*3.4. KETs*

Key enabling technologies have been defined as knowledge-intensive technologies associated with high-intensity research and development, rapid innovation cycles, high capital expenditure, and high-skilled employment [114]. They can also be classified as advanced manufacturing technologies, advanced materials and nanotechnologies, life-science technologies, artificial intelligence, micro/nano electronics and photonics, and security and connectivity tools.

Examples where KETS have a place include those of an organisational type (fractal, holonic, or bionic organisation), digital (cloud, big data, AR&VR, wearables, and mobile devices) and technological type (collaborative robots, additive manufacturing, etc.) [115] Data generation, data analytics and decision-making technologies are examples of KET-artificial intelligence. Human-machine interfaces and cyber-physical systems provide examples of KET-security and connectivity [116,117].

Information and communication technology (ICT) and KETs challenge traditional production structures and require the search for alternative and innovative solutions to those proposed so far, through the use of disruptive activities and compentencies by Operator 4.0. This requires engineers with the ability to solve problems based on different outlooks rather than those already raised by cultural heritage. At this point, the third generation of activity theory becomes useful by allowing the transcending of established frameworks and by generating creative and innovative solutions that break with the mental structures of established systems, thereby providing solutions to the new challenges that arise.

KETs are used as variety filters and amplifiers to support DfHFinI4.0. Connectivism, together with the law of requisite variety, activity theory and KETs, help SCMS modelling, and enable real-time support and assistance for the engineer, through the design of dynamic interfaces that co-evolve according to needs considered in engineering environments 4.0, which are characterised by the digitisation and virtualisation of products and processes. This involves an impact on improvement in the acquisition and training of competencies, as well as on the affective connectivity established between human and technological factors.

#### **4. DfHFinI4.0 Framework**

As shown above, the technological environment promoted by Industry 4.0 determines that the tasks of Operator 4.0 are mainly focused on solving complex problems, decision-making, and the ability to adapt to new scenarios and situations in which human and technological factors work collaboratively. It is therefore important to define, design, and build SCMS in a way that allows part of the knowledge of the workers to be supported, and to respond and support, giving response and support to articulate competencies of a wider scope at the cognitive, operational, affective, and co-evolutive levels with the best working conditions. The acquisition of the competencies required in Industry 4.0 for the development of day-to-day tasks and the accumulation of know-how from the lessons learned at the workstation must also be supported. The context in which the analysis of work must be placed is within that of the value chain as an instrument of analysis of added value that incorporates the activity into products and is demanded by customers. One example of this is related to lean manufacturing systems [41], in the form of a map of the value stream, otherwise known as value stream mapping (VSM).

The analysis of the value chain of smart factories must be carried out, while considering their primary and support activities in operating conditions (professional domain), and in learning factories (academic domain). To this end, the potential of digital enablers and their projection in the integration of the human factor, machines and robots for collaborative work performance must be considered, by interacting jointly, through interfaces for both primary and support activities. The latter are elements of the SCMS that enable the interaction and navigation of Operator 4.0, not only internally, but also externally, with other socio-technical systems from other value chains, such as suppliers, distributors, customers, and plants of reverse manufacturing. This interaction takes place at a physical and virtual level, and hence the interfaces must be capable of acting as links between both physical and virtual realities. Their study is required at the level of sustainable smart manufacturing value chain design, and of associated chains of suppliers and distributors.

The productive activity in the manufacturing systems is carried out from a set of activities represented in its value chain, or VSM, based on customer value maps, which are specified downstream with various lean manufacturing tools. Subsequently, the requirements of processes, material and information flows, scheduling, tasks, equipment, operators, and their associated competencies are established [118]. As indicated before, the value chain contains primary activities and support activities that host the KETs as a means of improving their e fficiency and sustainability. The process of incorporating the KETs into the activities of the value chain, focused on the human factor, requires their analysis, the potential of the KETs, the distribution of activities between those carried out by the human factor and the machines and collaborative robots, the knowledge required for this analysis, and strategies that support the entire manufacturing system, whose priority involves the development of strategies to empower the human factor and its a ffective coupling to the environment.

Based on the analysis of the added value, the analysis of the potential of the KETs and of the objective of SCMS focused on the human factor, a set of tools is proposed for the of the analysis. As illustrated in Figure 7, this analysis of the value creation activity is based on the conceptual frameworks, for the configuration of the value chain as an integrated and co-evolving SCMS to the highest degree of abstraction.

The set of proposed tools are related to the activity theory model proposed in a fractal way in terms of various degrees of granularity [119], based on the analysis of the levels of company, departments, activities and workstations. Subsequently, the variety required of Operator 4.0 in workstations is characterised, as is the variety required for the successful and satisfactory development of the task, by means of the addition of the necessary filters and amplifiers through adaptive interfaces, and by establishing the connectivist navigation strategies that allow the operator, assisted by the competencies, to manage the lessons learned in day-to-day work. This ensures that the process, machines and robots have at least the same variety as Operator 4.0, so that the cyber-physical socio-technical system reaches stability. The modelling of the SCMS employs the duality established by the digital twin by integrating the human and technological factors through adaptive communication interfaces, thereby allowing the generation of the variety of use required for the sustainable production process, through subrogate models that are dynamically built and managed from the cloud with tools of big data and artificial intelligence, such as machine learning, classification techniques, and deep learning.

**Figure 7.** DfHFinI4.0 for the configuration of the integrated cyber-physical and co-evolutionary system.

Given the potential of the digitisation of Industry 4.0, the company can establish new forms of relationships with suppliers, consumers and other value chains, by forming the global value chain through horizontal and vertical clustering with grea<sup>t</sup> innovative potential. This generates a horizontal integration that involves real-time cooperation between human and technological factors, as well as vertical integration between partners, suppliers, and customers, which, when brought to the field of cyber-physical systems, can be developed in a fractal way in easily replicable structures, throughout the DfHFinI4.0 framework that is proposed for the value chain of the company.

In the framework developed in this paper, the workflow is analysed through activity theory, and forms an activity system that produces and develops actions based on said theory, by breaking it down into elements at the individual level and elements at the collective level, whose explicit and implicit knowledge can be ascertained, modelled and divided between technology and the human factor. The contradictions that occur between the elements of the activity system appear once the forces that drive creativity and innovation within the smart and learning factories are resolved. This study of provoked interactions facilitates the modelling of the system itself, and explains the relationships that lead to the co-evolutive and a ffective coupling between the human and technological factors, which enables continuous improvement in the adaptive process between the two factors. This modelling results in a dynamic and multilateral flow of data associated with the information systems necessary for the adaptation of intelligent manufacturing to customer requirements.

Connectivism and the law of requisite variety enable the SCMS to be modelled, and characterize the particularity of the activity and the specific profile of the engineer or technician as the Operator 4.0 that carries out this activity, regarding experience, and cognitive and a ffective level, and adapts it through filters and variety amplifiers. Connectivism allows the establishment of navigation strategies and online assistance in problem-solving processes, through the information system, from subrogated models from the cloud, fog, and edge. This necessary connection is held on the semantic web that causes, shares, and connects content capable of being interpreted by all the elements of the cyber-physical socio-technical system. The information is collected through IIoT technologies that allow data sharing between smart devices that configure smart and learning factories, thereby fostering collaborative affective environments [120].

By establishing a cognitive design appropriate to the required variety through adaptive interfaces, its adaptive reconfiguration can take place based on the task specification and the competency model of Operator 4.0 [100]. In order to make this possible, variety adapters employed as either amplifiers or reducers are introduced in accordance with the requirements, so that the regulatory part has at least the same variety as the regulated subsystem, and the SCMS can therefore achieve stability. The design and assembly of the system, within the fractalized context of the company [121], is made up of the tasks carried out by Operator 4.0, the equipment with which it interacts, and the associated information system.

#### **5. Case Study: DfHFinI4.0 in PERA 4.0**

Companies, and their associated manufacturing systems 4.0, are becoming increasingly complex and dynamic. In order to reduce this complexity, the managemen<sup>t</sup> of knowledge and operational information is needed. Business architectures, such as PERA, GERAM, and CIMOSA, have hitherto been used, while more recently, di fferent architectures have been proposed for Industry 4.0, such as Holonic, RAMI 4.0, IIRA, SME, and IVI [122–124], which correspond to the di fferent ways of implementing the informational requirements of primary activities and support the smart and sustainable value chain. Among the proposed architectures, PERA is formed by the ecosystem of business entities [125], as shown in Figure 8, in which this methodology, which constitutes its life cycle engineering, can be implemented for each of its entities.

From among its characteristics, it is worth mentioning its orientation to the life cycle of the entities that make up the architecture, and considering the interaction and interfaces between technology and the human factors within its methodology. Li and Williams [126] highlighted the importance of considering good design in the communication interfaces between the diagrams that constitute the PERA model, in order to guarantee the correct exchange of information and the integration of the company, both vertically and horizontally. These characteristics determine that PERA constitutes a model of reference architecture and methodology on which to integrate the DfHFinI4.0 framework proposed under a cyber-physical conception of the entities contained therein, which evolves towards PERA 4.0. This enables research questions to be answered in life cycle engineering 4.0 formulated by Romero et al. [11], which establishes the need for the reference architecture to focus on the human factor.

The basic methodology of a PERA entity is illustrated in Figure 9. In the following, manufacturing PERA entity 3 of the architecture is employed, on which it is illustrated how the DFHinI4.0 framework proposed should be integrated together with the associated tools, which enable the cyber-physical manufacturing systems to be conceived based on the human factor.

**Figure 8.** Entities and processes for the company in Purdue Enterprise Reference Architecture (PERA) 4.0 architecture.

**Figure 9.** Basic methodology of a PERA 4.0 entity.

As indicated, PERA [127], in its methodological aspect as life cycle engineering, establishes the various regions, phases, and layers, into which the entities that constitute the company can be decomposed throughout its life cycle, while taking into account that the production equipment, the human factor, and the information and control system are involved in each element. Three separate elements are established in the design and implementation for entity three of PERA that correspond to the manufacturing system:


Certain vertical lines of grea<sup>t</sup> significance for the integration of Operator 4.0 can be observed among these elements, within which the DfHFinI4.0 framework is integrated:


As depicted in Figure 9, the PERA methodology, in its initial proposal, contemplated the human factor, the technology, and the interfaces for their operation on industrial equipment and the information system, hitherto with no set of tools derived from conceptual frameworks of other areas of research that would allow the integration and empowerment of the human factor into digital transformation processes characterised by the connectivity and smartisation of technology.

In the proposal regarding PERA 4.0, which integrates the human factor into SCMS, as illustrated in Figure 10, the aforementioned elements are maintained. The elements and tools belonging to the DFHinI4.0 framework are incorporated into the design of socio-technical systems for the integration of the human factor, the industrial equipment, and the information and control system, thereby giving rise to two types of interfaces. In the same model, Operator 4.0 is configured as one more cyber-physical system, whose subrogate model obtained in the cloud will serve to adapt the interfaces and technology of the occupational environment to the operator positioned at the workstation.

**Figure 10.** Integrating DfHFinI4.0 in PERA 4.0 operations phase.

The evolution of the PERA model of manufacturing entity three (analogously for the rest of the PERA entities) towards a PERA 4.0 entity three as a cyber-physical system under the possibilities of KETs, determines that the hierarchical architecture of the control system (analogous to ISA-85 and ISA-99) associated with manufacturing devices has been modified by distributed intelligent cyber-physical systems architecture, with real-time connectivity for monitoring and control in the edge and cloud. For this reason, the engineering or re-engineering of PERA towards PERA 4.0 architecture not only implies the integration and empowerment of the human factor with the DFHinI4 framework proposed, but also indicates the transformation of the hierarchical architecture, initially proposed under the PERA methodology of entity three, into a distributed architecture of cyber-physical systems based on micro-services under the PERA 4.0 methodology.

Entity three of the PERA manufacturing systems has established certain levels, characteristics and interrelations [128], which include:


Figure 11 shows the incorporation of the DFHinI4.0 framework in a fractalized way, in the design and development of the various entities of the PERA 4.0 architecture for all the phases of the life cycle engineering that integrates the PERA methodology. This framework enables interactions between the human and technological factors to be modelled, and establishes their integration and development dynamically for each of the stages of the life cycle. This situation, together with the conception of the manufacturing entity as an intelligent and distributed cyber-physical system, gives rise to PERA 4.0 as a distributed system and life cycle engineering 4.0 methodology, which empowers the human factor, as shown.

**Figure 11.** Application of the DfHFinI4.0 framework to the design and development process of entity three of PERA in the manufacturing system.

For the managemen<sup>t</sup> of the reconfiguration of the technological occupational environment, accordance with the characteristics and competence of the Operators 4.0, who can be interacting with the system at any given moment, it is proposed that the operator, as a cyber-physical system, possess a cloud model of its operational singularity. This model refers to experience, knowledge, capabilities, competences, and other characteristics, which, as parameters of a subrogated model, allow the technology to be adapted, as represented in Figure 11. To this end, the operator model is sent from the cloud to the edge when necessary. The operators are sensed and assisted by the KETs, which transmit the data of the operators to the cloud to configure a more refined subrogate model, that in turn, is sent to the edge, which leads to the adaptation of the interface and technologies to Operator 4.0, thereby empowering this operator and enabling affective coupling. Big data techniques, learning machines, and deep learning will be employed in the preparation of the subrogate model.

Under the PERA 4.0 approach of manufacturing system entity three, the hierarchical levels of control established by PERA must be embedded in a distributed system of its cyber-physical entities, with intelligence and local connectivity in the edge, and global intelligence and connectivity in the cloud through IIoT. Figure 12 presents a schematic of the way in which the information system of the manufacturing system can evolve for its integration into any of the distributed architectures of Industry 4.0. In our proposal, we opted for holonic architecture [122], and for the modelling of the different holons, the Arrowhead methodology is proposed, both locally and globally [129]. This approach can be carried out by using blockchain technology, through the open-source container orchestration and choreography software tool called Kubernetes, and the creation of container images by using Docker [130].

**Figure 12.** Evolution of the PERA control architecture to the PERA 4.0 architecture.
