, Volume 1, Issue 1–2, pp 36–47 | Cite as

Perspectives on Manufacturing Automation Under the Digital and Cyber Convergence

  • Shimon Y. Nof
  • Jose Reinaldo SilvaEmail author
Original Article


The evolution of industrial automation has been divided into four or five main cycles of “industrial revolutions,” also called “disruptive innovations” and “automation revolutions.” The most recent one, started around the 1990s and still on-going, points to the current perspectives envisioned for the twenty-first century and perhaps even beyond. In practice, however, it is difficult to comprehend the real value and impacts by the so called Digital Manufacturing, Smart Factory, Automation 5.0, or Industry 4.0. Furthermore, with frequent and rapid innovations, it is unclear how the emerging digital, smart, and cyber-augmented factories of the future can benefit from the digital and cyber convergence. Which are the dominant factors that motivate and justify the evolution of manufacturing through this current cycle? In this article, we review the relationships between digital, virtual, and cyber convergence, and recent manufacturing engineering challenges ranging from virtual enterprises to collaborative e-Manufacturing, and service orientation. We then point out new perspectives and opportunities for design and re-arrangement in production, highlighting the trend of fusion between knowledge, product, process, and service. The impact on methods of analysis, informatics, collaborative intelligence, and design of industrial systems is also analyzed under the new trends and achievements so far in digital and cyber convergence. With several case studies, we also illustrate the emerging challenges.


Manufacturing automation Digital and cyber convergence Cyber-physical manufacturing Product-service architecture Collaborative e-Manufacturing Best matching 

1 Introduction

Since the second half of the past century, the fruitful coupling between academic developments and industry progress has become more harmonized and synchronized, even if time delays inevitably persist between them. The endeavors of evolution on both sides opened a new stage, where change agents in industry and academy share leadership in innovation. That was the case with a set of new production concepts called (by industry) Industry 4.0 (Hermann et al. 2016; Moghaddam and Nof 2017b; Roblek et al. 2016) and Automation 5.0 (by industry and academy) as related to cyber-physical convergence (Conti et al. 2012; Monostori 2014; Rajkumar et al. 2010; Nof et al. 2015, 2017; Nof 2017). The full comprehension of the concept (Hermann et al. 2016) and its implementation over the legacy systems became one of the great engineering challenges in the current era. One of the difficulties is to rationalize and to fit the newly developed approaches in the industry development process. It began in the second half of the past century with the discovery of transistors and the advent of the computer. In fact, digital computation changed and disrupted practically everything in modern life. It has been especially evident after all communication and communication media turned to be digital, in a process called “digital convergence.” After that transformation, “computers” and “computation” became ubiquitous and present in all processes of human life and activities, from agriculture to space missions, from scientific explorations and healthcare to education. Industrial production is a fundamental human activity and has the characteristics of a legacy process that is actually aimed to support and sustain human life and activities in modern society.

As a result, it is highly important to derive an approach that takes into account the legacy production system, together with having a clear perspective of the future of industry production. Related to this objective is the need to develop and agree on new production architectures, enabling a common perspective of the emerging new approaches. These architectures could support intermediary steps towards benefiting from those new and converging targets, effectively, and particularly, without preventing further future progress. Obviously, recent innovations energized the emergence of cyber-physical systems (CPS), emerging in transportation, energy distribution, production, manufacturing, and critical infrastructure (Conti et al. 2012; Monostori 2014; Rajkumar et al. 2010; Nof et al. 2015, 2017; Nof 2017). Thus, it should be recognized that with convergence come, thanks to human ingenuity and competitiveness, associated new disruptive innovations and divergence.

Digital factories resulted from the confluence of digital convergence concepts and industrial production after the flexible manufacturing approach and era (Zhou et al. 2012). It was the first significant attempt to improve industrial operations with full computer communication and information. The use of open Internet instead of closed industrial network broke the walls of industrial facilities and positioned production in the sensible point of being open to IIot (Industrial Internet of Things) and IIoS (Industrial Internet of Services). But this useful progress has also become more vulnerable to risks in terms of security and information overloads (Brettel et al. 2016; Roblek et al. 2016). Industrial production has finally reached the same position that of social networks interactions and communication: more open to the world, relatively faster, and at the same time, subject to relatively more noise and more security risks.

On the other hand, from the perspective of benefits, factories with increasing levels of digital automation share certain advantages already available to other interactions, such as open and agile communication, sharing and exchanging data, and exploring and enabling interoperability and virtual operations. Those beneficial features have further increased the demand and improved the opportunity for innovation. They also influenced the need for and ability of merging between products and services (Nof 2013; Silva and Silva 2015; Silva 2014; Silva and Nof 2015). The old approach to virtual enterprises (Camarinha-Matos and Afsarmanesh 1999), based on restricted communication between companies, has been replaced by a strong and well-defined communication architecture (Camarinha-Matos and Afsarmanesh 1999; Moghaddam and Nof 2017a; Moghaddam and Nof 2017b; Silva and Nof 2015). An overview of the convergence examples already evident in industry is shown in Table 1 [updated from (Silva and Nof 2015; Nof et al. 2017; Nof 2017).
Table 1

Influence and challenges of the digital and cyber-physical divergence on factory work


Technology advancement

Human work augmented by

Fundamental work system principle

Engineering and management concerns

Work system goal1

Before the Industrial Revolution



Manual and animal power

Work methods

Enable work functions

Before computers

Engines: machines

Power; motion; moves

Human-machine systems

Work flow

Reduce muscle load

Computer Age

Computers; communication; robots

Control; data processing; automation

Computer-aided and computer-integrated systems

Human-computer interaction; human-robot interaction

Reduce information overload

Information Age

Telecom: Internet: mobility

Cognitive skills: collaboration

e-Work; e-Mfg

e-Work design; collaboration; parallelism

Reduce information and task2 overload

Cyber-physical Age

IIoT; IIoS; Al; Brain Models

Collaborative control protocols and collaborative intelligence

c-Work; c-Mfg service orientation

Humans-machines-robots harmony

Increase work quality3

1Including workers’ goal, organization’s goal, and customers’ satisfaction goal

2Including overloads of human and automation agents, and supply network participants

3Increasing work quality, including work processes and delivered products

Since merely about ten year have passed since the old industrial virtualization perspective to the current IIoT/IIoS/CPS era, it indicates a relatively fast evolution from one perspective. It also points out the fact that the industrial sector has been more resistant to the new approach, since ubiquitous computing had already reached many other services and operations in an increasingly networked and connected society. A possible explanation is the typical inertia of business processes, reluctant to assume more risks without proven and predictable returns.

Evidently, the collaborative factory of the future (Moghaddam and Nof 2017b) and Industry 4.0 raised several important questions about trust and security challenges; required intelligence of network control (Brettel et al. 2016); the practicable possibility of having reliable, satisfactory, and effective distributed production (Dutra and Silva 2016; Moghaddam and Nof 2017b; Silva and Nof 2015); and empowering the renewal of advanced planning and design methods (Hermann et al. 2016; Moghaddam and Nof 2017a; Silva and Silva 2015).

In what follows, we update the impacts of digital and cyber convergence for the current days to show that industry is, in fact, the last to incorporate changes but, on the other hand, the one that can make disruptive changes in society by distributing production. Following, e-Service and e-Manufacturing are explained as the basis for the factory of the future. The merging with service is focused on, to propose a new architecture, and finally the need for new methods to design this new industrial model, divided in stages: requirements identification and assignment by matching tools and solutions.

2 Digital and Cyber Convergence Scope for Manufacturing

Digital convergence (Bainbridge and Roco 2016; Blackman 1998) is the result of developments that occurred after the Second World War. With the invention of the transistor, its subsequent advantageous use in modern computers changed the face of the twentieth century. Initially, computers were accepted and meant as a tool to speed up calculations or decipher encrypted code. The challenge was how to manipulate the computer, and insert proper input and program instructions. Digitalization was a precondition for using a computer, and no other consequences were yet envisioned. The decade of the 1950s was noted for that.

The next decade, the 1960s, was marked by a huge evolution of software, besides the evolution in the hardware to achieve (at that time) greater computer capacity and power. At that time and still during the 1970s, there was a great concern about heat dissipation in computers, a barrier for any future faster computers. The use of computers was divided to the business market (banks, financial companies) and engineering applications, besides the military and academy. This decade and the next contributed to the emergence of a “software crisis”1 and to the reappearance of Artificial Intelligence (AI) now with a more pragmatic approach, consolidated in 1981 in the Handbook of Artificial Intelligence, by Avron Barr and Edward A. Feigenbaum. Then, digital computation received another boost, used in the “specialist, or expert systems” to solve problems. Since then, the explosion of use of digital computation became highly diversified, uncovering also the domain of computational, computer-based, and knowledge-based services.

It seems natural that the next step was the spread of computing services all over society, propelled by “personal computers.” Researchers and industry overcame the barriers of heat dissipation, and continuously discovered miniaturization of microelectronics for computer hardware. Innovations in computer communications, data base technology, better operating systems, and programming languages have been developed. The personal and smaller-size computers not only could bring to homes and offices the effective calculation tool and business processor, but also could reach anyone interested in new branches of application, from sophisticated agenda calendars to data processing and office automation platforms. The computer during this period of the 1980s keeps the protagonist role in the process; however, a consequence of its ubiquitous use exposes and prepares the convergence of all formats to digits: The eventual digital convergence.

In 1978, Nicholas Negroponte pointed to the inevitable convergence of “printing, computing and broadcasting,” opening a broad discussion about the engineering impacts of digitalization. On the other hand, the media aspects received even more attention from society and researchers (Kung et al. 1999; Yoffie 1996). Newspapers and television companies should be aware of the forthcoming change, disrupting their business process and foundations.

Milton Mueller (Mueller 1999), however, defined digital convergence as the overlapping of the media integration, putting together the following three ingredients:
  • “In a single application or service, information contents from telephony, sound broadcasting, television, motion pictures, photography, printed text publishing and electronic money” in a good anticipation of the blue media used in most automobiles and in other, similar services;

  • “The overlapping of functions performed by different networks;”

  • “Growing interactivity and interoperability in different networks and information appliances.”

Recent analysis of the digital convergence phenomenon detaches technological convergence, where several artifacts (products) are enhanced to fit several different functionalities exploring digital formats (Giachetti and Dagnini 2017). Smartphones are currently the best-known example of technological convergence. They have integrated powerful computing and communication and interaction and entertainment, now easily and readily available to the masses, to everyone, not just to computer geeks of yesteryears. The smartphones also illustrate a natural, almost intuitive tendency of servitization of several different products and became a strategy for different sectors of industry (Giachetti and Dagnini 2017). Furthermore, this tendency is now associated with an expectation that future and more exciting useful apps with service orientation can and should be integrated with and by the smartphone. It also seems natural to observe that this tendency is brought into the factory and supply chain, reaching the manufacturing environment, and comprising a branch of Manufacturing 4.0. Currently, manufacturing cells submitted to technological convergence and enhanced with digital communications attributes would and already result in cyber-physical systems (Nof et al. 2015, 2017; Nof 2017). Connected, autonomous cars, drones, robot, human, and machine teams already advance with cyber augmentation. They include cyber-physical systems built with intelligent machine vision and sensor networks, under intelligent collaborative control, and with brain-like guidance.

In fact, communication is a key issue. It is the foundation for information exchange and for interactions among distributed, selected, or massively distributed parties and services (human and artificial agents). Digital convergence and ubiquitous computing opened a new era where communications—that had already spread in all activities, social, cultural, and productive—were accelerated and widely intensified. Productive activities demanded more and broader communications, especially to achieve cooperation among the participating parties, humans, robots, software tools and agents, and machines. Today’s world could not be imagined without supply chains and supply networks, and their complete dependence on communications infrastructure, transmission, processing, and exchange of signals, information, knowledge, and intelligence.

Finally, there is also the integration of different engineering branches—which began in the late 1990s—and the intensification of multidisciplinary studies and applications. Nano-technological devices, processes, and materials can bring new results from engineering and physics, as well as the concepts of quantum computing, additive manufacturing, and collaborative robotics. Cognitive robotics celebrates the integration of cognitive science and engineering, mediated by artificial intelligence. Several tools show a new relationship between engineering and medicine, including new surgical, home-care, and rehabilitation automated systems. A recent report by the Organization for Economic Cooperation and Development (OECD-Paris) presented these multidisciplinary, polytechnic approaches, pointing out a new development challenge and opportunity to be explored in production and manufacturing.

In summary, the evident digital convergence, and the emerging digital and cyber-physical convergence are the engines of disruptive changes. They present powerful savings, competitive innovations, and opportunities for better quality of life, and at the same time, result in the current challenges for production and manufacturing. They bring new levels and power of communication and integration. In addition, they enable another key issue: computer-supported, communication-enabled, and cyber-augmented collaboration. Collaboration has historically enhanced the potential of production sites, which is a key point to reduce the size of current industrial facilities, to reduce pollution and save energy—both of which turn out to be expensive and destructive obstacles to supply, services, transportation, logistics, and delivery. In addition, new distributed architectures are demanded to explore the full solutions potential and possibilities, which has also become a serious, yet promising challenge.

2.1 Towards the Collaborative Factory of the Future

In a recent article, Moghaddam and Nof (Moghaddam and Nof 2017b) analyzed relevant and current definitions for the concept of factory of the future (FoF). Among them is the statement of the famous economist Warren Bennis in 1991, saying, “The factory of the future will have only two employees, a man and a dog. The man will be there to feed the dog. The dog will be there to keep the man from touching the equipment”. Somehow, that statement reflects the common thinking of the 1990s referred to in the previous section: That automation would mean a tremendous growth of autonomy, with full integrity, correctness, and completeness, reaching a point where humans should avoid “touching the machines”—a linear interpolation, evolving from the tendency launched by digital and cyber convergence.

Several other quotes obtained from experienced people in the industry (Moghaddam and Nof 2017b) point also to a continuous evolution of the factory keeping the same structure of the 1990s with a reflexing modernization of equipment by merely including better, more powerful ICT devices, services, and processes.

The paradigm shift in manufacturing, however, does not and cannot rely on just including more ICT, virtual, and augmented reality devices and processes. It does need to integrate their use for collaborative intelligence (CI) to support a new, distributed architecture that complies with the need to converge to sustainable production modes. Sustainable production and manufacturing requires small and/or virtual facilities, enabling drastic reduction of the amount of energy necessary to support each factory; a factory intelligent from rich computational and cyber-collaborative points of view, that is also sustainable, as depicted by Vernadat et al. (Vernadat et al. 2018).

The key concept is a change in distributed and collaborative architecture, towards a distributed set of cyber-physical entities with service orientation: enhanced cells, startups, and other production systems, supply sources, etc., all with proper identity and definitive purpose, and capability to collaboratively, including competitively, join different processes as a “service-component” (Moghaddam and Nof 2018). It is crucial to add that the service orientation in this scenario goes beyond the “servitization” as understood in the beginning of this century, but envisaged as a complete symbiosis between service and product. We will call that concept “product-service” in this precise sense (Dutra and Silva 2016; Melani et al. 2016; Moghaddam and Nof 2018; Moghaddam et al. 2015; Postigo et al. 2018; Silva and Silva 2015; Silva and Nof 2015).

2.2 Systems Design Applied to the Collaborative Factory of the Future

From the previous section, it becomes clear that the design of new production systems cannot remain the same. The new target is a more complex association of collaborative systems, each one considered as a service-component, meaning that each one is an individual intelligent service provider with its proper, defined purpose. A new production system of systems can combine—totally or partially—a set of these components to compose a production workflow and process mining (van der Aalst 2016; van der Aalst and van Hee 2004).

To design the digital factory of the future—in a version that goes beyond the limits of the current Industry 4.0 and touches the Automation 5.0—it is necessary to start with a method directed to systems of systems and propose the proper adaptations. The design mode can be summarized as a layered system where the first layer is the conception and requirement layer (CoL); the second layer is the architecture layer (ArL); and the last layer is the assignment layer (AsL). Layered approaches have been used in software engineering and multi-agent systems since the 1990s, but only currently reached the mainstream of Systems Design. Today, there are already commercial environments for Systems Design and Systems of Systems Design, applying this approach.

Figure 1 depicts a schema for the layered system approach to production systems in Industry 4.0 and also in Automation 5.0. It should be noticed that formal modeling and relational schemata would be present in each layer. For the first two layers, Petri Nets modeling techniques are used and for the third layer, a best matching formal approach (Moghaddam and Nof 2017a).
Fig. 1

Layered systems design for Industry 4.0 and Automation 5.0

The first layer, CoL, starts with the conception and requirements analysis and modeling to uncover the “what” in the design process—treated in the context of a system of systems. It belongs to this process to model the system/user coupling of symbiosis. Therefore, a functional approach would not be advisable, and that is why a goal-oriented approach is included. Consequently, conventional approaches in UML or similar semi-formal schema are replaced by KAOS (Silva et al. 2018; Silva and Silva 2015; Van Lamsweerde 2009).

An interesting and realistic example is shown in the following, where this KAOS approach is applied to the service of electric power distribution in urban environment. A distributed approach was developed in the conceptual layer that can fit with the legacy system and also enables the entire system to evolve to a more flexible configuration arrangement.

Figure 2 depicts an arrangement where power is distributed to clusters of residential/industrial microgrids, composing a co-generation distributed service. Requirements are depicted in the KAOS diagram shown below in Fig. 3 (Postigo et al. 2018).
Fig. 2

Microgrid general architecture

Fig. 3

Petri Net associated to the KAOS diagram for the microgrid design

Besides engineering planning in the conceptual design layer, a formalization of the main relations in the system in Petri Nets allows a sound analysis of requirements and also of the interaction between users and systems. A translation algorithm was developed and implemented to convert KAOS diagrams directly to Petri Nets (Silva et al. 2018; Silva and Silva 2015). Several works already available in the literature make a translation from a formal synthesis of the diagrams in LTL (linear tree logic).

The architecture layer (ArL) shown in the example is synthesized by a reference model based on the standard IEC61850 (Lu et al. 2014). A similar solution could be used by industrial groups and corporation to spread its culture and adherence to the different “components” belonging to the same group. Concepts of collaboration and composing e-Services should be included in the first two layers, to support decisions in the third layer of the assignment process.

3 Cyber-Augmented Collaboration, E-Services, and E-Manufacturing

Collaborative e-Work, e-Manufacturing, and e-Services have been developed over recent decades as a framework for design, engineering, and control of next-generation manufacturing systems, providing computer-supported, communication-enabled, and cyber-augmented collaborative, productive activities. They have fundamentally been transforming the ability of highly distributed and networked enterprises to improve e-Production and e-Service (Silva and Nof 2015; Nof et al. 2015; Nof 2003, 2006, 2007). The “e” in collaborative e-Work refers to all activities performed based on and executed through digital electronics e-Activities, including virtual e-Business, e-Commerce, e-Manufacturing, and e-Logistics, along with v- (virtual) Factories, and v-Enterprises, where virtual reality and augmented reality further support and enable cyber-augmented operations and collaboration among distributed agents.

All e-Activities rely on computer support and ICT (Information & Communication Technology), and require collaboration in their inherent interactions between machines, robots, humans, and computers. The e-Activities are supported by the four areas or wheels of collaborative e-Work, including (1) e-Work; (2) Integration, Coordination, and Collaboration; (3) Distributed Decision Support Systems; and (4) Active Middleware and their corresponding 15 e-Dimensions.

Technically and conceptually, e-Work is defined as follows [Nof 2003]: “As power fields, such as magnetic fields and gravitation, influence bodies to organize and stabilize, so does the sphere of computing and information technologies. It envelops us and influences us to organize our work systems in a different way, and purposefully, to stabilize work while effectively producing the desired outcomes. The emerging concepts of IIoT, IIoS, and CPS are now transforming the physical manufacturing world to cyber-physical and self-governing systems of embedded agents, e.g., sensors and actuators, linked and operating through the collaborative control protocols (CCP). The theory and models of e-Work (Nof et al. 2015) provide foundations for augmenting abilities of organizations and all agents, human, computer, robot, software, etc., to interact and collaborate through selective and adaptable protocols, in order to accomplish their goals (Table 1).

With the advent of cyber-physical systems and cyber-augmented collaboration, e-Work is being transformed to c-Work, just as e-Manufacturing is being transformed to c-Manufacturing, meaning cyber-collaborative work and manufacturing. In this emerging era, it is necessary to design, evaluate, and be able to predict and rationalize the factory of the future. The PSA architecture is a useful approach for this purpose.

4 PSA Architecture: a Distributed Approach to Industrial Production with Service Orientation

The emergence of manufacturing service orientation may be one of the most influential and beneficial trends in the digital and cyber-physical convergence (Dutra and Silva 2016; Moghaddam et al. 2015; Nof 2013; Schmenner 2009; Silva and Nof 2015; Tao et al. 2011; ISO/IEC 18384:2016 2016; Devadasan et al. 2013). A recent review of smart manufacturing architectures (Moghaddam et al. 2017) points out the convergence of manufacturing and industrial reference architectures and international standards towards service orientation, based on its benefits (Gao et al. 2011; Li et al. 2010; Jiang et al. 2016; IBM 2008; ISO, ISO/IEC/TR 30102 2012).

PSA architecture is an abstract reference model to the architecture layer (ArL), where new manufacturing is thought as a distributed composition of production units (PUs), or components. In the context of c-Work, these components are called Co-Us, collaborating units, each providing its services, as needed. Each component is an arrangement of persons, agents, machines, and other product/server units—in a recursive system of systems approach. Each component delivers a service or product-service, in a workflow that results in a final product-server, the main goal of the requirements established in the conceptual and requirements layers.

Final consumers are also modeled as a component and merged in the general system by a consumer provider interaction (Qiu 2014), which stands for the user coupling relations of the requirements layer. Thus, the general production system should couple the product-service with those user components, configuring what is called in service design a value co-creation relationship (Shimomura 2005; Sanders and Stappers 2008; Schmenner 2009; Spohrer et al. 2007).

A new manufacturing arrangement should then select components (production components or user components) and couple with each one in a service provider/consumer relation (Qiu 2014). Figure 4 illustrates an abstract coupling arrangement.
Fig. 4

Component coupling to compose a production in PSA

Naturally, those couplings should be done in a predictive, repetitive sequence—and sometimes in predictive real-time. Therefore, the result, which should end with a user component relation, must be able to accomplish a manufacturing process that can be formally represented by a Petri Net (Dutra and Silva 2016). Thus, the same formal representation can be used in requirements relations and for planning the sequence of couplings in a useful production arrangement.

In this case, the ArL layer can also be seen as a process tool or system that delivers tentative architectures to satisfy the requirements modeled in CoL layer. Figure 5 shows a general modeling schema to support the systematic modeling for PSA architecture. It starts with the annotation of a previous step to goal requirements, which is the intention model.
Fig. 5

Layered product-service architecture (PSA)

Intention model can precede requirements to capture the cognitive intention of the production designer—which should be present in all steps of the architecture modeling. Eric Yu first introduced the concept of intention model in his PhD thesis, by the end of 1990s. The evolution of the process creates a NIS project to transform intentions in a standard. A good description of intention models and its benefits can be found in the book Social Modeling for Requirements Engineering edited by Eric Yu, Paolo Giogini, Neil Maiden, and John Mylopoulos (Yu et al. 2010). We are not going deep in the formalism of intention models here, but it is important to point out that capturing intentions should be like capturing the essence of requirements or what is behind them—especially if we already know from CoL layer that requirements are consistent and fit those intentions. In that case, it is more efficient to keep the intentions to the next layer.

The value co-creation is responsible for the modeling of couplings between components and final users or to the integration of those couplings in the production process. Those production processes are modeled in the service (or product-service) planning. Process workflow can be checked (or model-checked if it is a real-time process) by the workflow management system, and finally, a protocol management should model the couplings with components and a process protocol that allow the supervision of the production process automatically. Workflow management based on Petri Nets is a good framework to verify the production model and process mining cold be used to investigate sound possibilities of assignment.

4.1 The Assignment and Best Matching Stage

The third and final layer, AsL, provides assignment modeling and best matching of components, tools, and services to fulfill the identified requirements. A technique for this purpose has been developed and implemented recently (Moghaddam and Nof 2018). It is based on the Best Matching Theory and Application recent manuscript (Moghaddam and Nof 2017a).

The assignment and best matching are anticipated to enable agile manufacturing processes in response to the growing demand for personalized products-services with shorter lifecycles. This trend has resulted in a gradual transformation of traditional tree-like and monolithic systems into complex networks of self-contained and autonomous components. Cloud manufacturing is also an emerging concept that enables modularization and service orientation in the context of c-Manufacturing, in which systematic orchestration, matching, and sharing of services and components are the objective. This approach is based on a framework for dynamic integration of manufacturing services and components in a collaborative network of factories, and physical or virtual organizations. The framework dynamically recommends the best matching of services, components, and supplier organizations, as well as the best collaboration decisions in terms of sharing shareable tools, services, and/or components between organizations.

The assignment and best matching problem is formulated as a bi-objective mixed-integer program (BOMIP). It is solved by an efficient socio-inspired tabu search. The objectives of the model are to increase service level and enhance collaboration through maximizing service fulfillment and minimizing unnecessary sharing of services/components, respectively. Numerical experiments have been conducted to demonstrate the benefits of the developed framework for efficient and optimal configuration and reconfiguration of collaborative networked factories and organizations, addressing the Industry 4.0 demand for agility through modularization and service orientation.

This approach contributes to the design of the collaborative factory of the future by Cloud Manufacturing (Wu et al. 2013; Tao et al. 2011). It defines a 3-dimensional matching problem associated with dynamic integration of services and components in a collaborative network of organizations (CNO), and developing a socio-inspired tabu search (SITS) for addressing the dynamicity and complexity of the design problem. In this definition, an “organization” refers a system of heterogeneous manufacturing components (e.g., assets; software; ideas; concepts) responsible for providing a set of services (Ko and Nof 2012), ranging from manufacturing services such as scheduling and sourcing to core services such as registry, search, and contracts (Tao et al. 2011). Specifically, the service-component assignment, matching, and integration problem is associated with the integration aspect of the Service-Oriented Architecture (SOA) Reference Model (ISO/IEC 18384-2:2016) (ISO/IEC 18384:2016 2016), which enables loose-coupling between the service requests and provisions by dynamically matching the requested services to the best available components (Zhong and Nof 2015; Lartigau et al. 2015; Tao et al. 2017).

In the CNO, each service may be processed by multiple components that belong to different organizations (Tao et al. 2009). That is, the organizations that are part of the same CNO share their services and/or components in order to create an agile network capable of addressing the emerging complexity, uncertainty, and dynamicity of processes (Huang et al. 2000) (see Fig. 6). In such settings, the key to sustain and evolve is to optimize collaboration among distributed organizations that share services/components to achieve individual and mutual benefits (Nof et al. 2015; Nof 2007; Camarinha-Matos et al. 2009). Advances in c-Work have facilitated cyber-physical collaboration among geographically dispersed organizations, from file exchange to direct access to computers, machines, tools, software, and databases (Foster et al. 2001). Hence, while some services require physical proximity to components, e.g., reading RFID tags, other services can be processed remotely through communication channels, e.g., remote control of tele-robots in virtual manufacturing. Thus, these services can be categorized as non-shareable and shareable, respectively.
Fig. 6

Service-component integration in a Cloud Manufacturing CNO composed of organizations A and B (Moghaddam and Nof 2018; ISO/IEC 18384:2016 2016; DIN SPEC 91345:2016-04 2016)

Due to the inherent physical limitations associated with the non-shareable services, integration and collaboration are enabled and augmented through two distinct yet interrelated types of matching decisions:
  1. 1)

    Service-component matching. Each organization owns a certain set of components that must provide specific services associated with the CNO. Due to the dynamic variations in demand and capacity, organizations may encounter capacity shortage or surplus over time. Hence, organizations may collaborate by sharing their shareable services in the case of insufficiency of local components, in order to balance the overall workload of the network and minimize idle resources, congestions, and delays. This set of decisions deals with intra-/inter-organization matching of services and components, with the objective of maximizing the overall fulfillment of requested services.

  2. 2)

    Component-organization matching. Unbalanced distribution of components among organizations may result in unnecessary and excessive rate of collaboration, high communication network load, congestion, delays, and cost, and increased possibilities of errors and conflicts. Such problems can be alleviated by enabling the organizations to dynamically share their shareable components. This set of decisions deals with dynamic matching of components and organizations, with the objective of minimizing the overall rate of overload and unwanted collaboration between organizations.


The collaborative service-component integration problem has two conflicting objectives: (1) maximizing the fulfillment rate of requested services; (2) minimizing unnecessary collaborations. These conflicting objectives are achieved by concurrent best matching between the sets of services and components, and the sets of components and organizations (Nof 2013). The BOMIP solution was developed for mathematical representation and computation of the problem. Due to the computational complexity of the problem, an original heuristic algorithm, Socio-Inspired Tabu Search, was developed for the optimization, considering potential changes in the system characteristics and the network topology. It was shown (Moghaddam and Nof 2018) that the achieved improvements are effective and in line with the emerging requirements for agility in terms of flexibility, scalability, adaptation, and resilience of CNO.

A challenge for further developments in the application of PSA as described above is the actual implementation in industrial case studies. Those case studies will enable refinement of the PSA procedures and algorithms, as well as the validation in practice.

5 Conclusions

The aim of this article has been to review, analyze, and explain perspectives on manufacturing automation under the digital and cyber convergence. The concepts attempted under the evolving definitions of Industry 4.0 (Brettel et al. 2016; Dutra and Silva 2016; Hermann et al. 2016; Roblek et al. 2016; DIN SPEC 91345:2016-04 2016) have been mostly motivated by the advantages, and challenged by the risks brought by the digital convergence.

Collaborative Control Theory (CCT) (Nof et al. 2015; Nof 2003, 2006, 2007) and the emerging cyber-physical convergence (Vernadat et al. 2018; Monostori 2014; Rajkumar et al. 2010; Nof et al. 2017; Nof 2017; Zhong and Nof 2015) attempt to augment, rationalize, and extend the goals of Industry 4.0 with the cyber-collaborative features of Automation 5.0.

In parallel, manufacturing automation advancements have enabled and at the same time benefited from the coupling of products and services in the framework of servitization (Shimomura 2005; Dutra and Silva 2016; Moghaddam and Nof 2018; Moghaddam et al. 2015; Nof 2013; Qiu 2014; Schmenner 2009; Silva et al. 2018; Silva and Silva 2015; Silva 2014; Silva and Nof 2015). Methods have been investigated and developed to design for servitization, both for Industry 4.0 and for Automation 5.0 goals and architectures. The PSA architecture/methodology and the best matching of product-service components and organizations under Cloud Manufacturing have been explained and illustrated.

Convergence, digital, cyber, and technological can be viewed as the confluence realized after significant technological disruptions, and leading to acceptance, stability, regulation, and even standardization. In the emerging c-Work and c-Manufacturing, and the collaborative factory of the future, however, careful assessment of risks and benefits using techniques, such as PSA and Best Matching, will have to guide researchers and industry in their challenging design for effective and sustainable solutions.

Two other challenges that can be recognized are as follows:
  1. 1.

    The fundamental need for Collaborative Intelligence (CI) that integrates collective and cumulative knowledge and intelligence from diversified sources for better design decisions, and timely and predictive responses during the life of the factory; See (Devadasan et al. 2013; Zhong et al. 2015) for initial work in this direction.

  2. 2.

    Cyber-augmented collaborative control protocols (C3P), that will be needed to harmonize the activities of the best matched product-service components, and the interactions among allied factories and organizations, once they are optimally assigned.


Further research and industry studies are also anticipated in the polytechnic approaches to advance the benefits of cyber-augmented collaborative interactions to accelerate and ameliorate learning, decision making, emergency response, and other social and cultural services.


  1. 1.

    The software crisis was created by the difference in the relatively slow evolution of hardware—inherited from the 1950s—compared to rapid software development evolution.



S.Y. Nof acknowledges support by colleagues, students, and projects under the PRISM Center for Production, Robotics, and Integration Software for Manufacturing and Management, at Purdue University, and the PGRN, PRISM Global Research Network.


  1. Bainbridge WS, Roco MC (2016) Handbook of science and technology convergence, Springer referenceGoogle Scholar
  2. Blackman C (1998) Convergence between telecommunications and other media. Telecommunication Policy 22 (Elsevier Science Ltd.): 163–170Google Scholar
  3. Brettel M, Friederichsen N, Keller M, Rosemberg M (2014) How virtualization, decentralization and network building change the manufacturing landscape: an Industry 4.0 perspective. International Journal on Mechanical, Industrial Science and Engineering 8(1):37–44Google Scholar
  4. Camarinha-Matos L, Afsarmanesh H (1999) Tendencies and general requirements for virtual enterprises. In: Infrastructures for virtual enterprises, chap. 2, pp. 15–30Google Scholar
  5. Camarinha-Matos LM, Afsarmanesh H, Galeano N, Molina A (2009) Collaborative networked organizations - concepts and practice in manufacturing enterprises. Comput Ind Eng 57:46–60CrossRefGoogle Scholar
  6. Conti M, Das SK, Bisdikian C, Kumar M, Ni LM, Passarella A, Roussos G, Tröster G, Tsudik G, Zambonelli F (2012) Looking ahead in pervasive computing: challenges and opportunities in the era of cyber–physical convergence. Per Mob Comp 8(1):2–21CrossRefGoogle Scholar
  7. Devadasan P, Zhong H, Nof SY (2013) Collaborative intelligence in knowledge based service planning. Expert Syst Appl 40(17):6778–6787CrossRefGoogle Scholar
  8. DIN SPEC 91345:2016-04 (2016) Reference Architecture Model Industrie 4.0 (RAMI4.0), DIN Deutsches Institut für Normung e. V., BerlinGoogle Scholar
  9. Dutra DS, Silva JR (2016) Product-service architecture (PSA): towards a service engineering perspective in Industry 4.0. IFAC-Papers OnLine 49(31):91–96. CrossRefGoogle Scholar
  10. Foster I, Kesselman C, Tuecke S (2001) The anatomy of the grid: enabling scalable virtual organizations. Int J High Perf Comp App 15:200–222CrossRefGoogle Scholar
  11. Gao J, Yao Y, Zhu V, Sun L, Lin L (2011) Service-oriented manufacturing: a new product pattern and manufacturing paradigm. J Intell Manuf 22(3):435–446CrossRefGoogle Scholar
  12. Giachetti C, Dagnini GB (2017) The impact of technological convergence on firm’s product portfolio strategy: an information-based imitation approach. R&D Mgt 47(1):17–35CrossRefGoogle Scholar
  13. Hermann M, Pentek T, Otto B (2016) Design principles for Industry 4.0 Scenarios. In: Proc. of 49th Hawaii International Conference on System Sciences, pp. 3928–3937. IEEE Press, HawaiiGoogle Scholar
  14. Huang CY, Ceroni JA, Nof SY (2000) Agility of networked enterprises – parallelism, error recovery and conflict resolution. Comput Ind 42:275–287CrossRefGoogle Scholar
  15. IBM (2008) Reaping the rewards of your service-oriented architecture infrastructure, IBM global servicesGoogle Scholar
  16. ISO, ISO/IEC/TR 30102 (2012) Information technology — Distributed Application Platforms and Services (DAPS) — General technical principles of Service Oriented ArchitectureGoogle Scholar
  17. ISO/IEC 18384:2016 (2016) Information technology — Reference Architecture for Service Oriented Architecture (SOA RA)Google Scholar
  18. Jiang P, Ding K, Leng J (2016) Towards a cyber-physical-social-connected and service-oriented manufacturing paradigm: social manufacturing. Mfg Let 7:15–21Google Scholar
  19. Ko HS, Nof SY (2012) Design and application of task administration protocols for collaborative production and service systems. Int J Prod Econ 135:177–189CrossRefGoogle Scholar
  20. Kung L, Kroll AM, Ripken B, Walker M (1999) Impact of the digital revolution on the media and communications industriesd. J Eur Ins for Comm and Cult 6(3):29–48Google Scholar
  21. Lartigau J, Xu X, Nie L, Zhan D (2015) Cloud manufacturing service composition based on QoS with geo-perspective transportation using an improved Artificial Bee Colony optimization algorithm. Int J Prod Res 53:4380–4404CrossRefGoogle Scholar
  22. Li B, Zhang L, Wang S, Tao F, Cao J, Jiang X, Song X, Chai X (2010) Cloud manufacturing: a new service-oriented manufacturing model. Comput Integr Manuf Syst 16(1):1–8Google Scholar
  23. Lu W, Zhao Y, Li W, Du H (2014) Design and application of microgrid operation control system based on iec 61850. J Mod Power Sys Cl Energy 3(2):256–263CrossRefGoogle Scholar
  24. Melani A, Silva J, Souza G, Silva J (2016) Fault diagnosis based on petri nets: the case study of a hydropower plant. IFAC Papers Online 49(31):1–6CrossRefGoogle Scholar
  25. Moghaddam M, Nof SY (2017a) Best matching theory and applications. SpringerGoogle Scholar
  26. Moghaddam M, Nof SY (2017b) The collaborative factory of the future. Int J Comput Integr Manuf 30(1):23–43Google Scholar
  27. Moghaddam M, Nof SY (2018) Collaborative service-component integration in Cloud Manufacturing. Int J Prod Res 56(1–2):677–691CrossRefGoogle Scholar
  28. Moghaddam M, Silva JR, Nof SY (2015) Manufacturing-as-a-service: from e-Work and service-oriented architecture to the Cloud Manufacturing paradigm. IFAC Papers OnLine 48(3):828–833. CrossRefGoogle Scholar
  29. Moghaddam M, Kenley CR, Colby JM, Berns MNC, Rausch R, Markham J, Skeffington WM, Garrity J, Chaturvedi AR, Deshmukh AV (2017) Next-generation enterprise architectures: common vernacular and evolution towards service-orientation. In Industrial informatics (INDIN), 2017 IEEE 15th International Conference on (pp. 32–37)Google Scholar
  30. Monostori L (2014) Cyber-physical production systems: roots, expectations and R&D challenges. Procedia Cirp 17:9–13CrossRefGoogle Scholar
  31. Mueller M (1999) Digital convergence and its consequences. Journal of the Institute of Communication and Culture 6(3)Google Scholar
  32. Nof SY (2003) Design of effective e-Work: review of models, tools and emerging challenges. PP&Control, Special Issue on e-Work: Models and Cases 15:8–681Google Scholar
  33. Nof SY (2006) Collaborative e-Work and e-Manufacturing: challenges for production and logistic managers. J Intell Manuf 17:689–701CrossRefGoogle Scholar
  34. Nof SY (2007) Collaborative control theory for e-Work, e-Production, and e-Service. Annu Rev Control 31:281–292CrossRefGoogle Scholar
  35. Nof SY (2013) Research advances in manufacturing with service-oriented e-work and production. In: Tsuzuki M, Silva J (eds) Proc. of the Int. 11th IFAC IMS, Intelligent Manufacturing Systems Workshop. Elsevier, Sao Paulo. CrossRefGoogle Scholar
  36. Nof SY (2017) Cyber-augmented collaboration of distributed e-Work and robotics, (plenary) Brazilian Symposium on Intelligent Automation, SBAI 2017Google Scholar
  37. Nof SY, Ceroni J, Jeong W-T, Moghaddam M (2015) Revolutionizing collaboration through e-Work, e-Business, and e-Service, Springer ACES Series for Automation, Collaboration, and E-ServicesGoogle Scholar
  38. Nof SY, Huang C-Y, Stachowiak A (2017) Cyber physical production and resilience, (Plenary), ICPR 24, PoznanGoogle Scholar
  39. Postigo M, Silva J, Silva J (2018) Formal requirements specification for microgrid based on architecture IEC 61.850. In: E.R.C. da Silva, E.C.T. de Macedo (eds.) proc. of the XXII Brazilian congress on automation. Brazilian society of automationGoogle Scholar
  40. Qiu R (2014) Service science: the foundation of service engineering and management. WileyGoogle Scholar
  41. Rajkumar RR, Lee I, Sha L, Stankovic J (2010) Cyber-physical systems: the next computing revolution. In Proceedings of the 47th design automation conference (pp. 731–736). ACMGoogle Scholar
  42. Roblek V, Mesko M, Kaprez A. (2016) Complex View of Industry 4.0. SAGE Open pp. 1{11.
  43. Sanders, E.N., Stappers, P.: Co-creation and the new landscapes of design. Co-Design 4(1), 5{18 (2008)Google Scholar
  44. Schmenner RW (2009) Manufacturing, service, and their integration: some history and theory. Int J Oper Prod Manag 29(5):431–443. URL CrossRefGoogle Scholar
  45. Shimomura Y (2005) Service modeling for service engineering. In: Knowledge and skill chains in engineering and manufacturing, chap. 4, pp. 31–38Google Scholar
  46. Silva JR (2014) New trends in manufacturing: converging to service and intelligent systems. IFAC PapersOnline 47(3):2628–2633. CrossRefGoogle Scholar
  47. Silva JR, Nof SY (2015) Manufacturing service: from e-work and service-oriented approach towards a product-service. IFAC-Papers OnLine 48(3):1628–1633. CrossRefGoogle Scholar
  48. Silva JM, Silva JR (2015) Combining KAOS and GHENeSys in the requirement and analysis of service manufac. IFAC-Papers OnLine 48(3):1634–1639. CrossRefGoogle Scholar
  49. Silva J, Salmon A, del Foyo P, Silva J (2018) Requirements engineering at a glance: comparing GORE and UNL methods in the design of automated systems. In: E.R.C. da Silva, E.C.T. de Macedo (eds.) proc. of the XXII Brazilian congress on automation. Brazilian Society of AutomationGoogle Scholar
  50. Spohrer J, Maglio PP, Bailey J, Gruhl D (2007) Steps toward a science of service systems. Computer 40(1):71{77. CrossRefGoogle Scholar
  51. Tao F, Zhao D, Hu Y, Zhou Z (2009) Resource service composition and its optimal-selection based on particle swarm optimization in manufacturing grid system. IEEE Trans Ind Inf 4:315–327CrossRefGoogle Scholar
  52. Tao F, Zhang L, Venkatech VC, Luo Y, Cheng Y (2011) Cloud manufacturing: a computing and service oriented manufacturing model. Proc Inst Mech Eng B J Eng Manuf 225:1969–1976CrossRefGoogle Scholar
  53. Tao F, Cheng J, Cheng Y, Gu S, Zheng T, Yang H (2017) SDMSim: a manufacturing service supply–demand matching simulator under cloud environment. Robot Comput Integr Manuf 45:34–46CrossRefGoogle Scholar
  54. van der Aalst W (2016) Process mining: data science in action. SpringerGoogle Scholar
  55. van der Aalst W, van Hee K (2004) Workflow management: models, methods and systems. MIT PressGoogle Scholar
  56. Van Lamsweerde A (2009) Requirements engineering: From system goals to UML models to software, vol 10. Chichester, WileyGoogle Scholar
  57. Vernadat F, Chan F, Molina A, Nof SY, Panetto H (2018) Industrial systems and knowledge management in industrial engineering: recent advances and new perspectives. Int J Prod Res 56(5):12Google Scholar
  58. Wu D, Greer MJ, Rosen DW, Shaefer D (2013) Cloud manufacturing: strategic vision and state-of-the-art. J Manuf Syst 32:564–579CrossRefGoogle Scholar
  59. Yoffie D (1996) Competing in the age of digital convergence. Calif Manag Rev 38(4):31–53CrossRefGoogle Scholar
  60. Yu E, Giorgini P, Maiden N, Mylopoulos J (2010) Social modeling for requirements engineering. MIT PressGoogle Scholar
  61. Zhong H, Nof SY (2015) The dynamic lines of collaboration model: collaborative disruption response in cyber-physical systems. Comput Ind Eng 87:370–382CrossRefGoogle Scholar
  62. Zhong H, Levalle RR, Moghaddam M, Nof SY (2015) Collaborative intelligence-definition and measured impacts on internetworked e-work. Mgt Prod Eng Rev 6(1):67–78Google Scholar
  63. Zhou Z, Xie S, Chen D (2012) Fundamentals of digital manufacturing science. SpringerGoogle Scholar

Copyright information

© Escola Politécnica - Universidade de São Paulo 2018

Authors and Affiliations

  1. 1.Purdue UniversityWest LafayetteUSA
  2. 2.Universidade de Sao PauloSao PauloBrazil

Personalised recommendations