, production waste
reduction started gaining momentum to reduce costs, and Just-In-
Time (JIT) philosophy gained traction. It aims at supplying the right Fig. 5. Manufacturing paradigms & manufacturing systems paradigms [Intelligent
Manufacturing Systems (IMS) Centre, U. Windsor].
components, in the right place, at the right time, with the correct and
exact quantity and order [208] and uses the Kanban system to control
the material flow. Dedicated manufacturing systems (DMS). This significant change in
Industrial product-service systems - IPSS model: Industrial product- production strategy goes back to 1908 when Henry Ford introduced
service systems business models can be either product-oriented affordable mobility by mass producing the Model T sedan car with
where suppliers provide customers with a product and associated only 2 variants, then rationalized limited product variety in mid-
service, or use-oriented where suppliers provide the product’s ser- 1980s. The moving assembly lines signalled the shift from craft
vice to customers through rental or leasing, and/or result-oriented manufacturing to standardization, interchangeability, automated
with agreed outcome [167]. IPSS business models also apply to material handling systems (MHS), and Taylor’s production principles.
manufacturing systems where focussed flexibility can be achieved by Dedicated manufacturing lines (DML) and systems are designed and
a) reconfiguration guarantee business model, in which the supplier optimized to mass produce single or very limited product variants
provides an initial manufacturing system with pre-designed flexibil- following the “Economy of Scale” strategy to maximize efficiency and
ity modules to be added as needed, and b) capacity guarantee busi- minimize cost. Factory automation and robotization are actively pro-
ness model, with built-in ability to change capacity if needed. These moted by companies such as Fanuc’s Oshino-mura factory, to achieve
business models require close cooperation between the supplier and long hours of continuous unmanned machining [59].
customer throughout the life of the manufacturing system [33,34]. Flexible manufacturing systems (FMS) consist of one or more inte-
Circular economy sustainability model: It is a business model that grated group of machines (NC, CNC, distributed numerical control
supports sustainability throughout products life cycle by re-using, re- (DNC)) and material handling equipment under central computer
pairing, re-manufacturing, and re-cycling technologies and impacts control for the automatic processing of palletized parts with flexible
manufacturing systems design and operation. Linear economy is routing.
characterized by a make-use-dispose business model [207]. Circular The concept of flexible manufacturing was developed by J. Lemel-
economy aims at “a manufacturing system in which there is no waste, son, an American industrial engineer and inventor in the early 1950s.
where the products of today are also the raw materials of tomorrow”. His 1956 automation patents included a "machine vision" and a
This business model has been actively researched and several proto- robot-based system that could weld, rivet, convey, and inspect manu-
types and implementations have been reported [31,32]. factured goods. Bar code scanning technology was developed around
Challenges exist in applying and monetizing the sustainable circu- the world and installed everywhere from supermarkets to automo-
lar economy business model due to the complexity of products, mul- bile assembly lines. Systems based on Lemelson's FMS inventions
tiplicity of materials, short life cycle of products and widely variable debuted on factory floors in the U.S. and Europe in the late 1960s and
conditions of returned products. The use of Industry 4.0, IoT and proliferated in the 1970s.
design for sustainability can extend the product’s life cycle and Flexible manufacturing systems continued to be used in discrete
enhance value creation towards a more innovative, resilient, and sus- manufacturing in response to the need for products customization
tainable economy. and greater responsiveness to changes in products, production tech-
It is evident that manufacturing systems business models have nology and market demands. They are suitable for mid-volume, mid-
evolved from providing more affordable products through mass pro- variety production of pre-planned parts/product families with simi-
duction (push) to more value through flexible, reconfigurable and larities in design features and/or production processes. They capital-
changeable manufacturing systems (pull) and continue to maximize ize on the commonalities and grouping methods to produce the
value by reducing waste (JIT and circular economy) and maximizing desired variety while achieving the efficiencies of high-volume
value to customers through customized and personalized products manufacturing. FMSs are designed a priori with built-in flexibility
(pull) while enhancing the economic viability of manufacturing sys- features, for anticipated variety within the product/part family,
tems and companies. including programmability, universal-adjustable fixtures, stream-
lined tools, changeable tool magazines, limited local buffers and cen-
2.1.4. Manufacturing and manufacturing systems paradigms tral automated retrieval system (ASRS). The product family-oriented
Craft Manufacturing or job shops feature general-purpose design, and use of group technology, and clustering in parts design,
machines, low-volume, high-variety production, and significant system layout, and process and tool planning are pre-requisites for
JID: CIRP
their success. FMSs are applied in different discrete manufacturing for Changeable manufacturing (CM). Changeability is key to dynamic
fabrication and assembly. adaptation of industrial production including FMS, RMS and beyond.
Flexibility can take many forms including machine, process, prod- It applies from machines/workstations level to cells, systems, seg-
uct, product mix, routing, material handling, production volume, ments, factories, and production networks. It includes characteristics
labor, and expansion flexibility. The main objective is adaptation to to accomplish early, foresighted and economic adjustments of the
changes in production volume and product variants, within a pre- structures and processes on all levels in response to change drivers
planned family, without interruptions in production for changeovers [204]. Manufacturing changeability is an umbrella which embodies
between models or penalty in time and cost. The concept of “focused changeover-ability, flexibility, reconfigurability, transformability and
flexibility” [183] refers to building flexibility into a limited section agility according to the level within the manufacturing organization
within an otherwise dedicated manufacturing line (DML) or system, [158,204].
which features flexible and programable automated machines Some definitions would be helpful to put these types of change-
and MHS where limited variations in the product is allowed while ability in perspective. Changeover ability is the operative ability of a
not incurring the capital investment needed for a completely flexible single machine or workstation to switch between work pieces at any
system. time with minimal effort and delay. Flexibility is the operative ability
The enablers of FMS are programmability for quick changeover of a manufacturing or assembly system to switch with minimal effort
between different part/product variants; computer integrated control and delay between parts/products within a pre-defined family of
and operation of system modules and production schedules for more products through soft/logical changes (programs, plans) and the
agility and responsiveness to change; parts pre-palletization, adjust- addition or removal of minor functional elements without changing
able adaptable and universal fixtures and tooling to reduce time the system physical structure. Reconfigurability is the tactical ability
waste during the production cycle; flexible routing to reduce down- of an entire manufacturing and logistics system to switch with rea-
time and increase machine utilization; built-in sensors, realtime con- sonable effort to new, albeit similar families of products by changing
trol and decision making for ease of fault detection and recovery; and the manufacturing processes, machines, material flows and logistical
adaptable process plans and production schedules. The key features functions as well as their structure, both physically and logically.
of FMS are adaptability, responsiveness, agility, waste reduction and Changeability has additional characteristics beyond FMS and RMS
lean manufacturing. Flexible manufacturing systems follow a pull including transformability and agility. Transformability is the tactical
business model. They are robust but have high initial capital cost and ability of an entire factory structure to switch to another product mix
their flexibility features are sometimes under-utilized. including production and logistics systems, facilities and buildings,
Reconfigurable manufacturing systems (RMS). The reconfigurable organization structure and process, and personnel. Enablers of trans-
manufacturing concept has emerged in the nineteen nineties [91] to formability include universality, modularity, scalability, mobility, and
achieve changeable functionality and scalable capacity. In RMS, compatibility. Global production networks (GPNs) design, operation,
machine components, machines, cells, and/or material handling units and their transformation into changeable GPNs have been reviewed
can be added, removed, modified, or interchanged to respond to [96]. Agility is the strategic ability of an entire company to open new
changing market requirements or technologies. RMSs provide “cus- markets, to develop the requisite products and services, to respond
tomized flexibility on demand” for a part/product family. They can be to change, and to build necessary manufacturing capacity within and
improved, upgraded, reconfigured, and extended rather than beyond the walls of the factory. Agility, therefore, is an enabler of
replaced. dynamic adaptation of an organization [204].
Reconfigurable manufacturing systems have distinguishing char- Manufacturing changeability, design, operation, enablers such as
acteristics and enablers which affect the ease and cost of re-configu- reconfigurable process plans (RPP) and adaptive production planning
ration. These include: a) modularity of physical and logical modules, and control (APPC), applications, performance indicators, cost and
b) integrability of system modules through standardized hardware economic justification, industrial adoption, benefits, and challenges
and software interfaces, c) customization of system capability to have been researched. A comprehensive framework of physical
match the variants within a planned product family, d) convertibility (hard) and logical (soft) capacity and capability adaptation in
to allow quick change-over between product family variants and manufacturing systems is illustrated in Figure 6.
adaptation for future product changes, e) diagnosability to identify
quickly the sources of quality and reliability problems requiring
repair or maintenance and f) scalability to alter production capacity
and capability by adding/removing system components.
The “plug and produce” reconfiguration may be physical or logical
and can take place at three levels: system level (machines, buffers,
MHS, parallel lines) [109], and machine level (machine modules/
axes, tool magazine, modular robots, entire machines) [4,123,146]
and controls level (open-architecture control modules, reconfigura-
ble control systems, CNC programs) [146]. Hardware reconfiguration
also requires major changes in the software used to control individ-
ual machines, complete cells, and systems as well as to plan and con-
trol individual processes and production; but it can potentially
increase the life and utility of a manufacturing system. Fig. 6. Manufacturing systems changeability includes flexibility and re-configuration
(modified from [42]).
A reconfigurable manufacturing system (RMS) architecture was
proposed [67] to produce personalized products by including parallel
subtractive and additive machine stages and allowing flow back- 2.2. Co-development of products and manufacturing systems
tracking, which can lead to less efficiency due to the complex flow,
more movement times, imbalanced production and complicated The co-platforming and biologically inspired co-evolution of prod-
scheduling. Examples of RMS research include design and configura- ucts and manufacturing systems are two important developments in
tion [13,22,82], optimal reconfiguration strategies [209], readiness of formalizing their bi-lateral association and co-development.
manufacturing firms for implementation [14], product family design Products platforms methodologies are useful in managing prod-
[3,90,116], performance [66,93], and production planning and control ucts variety and coping with dynamic and uncertain market
[68,126]. FMS and RMS were compared regarding pre-requisites, demands [170] due to the significant efficiencies in design, process
scope, features, intelligence, benefits, limitations, cost and life [42]. and production planning and manufacturing of families of parts/
Manufacturing paradigms evolved from mass production to mass products by capitalizing on the commonality between members of
customization and personalization [81]. the same product family. Modules are added/removed from the
JID: CIRP
platform to generate product variants. Product platforms are effec-

tive in achieving economy of scope by mass customizing products
in high production volumes while achieving economy of scale as
demonstrated in [60].
A methodology to develop manufacturing systems platforms was
introduced [44]. The manufacturing system platform consists of the
core machines, workstations, manufacturing processes required to
produce the core characteristics of the product family platform. Addi-
tional machines capabilities or machines needed for certain product
variants can be added through system re-configuration while the
core platform remains unchanged.
2.2.1. Co-platforming of products and manufacturing systems

A novel methodology for designing manufacturing systems by co-
Fig. 7. Co-evolution model for products and manufacturing system [9].
platforming them with product families was introduced [44]. It used
matrix-based mapping for machining and assembly and was vali-
2.3. Biologicalisation in manufacturing systems
dated using automobile cylinder engine blocks families of 4, 5 and 6-
cylinder engines machined using the same system [1,2].
A new emerging frontier in the evolution of the digitalization and
Other methods for integrating products and manufacturing sys-
the 4th industrial revolution is named biologicalization in manufactur-
tems development include: conceptual design [111], product archi-
ing which is defined as “The use and integration of biological and bio-
tecture commonality [89], simultaneous assembly and disassembly
inspired principles, materials, functions, structures and resources for
for customizing products platforms [18], product platform optimal
intelligent and sustainable manufacturing technologies and systems
configuration and co-planning, and relationships between products
with the aim of achieving their full potential” [27].
and systems, knowledge discovery using Bayesian networks [71],
The underlying concept of biologicalization in manufacturing can-
matching product components with specific process capabilities [24],
not be considered as new. What is novel, however, is the acceleration
as well as a cladistics model to determine the relationship between
of its realization, which builds on the capabilities available today and
product features and the associated machining capabilities [11]. The
prospectively in the future through the Industry 4.0 developments.
developed co-platforming concept and methodology were also used
Although not exactly under this name, the principles of biologically
to develop a methodology that integrates the product platform syn-
inspired manufacturing systems have a remarkable history [26]. The
thesis with the selection of suppliers to form a suppliers platform
concept of biological manufacturing systems (BMS) was introduced in
[132].
[189] with the aim to deal with the dynamic changes in the external
and internal environments in the whole product life cycle from plan-
2.2.2. Co-evolution of products and manufacturing systems ning to disposal, relying on biologically-inspired ideas such as self-
Engineers have been fascinated with the powerful biological growth, self-organization, adaptation and evolution [190,191,193].
transformations and their mechanisms and biological mimicry has Cyber-physical manufacturing is evaluated in the light of Professor
long inspired products designs, algorithms and optimization techni- Kanji Ueda's legacy in [195].
ques for planning, controlling, and scheduling of manufacturing sys- The cyber-physical era [119], i.e. the unprecedented integration of
tems [169]. A framework for the coordinated evolution the physical and the cyber spheres in industry, creates opportunities
(co evolution) of products, processes, and production systems (Spe- to realize biology inspired solutions in practice, including production
cies) was elaborated [187]. systems and organizations.
Holonic systems, bionic systems and fractal manufacturing sys- Fig. 8 maps the various manufacturing topics with similar biologi-
tems, multi-agent systems applied to manufacturing systems have cal elements (including botany) including “top-down” procedures, or
been investigated [99]. Intelligent adaptations which possess self- problem-driven, or manufacturing-driven, or technology pull, and
properties, such as self-configuration, self-organization, self-optimi- “bottom-up” approaches, i.e. from biology to manufacturing.
zation, and self-healing, are useful to apply to ensure robustness,
scalability, flexibility and re-configurability, and support developing
adaptive systems. These are discussed further in section 3.
This section focuses on the important phenomenon of symbiotic
co-evolution observed among species in nature and its use to model
co-evolution of products and manufacturing systems. The concepts
of dynamic evolving families of parts and products inspired by evo-
lution in nature, and co-evolving products and manufacturing sys-
tems were introduced [9]. Biologically inspired symbiosis, in the
world of artefacts, between products and their manufacturing sys-
tems using cladistics models was introduced and further exploited
as a powerful classification method to suggest future trends in the
design of products and their manufacturing systems [45]. This pow-
erful co-evolution model was applied to the co-evolution of milling
and turning machine tools and products using data from Morei-
Seiki as shown in Figure 7. An extended multi-domain evolution
and cladogram model was later introduced and demonstrated for 3
domains [10].
A new biologically inspired co-evolution model of products design
and manufacturing systems capabilities was introduced [9] using Fig. 8. Potential applications of biological transformation in manufacturing [27].
parsimony analysis of cladograms. The model is validated using sev-
eral case studies including assembly of automotive engine accesso- 2.4. Manufacturing systems life cycle and sustainability
ries. Bio-inspired phylogenetics for designing product platforms and
delayed differentiation utilizing hybrid additive/subtractive Sustainability has three pillars; environmental, social, and eco-
manufacturing was developed [46]. nomic all of which affect manufacturing systems. Environmental
JID: CIRP
sustainability emphasizes environmentally conscious manufacturing The pilot plant includes technologies to support products disassem-
processes, and practices, and reduction of energy usage, resources bly, remanufacturing and recycling of materials, including mechani-
consumption and harmful emissions [73]. Social sustainability is con- cal pre-treatments, and implementing the most valuable End-of-Life
cerned with the wellbeing of the humans who work in these systems (EoL) strategy for the various parts [25]. Model predictive control
and the quality of their work as they are increasingly called upon to (MPC), knowledge-based modules for adaptive distributed control
interact and share tasks with technology, machines, and robots. The systems, and a generic evolutionary control knowledge-based mod-
present I4.0 industrial revolution has the capabilities required to ule (GECKO) multi-agent distributed control approach were devel-
focus on the use of technology to support humans and improve their oped. Online part routing problem (OPRP) where the reconfigurable
quality of life and their jobs including within manufacturing systems transportation systems (RTSs) mechatronic modules transport parts
[177]. according to their destinations efficiently and collision-free was
Economic sustainability implies a good balance between the cost implemented. A hyper spectral imaging (HSI) system for in-line rec-
of manufacturing and profits to ensure business continuity. This in ognition and classification of shredded products mixture composition
turn drives efficiency, waste reduction and productivity affecting the (i.e. percentage of metal and non-metal fractions, shape, and dimen-
design and operation of manufacturing systems and the whole enter- sional distribution) enables application of smart waste classification.
prise [203]. Important strategies for achieving economic sustainabil- A virtual system model (Digital Twin) connected to the real plant was
ity of manufacturing systems are highlighted next. developed for implementing concepts of the digital de-manufactur-
Co-evolution and co-platforming of products and manufacturing ing factory. In addition, the plant supports technology services to
systems proved to have a significant effect on the life and economic companies for assessing new integrated technological solutions for
sustainability of manufacturing systems. Having an optimal stable circular economy, mainly in the automotive, white goods, and tele-
core and non-core of machines capable of producing all product fam- communication sectors. It is also used for training and education as a
ily variants as the product and manufacturing system evolve prolongs learning factory [186].
the life and utility of a manufacturing system beyond one product
generation. They minimize the cost and need for factories re-tooling 3. Smart manufacturing systems (SMS) paradigm
or re-building manufacturing systems every time a substantially new
product is introduced, reduce repeated ramp-up cost, and improve Smart manufacturing is defined as “fully-integrated, collaborative
quality. Co-evolution and co-platforming enhance manufacturing manufacturing systems that respond in realtime to meet changing
systems economic sustainability and prolongs their useful life. demands and conditions in the factory, in the supply network, and in
The various stages of a manufacturing system life from its design, customer needs” [128]. There are a few similar paradigms which rely
modelling, planning, construction, operation, reconfiguration, and re- on previous and foreseeable further developments of computer sci-
design when needed are illustrated in Fig. 9 including end of life end ence, information, and communication technologies, as well as
of life reusing, recycling, and retiring. manufacturing science and technology, and promise higher, sustain-
able performance of manufacturing systems.
The cooperative and responsive manufacturing enterprises (CORME)
concept emphasizes abilities for cooperation and responsiveness of
future manufacturing enterprises, which are vital in competitive, sus-
tainable manufacturing. The compelling challenges due to generic
conflicts between cooperation versus competition, local autonomy
and emergence versus global behavior, adaptiveness and robustness
versus optimization, plethora of information versus responsiveness
are also discussed [196].
Cyber-physical production systems (CPPS) consist of autonomous
and cooperative elements and sub-systems that, based on the con-
text, are connected within and across all levels of production, from
processes through machines up to production and logistics networks
[118,119]. Three main characteristics of CPPS are underlined:
Intelligence (smartness) where elements can acquire information

Fig. 9. Manufacturing systems life cycle for sustainability (modified from [43]).
from their surroundings and act autonomously.
Connectedness, such as the ability to set up and use connections to
A manufacturing systems classification code was developed [41] the other system elements, including human beings, for coopera-
for characterizing the components of manufacturing systems includ- tion and collaboration, and to the knowledge and services avail-
ing machines, robots, material handling equipment in great details. able on the Internet.
This classification system represents the characteristics of all system Responsiveness towards internal and external changes.
components in strings akin to genetics codes, much the same as
OPITZ parts classification system. This novel classification and coding Cyber-physical production systems (CPPS) enable the 4th Indus-
system was demonstrated and applied to machining and assembly trial Revolution (Industry 4.0) [6,8,16,88,154,173,184]. “Smart
systems [48] including industrial applications [172]. The developed manufacturing integrates manufacturing assets of today and tomor-
manufacturing systems coding system is used during the design row with sensors, computing platforms, communication technology,
phase in identifying similarities, grouping of system components, data intensive modelling, control, simulation and predictive engi-
streamlining the acquisition of machinery, and assessing the system neering. Smart manufacturing utilizes the concepts of CPS, Internet of
design and operation complexity with a view to re-designing it to Things (IoT) (and everything), cloud computing, service-oriented
reduce complexity and hence cost. computing, artificial intelligence, and data science. Once imple-
Example of de- and re-manufacturing systems mented, these overlapping concepts and technologies will make
A laboratory scale “De- and Remanufacturing” pilot plant is manufacturing the hallmark of the next industrial revolution.” [95].
installed at CNR-STIIMA, Milan, Italy to integrate and validate multi-
disciplinary methodologies, tools, and technologies for the smart de- 3.1. Smart communication and connectivity
and re-manufacturing systems of the future, with specific focus on
mechatronic products. Research supporting this system is pursued at Throughout the development towards Smart Manufacturing,
three levels; single process/technologies innovations, integrated pro- important key issues include IIoT, cloud computing, edge computing
cess chain, and sustainable circular economy business models [31]. and fog computing, digital twins [181,198], digital shadows [174],
JID: CIRP
and service-oriented technologies. The 5th generation mobile net- used deep learning algorithms is presented and their applications
work (5G) promises high transmission rate, low latency and high towards making manufacturing smart are discussed [198].
security [29]. A critical review of standards applicable to smart auto-
mation is found in [103]. Architecture of intelligent perception, IoT, 4. Smart manufacturing systems implementations
cloud manufacturing is discussed in [182] and [179] respectively.
Smart manufacturing systems offer competitive advantages for
3.2. Smart products companies when technical progress is transferred to industry. Assess-
ment of the gap between the state of research compared to actual
Industrial product-service systems (IPS2) [110] and hybrid products industrial practice is accomplished using maturity models as indica-
[88] consider the dynamic interdependencies of products and serv- tors of progress. Ongoing innovation implementation is evaluated by
ices throughout the entire product life cycle. Smart products using examining various maturity models and analyzing the state of indus-
CPS represent the new generation of intelligent, agile, flexible and trial practice based on latest research trends regarding SMS. The
networked products [5,166,168]. Smart products are considered CPSs transformation path of different countries, and governments initia-
which integrate internet-based services [188]. tives are described.
4.1. Maturity models and benchmarks for manufacturing systems

3.3. Augmented reality for smart manufacturing systems design
Maturity model is a tool to assist manufacturing companies in
Augmented reality (AR) has matured and proven to be an effective
comparing their status quo with a defined target state and develop
and innovative solutions in design and manufacturing [129]. It is
an implementation roadmap [155,163]. Comparing relevant maturity
related to the more general concept of mixed reality (MR) that merges
models for manufacturing systems regarding CPS, Industry 4.0 and
real and virtual (digital) information into the user’s view [105]. Prod-
sustainability provides an overview of their purpose and scope [156].
ucts and manufacturing processes can be simulated, assisted, and
The capability maturity model integration (CMMI) and the process
improved using AR before implementation. Augmented reality
and enterprise maturity model (PEMM) are universally applicable as
enhanced human-machine interfaces are enabling users to manipu-
a basis for production-specific maturity models [69,197]. The smart
late components realistically using hand gestures and receive tactile
manufacturing system readiness assessment model (SMSRL) exam-
feedback (forces and torques) for more realistic interaction [201].
ines four dimensions: organizational, IT, performance management
and information connectivity maturity. The measurement categories
3.4. Smart manufacturing systems control are processes, personnel, software systems, output data format, key
performance indicators (KPI) and KPI relationship [87].
Some earlier concepts, often considered revolutionary, can now The smart manufacturing maturity model for small and medium-
find real industrial applications by utilizing the latest developments sized enterprises (SMEs) (SM3E) developed in the USA and Mexico
of information and communications technologies. contains five categories: finance, people, strategy, process and prod-
One such concept is the holonic (or agent-based) manufacturing uct and evaluates smart manufacturing and the organizational struc-
systems (HMSs) which consist of autonomous, intelligent, flexible, dis- ture [115]. PEMM, “Leitfaden Industrie 4.0” and the “Reifegrad fu €r
tributed, cooperative agents or holons [21,107,122,192,193,194]. One Industrie 4.0” examine an organization’s maturity level by analyzing
of the most promising features of HMSs is that they represent a tran- methodological competence and corporate culture [84,165]. They
sition between fully hierarchical and heterarchical systems [72]. focus on information flow and use of data in production such as infor-
Digital twins represent a way for realizing these earlier concepts as mation generation and processing, networking, interaction of CPS
they provide two-way interaction between the real and virtual and intelligent and self-controlling processes [84,152]. The “Leitfaden
worlds of manufacturing. “A Digital Twin is the digital representation Industrie 4.0” and the “IMPULS” maturity models focus on technical
of a unique asset (product, machine, service, product-service system assistance systems and production networks [12,102] for SMEs and
or other intangible asset) that compromises its properties, condition classifies the company competencies in production data processing,
and behavior by means of models, information and data” [174]. In machine-to-machine communication, company-wide networking,
the literature, digital twins and digital shadows are usually distin- ICT infrastructure, human-machine-interfaces and efficiency for
guished. Both incorporate data and information collected during the small batch sizes. Other models such as the Industry 4.0 / digital oper-
usage / operation phases of the product or production system. With ations self-assessment are cross-industry self-assessments with sub-
the use of digital twins and digital shadows, together with predictive sequent recommendations for action to achieve a higher degree of
engineering [95] anticipatory rather than reactive enterprises can be maturity and add perspectives on market and customer access, com-
realized. With the construction of high-fidelity models (digital repre- pliance, legal, risk, security and taxes [144]. The SIMMI 4.0 measures
sentations) of the phenomena of interest, future spaces can be maturity in information flow along the supply chain and across all
explored and appropriate decisions made [181]. hierarchical levels, cross-sectional technologies and digital product
development [100].
3.5. Data-driven smart manufacturing Another important aspect of maturity models with focus on CPS is
customer orientation. For example, the “Digitalisierungs” index
The volume of data collected is rapidly increasing due to the digi- assesses the digital maturity regarding customer relationship, pro-
talization of manufacturing and growing number of sensors and dif- ductivity, digital offers and IT-security [38]. The Industry 4.0 “Reife-
ferent IoT devices. Cloud computing enables networked data storage, grad Test” assesses the maturity regarding research and
management and off-site analysis [180]. Data-driven strategies are development, production, logistics and warehouse management and
important for companies to remain competitive. In manufacturing administration as well as sales and customer service. Both the Leitfa-
systems, multi-source, heterogeneous data are generated throughout den Industrie 4.0 and the Brazilian instrument to measure lean
the product life cycle. They can be characterized by 5Vs: high volume, manufacturing maturity include customer orientation and supplier
variety, velocity veracity and value [28] as data is generated and col- integration. The instrument to measure lean manufacturing maturity
lected [180]. Machine learning approaches can be used for processing evaluates information flows, corporate culture, suppliers and custom-
and analyzing big manufacturing data [120]. Algorithms such deep ers regarding quality of data source, problem solving, processes and
learning are increasingly being used. They represent extremely pow- tools, strategic planning, continuous improvement, supplier integra-
erful techniques for many applications such as pattern recognition, tion and customer orientation [153]. The Korean assessment frame-
however, their applicability to a specific problem, availability of work analyzes the maturity level of intelligent manufacturing
appropriate training patterns, and whether incremental learning is regarding the integration of data analytics in production as well as
important should be assessed. A comprehensive survey of commonly finances. The smartness assessment framework for smart factories
JID: CIRP
uses analytic network process and focuses on leadership and perfor-

mance. It is the only maturity model that explicitly considers perfor-
mance factors such as productivity, finance, quality and lead time as
evaluation criteria as well as leadership, processes, and system and
automation [98].
Most German and the Korean maturity models generally focus on
technology, automation and production and many German models
specifically cater to SMEs. The American and international models
often focus on connectivity, corporate culture and finance or the
company indicating the current technological mid-term outlook.
The Singapore smart industry readiness index (SIRI) assesses Fig. 11. Comparison of USA, EU, China, Japan, Germany and Korea.
Industry 4.0 based on processes, technology, and organization crite-
ria. The integration of operations, supply chain and product lifecycle
are evaluated. Criteria for assessing technology readiness are the research programs (e.g. with UK and Germany) to support small and
degree of automation, connectivity, and intelligence. In addition, the medium size innovation and manufacturing enterprises and technol-
organization and employees’ readiness for Industry 4.0 are evaluated ogy transfer of academic research in smart manufacturing. The Chi-
[171]. nese government has initiated plans to develop artificial intelligence,
Benchmarking studies assess the state of industrial practice, readi- robotics, better handling of big data and above all their initiative of
ness and implementation and underlying success factors. A recent “Made in China 2025” [138].
benchmarking of Industry 4.0 identified successful and proven The Japanese government supports initiatives regarding robotics,
approaches for German and European small, medium, and large com- new IT, IoT, industrial value chains or connected industries to build
panies. Manufacturing companies that specifically address smart up new industries and strengthen existing ones and enhance data
data and digital assistance of Industry 4.0 in addition to culture and exchange, use of AI in SMEs and cloud services [64]. Besides
methods tend to be ahead with their systematic implementation of manufacturing benefits, they focus on problems like the aging society
modern manufacturing systems (Fig. 10). with their Society 5.0 initiative to achieve high acceptance of human-
machine-interaction including in manufacturing. In Korea, the term
fourth industrial revolution is more used than smart manufacturing
systems. Government initiatives include Manufacturing Industry
Innovation 3.0 Strategy as part of the Creative Economy Initiative or
the Connected Smart Factory, along with company collaborations
regarding smart factories, intelligent manufacturing or smart engi-
neering [7]. Korea is making huge efforts to support their companies
through other initiatives from different ministries ranging from Sci-
ence and ICT to interior and safety [176]. The European community
supported many multi-national collaborative programs and projects
Fig. 10. Results of benchmarking Production Systems 4.0 [160].
focused on Industry 4.0 and its enabling technologies. Europe is het-
erogenous regarding programs, funding, and challenges. For example,
Successful companies consistently apply lean methods in Industry Scandinavian countries, the Netherlands and the UK are dominating
4.0 to increase agility and productivity, develop strategically impor- the digital economy and society index (DESI), which tracks digital
tant assistance systems and integrate users [171]. performance and competitiveness in Europe, whilst Malta has the
Smart data including effective analytics applications, consistent highest percentage of companies analyzing big data [57]. Regarding
semantics along the entire order processing chain, a middleware con- Industry 4.0, less than 35% of European companies have implemented
necting data of different domains as well as cooperation of data scien- two or more key technologies, which are social media, big data & data
tists and production employees are important success factors for analytics, cloud technologies, IoT, mobile services, robots and auto-
increasing productivity. mated machinery, cyber-security solutions, 3D printing, and artificial
intelligence, while only 25% is using big data analytics countries have
4.2. Modern manufacturing in the global context their own initiatives for implementing Industry 4.0 combined with
the digitalizing European Industry initiative in 2016 [147]. In Ger-
It is important to understand the economic and social impact of many, more than 20 Industry 4.0 competence centers are developed,
ongoing digital transition on the entire manufacturing economic sec- and German companies are investing €40 bn annually in Industry 4.0
tor in different countries, the role of manufacturing within an econ- technologies until 2020 [145]. The government supports initiatives to
omy and the measures taken to secure or strengthen their relevance strengthen Germany’s position including Industry 4.0, smart service
in global markets. world, high-tech-strategy, and collaborations with companies and
Some countries are selected based on rankings like the global research organizations on digital factories and Internet of Things
competitiveness index (GCI) and the global manufacturing competi- [87,163]. Together with research related to sustainability and effi-
tive index (GMCI) which reveal their economic and scientific rele- ciency [151], it is expected to achieve an overall efficiency (productiv-
vance [37,58] to examine their efforts and spending related to GDP, ity, energy and resource utilization) gain of 18% in five years [145].
Industry 4.0, regulations, tax policies, energy, transportation, health The gap between companies is mainly seen in Asia where some
cost, workforce quality, infrastructure and innovation of different corporations are huge and innovative while followers are usually
countries [58,130,131,202,205,206]. Fig. 11 summarizes the results small without advanced manufacturing systems. While governments
[130,131,175,184,205,206]. Many countries introduced initiatives to support companies significantly in achieving Industry 4.0 standards,
support the digital transition of small, medium, and large manufac- they often face obstacles such as changes in regulations which hinder
turers. Some examples include the USA smart manufacturing pro- commercialization of innovative solutions [137].
grams and clusters of excellence in IoT, advanced manufacturing, In addition to governments supportive initiatives, it is up to the
new materials and software [7,74]. Germany introduced the Industry sector’s ingenuity to invent and innovate new concepts, ecosystems
4.0 initiative and funded related academic and industrial research and solutions and implement them successfully in practice, for exam-
and development programs. Canada established an advanced ple in smart factories. To apply CPS to manufacturing, the concept of
manufacturing program through the national research council of the smart factory as a hyper-connected network-based integrated
Canada (NRC), next generation manufacturing supercluster (NGen), a manufacturing system has been developed. It uses IoT to connect the
network of super research clusters, and collaborative international real shop floor equipment such as machines and assembly lines by a
JID: CIRP
CPPS Platform that monitors, plans and controls manufacturing steps The idea of the IoP is pursued within Industry 4.0. Linkage hetero-
by MES which are supported by a cyber model of the real production geneously available data from different IT systems and sensors (as
- digital twin. Based on this model, discrete event simulation and done in IoT); the IoP generates a digital representation of the
data analytics methods help improve the factory performance [15]. manufacturing system at different levels of aggregation, called digital
This vertical integration of sensors, MES and ERP helps the smart fac- shadow. Application-specific apps with detailed production engi-
tory to achieve product life cycle integration and a horizontal integra- neering models can be created based on the digital shadow as shown
tion of several smart factories into a smart supply chain. The product in Fig. 12. They aim to enable production managers to react faster to
life cycle integration allows an early integration of the optimal pro- problems and gain new insights to increase productivity. Therefore,
duction structure for upcoming new and changing products and their the IoP is an extension of the IoT but with specific production tech-
manufacturing requirements in the factory development planning. nology models and associated data structures [162].
PLM software is used to link R&D data with production and user-
data. The connection of several smart factories in an inter-company
value chain finally creates a hyper-connected network that allows
new forms of cooperation and business models [141].
Smart manufacturing (Industry 4.0) industrial adoption
The smart manufacturing (Industry 4.0) adoption report [85] anal-
yses the status in the manufacturing industry with regard to Industry
4.0. Overall, less than 30% of manufacturing companies use Industry
4.0 technologies to a great extent. In the regional comparison, North
American companies have the highest adoption of Industry 4.0. The
adoption of technologies and use cases varies between different
industry sectors, with companies in the automotive sector using
Industry 4.0 most extensively. For this also, the average return on
investment and the likelihood to increase budget for Industry 4.0
technologies is highest in the automotive sector. Fig. 12. Digital shadow and its use in different manufacturing cycles.
The e.GO Life electric car is produced in an Industry 4.0 plant

4.3. Industrial applications of smart manufacturing systems
where technologies like sensors, connected material handling equip-
ment, automated guided vehicles (AGVs) and robots are used. The
The different smart manufacturing paradigms have already been
data collected in production is transmitted via 5G network, which
springing to life in industry in the past years with many different sol-
guarantees high data transfer rates and reliable low latencies sup-
utions. Examples of implemented smart manufacturing systems and
porting realtime communication. Production and other data from IT
technologies, advantages and challenges are presented.
systems are integrated to enable cross-domain collaboration. The
change request procedures (Fig. 13) are used. Data is selected and
4.3.1. Smart digital manufacturing and assembly systems aggregated on an application-specific basis to support daily work and
The e.GO Life is an electric car designed and manufactured in decision-making.
Aachen, Germany by the e.GO Mobile AG. This example is an excel-
lent demonstration of successful transfer of technical knowledge
from academia to industrial application as well as commercial prod-
ucts and demonstration of several Industry 4.0 enablers.
Long time to market and low agility limit existing manufacturing
systems in automotive industry. At the e.GO Mobile AG, the time to
market for new products is drastically decreased by using a highly
iterative product development approach. Horizontal integration of
the manufacturing system is essential to enable industrialization
while the product being developed is constantly changing without
defined design freezes. Hence, company-wide agility, including han-
dling of change requests, is required. A change request is a process
which initiates a change of the product after releasing it for further
development or manufacturing. It can be categorized into internal,
such as due to construction mistakes, and external, e.g. induced by
changes in customer needs and requests. These changes have a wide Fig. 13. Change request system at the e.Go LIFE electric car factory.
range of effects on the company as they influence employees, prod-
ucts, processes, and cost. Establishing an effective change request
process ranges from increasing the customer surplus and integrating One example of implemented connected systems is the end of line
technological innovations to shortening the time needed for a devel- control. The employee uses an augmented reality app to report
opment process in an agile environment. These two main goals are errors. When entering a light tunnel, the vehicle is automatically rec-
contradictory for most manufacturing processes. Certain steps are ognized via RFID. The product life cycle management system provides
needed to reduce the conflict, by initiating a combination of organiza- information about the vehicle configuration and necessary checks.
tional actions and other manual actions regarding the information Cameras scan the vehicle and the image data is processed using artifi-
and communication technology infrastructure. cial intelligence (AI) application trained to detect errors. Quality
Other key enablers of efficient change requests are filtering rele- inspectors check the errors and manually enter errors in the app. All
vant data, transparency and flexibility of data distribution, defined errors are documented in a manufacturing execution system (MES)
data structures and high-quality data. The main challenge in realizing which provides necessary data for rework. After rework, all informa-
these key enablers is insufficient working infrastructure, which can tion is added to a digital vehicle file. This example shows how differ-
effectively be overcome by implementing the IoP framework which ent IT systems, sensors, AI, and augmented reality are connected in
combines the use of apps and smart data with integrated data man- one system to simplify and improve the quality inspection. All neces-
agement acting between raw and smart data, hence, enabling effi- sary information is provided by an app, hence, reducing information
cient change processes and highly iterative product development search and increasing process stability. All data for a vehicle is imme-
[23]. diately stored in the vehicle file creating continuous documentation
JID: CIRP
and transparency where all assembled parts and performed work $10,000. This smart factory project resulted in a 50% ROI, with a
steps are traceable. In addition, the pre-evaluation of errors using AI breakeven duration of about two years [54,55].
enables a focused and faster inspection and improves quality. As a
result, the time to market is decreased compared to traditional OEMs. 4.4. Research use cases
The e.Go electric cars, now in serial production (Fig. 14), are avail-
able in different configurations/variants e.g. size and wheels material, Extensive research in academia and industry is being devoted to
and depending on the model can feature infotainment system, seat developing enablers of smart manufacturing systems including
heating, parking sensors and LED headlights. human-machine collaboration. The following are but a few examples
of ongoing joint academic-industrial research projects.
4.1.1. Autonomous, mobile, and ad-hoc cooperating robot teams

High integration costs and complex programming often limit the
use of industrial robots to identical repetitive tasks behind protective
fences and separate from workers. Advanced sensors and increasing
miniaturization of control systems allow mobile industrial robots to
navigate autonomously through the factory while using end effectors
to perform a variety of tasks with sensory feedback. This was demon-
strated by the Technical University of Munich (TUM) with three
industrial partners as part of the FORobotics research network for uti-
lizing autonomous mobile robot platforms in smart production. This
use case features modular software architecture for flexible tasks
execution through reconfiguration of robot tasks, and interaction
Fig. 14. e.GO LIFE electric car production line in Aachen, Germany [e.GO Mobile AG]. between humans and robots. It includes interpretation of human ges-
tures and 6 degrees of freedom pose estimation for assembly and
heavy objects bin picking, a world model representing multi-user
4.3.2. Internet of things (IoT) platform at Hirotec
capability, combined consideration of geometric, topological, and
Hirotec is a US$1.6 billion automotive part and tooling manufac-
hierarchical environment information, and integration of predicted
turer with 23 locations around the world. To improve quality, reduce
and planned object dynamics. It enables motion and task planning in
downtime and optimize production planning, Hirotec implemented
unstructured environments. For Human-Robot cooperation, the robot
PTC’s ThingWorx IoT Platform and Kepware's IoT Gateway in its
uses cameras to enable detection of human hand, eye, or speech ges-
Detroit, Michigan factory.
tures. A projector displays information on the floor about the motion
The company recognized early on that access to operating data
direction and target for the human to understand the robot’s activi-
from their machines had an enormous impact on planning, avoiding
ties. Various psychological and ergonomic aspects such as trust in
reactive maintenance, and missed opportunities. Today, production
automation, stress, safety perception or workers’ attitudes and their
management can use realtime data from the factory linked to the ERP
experience were studied in this user-centred application using field
system for planning and optimizing CNC modules, systematic analy-
observations, laboratory experiments with participants, employees’
sis and decision making. To support the long-term IoT vision, Hirotec
surveys, and interviews to use in further development. This combina-
developed an IoT framework with short six-week agile sprints. While
tion of human and robot capabilities resulted in new forms of ad-hoc
a full IoT implementation would have taken several years for the first
robot human cooperation in a smart manufacturing system (Fig. 15)
results to emerge, the Scrum model delivered early visible and quan-
[76,77].
tifiable progress. In six weeks, the company went from no realtime
visibility in its operations to full visibility of the operational states at
any time, with data analytics capabilities for trend forecasts of uptime
and efficiency.
Hirotec expects that the IoT measures will affect every aspect of
the company, from operating the business and IT to financial fore-
casts, customer relations and sales. As the sprint projects progress,
Hirotec will obtain various contextual data to develop new improve-
ment case [104,148,149].
Fig. 15. Human robot collaboration autonomous production control [Fraunhofer
IGCV/Institute for Machine Tools and Industrial Management] [76, 77].
4.3.3. Smart factory by Ericsson

Ericsson, a global distributor, provider and manufacturer of com- 4.4.2. Flexible reconfigurable allocation of work tasks
munication technologies goods and services in Sweden, invested in a This use case conducted at iwb, TUM introduces flexible reconfig-
smart factory in the Jiangsu province, China. It is located in Nanjing urable work tasks allocation using “jumpers” or “auxiliary workers”.
and focuses on the manufacture of 5G and 4G radio technology prod- It is implemented on a balanced mixed-model assembly line at a
ucts. With an investment of about US$50 million, smart manufactur- brakes and transmission systems manufacturer, which is character-
ing capabilities were enabled and automated streamlined production ized by a one-piece flow of three product variants, running through
systems were implemented. The new technologies include a modular the same assembly process and is fed by three pre-assembly stations.
automatic assembly line for 5G radios as well as an upgraded auto- Hybrid combination of automated and manual processes at the same
matic packing line for higher speed and efficiency. Modernized 5G station is conducted with proportionately higher manual activities.
testing equipment allow more efficient processes and higher flexibil- Initially, the workstations were equipped with touch screens for digi-
ity in reacting to changes. Data analytics capabilities were imple- tally displaying work instructions. To avoid interruptions in produc-
mented for components and objects recognition in production using tion caused by bottlenecks, knowledge deficits or imbalanced
AI and machine learning. IoT-technologies enabled an automated distribution of the workers tasks; smart watches (wearables) were
alert system for critical issues and faults. The decreased latency of used by workers to initiate requests for “jumpers” and coordinate
error in production to initiate counter measures allows accelerated their deployment (Fig. 16). The smart watch app provides three
production and increased efficiency. Reduced cost of material and jumper request options; namely replacement, assistance or coach
manual machine maintenance due to reduced human errors and pro- jumper [40]. The use of smart watches and apps to coordinate
duction system downtime, amounted to annual savings of up to US replacement workers (jumpers) increased production by 15% by
JID: CIRP
is researched at the laboratory for manufacturing systems and auto-

mation (LMS) at University of Patras, Greece. An assembly line
inspired by assembly of the front axle of a passenger vehicle where
dampers are assembled on the disks with 480 parts for three axle var-
iants/models in 8-h shifts. Mobile dual-arm robot workers can auton-
omously navigate on the shop floor and perform multiple operations
such as screwing, handling, and drilling while acting as assistants to
human operators. Human operators’ behavior is estimated using
multiple sensors data to enable human robot interaction, task plan-
ning and robot behavior adaptation to surroundings. Robot percep-
tion libraries allow MRPs to avoid collisions, dock at different
workstations accurately and detect parts orientation for assembly
[112,140].
Fig. 16. Work tasks assignment with an App for jumpers’ allocation [40].
Application of the new system reduced weight lifted by operators;
increased assembled variants from 3 to 6 models and operator utili-
zation; reduced number of operators from 3 to 1 as well as part flow
maximizing the operation efficiency and capacity utilization of the time.; improved quality and productivity by automating repetitive
assembly line, and reduced reaction time for handling interruptions strenuous tasks, and reduced set-up time for changing models
by 20%. Insufficient internet speed, which caused the calls for jump- through robot mobility. The return on investment (ROI) was 12
ers to be interrupted frequently, presented a challenge. This confirms months.
the importance of fast, high capacity integrated IT-infrastructure as a Some limitations of robot perception accuracy due to lighting con-
prerequisite for successfully implementing smart technologies in fac- ditions, realtime continuous object detection, and networking issues
tories. The overall feedback from workers was very positive, as they were encountered when multiple sensing devices were encountered
could receive support promptly as needed. One downside for some and required human participation and support. This shows that the
jumpers was the uncertainty caused by not knowing the next assign- future of AI in robotics requires combined AI applications and crea-
ment. tive human operators.
Summary of observations on industrial and academic use cases
4.4.3. Autonomous matrix production control Different applications and use-cases were presented which exem-
Agile production system for remanufacturing using artificial intel- plified various aspects of smart manufacturing systems such as flexi-
ligence (AgiPROBOT project 2019-2024) is carried out by nine ble operation and production control using intelligent mobile robots,
research institutes at the Karlsruhe Institute of Technology (KIT) and connectivity and integration using advanced IT systems, use of smart
funded by Carl-Zeiss-Stiftung. Integrated semi-automated demon- automation and control technologies, and adjustable flow and adap-
strator factory for remanufacturing capable of autonomously disas- tive intelligent control of equipment and systems to improve perfor-
sembling automotive electric drives manufactured by Bosch after mance and increase responsiveness.
their usage phase characterized by a high degree of variety and Digitalization which is an important foundation for smart
uncertainty regarding their condition and specification is used. This manufacturing systems and the important role of humans in future
agile system is structured with a matrix layout to enable highly flexi- manufacturing systems and facilitating their cooperation with
ble and adaptable material flows (Fig. 17). It features several autono- machines is abundantly clear. These are only few of the numerous
mous capsuled stations and a variable material flow using examples of smart manufacturing research carried out in academia
independent AGVs and robots for inspection and disassembly in col- and industry around the world. They illustrate the necessity of inves-
laboration with humans. ting in research and development as well as effective technology
transfer and implementations.
Increased attention is being paid to making the concepts and
implementation of augmented human machine interaction in smart
factories more accessible and clearer to industry and academia.
Research centres and laboratories provide opportunities to explore
innovations regarding the digital transformation, impact of Industry
4.0 on manufacturing, and making digitalization and automation sol-
utions tangible to interested stakeholders. In different demonstrators
the human-robot collaboration, use of AI and mobile apps in
manufacturing and autonomous production as part of SMS have been
illustrated.
5. Adaptive cognitive manufacturing systems paradigm

Fig. 17. Autonomous adaptable matrix production control [Karlsruhe Institute of Tech-
nology. KIT]. As many innovations and disruptors along the axes of evolution
continue to appear, it is important to carefully consider, and be cogni-
The system performance is enhanced using reinforced machine zant of, those changes that would truly lead to paradigm shift(s) in
learning for adaptive dispatching, co-learning with humans in disas- manufacturing systems causing them to be designed, operated, con-
sembly operations, and decentralized hierarchical intelligent control trolled and/or used differently. Disruptive innovation, a term used in
via autonomous agents to utilize the potential of the modular matrix business and technology, is one that helps create a new market and
or grid manufacturing systems [78]. This use case demonstrates a value network, and eventually disrupts existing ones and displaces
“plug-and-work” functionality of manufacturing systems designed an earlier technology [30,80].
with flexible matrix layout and intelligent production control all of Paradigm shift, a concept identified and coined by the American
which enable adaptable remanufacturing of electric drives with physicist and philosopher Thomas Kuhn [94], indicates a change in
uncertain product specifications using adjustable automation. the basic assumptions within the ruling theory and represents a fun-
damental change in the basic concepts and experimental practices of
4.4.4. Reconfigurable intelligent robotic assembly system a scientific discipline. In manufacturing systems this means funda-
Dynamic reconfiguration of a factory by employing cooperating mental change in the ways products are made. This paradigm shift
mobile robot platforms (MRPs) and mobile product platforms (MPPs) criteria have been used in evaluating manufacturing systems
JID: CIRP
evolution trends in this keynote paper and in identifying future new industrial practice. The vision and recommendations are based on
manufacturing systems paradigms. the work of the five EU funded H2020 research projects
(2016 2020): A4BLUE, Factory2Fit, INCLUSIVE, HUMAN and MANU-
5.1. Towards bio-intelligent manufacturing WORK. The ACE Factories cluster has identified several lessons
learned and recommendations for successful technology and best
In the long term, transition is expected from the old “lifeless practices adoption including: a) augmented reality (AR) and virtual
manufacturing systems” to the manufacturing systems being alive: reality (VR) are efficient tools for on the job training, which increase
self-learning, cognitive, communicative, self-healing, and self-assem- productivity and enhance the workers’ well-being; b) making oper-
bling towards a “living manufacturing system”. Future manufacturing ators’ tacit knowledge, such as best work practices and problem
systems are expected to incorporate components, features, and capa- solving, visible and accessible with social media-based tools can be
bilities that enable the convergence towards living systems as a very effective complement for workers support and training; c)
hypothesized in [27] and further elaborated in [121]. It is observed the usage of wearable apparatus like exoskeleton devices has shown
that the ongoing Industry 4.0/Smart Manufacturing is increasingly their potential to reduce operator’s physical fatigue and increase
bringing together the physical, virtual, and biological worlds. Next their overall safety and productivity; d) ACE pilot cases have shown
phase of the evolution is envisioned to be the emergence of what can that human-centered factory solutions have positive impacts both
be labeled “human-centric” “bio-intelligent” manufacturing [113] on the productivity and well-being of the operators; e) the know-
integrating automation, information, bio-production, smart products, how of industrial workers must be protected from unauthorized use
and materials technologies. especially by data and analytics companies; f) the human-centered
paradigm shift will only be successful if work processes are
5.2. Human-centric adaptive manufacturing reshaped and new training approaches are introduced to support
continuous development of skills taking into account personal capa-
The manufacturing industry continues its evolution towards per- bilities, skills and situational preferences of individual operators; g)
vasive automation, while human-machine collaboration is advanc- new technical solutions for the realtime measurement of the opera-
ing by placing human operators in the center of attention. Even tor’s capacities, mental strain and adaptation to automated pro-
with the advances towards more intelligent automation, the trend cesses can be used to improve productivity and workers’ well-being
is an increased attention to the central role played by human work- and increase the value of humans role; h) providing factory workers
ers and their well-being both physically and psychologically as well with ways to influence and improve their work will increase work
as considering the environmental issues. This socio-technical motivation and productivity; e) changing work roles should be
approach to the evolution of the manufacturing system, where auto- implemented with consideration of the needs of elderly workers
mation is human-centric, cognitive, intelligent and environmentally such that no one is left behind; j) criteria related to enhancing trust
friendly has led an IEEE technical committee to call this “bio-auto- in the collaboration between automation and advanced technologi-
mation”, others have claimed that these are characteristics of the cal applications such as human-machine and human-robot collabo-
fifth industrial revolution (Industry 5.0), however, these terminolo- ration (HRC) should be considered; and k) small and medium
gies and associated implementations have not yet been widely enterprises (SMEs) should be supported in adopting human-cen-
adopted. tered factory solutions.
The idea of self optimizing machining systems (SOMS) in the For adaptive factory automation and management solutions inte-
context of Industry 4.0 was investigated [117]. Enabling technologies, grating the man in the loop, a methodology was proposed [35], vali-
principles, and methods are described that would potentially allow dated in two industrial cases, to integrate cognitive workload into
for the implementation of machining systems which are capable of the design of workplaces to match the human safety and well-being
adapting their parameters and settings autonomously, in order to necessities and the tasks cognitive requirements. The proposed
optimize for productivity, quality, and efficiency and concluded that approach allows for the human-in-the-loop within factory automa-
“last but not least, the higher complexity of SOMS requires new solution through seamless human and automation collaborative deci-
tions for human machine interaction”. sion-making, while monitoring production performances and
Industry 4.0 brought a great change in the interaction between workers well-being indicators.
workers and machines; the latter includes every kind of dynamic The symbiotic human-robot collaborative assembly issues were
technical systems such as automation, robots, decision support, discussed [199] Human-robot collaboration (HRC) in a manufacturing
equipment, and software [127]. Industry 4.0 allows communication context has been researched in the last few years, with a view to
between humans and machines throughout a highly networked facilitating multimodal communication, dynamic assembly planning
environment, using automation technologies like CPS, IoT and cloud and tasks assignment assisted by deep learning. Insights on program-
computing and the various levels of the supervision and control sys- ming-free adaptive robot control through algorithm embedding and
tems. Furthermore, human machine interaction deeply changed brainwave-driven methods; and different techniques for mobile
over the years, and reached a new level of innovation in Industry worker assistance were discussed. Challenges and twelve future
4.0 due to some additional pillars including big data analytics; research directions were identified for further advancement in the
robot-assisted production; self-driving logistics vehicles; and aug- years to come. It envisioned that “with the support of the latest tech-
mented reality. It is possible to think of the resulting systems as the nologies of sensing, communication, AI, AR and robot control, HRC
new ‘internet-of-people-and-things’ in which the cyber-human sys- will find its way to practical applications on shop floors in factories of
tem (CHS) complements the activities deemed to be difficult for the the future”.
CPS and vice-versa, with the CHS having the supervisory control to Human-centric bi-directional interaction between hardware and
naturally leverage the needed cognitive, adaptive, and corrective software components in the system and the people associated with
actions. In the Industry 4.0 era, companies are required to use a its functioning will benefit from an effective blend and symbiotic
socio-technical strategy. In addition to investing in technological relationship between the principles and drivers of Industry 4.0 and
infrastructures, it is essential to value the human factor and workers Society 5.0 to maximize the effectiveness and contributions of the
well-being before technology, and drive and anticipate change. The humans in future manufacturing systems and enhance their well-
ability to solve complex problems and use critical thinking to help being.
organizations adapt quickly to changes in perspective is critical.
Database analysis must turn data into knowledge and strategic busi- 5.3. The multiple facets of adaptability
ness suggestions [127].
The HumAn-CEntred (ACE) factories [83] cluster, shared in a key Three important related mechanisms of effecting changeability
white paper, the understanding of future human-centered factories are resilience, robustness, and adaptability. Resilience is the capacity
and provided recommendations on how to bring this vision into to recover quickly from disruptions and spring back into the original
JID: CIRP
designed system state or normal operation. Robustness is fault-toler- normal/near normal designed system state or operation with the
ance, i.e. the ability to withstand disruptions without the need for least delay and losses. The key operative words in extreme adaptabil-
adaptation. Adaptability, however, is the ability to adjust to new con- ity are unanticipated, extreme disruptions, and least losses. The antic-
ditions and to be modified for a new goal, use or purpose. It is the ipation and prediction capabilities are therefore particularly
most relevant characteristic of evolving manufacturing systems and a important in forecasting impending disruptive changes and in timely
core foundation for any new system paradigm. planning and deciding optimal and economically feasible adaptation
System adaptability has been utilized in earlier paradigms such as responses and action.
flexible and reconfigurable manufacturing systems. A new classifica- Manufacturing systems response to the unprecedented disruption
tion is presented next. The four classes of adaptability are informative during 2020 Pandemic brought to the fore the importance of this
in differentiating the new adaptive cognitive manufacturing system class of extreme adaptability.
(ACMS) paradigm. “The COVID-19 pandemic is challenging politics, society and the
economy to an unprecedented extent. Its effects are so drastic that
5.3.1. Static adaptability it requires companies and industries not only to manage the crisis
The static adaptability refers to built-in pre-planned flexibility by in the short term, but also to develop strategic options for the
design (Section 2.1.4) enabled by the design of the system compo- future” [143]. Indeed, this multi-domain disruption which occurred
nents, modules, machines, configuration, and operating rules. It in 2020 significantly afflicted countries around the world. Drastic
allows flexible, resilient, and robust behavior within the pre-defined changes in production volume (increase and decrease depending on
parameters and boundaries such as pre-planned product family and the product) often resulted in shutdowns for extended periods.
well-defined range of production capabilities (scope) and capacity Increased demands for essential products such as medical supplies
(scale). This type of adaptability relies on universality and program- and protective personal equipment depleted existing stock in very
mability features of machines, robots, fixtures, and manufacturing short order and outstripped any planned production scope and
system, to cope with anticipated ranges of change in products and in capacity/volume. Supply chains were crippled or broke down
production volume. completely by travel and transportation restrictions as well as
national protectionism. Companies which produce pharmaceuticals,
5.3.2. Dynamic adaptability medical gowns, face shields and masks were asked to double, triple,
This class of adaptability is labeled dynamic because it involves and quadruple their production; others like auto-parts manufac-
action-oriented changes that affect the manufacturing system and its turers and OEMs were called upon to produce the essential products
constituents and involves external efforts by the technical specialists that are far from their normal products which presented many chal-
such as engineers, technicians, and workers to implement and realize lenges. Other companies faced drastically reduced demands and
the intended adaptation. It involves both physical hardware adapta- were forced to consider producing significantly different products
tion, and logical soft adaptation by reprograming devices, changing to stay afloat. Furthermore, the supply chains of just about all goods
controls of machines and/or system, and revising operating and and materials came to a near standstill. In summary, the pre-
sequencing rules as discussed in Section 2.1.4. It relies on built-in planned defensive strategies of flexibility, reconfigurability, agility,
flexibility and reconfigurability enablers, such as modularity, stan- changeability, resilience, and robustness of manufacturing systems
dard interfaces, mobility, integrability, diagnosability, and program- were all put to the test compounded by the immediacy of the
mability to allow agile changes in function (scope) and capacity required responses. As discussed in Section 4, the degree of pre-
(scale) between anticipated flexibility corridors within manageable paredness and implementation of flexibility, reconfigurability, agil-
variations above initially designed boundaries such as extending ity, changeability and smart manufacturing paradigms varies
products family, and production scope and scale. among companies which affected their ability to respond to these
extreme changes and in a timely manner.
5.3.3. Cognitive adaptability Response to extreme disruption
Cognitive adaptability is built on top of, but is differentiated from, Some manufacturing systems were able to make changes quickly
static and dynamic adaptability in that the adaptive responses are trig- to keep the business running and protect jobs. Wineries, liquor and
gered and/or executed autonomously by important cognitive character- perfumes makers and drinks bottlers were able to produce new
istics. Human-centric adaptive cognition includes context- and self- product variants in an expanded product family. For example, win-
aware as well as self-optimizing behavior in the two-way interac- eries already using alcoholic liquids were able to switch to making
tions between a) machines and other hardware components in the disinfectants and hand sanitizers by changing the fluids formula-
manufacturing system using sensors and IT, IoT and IoP capabilities, tion, the bottles, and labels, and reprogramming the material han-
and b) human operators and intelligent technological applications. dling systems while using the same processes. Parts manufacturers
Therefore, cognitive adaptation utilizes built-in changeability ena- and tool and die makers which produced small and medium size
blers to allow agile and optimal changes in function (scope) and batches of custom orders using versatile multi-purpose and pro-
capacity (scale) beyond previously anticipated and planned total grammable flexible machines switched easily with minor changes
changeability ranges/corridors using the autonomously synthesized to producing face shields and masks in large size lots. Use of
cognitive adaptation response. It is enabled by employing elements advanced digital design and 3D printing technologies made rapid
of artificial and hybrid human-machine intelligence such as sensing, switching to new products feasible. Other urgent virus-related
perception, anticipation, prediction, planning, action, and autono- products include clips to attach to paediatric face masks, sheet
mous decentralized decision making and control of machines and metal components for automated COVID-19 lab test equipment, and
production. In addition to self-awareness and self-optimizing fea- different moulds for ventilators production. A vacuum-maker
tures, the cognitive adaptability includes self-planning, self-healing switched to making ventilators by switching from making suction
such as maintenance and repair, and generally knowledge and cogni- machines to ones that blow air. Even Mints, known for making
tion-based adaptive responsiveness. money bills and coins, tuned their focus to helping protect people
against COVID-19 by making valuable plastic visors for healthcare
5.3.4. Extreme adaptability staff. Major electronics manufacturers adapted existing clean-room
Extreme adaptability relies on the manufacturing system resil- production facilities for LCD display panels to make surgical masks
ience and capacity to recover (partially or fully) from major unex- in large quantities. Several automotive OEMs began producing face
pected multi-dimensional extreme disruptions and return to the masks using medical-grade textiles previously used for car seats
JID: CIRP
and interior details. Setting up new or re-configured production more environmentally, socially, and economically sustainable and
lines by adding/removing and reconfiguring portions of the more human-centered. Re-thinking the value of work and workers
manufacturing equipment and systems also worked for some other as most important and flexible assets is already on-going but work-
manufacturers. Many companies used time flexibility to increase erless manufacturing systems are not part of the future manufactur-
the number of worked shifts as needed. Companies re-purposed ing systems paradigms.
their production lines for different reasons, such as government The experienced products shortages during the 2020 pandemic
incentives or invoking certain laws, but the majority switched pro- emphasized the importance of supporting local manufacturing,
duction willingly to maintain some output and revenues when reg- research and development, and promoting self-sufficiency that will
ular orders dried up. Such drastic production changes are not slowly lead to de-globalization. It will intensify countries investment
without their hurdles, such as securing new materials at a time of in innovation in product design and manufacturing technologies to
scarcity and developing new/modified product designs and quickly maximize locally owned intellectual property (IP). Supply chains will
getting regulatory approvals. Sharing design information and intel- be re-designed for the unexpected with optional scenarios that
lectual property across business sectors are other issues. Digitaliza- embrace uncertainty to increase their adaptability, robustness, and
tion, communication, connectivity, and availability of open-source resilience.
tools allowed effective sharing of knowledge and resources. The
urgency and scale of disruption due to the pandemic could change 5.4. ACMS paradigm characteristics, drivers, and enablers
the way collaboration is done in the future. The business paradigm
may also change from “just-in-time” to “just-in-case”, and the The evolution of manufacturing systems and future trends
manufacturing system and supply chains may be designed for emer- towards smart cognitive manufacturing is discussed in this section
gence and to accommodate black swan events. This need was within the context of the evolution of manufacturing throughout the
addressed [143]. industrial revolutions from craft production to the current smart
Observations manufacturing (Industry 4.0) era and into a future bio-intelligent
The above examples demonstrate successful static and dynamic manufacturing era, in which the augmented human abilities will play
adaptability. Manufacturers with reasonably diversified and robust a central role in enhanced decision making.
supply chains fared better in securing the materials and parts needed The human cognition capabilities can receive visual cues from the
for their operations. However, in dire situations such as the 2020 environment and combine them with other sensory information
pandemic, increased nationalisms, and a continued gradual move such as sounds, smell and tactile feedback to create perceptual expe-
away from globalization were clearly observed and affected the eco- riences. Perceptual processes depend on the perceiver’s expectations
nomic recovery efforts. Built-in passive and dynamic adaptability and previous knowledge as well as the information available in the
enablers in manufacturing systems and the extent of implementing stimulus itself. Processing all this information in a lapse of millisec-
supportive advanced technologies played an important role in the onds makes the humans a very powerful “cognition machine”. Fur-
ability of manufacturing systems to cope with the drastic challenges thermore, humans are very adaptable to the environmental stresses,
posed by the pandemic and the required immediacy of response, but changes, and complexities. In this context, the cognitive system has
both are effective only within boundaries of anticipated changes and emerged to meet human capabilities and has been defined as “a sys-
limited by the design of the systems and their components. Defensive tem that can modify its behavior on the basis of experience” [79]. In
passive responsiveness and adaptability proved insufficient in situa- general, it can be said that the term “cognitive system” has been used
tions of extreme disruptions in manufacturing scope and/or scale. to define a new solution, software or hardware that mimic in some
Offense strategies are required to ensure cognitive adaptation beyond ways human intelligence.
the planned scenarios in short order. Just-in-Case supply chains sce- Manufacturing systems are continuing to evolve in response to
narios not only Just-in-Time are essential [133,134,135,136]. A con- may disruptive products, processes and market drivers, and the need
ceptualization of a decision-making environment of integrated to adapt to these changes. The evolution and co-evolution trends of
supply chain (ISN) viability formation through a dynamic game-theo- products, technologies, business models, and production paradigms;
retic modeling of a biological system that resembles the intertwined the accelerated rate of adaptation to change; and research and devel-
supply network was proposed [86]. opment of new and disruptive game changing technologies all point
Lessons learned to a fundamental change in the ways products are made.
Effective human-machine collaboration in all aspects of It is envisioned that the next manufacturing systems paradigm
manufacturing is needed as workers proved pivotal and most flexi- will be an adaptive cognitive manufacturing system, coined as ACMS,
ble in making the transition to new manufacturing systems and and characterized by its cognitive adaptability. It differs from static
operation strategies during this time of extreme changes. There is a and dynamic adaptability in the manner in which the need to adapt,
need for a combination of proactive and reactive manufacturing sys- e.g. due to different products/variants, change in production volume,
tems adaptation, however, it is the cognitive adaptation that will supply shortages, technological advances, and online changes is rec-
play a crucial role in anticipating extreme changes and planning ognized and response is triggered, as well as how adaptation gets
optimal adaptation plans and implementation strategies. These new implemented; will all be enabled and supported by AI modules, smart
norms are significantly influencing future manufacturing systems. sensors, extensive information and data analytics, and the auto-
The pandemic experience will prove to be a turning point with mated, cognitive and hybrid human-machine adaptation actions and
significant impact for manufacturing and manufacturing systems. execution methods and human experience and wisdom. It is a new
For instance, many of the artificial barriers to moving more of the paradigm where the power of the 4th industrial revolution and
manufacturing activities online will be removed. Not everything can beyond is deployed to achieve a more responsive as well as a more
be virtual, of course, but in many areas remote work will become humane and human-centric manufacturing systems driven by eco-
not only feasible but also necessary. Once companies sort out nomic and environmental sustainability and social responsibility.
related technicalities, it will be harder and more expensive to deny Such new ACMS paradigm will be enabled and made possible by
employees those options. Indeed, a great deal of design, planning, predictive analytics, AI enhanced decision making and cognitive
support functions and management meetings can be effectively behavior such perception, planning, and smart actions as well as
done virtually and remotely saving travel, reducing pollution, and effective connectivity and seamless integration. Features of the
allowing more flexible work environment so workers can better ACMS will include ability to anticipate changes by continuously ana-
support their families. It is anticipated that remote will become per- lyzing wide range of data collected at all levels internally within the
manent with more people working from a distance. All these new system and externally from other sites, partners, markets, and
modes of work in the future will make manufacturing and systems global trends; and planning and constructing sound strategies for
JID: CIRP
the most appropriate type of adaptability to be implemented and The future cognitive digital twin (CDT) will become not only an
the right timing supported by scenarios for just-in-case disruptions accurate digital representation but also an augmentation and intelli-
and commensurate physical, logical and strategic mitigation gent companion of the physical system, including sub-systems,
responses. The ACMS will ultimately also be characterized by con- throughout its life cycle and evolution. In the fourth stage of transfor-
text and self-awareness, self-diagnosing, self-healing and repairing mation, CDTs will become highly interconnected, distributed cogni-
behavior as well as self-optimizing control and operation strategies. tive adaptive systems evolving as their physical counterpart grows in
These characteristics will be designed to make it possible to support complexity and smartness [63]. The adaptive cognitive digital twin
a healthy symbiotic relationship between human workers and will also have built-in models of human operators in the system not
smart automation in an integrated collaborative workspace envi- only physically but also behaviourally to capture their actions and
ronment which utilizes essential humans’ input to critical tactic and guide their interaction with increasingly smart, adaptive, and cogni-
strategic decision making at all levels while also making the nature tive collaborating robots and machines.
of work not only safer and more ergonomic but also more fulfilling
and rewarding to maximize the satisfaction of the people in the sys-
5.4.2. Prognostics predictive maintenance using CDT
tem.
Digital twins offer a great amount of business potential by pre-
It should be emphasized that the time frame for when ACMS with
dicting the future instead of just analyzing the past of the
all its features and capabilities will be developed and implemented in
manufacturing process. General Electric (GE) currently operates
factories was intentionally left out as many of its enabling technolo-
more than 500,000 alive cognitive digital twins [63]. They are used
gies are still being developed and evolving.
to eliminate guesswork for service and to prevent catastrophic fail-
ures because they continuously learn and update themselves from
5.4.1. Cognitive digital twin (CDT) transformation
multiple sources representing their near realtime status, working
The digital twin (DT) of manufacturing systems has evolved
conditions or environmental factors. This learning system learns
greatly, since the term was first coined in 2002 [65]. In the first stage
from itself, via artificial intelligence and/or machine learning algo-
of digital transformation, DT transitioned from standalone simula-
rithms using a trove of data from sensors that convey various
tion model to a more detailed digital mock-up. The next stage of
aspects of its operating conditions, and from human operators mak-
digital transformation till present saw the introduction of IoT, IoP,
ing functional decisions and other human specialists with deep and
sensors and data analytics allowing the digital simulation to become
relevant industry domain knowledge. A cognitive digital twin also
more representative of the physical system and more connected in
integrates historical data from past usage to compare deviation
realtime to its operation. This expanded its use from off-line deci-
from a baseline.
sion support tool during the design and planning of a manufacturing
system to an integrated multi-physics, multi-scale simulation sys-
tem that uses the most appropriate model, data history and sensor 5.5. Future manufacturing systems perspectives
updates to mirror the operation of its corresponding physical sys-
tem throughout its life from design to implementation and actual Highlights of expected features of future manufacturing systems
operation. This is when it was labeled a digital twin. It is worth men- considering the four axes of evolution include:
tioning that even in real life human twins are not always identical, Products: will be more intelligent, more complex, and more
hence, the level of granularity and accuracy of a digital twin in environmentally friendly, include embedded systems and embedded
model representation, analysis and simulation is a matter of trade- intelligence, and use bio-degradable smart and self-healing materi-
off between the desire for a high fidelity and need for realtime inter- als.
action performance depending on the application. Digital shadow is Technology: will witness accelerated progress in exponential tech-
a term used to refer to reduced but sufficient level of detailed repre- nologies including computing, information technology, communica-
sentation in favor of delivering time sensitive feedback for realtime tion, artificial intelligence applications, machine learning, and deep
adaptive machine and system control. learning methodologies; advances in transformative manufacturing
Introduction of smart sensors, artificial intelligence, and simple technologies; development of resilient, communicating, cognitive,
machine learning technologies such as pattern recognition to a DT and more autonomous machines; and deployment of biologically
saw the beginning of its cognitive transformation to stage 3 of its inspired technologies.
evolution. This is supported by the increased application of CPS, Business Models: will employ digital business strategies and more
which is at the core of smart manufacturing systems (Industry 4.0), diversity in operating models; augment “just-in-time” model with
that resulted in digitally and adaptively controlled machines with scenarios for “just-in-case”; utilize new strategic collaboration and
embedded sensors and software, and high connectivity within, and partnership networks; implement pay-per-use business models,
between, machines in the system to collect and analyze pertinent such as leasing and subscription that will likely disrupt manufactur-
data and control various functions. Digital twins are evolving and ing systems with incremental payments for performance and guar-
growing in sophistication and abilities mirroring the evolution of antee of usage level a priori. They will ensure more effective
manufacturing systems as illustrated in Fig. 18. scalability of systems capacity and capability by sharing distributed
resources among many customers; derive increased value from dig-
ital services; use more resilient supply chains and value networks
enabled by AI will contribute significantly to competitiveness and
offer more support for local manufacturing and local innovation
[164].
Manufacturing Systems: will feature maximum flexibility, physi-
cal and logical scalability, and agility; and utilize more static,
dynamic, and cognitive adaptability enablers to improve produc-
tivity and emphasize all three facets of sustainability; increase
shared human-machine collaboration and decision making,
replace implicit interactions with explicit tasks sharing, and enjoy
greater visibility throughout. Future manufacturing systems will
Fig. 18. Digital transformation towards the adaptive cognitive digital twin [Intelligent
use hybrid augmented natural and artificial intelligence in systems
Manufacturing Systems (IMS) Centre, U. Windsor]. operation and control. The use of autonomous machines, robots,
JID: CIRP
production planning and control, enhanced diagnostics, predictive symbiosis. Undoubtedly, the significant advances in manufacturing,
maintenance, and quality verification will be extensive. Using information technologies and industrial revolutions will bring trans-
intelligent management and business functions; expanded data formational change to all aspects of manufacturing and manufactur-
and knowledge sharing with cyber-security measures in place will ing systems, and the impact on society and humanity will be
be commonplace. profound.
Digital and physical twins will become inseparable for more effi- The following sections review how manufacturing systems are
cient and optimum operation, but humans will continue to be an being re-imagined, highlight important conclusions, and indicate
essential part of interactive decision-making on the operational, tac- some directions of future research in the field.
tical, and strategic levels.
People are the most adaptable and valuable assets in manufactur- 6.2. Manufacturing systems re-imagined
ing systems. Integrating human experience and insights with
machine learning visibility and cyber-physical digital and cognitive Future manufacturing systems are being re-imagined in many
transformation requires new skills and upgraded multi-disciplinary aspects as categorized and summarized in Fig. 19.
education. More versatile and flexible work and workers will be
essential. Remote work will increase enabled by enhanced digital
operation transparency. The nature of work in manufacturing sys-
tems will change, and different jobs will appear to support the new
technologies.
The evolving adaptive cognitive manufacturing systems (ACMS) par-
adigm will become more predictive, adaptive, human-centric, and
transparent and will enjoy increased industrial adoption.
6. Conclusions and future research
6.1. Insights and reflections

Fig. 19. Aspects of re-imagined manufacturing systems.
Manufacturing has witnessed many major changes throughout
the first, second and third industrial revolutions. The fourth indus-
trial revolution is characterized by more distributed, collaborative, The system design will include features to enable static, dynamic,
connected, networked and global manufacturing. Sophisticated and and cognitive adaptability. It will continue to evolve and co-evolve
powerful sensors and sensing techniques are introduced allowing with the four axes of evolution and benefit from biologically inspired
better communication between manufacturing entities on the shop designs, tools, and materials. The system design will also be influ-
floor and throughout the enterprise and beyond. The ever-increas- enced by emerging business models such as considering manufactur-
ing computing power, speed and storage capacity and novel coming systems as providers of valuable service by producing certain
munication techniques created a very fast wide band information goods as needed. This subscription and pay-per-output model will
highway making it possible to more effectively utilize IoT, IIoT and lead to re-thinking the relationship between customers and manu-
IoP and enabled better knowledge-based decisions in realtime. This facturers and affect how manufacturing systems are designed, oper-
in turn led to generating huge amount of data (big data) about all ated, sized, located, and owned.
aspects of manufacturing and products. Powerful data analytics, The system configuration is all the physical modules that make-up
application of artificial intelligence, expert systems and machine the system, their arrangement, and physical and logical relationships
learning methods made intelligent knowledge-based decision mak- between them which define the parts flow. Increased adaptability
ing in realtime feasible. New products with embedded intelligence requires more modularity at the machines, stations and system lev-
and the design and control of more intelligent manufacturing sys- els, and sufficient decoupling of function between modules to allow
tems allow autonomous planning, operation, and execution. Advan- freedom of mobility, reconfiguration, and scalability of the equip-
ces in intelligent automation and robotics and their use in ment. This level of modular functionality leads to significantly
manufacturing, along with human workers, presents novel ways of streamlined production scheduling, flow control, changeability, and
human-machine collaboration. better cost.
Advanced software applications and increased computing capa- The system control and operation will become more de-central-
bilities are enabling high fidelity simulation and digitalization of all ized, reconfigurable, agile, responsive, and adaptive. It will be data-,
manufacturing aspects including products design, making, use and knowledge-, and AI-driven with many cognitive features such as self-
recycling/re-use throughout their life cycle, as well as systems awareness, self- repairing, self-organizing [36,157] and self-optimiz-
design, implementation, control/operation and redesign/recycle, ing behavior. The control and operation will likely be semi-autono-
and realistic useful mathematical and simulation models (DT, CDT, mous or autonomous.
and ACDT). Smart products and systems which are increasingly The human capital and workers in the system will remain in the
multi-disciplinary coupled with increased variety add new layers of loop as important elements of future manufacturing systems but
complexities and present challenges in managing such complexity they will have to become multi-skilled through re-skilling and up-
while ensuring products and systems robustness and resilience skilling for maximum versatility and adaptability in a smarter envi-
[52,53]. ronment; their work will be AI supported and augmented by collabo-
There have been successes and failures along the evolution of rating robots and machines to increase efficiency and reduce errors.
manufacturing systems path, which have been used to inform future Digital twins with imbedded cognitive abilities will include physical,
evolutions. For instance, the early enthusiasm for computer inte- behavioral, and cognitive models of human workers abilities which
grated manufacturing (CIM), artificial intelligence and machine learn- can be used for planning and training. Many work categories will
ing did not achieve their intended purpose in the past, are now become possible to perform remotely, and work weeks will be
realized with the advent of the Industry 4.0 transformational technol- shorter. Automation applications in manufacturing systems will be
ogies. The early vision of a fully automated unmanned factory is now more socially responsible with humans’ safety and security at the
being realized in autonomous systems and in human machine fore front.
JID: CIRP
Multiple scales of manufacturing systems will exist with a wide augment humans; b) Multi-experience that aims at replacing tech-
range of sizes to suit the needs. Full scale large manufacturing sys- nology-literate people with people-literate technology, c) Human
tems will continue for certain types of manufacturing but with augmentation which uses technology to enhance a person’s cogni-
added agility, adaptability, and smart cognitive features. In additive and physical experiences, for example by using smart wear-
tion, smaller highly agile niche or boutique manufacturing systems ables; d) The empowered edge which explores how increasing smart
will increasingly be used for smaller production of specialty prod- devices are forming the foundations for smart spaces, and moving
ucts. Demand will increase for mini-manufacturing systems with key applications and services closer to the people and devices that
full and versatile capabilities and better ability to adapt and recon- use them; and e) Autonomous technology which operates on a spec-
figure as needed. Small, mini- and desk-top systems will surge in trum of intelligence ranging from semi-autonomous to fully autono-
response to increased demand for more products variety, customi- mous, and from stand-alone to collaborative swarms. Additional
zation, and personalization, and remote manufacturing enabled by trends include democratization of technology, transparency, and
advances in additive 3D manufacturing for producing relatively traceability.
simpler and smaller size products, will become more common
place.
The location of future manufacturing systems also will vary from 6.3.2. Adaptive cognitive manufacturing systems research
being in industrial sites as usual to more distributed yet well con- As manufacturing systems evolve to the next stage of adaptive
nected and integrated networks of manufacturing systems nationally cognitive manufacturing systems (ACMS), there are basic technolog-
and internationally. The location and relocation of manufacturing ical challenges ahead while incorporating concepts, enablers and
systems will be influenced by the need for more urban, sub-urban, technologies developed in earlier paradigms. New research direc-
close to home or at home manufacturing [75]; more diversified, resil- tions are needed to support the evolution of future manufacturing
ient and adaptive supply/value chains closer to home base will systems through its digital and cognitive transformations including
increase; and increased tendency for de-globalization and protec- manufacturing systems physical, sensorial, and cognitive support;
tionism will become evident. static, dynamic, cognitive and extreme adaptation methodologies;
modularity, flexibility, reconfigurability, changeability, and respon-
siveness; more intelligent, cognitive, knowledge-intensive, data-
6.3. Challenges and future research
driven cyber-physical and biologically inspired manufacturing sys-
tems; and better connected, integrated, and networked autonomous
The impact of manufacturing throughout the industrial revolu-
systems.
tions up to the ongoing Industry 4.0 is well recognized and docu-
Smart adaptive automation systems design and operation
mented. The contributions of CIRP researchers and others to the
An imminent change to a future where fully integrated and inher-
development of manufacturing systems paradigms, drivers, and
ently intelligent systems, subsystems, and components shall define
enablers from changeability (flexible and reconfigurable
the next generation of intelligent machines, systems, and enterprises.
manufacturing systems (FMS/RMS)) to cyber-physical systems
A great more research is needed to bring this closer to practical appli-
(CPS) and smart manufacturing systems (SMS) are extensively dis-
cations. Related research topics include: new production system con-
cussed in the literature. Manufacturing systems have been trans-
cepts through the study of adaptability, emergence,
formed from isolated optimized cells to fully integrated data and
self organization, and cooperation; autonomous production sys-
product flows within a factory and between distributed locations,
tems; bio inspired manufacturing; human centric dynamic adapta-
with vertical and horizontal communication along the entire value
tion; manufacturing as a service; manufacturing on demand; and
chain. The objective of the earlier computer integrated
subscription models for production facilities.
manufacturing (CIM) is finally becoming realizable with the tech-
Adaptive cognitive digital twin (ACDT)
nological enablers and pillars of smart manufacturing (Industry
Future research includes robust multi-scale mathematical models
4.0). There is still a myriad of related research topics in which
to increase the accuracy and fidelity of digital twins of machines and
manufacturing researchers are actively engaged, and industry
manufacturing systems; engineering CPS and IT powerful architec-
(large, medium, and small) continue to develop and implement to
tures to increase efficiency and reliability of digital twins and shorten
achieve the business objectives.
their development cycle; and development of the new cognitive digi-
Nevertheless, the future productivity and growth in the
tal twins, and adaptive cognitive digital twins of humans for use in
manufacturing industries require careful long-term strategic plan-
planning of human-centric manufacturing systems and in workers
ning of future research directions to reap the desired benefits from
training.
their industrial implementation and increase sustainability and com-
Data processing, perception, and knowledge discovery
petitiveness into the future.
A main challenge is to develop systems capable of processing all
It is recognized that each stage of the evolution makes full use of
the needed information and data from various sensors, devices and
the results of and experience gained from earlier stages. Therefore,
machines and any other contextual information available to charac-
this section focuses on the research targets and topics that are moti-
terize settings in analysis, and retrieve knowledge and past physical,
vated by the new adaptive cognitive manufacturing system (ACMS)
virtual, or human experiences for creating perceptions and aug-
paradigm, while recognizing that the earlier research agendas will
menting the human experience and expertise and knowledge. There
naturally continue, probably at an accelerated pace. It is also known
is also need for implementing a human-centric decision making in
that the implementation in practice normally lags research results
meshed collaboration with intelligent systems. Therefore, knowl-
to various degrees depending on the industry sector and company
edge representations capable of building a multi-modal space com-
size.
posed of information from different sources, in the form of
experiential knowledge, would be a very useful tool to facilitate this
6.3.1. ACMS top strategic technology trends and challenges process.
Hyper-automation, blockchain, AI security, and autonomy drive New smart strategies for vertical and horizontal integration
disruption and create opportunities in strategic technology areas Physical and logical enablers need to be researched and enhanced
[139]. ACMS includes many of the Gartner 2020 technology trends to implement further collaboration between the hard and soft ena-
such as: a) Hyper-automation which deals with the application of blers, the physical and virtual domains, and the humans in the sys-
advanced technologies, including artificial intelligence (AI) and tem. Connecting existing machines and systems, attaching sensors,
machine learning (ML) to increasingly automate processes and and collecting large volumes of data are insufficient to make
JID: CIRP
manufacturing systems cognitive. New concepts, theories and practi- tainability strategies can help guide manufacturers to contribute to
ces regarding engineering and design of future systems should be a prosperous future for themselves and the society. New models
fully researched and developed. and justification methods should be developed for making smart
Smart manufacturing systems prognostic maintenance reverse logistics and circular economy not only more economically
Prognostics in maintenance intended to predict failure before it feasible but also to reap the full benefits of the adaptive cognitive
happens is an enhanced predictive maintainability strategy which principles.
should be further researched and developed as well as AI-based algo- Future manufacturing jobs, learning and training for ACMSs
rithms and supportive cognitive and adaptive cognitive digital twins The next phase of the evolution of manufacturing systems, dis-
of systems, equipment and subsystems. cussed in detail in this paper, is a cognitive transformation that aims
Complexity and trust management to develop smarter more sustainable factories and business pro-
Methodologies for ACMS complexity management, transparency, cesses. Therefore, the future of manufacturing jobs and the contin-
and traceability; blockchain and AI security with emphasis on ensur- ual learning and training of the workforce is of paramount
ing human-centric decision making; and collaboration and mutual importance.
trust need further investigation. Careful attention should be paid to Several publications [106,108,198] have noted that AI is intro-
the ethical issues arising from use of smart machines, hybrid human- ducing asymmetries that are transforming the job market and creat-
machine collaboration and intelligence. Specific topics include: ing misalignments with the effective technical readiness levels,
cyber-security mechanisms and governing regulations and standards which is typically the case with any new technology. It is under-
as the pervasive use of smart sensors, and data collection with 5G stood that AI will affect some aspects of all jobs to various degrees.
communication networks increase; better built-in protections and Existing literature project a loss in traditional manufacturing jobs,
safeguards, and developing appropriate guidelines and legislations with the possibility that new jobs will be created. Research suggests
for operators in the new work environment under smart digitalized that organizations adopting smart manufacturing technologies will
operating schemes; legislations for protection of data collected on need a workforce with increased variety of technical skills, auton-
human performance with digital supports, and regarding human- omy, and interdependence, as well as increased cognitive, creative,
machine interaction; and socio-technical research regarding imple- technical, and social skills. While automation and AI will likely dis-
menting socially responsible manufacturing and artificial intelli- place some manual work and entry level jobs, the engineering, plan-
gence. ning, and managing tasks as well as all operation activities will
Supply chain design and operation for the unexpected remain human-centric, albeit with augmented machine intelligence
Proactive and reactive strategies should be developed for ACMS to capabilities, for the foreseeable future due to their relative complex-
plan for inevitable disruptions in a multi-echelon supply chain, since ity.
unexpected or black swan events can highly deteriorate supply It should be noted, however, that the future is not inevitable.
chains performance. The drastic effects of the COVID-19 pandemic While the existence of the necessary technologies is a precondition
highlighted the need for developing more effective methods and for automation, it does not necessarily mean that all manufacturing
models not only to manage crises in the short term, but also to activities will be automated, because automation depends on several
develop strategic options for future more resilient and robust supply other factors including a) cost of automation; b) cost and relative
chains. scarcity of trained labor and required skills; c) benefits of automation
Innovative dynamic cost models and return on investment; d) social acceptance, and e) regulatory
Companies can achieve competitive advantage by reducing oper- issues.
ating cost while investing in automated cognitive manufacturing sys- Nevertheless, everyone must embrace change and develop a
tems. One of the important factors affecting the adoption and mindset of continuous evolution, and workers will require contin-
implementation of next generation manufacturing systems is cost. uous retraining in advanced technological skills such as intelligent
Better business and cost estimation models and comprehensive automation technologies, programming and big data analytics
methods for justification of investments in ACMS and related which will grow rapidly. Furthermore, it should be noted that the
advanced technologies are needed to accelerate the introduction and adaptive cognitive transition will favor social, emotional, and
implementation of smart manufacturing systems. higher cognitive skills, such as creativity, critical thinking, and
Wireless power transfer for improved equipment mobility teamwork all of which machines find hard to replicate. Additional
The introduction and application of wireless power transfer sys- pressure on the already existing workforce skills challenge
tems in manufacturing can bring about not only convenience but also includes the need for new credentials and certification systems for
improvement in safety and reliability as well as cost savings due to training people to do the jobs that cannot be replaced by robotics
the automatic recharging of AGVs, mobile robots, mobile inspection and smart automation.
stations and other manufacturing equipment. In automated produc- Finally, it agreed that AI is doing a lot of good in many fields
tion systems and warehouses, AGVs are heavily used in material han- and will continue to provide several benefits for manufacturing
dling and transportation of material and goods. Optimally placed while allowing people to enhance their human contributions.
power transmission pods throughout the factory would help AGVs However, along with the good, there will inevitably be some neg-
become self-charging while moving or when they are idle while ative consequences. That is why humans should remain in control
goods are loaded/unloaded. A battery with a substantially reduced and develop and introduce the appropriate level of automation
capacity can be used and recharged, thus considerably saving the and intelligence to maintain the overall good. With careful plan-
operation and maintenance costs. New methods, hardware solutions ning, the worst fear by some about "superintelligence" - the point
and software tools to encourage using this emerging technology in at which computers become more intelligent than humans - can
manufacturing systems need further research. This technology can be avoided as humans maintain control of their competitive crea-
also facilitate the mobility of mini factories between several locations, including adaptive cognitive manufacturing systems
tions. (ACMS).
Smart reverse logistics and circular economy
The prospect of nearly 9 billion people on the planet by 2030 is Declaration of Competing Interest
driving leaders to retool their business models to enable their long-
term growth and prosperity. Sustainability mega forces continue to The authors declare that they have no known competing financial
change the operating conditions in which companies can succeed interests or personal relationships that could have appeared to influ-
and thrive. Smart cognitive systems and ambitious long-term sus- ence the work reported in this paper.
JID: CIRP
Acknowledgments [25] Brusaferri A A, Colledani M, Copani G, N Pedrocchi MS, Tolio TAM (2012) Inte-
grated De-Manufacturing Systems as New Approach to End-of-Life Management
of Mechatronic Devices. 10th Global Conference on Sustainable Manufacturing
The authors wish to thank the following colleagues and their Towards Implementing Sustainable Manufacturing, 332-339.
research groups for providing industrial implementations, use [26] Byrne G, Damm O, Monostori L, Teti R, van Houten F, Wegener K, Wertheim R,
cases, reports and case studies: Professor G. Reinhart and Professor Sammler F (2021) Towards High Performance Living Manufacturing Systems - a
New Convergence between Biology and Engineering. CIRP Journal of Manufactur-
M. Zaeh, Technical University of Munich, Germany; Professor G. ing Science and Technology.
Lanza, Karlsruhe Institute of Technology (KIT), Germany; Professor [27] Byrne G, Dimitrov D, Monostori L, Teti R, van Houten F, Wertheim R (2018) Bio-
F. Piller, Technical University of Aachen (RWTH), Germany; and logicalisation: Biological Transformation in Manufacturing. CIRP Journal of
Manufacturing Science and Technology . 21/1-32.
Professor D. Mourtzis and Dr. S. Makris, University of Patras (LMS),
[28] Chen M, Mao S, Liu Y (2014) Big Data: A Survey. Mobile Networks and Applica-
Greece. tions 19/2:171–209.
[29] Cheng J, Chen W, Tao F, Lin C-L (2018) Industrial Iot in 5g Environment Towards
Smart Manufacturing. Journal of Industrial Information Integration . 10/10-19.
References [30] Christensen CM (2013) The Innovator's Dilemma: When New Technologies Cause
Great Firms to Fail, Harvard Business Review Press. ISBN: 1422197581.
[31] Colledani M, Copani G, Tolio T (2014) De-Manufacturing Systems, Procedia CIRP.
[1] Abbas M, ElMaraghy H (2016) Design Synthesis of Machining Systems Using Co- 17/14-19.
Platforming. Journal of Manufacturing Systems . 41/299-313. [32] Copani G, Picone N, Colledani M, Pepe M, Tasora A (2019) Highly Evolvable E-
[2] Abbas M, ElMaraghy H (2018) Synthesis and Optimization of Manufacturing Sys- Waste Recycling Technologies and Systems, in Factories of the Future, Springer,
tems Configuration Using Co-Platforming. CIRP Journal of Manufacturing Science Cham109–128.
and Technology. 20/51-65. [33] Copani G, Urgo M (2015) Innovative Flexibility-Oriented Business Models and
[3] Abdi MR, Labib AW, Delavari Edalat F, Abdi A (2018) Product Family Formation System Configuration Approaches: An Industrial Application. CIRP Journal of
and Selection for Reconfigurability Using Anp, in Integrated Reconfigurable Manufacturing Science and Technology. 11/10-17.
Manufacturing Systems and Smart Value Chain: Sustainable Infrastructure for [34] Copani G, Urgo M (2012) New Business Models and Configuration Approaches
the Factory of the Future, Springer International Publishing: Cham. https://doi. for Focused-Flexibility Manufacturing Systems. Procedia CIRP. 2/10-15.
org/10.1007/978-3-319-76846-5_6 125-142 ISBN: 978-3-319-76846-5. [35] D’Addona DM, Bracco F, Bettoni A, Nishino N, Carpanzano E, Bruzzone AA (2018)
[4] Abele E, Wo € rn A (2009) Reconfigurable Machine Tools and Equipment, in Change-
Adaptive Automation and Human Factors in Manufacturing: An Experimental
able and Reconfigurable Manufacturing Systems, ElMaraghy HA, Editor, Springer Assessment for a Cognitive Approach. CIRP Annals 67/1:455–458.
London. London. https://doi.org/10.1007/978-1-84882-067-8_6 111-125 ISBN: [36] de Oliveira CS, Sanin C, Szczerbicki E (2020) Context-Aware Indexing and Retrieval
978-1-84882-067-8. for Cognitive Systems Using Soeks and Ddna, Springer International Publishing,
[5] Abramovici M (2014) Smart Products. CIRP Encyclopedia of Production Engineer- Cham7–16.
ing: 1–5. [37] Deloitte Touche Tohmatsu Limited (DTTL), 2016, Global Manufacturing Compet-
[6] acatech. Cyber-Physical Systems: Driving Force for Innovation in Mobility, Health, itiveness Index Gmci. Global Consumer & Industrial Products Industry Group
Energy and Production, . https://doi.org/10.1007/978-3-642-29090-9 ISBN: 978- and the Council on Competitiveness.; https://www2.deloitte.com/content/dam/
3-642-29089-3. Deloitte/global/Documents/Manufacturing/gx-global-mfg-competitiveness-
[7] acatech. Industry 4.0 - International Benchmark, Future Options and Recommenda- index-2016.pdf (accessed 11 August 2020).
tions for Action in Production Research, WLZ- RWTH Aachen University https:// [38] Deutsche Telekom, Digitalisation Index. www.digitalisierungsindex.de. (accessed
en.acatech.de/publication/industrie-4-0-international-benchmark-options-for- 11 August 2020).
the-future-and-recommendations-for-manufacturing-research/. [39] Ding Y, Xu W, Liu Z, Zhou Z, Pham DT (2019) Robotic Task Oriented Knowledge
[8] acatech. Integrierte Forschungsagenda Cyber-Physical Systems, Springer-Verlag Graph for Human-Robot Collaboration in Disassembly. Procedia CIRP . 83/105-
Berlin Heidelberg. https://doi.org/10.1007/978-3-642-29099-2 ISBN: 978-3- 110.
642-29099-2. [40] Dollinger C, Reinhart G (2016) A Competence Based Approach to Support the
[9] AlGeddawy T, ElMaraghy H (2010) Co-Evolution Hypotheses and Model for Working Force within Assembly Lines, Springer International Publishing,
Manufacturing Planning. CIRP Annals 59/1:445–448. Cham557–567.
[10] AlGeddawy T, ElMaraghy H (2020) A Holistic Multi-Domain Association Model [41] ElMaraghy H (2006) A Complexity Code for Manufacturing Systems. ASME 2006
for Industrial Data. Procedia Manufacturing, 51/920-925. International Manufacturing Science and Engineering Conference, 625-634, .
[11] AlGeddawy T, ElMaraghy H (2011) Manufacturing Systems Synthesis Using [42] ElMaraghy H (2005) Flexible and Reconfigurable Manufacturing Systems Para-
Knowledge Discovery. CIRP Annals : 437–440. 60/1. digms. International Journal of Flexible Manufacturing Systems 17/4:261–276.
[12] Anderl R, Picard A, Wang Y, Fleischer J, Dorsch S, Klee B, and Bauer J, 2015, [43] ElMaraghy H (2007) Reconfigurable Process Plans for Responsive Manufacturing
Guideline Industrie 4.0 Guiding Principles for the Implementation of Industrie Systems, Springer US, Boston, MA35–44.
4.0 in Small and Medium Sized Businesses. [44] ElMaraghy H, Abbas M (2015) Products-Manufacturing Systems Co-Platforming.
[13] Andersen A-L, Brunoe TD, Nielsen K, Ro € sio
€ C (2017) Towards a Generic Design
CIRP Annals 64/1:407–410.
Method for Reconfigurable Manufacturing Systems: Analysis and Synthesis of [45] ElMaraghy H, AlGeddawy T (2012) Co-Evolution of Products and Manufacturing
Current Design Methods and Evaluation of Supportive Tools. Journal of Capabilities and Application in Auto-Parts Assembly. Flexible Services and
Manufacturing Systems . 42/179-195. Manufacturing Journal 24/2:142–170.
[14] Andersen A-L, Larsen Jesper K, Brunoe Thomas D, Nielsen K, Ketelsen C (2018) [46] ElMaraghy H, Moussa M (2019) Optimal Platform Design and Process Plan for
Critical Enablers of Changeable and Reconfigurable Manufacturing and Their Managing Variety Using Hybrid Manufacturing. CIRP Annals 68/1:443–446.
Industrial Implementation. Journal of Manufacturing Technology Management: [47] ElMaraghy H, Schuh G, ElMaraghy W, Piller F, Scho €nsleben P, Tseng M, Bernard A
983–1002. 29/6. (2013) Product Variety Management. CIRP Annals 62/2:629–652.
[15] Assad F, Alkan B, Chinnathai MK, Ahmad MH, Rushforth EJ, Harrison R (2019) A [48] ElMaraghy HA, Kuzgunkaya O, Urbanic RJ (2005) Manufacturing Systems Config-
Framework to Predict Energy Related Key Performance Indicators of uration Complexity. CIRP Annals - Manufacturing Technology 54/1:445–450.
Manufacturing Systems at Early Design Phase. Procedia CIRP . 81/145-150. [49] ElMaraghy W (2009) Knowledge Management in Collaborative Engineering.
[16] Bauernhansl T, Ten Hompel M, Vogel-Heuser B (2014) Industrie 4.0 in Produktion, International Journal of Collaborative Engineering 1/1-2:114–124.
Automatisierung Und Logistik: Anwendung-Technologien-Migration, Springer. [50] ElMaraghy W, ElMaraghy H, Tomiyama T, Monostori L (2012) Complexity in
ISBN: 3658046813. Engineering Design and Manufacturing. CIRP Annals 61/2:793–814.
[17] Bejlegaard M, ElMaraghy W, Brunoe TD, Andersen A-L, Nielsen K (2018) Meth- [51] ElMaraghy W, Nada OA, ElMaraghy H (2008) Quality Prediction for Reconfigura-
odology for Reconfigurable Fixture Architecture Design. CIRP Journal of ble Manufacturing Systems Via Human Error Modelling. International Journal of
Manufacturing Science and Technology . 23/172-186. Computer Integrated Manufacturing 21/5:584–598.
[18] Ben-Arieh D, Easton T, Choubey A (2009) Solving the Multiple Platforms Config- [52] ElMaraghy W, Urbanic RJ (2004) Assessment of Manufacturing Operational
uration Problem. International journal of production research 47/7:1969–1988. Complexity. CIRP Annals 53/1:401–406.
[19] Bikas H, Stavropoulos P, Chryssolouris G (2016) Additive Manufacturing Meth- [53] ElMaraghy W, Urbanic RJ (2003) Modelling of Manufacturing Systems Complex-
ods and Modelling Approaches: A Critical Review. The International Journal of ity. CIRP Annals 52/1:363–366.
Advanced Manufacturing Technology. 83/1-4:389-405. [54] Ericsson LM (2019) Ericsson Automated Smart Factory Operational in China,
[20] Bogue R (2014) Smart Materials: A Review of Capabilities and Applications. https://www.ericsson.com/en/press-releases/2019/9/ericsson-automated-
Assembly Automation 34/1:16–22. smart-factory-operational-in-china (accessed 11 August 2020).
[21] Bongaerts L, Monostori L, McFarlane D, Kada r B (2000) Hierarchy in Distributed
[55] Ericsson LM, 2018, The World's First Cellular Iot-Based Smart Factory. https://
Shop Floor Control. Computers in Industry 43/2:123–137. www.ericsson.com/en/cases/2018/industry-4-0. (accessed 11 August 2020).
[22] Bortolini M, Galizia FG, Mora C, Pilati F (2019) Reconfigurability in Cellular [56] Esmaeilian B, Behdad S, Wang B (2016) The Evolution and Future of Manufactur-
Manufacturing Systems: A Design Model and Multi-Scenario Analysis. The Inter- ing: A Review. Journal of Manufacturing Systems . 39/79-100.
national Journal of Advanced Manufacturing Technology 104/9:4387–4397. [57] Europa €ische Kommission, 2018, The Digital Economy and Society Index (Desi).
[23] Brecher C, Klocke F, Schmitt R, Schuh G (2017) Internet of Production Fu € r Agile
https://ec.europa.eu/digital-single-market/en/digital-economy-and-society-
Unternehmen: Awk Aachener Werkzeugmaschinen-Kolloquium, Apprimus Verlag. index-desi. (accessed 16 March 2021).
ISBN: 3863595122. [58] European National Commission (The), 2019, Eurostat- Your Key to European
[24] Brunoe TD, Hellerup Sørensen DG, Andersen A-L, Nielsen K (2018) Framework Statistics 2019 - National Accounts and Gdp. https://ec.europa.eu/eurostat/web/
for Integrating Production System Models and Product Family Models. Procedia national-accounts. (accessed 11 August 2020).
CIRP . 72/592-597.
JID: CIRP
[59] FANUC, 2021; https://www.fanuc.co.jp/en/profile/production/index.html. [94] Kuhn TS (2013) The Structure of Scientific Revolutions. 50th Anniversary. Argu-
(accessed 3 April 2021). ment: Biannual Philosophical Journal 3/2. 539 543-539 543.
[60] Galizia FG, ElMaraghy H, Bortolini M, Mora C (2020) Product Platforms Design, [95] Kusiak A (2018) Smart Manufacturing. International Journal of Production
Selection and Customisation in High-Variety Manufacturing. International Jour- Research : 508–517. 56/1-2.
nal of Production Research 58/3:893–911. [96] Lanza G, Ferdows K, Kara S, Mourtzis D, Schuh G, Vancza J, Wang L, HP Wiendahl
[61] Gao RX, Wang L, Helu M, Teti R (2020) Big Data Analytics for Smart Factories of (2019) Global Production Networks: Design and Operation. CIRP Annals 68/
the Future. CIRP Annals 69/2:668–692. 2:823–841.
[62] Gardan J (2019) Smart Materials in Additive Manufacturing: State of the Art and [97] Laperrie re L, Reinhart G (2014) Cirp Encyclopedia of Production Engineering,
Trends. Virtual and Physical Prototyping 14/1:1–18. Springer Berlin. ISBN: 3642206174.
[63] GE, Digital Minds and Machines Meet a Digital Twin. https://adesignconsult. [98] Lee J, Jun S, Chang T-W, Park J (2017) A Smartness Assessment Framework for
com/projects-technology/having-digital-twins/. (accessed 7 August 2020). Smart Factories Using Analytic Network Process. Sustainability 9/5:794–808.
[64] Granrath L, 2019, Connected Industries - Japan's Approach to Industry 4.0 on the [99] Leit~ao P, Barbosa J, Trentesaux D (2012) Bio-Inspired Multi-Agent Systems for
Way to Society 5.0. https://www.japanindustrynews.com/2019/04/connected- Reconfigurable Manufacturing Systems. Engineering Applications of Artificial
industries-japans-approach-to-industry-4-0-on-the-way-to-society-5-0/. Intelligence 25/5:934–944.
(accessed 11 August 2020). [100] Leyh C, Scha €ffer T, Forstenha
€usler S (2016) Simmi 4.0 - Proposal of a Maturity
[65] Grieves M, Vickers J (2017) Digital Twin: Mitigating Unpredictable, Undesirable Model for the Classification of the Enterprise Application System Landscape
Emergent Behavior in Complex Systems, in Transdisciplinary Perspectives on Com- with Focus on Industry 4.0. Multikonferenz Wirtschaftsinformatik (MKWI) : 981–
plex Systems, Springer, 85–113. 992.
[66] Gu X, Jin X, Ni J, Koren Y (2015) Manufacturing System Design for Resilience. [101] Li J, Tao F, Cheng Y, Zhao L (2015) Big Data in Product Lifecycle Management. The
Procedia CIRP . 36/135-140. International Journal of Advanced Manufacturing Technology 81/1:667–684.
[67] Gu X, Koren Y (2018) Manufacturing System Architecture for Cost-Effective [102] Lichtblau K, Stich V, Bertenrath B, Blum M, Bleider M, Millack A, Schmitt K,
Mass-Individualization. Manufacturing Letters . 16/44-48. Schmitz E, Schro €ter M, 2015, Impuls - Industrie 4.0 Readiness.
[68] Gyulai D, K adar B, Monostori L (2017) Scheduling and Operator Control in [103] Lu Y, Xu X, Wang L (2020) Smart Manufacturing Process and System Automation
Reconfigurable Assembly Systems. Procedia CIRP . 63/459-464. a Critical Review of the Standards and Envisioned Scenarios. Journal of
[69] Hammer M (2007) The Process Audit. Harvard business review 85/4:111. Manufacturing Systems . 56/312-325.
[70] Hanafy M, ElMaraghy H (2015) Developing Assembly Line Layout for Delayed [104] Maddox T, 2017, How Hirotec Added Industrial Iot and Improved Its Scheduling
Product Differentiation Using Phylogenetic Networks. International Journal of Capabilities. ZDNet; https://www.zdnet.com/article/how-hirotec-added-indus-
Production Research 53/9:2633–2651. trial-iot-and-improved-its-scheduling-capabilities/. (accessed 7 August 2020).
[71] Hanafy M, ElMaraghy H (2017) Integrated Products Systems Design Environ- [105] Makris S, Karagiannis P, Koukas S, Matthaiakis A-S (2016) Augmented Reality
ment Using Bayesian Networks. International Journal of Computer Integrated System for Operator Support in Human Robot Collaborative Assembly. CIRP
Manufacturing 30/7:708–723. Annals 65/1:61–64.
[72] Hatvany J (1985) Intelligence and Cooperation in Heterarchic Manufacturing [106] Malone TW (2018) Superminds: The Surprising Power of People and Computers
Systems. Robotics and Computer-Integrated Manufacturing 2/2:101–104. Thinking Together, Little. Brown Spark ISBN: 0316349100.
[73] Hauschild MZ, Kara S, Røpke I (2020) Absolute Sustainability: Challenges to Life [107] Ma rkus A, Kis Va ncza T, Monostori L (1996) A Market Approach to Holonic
Cycle Engineering. CIRP Annals 69/2:533–553. Manufacturing. CIRP Annals 45/1:433–436.
[74] Heimann D, Eickemeyer L, Kleibrink J (2016) Study Industry 4.0: International [108] McKinsey, 2018, Ai, Automation, and the Future of Work: Ten Things to Solve
Comparison. Huawei, and Handelsblatt Research Institute . For, McKinsey Global Institute briefing note prepared for the Tech4good Sum-
[75] Herrmann C, Juraschek M, Burggra €f P, Kara S (2020) Urban Production: State of mit, organized by the French Presidency, 7.
the Art and Future Trends for Urban Factories. CIRP Annals 69/2:764–787. [109] Mehrabi MG, Ulsoy AG, Koren Y (2000) Reconfigurable Manufacturing Systems:
[76] Heuss L, Lux-Gruenberg G, Hammerstingl V, Schno € s F, Rinck P, Reinhart G, M Z Key to Future Manufacturing. Journal of Intelligent Manufacturing 11/4:403–419.
(2018) Dynamic Planning and Adaption of Mobile Robots for Flexible Produc- [110] Meier H, Roy R, Seliger G (2010) Industrial Product-Service Systems—Ips2. CIRP
tion. Moblie autonomous robots in the Smart Factory Wt Werkstattstechnik 108/9 . Annals 59/2:607–627.
[77] Heuss L, Blank A, Dengler S, Zikeli GL, Reinhart G, Franke J (2019) Modular Robot [111] Michaelis MT, Johannesson H, HA ElMaraghy (2015) Function and Process
Software Framework For the Intelligent and Flexible Composition of Its Skills, Modeling for Integrated Product and Manufacturing System Platforms. Journal
Springer International Publishing, Cham248–256. of Manufacturing Systems . 36/203-215.
[78] Hofmann C, Krahe C, Stricker N, Lanza G (2020) Autonomous Production Control [112] Michalos G, Spiliotopoulos J, Makris S, Chryssolouris G (2018) A Method for
for Matrix Production Based on Deep Q-Learning. Procedia CIRP . 88/25-30. Planning Human Robot Shared Tasks. CIRP Journal of Manufacturing Science and
[79] Hollnagel E, Woods DD (2005) Joint Cognitive Systems: Foundations of Cognitive Technology . 22/76-90.
Systems Engineering, CRC press. ISBN: 1420038192. [113] Miehe R, Full J, Scholz P, Demmer A, Bauernhansl T, Sauer A, Schuh G (2019) The
[80] Holweg M (2008) The Evolution of Competition in the Automotive Industry, in Build Biological Transformation of Industrial Manufacturing-Future Fields of Action in
to Order: The Road to the 5-Day Car, Parry G, and Graves A, Editors, Springer Lon- Bioinspired and Bio-Based Production Technologies and Organization. Procedia
don. London 13-34 ISBN: 978-1-84800-225-8. doi: 10.1007/978-1-84800-225- Manufacturing, 39/737-744 .
8_2. https://doi.org/10.1007/978-1-84800-225-8_2. [114] Miranda J, Pe rez-Rodríguez R, Borja V, Wright PK, Molina A (2019) Sensing,
[81] Hu SJ (2013) Evolving Paradigms of Manufacturing: From Mass Production to Mass Smart and Sustainable Product Development (S3 Product) Reference Frame-
Customization and Personalization, Elsevier B.V., 3–8. work. International Journal of Production Research 57/14:4391–4412.
[82] Huang S, Wang G, Shang X, Yan Y (2018) Reconfiguration Point Decision Method [115] Mittal S, Romero D, Wuest T (2018) Towards a Smart Manufacturing Maturity
Based on Dynamic Complexity for Reconfigurable Manufacturing System (Rms). Model for Smes (Sm3e), Springer International Publishing, Cham155–163.
Journal of Intelligent Manufacturing 29/5:1031–1043. [116] Moghaddam SK, Houshmand M, Saitou K, Fatahi Valilai O (2020) Configuration
[83] HumAn CEntred Factories Cluster (ACE). Human-Centred Factories from Theory Design of Scalable Reconfigurable Manufacturing Systems for Part Family. Inter-
to Industrial Practice. White Paper Lessons Learned and Recommendations : 51. national Journal of Production Research 58/10:2974–2996.
[84] IHK Mu € nchen und Oberbayern, Guide Industry 4.0. https://ihk-industrie40.de/. [117] Mo € hring HC, Wiederkehr P, Erkorkmaz K, Kakinuma Y (2020) Self-Optimizing
(accessed 19 September 2020). Machining Systems. CIRP Annals 69/2:740–763.
[85] IoT Analytics, 2020, Industry 4.0 Adoption 2020 Who Is Ahead?; https://iot- [118] Monostori L (2014) Cyber-Physical Production Systems: Roots, Expectations and
analytics.com/industry-4-0-adoption-2020-who-is-ahead/. (accessed 13 August R&D Challenges, Procedia CIRP, 17/9-13.
2020). [119] Monostori L, Kada r B, Bauernhansl T, Kondoh S, Kumara S, Reinhart G, Sauer O,
[86] Ivanov D, Dolgui A (2020) Viability of Intertwined Supply Networks: Extending Schuh G, Sihn W, Ueda K (2016) Cyber-Physical Systems in Manufacturing. Cirp
the Supply Chain Resilience Angles Towards Survivability. A Position Paper Annals 65/2:621–641.
Motivated by Covid-19 Outbreak. International Journal of Production Research 58/ [120] Monostori L, Markus A, Van Brussel H, Westk€ampfer E (1996) Machine Learning
10:2904–2915. Approaches to Manufacturing. CIRP Annals 45/2:675–712.
[87] Jung K, Kulvatunyou B, Choi S, Brundage MP (2016) An Overview of a Smart [121] Monostori L, Va ncza J (2020) Towards Living Manufacturing Systems. Procedia
Manufacturing System Readiness Assessment, Springer International Publishing, CIRP . 93/323-328.
Cham705–712. [122] Monostori L, Vancza J, SRT Kumara (2006) Agent-Based Systems for Manufactur-
[88] Kagermann H, Wahlster W, and Helbig J, 2013, Securing the Future of German ing. CIRP Annals 55/2:697–720.
Manufacturing Industry: Recommendations for Implementing the Strategic Initia- [123] Mori M, Fujishima M (2009) Reconfigurable Machine Tools for a Flexible
tive Industrie 4.0, Final report of the Industrie, 4/0: Manufacturing System, in Changeable and Reconfigurable Manufacturing Sys-
[89] Kampker A, Schuh G, Burggr€af P, Nowacki C, Swist M (2012) Cost Innovations by tems, ElMaraghy HA, Editor, Springer London. London101–109. https://doi.org/
Integrative Product and Production Development. CIRP Annals 61/1:431–434. 10.1007/978-1-84882-067-8_5.
[90] Kashkoush M, ElMaraghy H (2014) Product Family Formation for Reconfigurable [124] Mourad MH, Nassehi A, Schaefer D, ST Newman (2020) Assessment of Interopera-
Assembly Systems. Procedia CIRP . 17/302-307. bility in Cloud Manufacturing, Robotics and Computer-Integrated Manufacturing.
[91] Koren Y, Heisel U, Jovane F, Moriwaki T, Pritschow G, Ulsoy G, Van Brussel H 61/101832.
(1999) Reconfigurable Manufacturing Systems. CIRP Annals 48/2:527–540. [125] Moussa M, ElMaraghy H (2020) A Genetic Algorithm-Based Model for Product
[92] Koren Y, Hu SJ, Gu P, Shpitalni M (2013) Open-Architecture Products. CIRP Platform Design for Hybrid Manufacturing. Procedia CIRP . 93/389-394.
Annals 62/2:719–729. [126] Moussa M, ElMaraghy H (2019) Master Assembly Network for Alternative
[93] Koren Y, Wang W, Gu X (2017) Value Creation through Design for Scalability of Assembly Sequences. Journal of Manufacturing Systems . 51/17-28.
Reconfigurable Manufacturing Systems. International Journal of Production [127] Nardo M, Forino D, Murino T (2020) The Evolution of Man Machine Interaction:
Research 55/5:1227–1242. The Role of Human in Industry 4.0 Paradigm. Production & Manufacturing
Research 8/1:20–34.
JID: CIRP
[128] National Institute of Standard and Technology (NIST), 2014, Smart Manufactur- [157] Schuh G, Gottschalk S (2008) Production Engineering for Self-Organizing Com-
ing Operations Planning and Control. https://www.nist.gov/system/files/docu- plex Systems. Production Engineering 2/4:431.
ments/2017/05/09/FY2014_SMOPAC_ProgramPlan.pdf. (accessed 17 March [158] Schuh G, Lenders M, Nussbaum C, Kupke D (2009) Design for Changeability, in
2021). Changeable and Reconfigurable Manufacturing Systems, ElMaraghy HA, Editor,
[129] Nee AYC, Ong SK, Chryssolouris G, Mourtzis D (2012) Augmented Reality Appli- Springer London. London251–266. https://doi.org/10.1007/978-1-84882-067-
cations in Design and Manufacturing. CIRP Annals 61/2:657–679. 8_14.
[130] OECD, 2019, Gdp Per Hour Worked (Indicator). https://doi.org/10.1787/ [159] Schuh G, Monostori L, Csa ji BC, Do
€ ring S (2008) Complexity-Based Modeling of
1439e590-en. (accessed 10 September 2020). Reconfigurable Collaborations in Production Industry. CIRP Annals 57/1:445–
[131] OECD, 2019, Gross Domestic Spending on Research and Development (Indica- 450.
tor). https://data.oecd.org/rd/gross-domestic-spending-on-r-d.htm. (accessed [160] Schuh G PJ, Cremer S, Liu Y (2020) Production System 4.0 Ergebnisse Aus Dem
10 September 2020). Konsortial-Benchmarking, Werkzeugmaschinenlabor WZL der RWTH Aachen,
[132] Olivares Aguila J, ElMaraghy H (2020) Co-Development of Product and Supplier ISBN: 978-3-9822216-0-1 https://www.wzl.rwth-aachen.de/global/show_docu-
Platform. Journal of Manufacturing Systems . 54/372-385. ment.asp?id=aaaaaaaaaqnncop&download=1.
[133] Olivares Aguila J, ElMaraghy W (2018) Structural Complexity and Robustness of [161] Schuh G, Potente T, Wesch-Potente C, Weber AR, Prote J-P (2014) Collaboration
Supply Chain Networks Based on Product Architecture. International Journal of Mechanisms to Increase Productivity in the Context of Industrie 4.0. Procedia
Production Research 56/20:6701–6718. CIRP . 19/51-56.
[134] Olivares Aguila J, ElMaraghy W (2019) Supply Chain Resilience and Structure: [162] Schuh G, Prote J-P, Gu € tzlaff A, Thomas K, Sauermann F, Rodemann N (2019)
An Evaluation Framework. Procedia Manufacturing . 28/43-50. Internet of Production: Rethinking Production Management, Springer Berlin Hei-
[135] Olivares Aguila J, ElMaraghy W (2020) System Dynamics Modelling for Supply delberg, Berlin, Heidelberg533–542.
Chain Disruptions. International Journal of Production Research : 1–19. [163] Schumacher A, Erol S, Sihn W (2016) A Maturity Model for Assessing Industry
[136] Olivares Aguila J, ElMaraghy W, ElMaraghy H (2019) Impact of Risk Attitudes on 4.0 Readiness and Maturity of Manufacturing Enterprises. Procedia CIRP . 52/
the Concurrent Design of Supply Chains and Product Architectures. Procedia 161-166.
CIRP . 81/974-979. [164] Schwab K (2018) The Global Competitiveness Report 2017-2018. World Eco-
[137] PAiCE Begleitforschung, 2018, Potentials of Artificial Intelligence in the nomic Forum .
Manufacturing Industry in Germany. https://www.tip-innovation.de/fileadmin/ [165] Schwarz C, Schmitt M (2017) Maturity Model for Lean Production. ZWF 7:506–
inhalte/News/potenziale-kuenstlichen-intelligenz-im-produzierenden- 509.
gewerbe-in-deutschland.pdf. (accessed 19 September 2020). [166] Shimomura Y, Akasaka F (2013) Toward Product-Service System Engineering: New
[138] PAiCE Bundesministerium fu € r Wirtschaft und Energie, 2018, Potentials of Artifi- System Engineering for Pss Utilization, Heidelberg: Springer Berlin Heidelberg,
cial Intelligence in the Manufacturing Industry in Germany. https://www.bmwi. Berlin27–40.
de/Redaktion/DE/Publikationen/Digitale-Welt/paice-monitor-2018.html [167] Shimomura Y, Hara T, Arai T (2009) A Unified Representation Scheme for Effec-
(accessed 19 September 2020). tive Pss Development. CIRP Annals 58/1:379–382.
[139] Panetta K, 2019, Gartner Top 10 Strategic Technology Trends for 2020. https:// [168] Shimomura Y, Nemoto Y, Kimita K (2014) State-of-Art Product-Service Systems
www.gartner.com/smarterwithgartner/gartner-top-10-strategic-technology- in Japan the Latest Japanese Product-Service Systems Developments. Procedia
trends-for-2020/. (accessed 7 August 2020). CIRP . 16/15-20.
[140] Papanastasiou S, Kousi N, Karagiannis P, Gkournelos C, Papavasileiou A, Dimou- [169] Shu LH, Ueda K, Chiu I, Cheong H (2011) Biologically Inspired Design. CIRP Annals
las K, Baris K, Koukas S, Michalos G, Makris S (2019) Towards Seamless Human 60/2:673–693.
Robot Collaboration: Integrating Multimodal Interaction. The International Jour- [170] Simpson TW, Jiao J, Siddique Z, Ho € ltta
€-Otto K (2014) Advances in Product Family
nal of Advanced Manufacturing Technology 105/9:3881–3897. and Product Platform Design, SpringerNew YorN.
[141] Park S (2016) Development of Innovative Strategies for the Korean Manufactur- [171] Singapore Economic Development Board. The Singapore Smart Industry Readiness
ing Industry by Use of the Connected Smart Factory (Csf). Procedia Computer Sci- Index - Catalysing the Transformation of Manufacturing, https://speta.org/wp-
ence 91:744–750. content/uploads/2019/08/SIRI-White-Paper.pdf.
[142] Pellegrinelli S, Orlandini A, Pedrocchi N, Umbrico A, Tolio T (2017) Motion Plan- [172] Sorensen DG, ElMaraghy H, Brunø TD, Nielsen K (2020) Classification Coding of
ning and Scheduling for Human and Industrial-Robot Collaboration. CIRP Annals Production Systems for Identification of Platform Candidates. CIRP Journal of
66/1:1–4. Manufacturing Science and Technology . 28/144-156.
[143] Piller FT, 2020, Ten Propositions on the Future of Digital Business Models for [173] Spath D, Ganschar O, Gerlach S, H€ammerle M, Krause T, Schlund S (2013) Pro-
Industry 4.0 in the Post-Corona Economy, Available at SSRN 3617816. duktionsarbeit Der Zukunft-Industrie 4.0, Vol. 150, Fraunhofer Verlag Stuttgart.
[144] PricewaterhouseCoopers, 2016, The Industry 4.0 /Digital Operations Self Assess- [174] Stark R, Kind S, Neumeyer S (2017) Innovations in Digital Modelling for Next
ment. https://i40-self-assessment.pwc.de/i40/landing/. (accessed 3 April 2021). Generation Manufacturing System Design. CIRP Annals 66/1:169–172.
[145] PricewaterhouseCoopers (PwC), 2014, Industry 4.0 - Chances and Challenges of [175] Statista Research Department, 2020, R&D Spending as a Share of Gdp in G7
the Fourth Industrial Revolution. https://store.pwc.de/de/pdf/2014/october/ Countries and China 2000-2018. https://www.statista.com/statistics/1102457/.
industrie-4-0-chancen-und-herausforderungen-der-vierten-industriellen-revo- (accessed 16 March 2021).
lution. (accessed 16 March 2021). [176] Sung T (2017) Industry 4.0: A Korea Perspective. Technological Forecasting and
[146] Pritschow G, Wurst K-H, Kircher C, Seyfarth M (2009) Control of Reconfigurable Social Change . 132/40-45.
Machine Tools, in Changeable and Reconfigurable Manufacturing Systems, ElMara- [177] Sutherland JW, Richter JS, Hutchins MJ, Dornfeld D, Dzombak R, Mangold J, Rob-
ghy HA, Editor, Springer London. London71–100. https://doi.org/10.1007/978-1- inson S, Hauschild MZ, Bonou A, Scho € nsleben P, Friemann F (2016) The Role of
84882-067-8_4. Manufacturing in Affecting the Social Dimension of Sustainability. CIRP Annals -
[147] Proctor M, and Wilkins J, 2018, Eu Automation - 4.0 Sight: Digital Industry Manufacturing Technology 65/2:689–712.
around the World. https://www.euautomation.com/4sight?utm_source=Stone- [178] Tao F, Cheng Y, Zhang L, Nee AYC (2017) Advanced Manufacturing Systems:
JunctionEUI017&utm_medium=PR. Socialization Characteristics and Trends. Journal of Intelligent Manufacturing 28/
[148] PTC, 2016, Automotive Parts Supplier Launches Iot Initiative in Sixweek Sprints 5:1079–1094.
Powered by Ptc Iot Manufacturing Solutions. https://www.ptc.com/-/media/ [179] Tao F, Qi Q (2019) New It Driven Service-Oriented Smart Manufacturing: Frame-
Files/PDFs/Case-Studies/J8229-Hirotec-America-CS_EN.pdf?la=de- work and Characteristics. IEEE Transactions on Systems, Man, and Cybernetics:
DE&hash=D7DDC76755A5D5010203352BC1ED29B8. (accessed 16 September Systems 49/1:81–91.
2020). [180] Tao F, Qi Q, Liu A, Kusiak A (2018) Data-Driven Smart Manufacturing. Journal of
[149] PTC, 2016, Hirotech Startet Iot-Initiative. https://www.ptc.com/de/case-studies/ Manufacturing Systems . 48/157-169.
hirotec. (accessed 16 September 2020). [181] Tao F, and Zhang M, 2017, Digital Twin Shop-Floor: A New Shop-Floor Paradigm
[150] Ren L, Zhang L, Wang L, Tao F, Chai X (2017) Cloud Manufacturing: Key Charac- Towards Smart Manufacturing,IEEE Access, 5/20418-20427.
teristics and Applications. International Journal of Computer Integrated [182] Tao F, Zuo Y, Xu LD, Zhang L (2014) Iot-Based Intelligent Perception and Access
Manufacturing 30/6:501–515. of Manufacturing Resource toward Cloud Manufacturing. IEEE Transactions on
[151] centre Resource efficiency (2017) Resource Efficiency through Industry 4.0: Poten- Industrial Informatics 10/2:1547–1557.
tial for Manufacturing Smes, https://www.resource-germany.com/publications/ [183] Terkaj W, Tolio T, Valente A (2009) Focused Flexibility in Production Systems,
studies/ (accessed 16 March 2021). Changeable and Reconfigurable Manufacturing Systems, ed, ElMaraghy HA,
[152] Reuter C, Gartzen T, Prote J, Fra €nken B (2016) Industry 4.0 Audit, in VDI-Z Integ- Springer London. London. https://doi.org/10.1007/978-1-84882-067-8_3 ISBN:
rierte Produktion, Springer-VDIDu € sseldorf. 978-1-84882-067-8. doi: 10.1007/978-1-84882-067-8_3.
[153] Bento Gd Santos, Tontini G (2018) Developing an Instrument to Measure Lean [184] Textor C (2020) Labor Productivity Per Hour in China 2000-2018, https://www.
Manufacturing Maturity and Its Relationship with Operational Performance. statista.com/statistics/878164/china-labor-productivity-per-hour/ (accessed 16
Total Quality Management & Business Excellence : 977–995. 29/9-10. March 2021).
[154] Schmitt R, Große-Bo € ckmann M (2014) Kollaborative Cyber-Physische Produk- [185] Thompson MK, Moroni G, Vaneker T, Fadel G, Campbell RI, Gibson I, Bernard A,
tionssysteme: Ausbruch Aus Der Produktivita €tsfalle. Integrative Produk- Schulz J, Graf P, Ahuja B, Martina F (2016) Design for Additive Manufacturing:
tion Industrie : 365–374. Trends, Opportunities, Considerations, and Constraints. CIRP Annals 65/2:737–
[155] Schuh G, Aghassi S, Orilski S, Schubert J, Bambach M, Freudenberg R, Hinke C, 760.
Schiffer M (2011) Technology Roadmapping for the Production in High-Wage [186] Tolio T, Bernard A, Colledani M, Kara S, Seliger G, Duflou J, Battaia O, Takata S
Countries. Production Engineering 5/4:463–473. (2017) Design, Management and Control of Demanufacturing and Remanufac-
[156] Schuh G, Anderl R, Gausemeier J, ten Hompel M, and Wahlster W, 2017, Indus- turing Systems. CIRP Annals 66/2:585–609.
trie 4.0 Maturity Index - Managing the Digital Transformation of Companies. [187] Tolio T, Ceglarek D, ElMaraghy HA, Fischer A, Hu SJ, Laperrie re L, Newman ST,
https://en.acatech.de/publication/industrie-4-0-maturity-index-managing-the- Vancza J (2010) Species-Co-Evolution of Products, Processes and Production Sys-
digital-transformation-of-companies/. tems. CIRP Annals - Manufacturing Technology 59/2:672–693.
JID: CIRP
[188] Tomiyama T, Lutters E, Stark R, Abramovici M (2019) Development Capabilities [200] Wang P, Liu H, Wang L, Gao RX (2018) Deep Learning-Based Human Motion Rec-
for Smart Products. CIRP Annals 68/2:727–750. ognition for Predictive Context-Aware Human-Robot Collaboration. CIRP Annals
[189] Ueda K (1992) A Concept for Bionic Manufacturing Systems Based on DNA-Type 67/1:17–20.
Information, in Human Aspects in Computer Integrated Manufacturing, Olling [201] Wang X, Ong SK, Nee AYC (2016) Real-Virtual Components Interaction for
GJ, and Kimura F, Editors, Elsevier. Amsterdam853–863. https://doi.org/10.1016/ Assembly Simulation and Planning. Robotics and Computer-Integrated
B978-0-444-89465-6.50078-8. Manufacturing . 41/102-114.
[190] Ueda K, Hatono I, Fujii N, Vaario J (2000) Reinforcement Learning Approaches to [202] West DM, and Lansang C, 2018, Global Manufacturing Scorecard. https://www.
Biological Manufacturing Systems. CIRP Annals 49/1:343–346. brookings.edu/research/global-manufacturing-scorecard-how-the-us-com-
[191] Ueda K, Vaario J, Fujii N (1998) Interactive Manufacturing: Human Aspects for pares-to-18-other-nations/. (accessed 7 September 2020).
Biological Manufacturing Systems. CIRP Annals 47/1:389–392. [203] Westka €mper E (2003) Assembly and Disassembly Processes in Product Life Cycle
[192] Valckenaers P, Van Brussel H (2015) Design for the Unexpected: From Holonic Perspectives. CIRP Annals - Manufacturing Technology 52/2:579–588.
Manufacturing Systems Towards a Humane Mechatronics Society. Butterworth- [204] Wiendahl HP, ElMaraghy HA, Nyhuis P, Za €h MF, Wiendahl HH, Duffie N, Brieke
Heinemann . ISBN: 0128036966. M (2007) Changeable Manufacturing - Classification, Design and Operation. CIRP
[193] Valckenaers P, Van Brussel H (2005) Holonic Manufacturing Execution Systems. Annals 56/2:783–809.
CIRP Annals 54/1:427–432. [205] World Bank (The), 2019, Gross Domestic Product (Gdp). https://data.worldbank.
[194] Van Brussel H, Wyns J, Valckenaers P, Bongaerts L, Peeters P (1998) Reference org/indicator/NY.GDP.MKTP.CD. (accessed 10 September 2020).
Architecture for Holonic Manufacturing Systems: Prosa. Computers in Industry [206] World Bank (The), 2019, Manufacturing, Value Added. https://data.worldbank.
37/3:255–274. org/indicator/NV.IND.MANF.CN. (accessed 10 September 2020).
[195] Vancza J, Monostori L (2017) Cyber-Physical Manufacturing in the Light of Pro- [207] World Economic Forum (WEF), 2019, Harnessing the Fourth Industrial Revolution
fessor Kanji Ueda's Legacy. Procedia Cirp . 63/631-638. for the Circular Economy Consumer Electronics and Plastics Packaging. (White
[196] Vancza J, Monostori L, Lutters D, Kumara SR, Tseng M, Valckenaers P, Van Brussel Paper). Platform for Accelerating The Circular Economy (PACE). http://www3.
H (2011) Cooperative and Responsive Manufacturing Enterprises. CIRP Annals weforum.org/docs/WEF_Harnessing_4IR_Circular_Economy_report_2018.pdf.
60/2:797–820. [208] Yin Y, Stecke KE, Li D (2018) The Evolution of Production Systems from Industry 2.0
[197] Vercruysse J, 2018, Business Process Maturity Models. https://www.effic.be/en/ through Industry 4.0. International Journal of Production Research : 848–861. 56/1-2.
business-process-maturity-models/. (accessed 19 September 2020). [209] Youssef AM, HA ElMaraghy (2007) Optimal Configuration Selection for Reconfig-
[198] Wang J, Ma Y, Zhang L, Gao RX, Wu D (2018) Deep Learning for Smart urable Manufacturing Systems. International Journal of Flexible Manufacturing
Manufacturing: Methods and Applications. Journal of Manufacturing Systems . Systems 19/2:67–106.
48/144-156. [210] Zhang J, Xue D, Gu P (2015) Adaptable Design of Open Architecture Products
[199] Wang L, Gao R, Va ncza J, Kru€ ger J, Wang XV, Makris S, Chryssolouris G (2019) with Robust Performance. Journal of Engineering Design : 1–23. 26/1-3.
Symbiotic Human-Robot Collaborative Assembly. CIRP Annals 68/2:701–726.

Evolution and Future of Manufacturing Systems

Uploaded by

Copyright:

Available Formats

Evolution and Future of Manufacturing Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Evolution and Future of Manufacturing Systems

Uploaded by

Copyright:

Available Formats

JID: CIRP

ARTICLE IN PRESS [m191;July 20, 2021;18:02]

CIRP Annals - Manufacturing Technology 00 (2021) 1 24

Contents lists available at ScienceDirect

CIRP Annals - Manufacturing Technology

Evolution and future of manufacturing systems

Manufacturing systems continue to evolve in design, conﬁgura-

1.1. Challenges and motivation

2 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

machines, workstations, robots, and other equipment arranged in

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 3

4 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 5

6 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 7

platform to generate product variants. Product platforms are effec-

2.2.1. Co-platforming of products and manufacturing systems

8 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

 Intelligence (smartness) where elements can acquire information

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 9

4.1. Maturity models and benchmarks for manufacturing systems

10 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

uses analytic network process and focuses on leadership and perfor-

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 11

The e.GO Life electric car is produced in an Industry 4.0 plant

12 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

4.1.1. Autonomous, mobile, and ad-hoc cooperating robot teams

4.3.3. Smart factory by Ericsson

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 13

is researched at the laboratory for manufacturing systems and auto-

5. Adaptive cognitive manufacturing systems paradigm

14 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 15

16 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 17

18 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

6. Conclusions and future research

6.1. Insights and reﬂections

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 19

20 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 21

22 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24 23

24 H. ElMaraghy et al. / CIRP Annals - Manufacturing Technology 00 (2021) 1 24

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Intelligence (smartness) where elements can acquire information