Metamodeling Approach to Sociotechnical Systems’ External Context Digital Twins Building: A Higher Education Case Study

Abstract

Sociotechnical systems (STSs) are generally assumed to be systems that incorporate humans and technology, strongly depending on a sustainable equilibrium between the following nondeterministic social context ingredients: social structures, roles, and rights, as well as the designers’ Holy Grail, the deterministic nature of the underlying technical system. The fact that the relevant social concepts are more mature than the supporting technologies qualifies the digital transformation of sociotechnical systems as a reengineering rather than an engineering endeavor. Preserving the social mission throughout the digital transformation process in varying social contexts is mandatory, making the digital twins (DT) methodology application a contemporary research hotspot. In this research, we combined continuous transformation STS theory principles, an observer-based system-of-sociotechnical-systems (SoSTS) architecture model, and digital twinning methods to address common STS context representation challenges. Additionally, based on model-driven systems engineering methodology and meta-object-facility principles, the research specifies the universal meta-concepts and meta-modeling templates, supporting the creation of arbitrary sociotechnical systems’ external context digital twins. Due to the inherent diversity, significantly influenced by geopolitical, economic, and cultural influencers, a higher education external context specialization illustrates the reusability potentials of the proposed universal meta-concepts. Substituting higher-education-related meta-concepts and meta-models with arbitrary domain-dependent specializations further fosters the proposed universal meta-concepts’ reusability.

1. Introduction

Sociotechnical systems (STSs) generally assume system-of-systems (SoS) incorporating humans and technology, with three essential characteristics: emergence (the system as a whole is not a simple sum of its components), non-determinism (the system does not always produce the same output as the response to the same input), and subjective behavior (the quality and usability assessments rely solely on the system stakeholders), demanding joint optimization equally respecting technical excellence and inherent diversity of human activities [1,2]. They strongly depend on a sustainable equilibrium between the nondeterministic social context ingredients: social structures, roles, and rights, and the designers’ Holy Grail, the deterministic nature of the underlying technical system.

Concerning maturity, the relevant social concepts are more mature than the supportive technologies, causing the digital transformation of sociotechnical systems to be a reengineering instead of an engineering endeavor. Following the maturation dimension of STS design principles, in [3], the authors point out a still dominant linear approach to organizational transitions, with clearly defined current and future states, rather than keeping up with a nondeterministic and continuously evolving nature of the future state. Concerning the STS theory, the extent to which an organizational system adheres to the knowledge management (KM) drivers determines its ability to mitigate the inherent non-linear social uncertainty while incorporating linear and deterministic technological innovations [4].

Consequently, engineering or reengineering of STSs represents a wicked problem usually perceived as difficult to define, to select the appropriate resolving methodology, to achieve a definitive successful solution, or even to assess the solution’s appropriateness without at least one previous unsuccessful solving attempt [5,6]. The relations between the STS’s functions and human-task allocation determine the importance of requirements elicitation, specification, and modeling activities [7,8]. On the other hand, the nature of information flow across the STS architecture demands mitigation of the inherent information architecture (IA) duality (social and technology-related) [9]. In keeping the STSs’ resilience at a maximum level, adaptive strategies in a varying context and possibly unpredictable human behavior open a challenging approach to the dynamic game theory application [10]. According to [11], agent-based modeling and simulation (ABMS) is a valuable tool for assessing complex STSs and is becoming increasingly advanced through integrating data-driven capabilities.

The inherent dichotomy of sociotechnical systems is augmented by segregating humans from social context and introducing the observer-based (human, social, and technology) approach, mitigating the complexity emerging from biological, organizational, and technology systems interactions through structural coupling that forms the human–social–technology (HST) recursive trichotomy [12]. Each observer possesses the evaluation criteria that ensure the preservation of its internal logic in potentially disruptive context (codes), rules and criteria guiding the application of codes while architecting and operating the observer system to mitigate the context-related influences (programs), and mechanisms that enable the observer-system to distinguish itself from the contemporary context (medium/form distinction). Assuming humans are psychical systems with inherent biological autopoiesis (producing and maintaining itself by creating its parts), in [12], the authors perceive the technology as an autopoietic system and leave the inherent nature of the social dimension intact.

The concept of transpoiesis in social systems theory emerges from the need to maintain strategic coherence within decentralized organizations. The STSs fulfill the proclaimed missions in the broader social and technology context, raising the importance of context management foundations. Independent of the status of the digital transformation process, preservation of a proclaimed social mission throughout the entire lifecycle of STSs is mandatory. Concerning contemporary trends in social systems theory, mechanisms fostering sustainability determine the dynamic balance between the concept of decentralized organizations and the necessity of strategic coherence inherent to the context in which the decentralized organizational systems evolve [13].

The previous elaboration on key STS concepts forms the foundations for one of our key research directions, the generic representation of STS leveraging the human and technology autopoiesis and social systems transpoiesis. At the same time, concerning the challenging view of STSs as complex living entities [14], through two inspiring principal concepts, simplexity (intrinsic complexity of a system artfully masked by an overlay of simplicity) and complixity (dynamics of simplicity emerging from the STS behavior when driven by environmental constraints and other exogenous factors), we adopt the essential need to creatively integrate the internal structural and behavioral complexity of STSs with their operational simplicity in STS’s context representation.

The synergy between operational context and the corresponding model of an arbitrary STS makes the digital twins (DT) methodology application in varying social contexts a contemporary research hotspot. The controversies related to the creation of organizational digital twins are elaborated on in [15]. They originate from the different aspects determining the organizational science perspectives, key technical challenges when moving from models to digital twins and vice versa, and the nature of the relation established between the real-world entity and its twin.

The process of building DT for a physical asset assumes the existence of a corresponding configurable and reusable digital template with an open set of tunable attributes, enabling its ad libitum convergence to the desired twin. An attempt to apply the same methodology while creating a digital twin of an organizational system (DTOS) fails because of the lack of a universal digital template available due to inherent uniqueness and the emergent nature of each organizational system [16]. Consequently, building a concrete DTOS demands repetitive redeveloping instead of simply purchasing one. DTOSs require the existence of the actual system models, whose initial creation represents a challenging endeavor. The initial quality of these models and their alignment with the structure and behavior of the real-world system is paramount in the agile approach to the digital transformation of complex STSs, composed of interrelated DTOSs, demanding the adoption of the philosophical foundations of digital twinning [17], together with the systematic application of principles and techniques of model-driven engineering [18], and model-based systems engineering [19].

The previous elaboration on key DT concepts in the STS context forms the additional research direction concerning the meta-concept-based foundation for STSDT representation.

This research article’s motivation is twofold. First, it aims to mitigate STSDT complexity through the recursive decomposition of STS architecture, enable transparent collaboration of component STSs, and foster the absolute autonomy of component internal operations in a varying STS context. Second, it aims to specify template meta-models of STS context digital twin (STSCDT) and contribute to mitigating challenges emerging from the DT model derivation process. Therefore, the sustainable representation of STSCDT demands a context-adaptable metamodel-based, knowledge-driven, and simulation-enabling framework.

The most challenging segment of digital transformation is related to domain-specific components that build the identity of the system’s mainstream, usually addressed as a work-process technology. Within an arbitrary STS, we claim that it is possible to reuse general patterns that direct architecture, functionality, information resources, communication, and behavior of the domain-independent supportive components.

The digital twinning approach significantly depends on the creative balance between structural and behavioral aspects of the supportive framework. Structural challenges mitigation usually assumes the existence of an open library of reusable virtual twin models. The behavioral dimension assumes either the existence of a configurable simulation framework capable of processing model instances or the development of dedicated software tools that interpret dataset instances to gain data-driven synchronization of real and virtual twins.

This research article focuses solely on the structural dimension of digital twinning. It specifies a novel approach to hyper-document-based models integration and formulates a reusable set of meta-concepts and derived meta-model templates. Nevertheless, the initial, observer-based, object model announces the behavioral dimension of digital twins building directions.

To frame the research methodology and the solution context, we specify a set of guidelines, expressed in the form of research hypotheses (RH), representing the research’s essential pillars justifying the proposed novelties.

RH1—The overall STSCDT representation needs to be organized into a hyper-document, integrating all of the abstraction layers of the proposed STSCDT representation (meta-model, model, and model-instance layers).

RH2—The proposed STSCDT representation must provide an open, data-driven, domain-independent, and generally reusable set of context meta-concepts (atomic building blocks) and meta-models (derived, coarse-grained meta-structures) supporting the representation of an arbitrary STS external context.

RH3—The reusability potential of RH2-based building blocks, at the meta-modeling abstraction level, must be verified by deriving domain-dependent STSCDT meta-concepts of the appropriately selected real-world STS throughout the specialization of domain-independent meta-concepts.

The domain-dependent context-related meta-concepts internal to the domain-specific STS, being stand-alone or derived from universal, generic, and external domain-specific context meta-concepts, are beyond the scope of this research and serve as the foundation for further specification of the domain-dependent micro-core.

With this in mind, the rest of the article is composed of five additional sections. Section 2, Materials and Methods, elaborates on the research methodology and the foundations of the stated hypotheses. Section 3, Results, introduces the main characteristics of the proposed meta-concepts and meta-models used in building a domain-independent representation of the STS context digital twin (STSCDT). Section 4, Domain-dependent verification of STSCDT reusability potentials, illustrates the reusability potentials of the proposed STSCDT meta-concepts of a higher education external context (HEEC) digital twin model specialization. Section 5, Discussion, cross-relates the analyzed references, justifies the appropriateness of the proposed meta-models and discusses the generic and specific limitations of the proposed solution. Section 6, Conclusions, contains the concluding remarks and future research directions.

2. Materials and Methods

As elaborated in the Section 1, the essential universal challenge in managing contemporary STSs is unavoidable digital transformation. In [20], the authors elaborate on the six key success factors of digital transformation arbitrary industry domains affecting the development of the bionic organization capable of standing up the continuous transformation activities.

In our previous research publication [21], the digital transformation hyper-framework model frames a digital transformation life cycle model’s impact on the successful outcome. Digital transformation, as a lean and evolutionary endeavor, needs disturbance minimization mechanisms that gradually transform an organizational system from the current level of digital maturity to the proposed, presumably higher, one.

The rest of the section elaborates on the essential research foundations and the significance of individual research hypothesis impact on the proposed meta-concepts and meta-models.

2.1. Research Foundations for the Hypothesis RH1

The inherent complexity of an STS reflects the overall complexity of the related digital twin. To mitigate it, we adopt the constant transformation principle from STS theoretical foundations and the SoS nature of the related digital twin. Within the continuous transformation theoretical principles, an STS is in a transition stage throughout the entire life cycle and assumes continuous and leaned evolutionary reengineering improvement of its structure and behavior. Additionally, with the digital twinning approach, lacking a physical asset challenges the traditional separation of a physical and a virtual twin due to the intangible nature of STS-related assets.

From the nomenclature aspect, we adopt an agile approach facilitating the isolation of manageable sprints addressing STSXDT, where X stands for a particular sprint related to X-SoS refinement of the STSDT. In this research article, X assumes a Context (C).

Concerning the stated research hypotheses RH1, we have specified a layered methodology framework with four transformation layers enabling STS digital twin model specification and data-driven synchronization between virtual and physical twin of an arbitrary STS component SoS (Figure 1).

From a broader context, a system of STS’s ontology model, presented in Figure 2, is adopted from our previously published article [22].

The presented ontology model defines the essential concepts related to an arbitrary system: a Mission (goal or a set of goals that generally justifies the rationality defining the existence of a specified system), the two-folded Organizational Structure specified by the composite structure of the organization (the internal components (OrganizationUnit) of system under consideration) forming the internal topology (InternalTopology) and the composite structure of related autonomous systems forming the ExternalColaboration (SystemOfSystems), the FunctionalStructure (system-related functions further specialized as InternalFunctions and ExternalServices), the information structure (the overall collection of InformationResources of arbitrary type and complexity forming the persistency (PersistentIR), dynamic (DynamicIR), and exchanging (ExchangeIR) layers), and the ControlStructure (internal (IntermalDriver) and external (ExternalDriver) drivers that pertain other components in performing allocated roles).

The observer-based object model framing the behavioral aspects of complex STS (Figure 3) augments the adopted ontology model (Figure 2) with the explicit specification of the Context SoS corresponding to all related individual STSs instead of the one under consideration.

With the explicit specification of the SystemConfiguration collection, it is possible to describe different configurations related to the observed STS under the same identity. The duality of the abstract System role, represented by the ObservableSystem and ObrserverSystem abstract specializations, enables the decoupling of an AutonomousSoS from the ContextSoS and hides their internals behind the Observable and the Observer interfaces (see Figure 3).

The historical time frame of STSs correlates with the parallel existence of different supporting mechanisms engaged in the work process technology. In [23], the author elaborates on the historical context and the advantages of integration in contemporary business and industry contexts. The holistic approach to the digital transformation of the transdisciplinary systems assumes interoperable complexity-reducing mechanisms [22] and digital twinning support with embedded verification and validation principles scalable and incorporated in each systemic level (macro, mezzo, and micro) [21], as well as demanding modeling and simulation supportive frameworks [23].

The digital transformation problem-solving assumes expert knowledge [24], experience, and wisdom glued and supported by the software tool. The supportive software tools do not solve the underlying problem. They support domain experts in solving the problem by augmenting their expertise throughout the routine segment automation and decision-making and evaluation processes.

We claim that the external contexts frame STSs in the addressed domain and level and depend on the main structuring and organizing principles that guide the domain-dependent architecture and services. The maintenance of model integrity and the full traceability over the continuous digital transformation of complex STS significantly depends on the hyper-documenting features of the modeling tools and the quality of model structuring justifying the significance of the RH1 hypothesis.

2.2. Research Foundations for the Hypothesis RH2—The Need for a Supportive Meta-Model with Universal and Generic Meta-Concepts

As stated in the Section 1, DTs for STSs represent a comprehensive framework that includes models and simulation services supporting the evolution of complex STSs throughout their entire lifecycle. They exhibit great potential in achieving common social goals, considering the continuous changing of the highly interconnected STS and the need for establishing an automatic change propagation that synchronizes the STSDTs at different STS levels [25].

In [20], the authors conclude that only by the virtue of time-based interaction between models and their dynamic refinements is it possible to mitigate the effective system changes and propose a research roadmap to realize DTs for STS ecosystems with the following milestones: Multilayered and Multi-view Modeling Framework, Capturing Data of Social Aspects and Model-to-Data Connections, Adaptive Simulation Framework for Prediction, Motivational Techniques to Drive Societal Change, and Digital Twin Evolution of Highly Connected Systems.

According to research hypothesis RH2, universal, domain-independent meta-concepts follow the abstraction level-raising activities and enable complexity mitigation of derived meta-concepts. Universal meta-concepts are stand-alone or exclusively derived from other universal meta-concepts and serve as a foundation for other domain-specific meta-concepts.

Interpreted in the research domain, the general hypothesis RH2 claims it is possible and highly desirable to specify and develop a generally usable STS context management meta-model based on the object management group (OMG) meta-object-facility (MOF) four-layer architecture, commonly used in organizing and structuring models in object-oriented systems [26,27]. Additionally, we claim that such a meta-model is a foundation for automatic software service generation based on model-to-code transformation mechanisms [28,29].

2.3. Methodology Aspects of an STS’s External Context Digital Twin (RH1- and RH2-Related)

The digital-twinning-based approach to digital transformation of complex sociotechnical systems faces two antagonistic foundation principles: raising the abstraction level to gain the highest possible context independency (non-contextualization), or facilitating the highest possible reusability level through transparent context-dependent specializations.

Non-context foundation is based on the rational approach to the representation of complex organizational systems, has a dominantly mechanistic nature, and favors homogenization principles in gaining global interoperability [22]. On the other hand, the importance of context-dependency foundations highlights the fact that the tacit aspects of the different organizational systems belonging to the same or different domains share the same external context affecting the internal structure and behavior [30,31,32] and consequently favor heterogeneity in interoperability mitigation [22]. The absolute heterogeneity is the main methodology motivation directing this research methodology pillars.

An individual STS operates in a dynamic context that encapsulates external drivers (local or global) affecting the STS’s structure and behavior; regulating legal, technical, and technology influencers; and following assessing criteria, norms, and metrics. STS context is an SoS recursively compliant with the domain-specific STS meta-model. It is essential in fostering interoperability and establishing the trustworthiness of a quality measuring and macro-grained assessment concerning the inherent characteristics of individual STSs.

The related foundation’s repertoire is enormous and far beyond a rational presentation in a single article. Nevertheless, it is necessary to elaborate on the most significant sources of unavoidable hotspots influencing STS’s contemporary context.

The essential components of the STS’s external context are globally determined by a set of general classifiers, consisting, but not limited to, the territory (geographical, geo-topology), political and legal systems (ministry for education, national education council, and parliament assembly), language, religion, normative systems (national accreditation and licensing), occupational systems (national occupation nomenclatures), professional associations, non-governmental sector, and industry and services providers, that form the context layer’s topology.

Each general classifier may establish an extendible, hyperdimensional, and quantifiable set of composite sub-dimensions. Therefore, the elements of the STS context form a multilayered hyperdimensional classification system that needs to be specified by the meta-model, model, and instance abstractions.

2.4. Methodology Aspects of a Higher Education External Context Digital Twin (RH3-Related)

Following a particular approach to domain complexity mitigation, hypothesis RH3 suggests the isolation of reusable domain-dependent meta-concepts, similar to the universal meta-concepts specification, but applied to the selected domain (in this research approach, higher education). The generic domain-related meta-concepts are either stand-alone in the specialized domain or derived from universal meta-concepts.

The education system is highly mature with an astonishing theoretical, methodology, and practical foundation, materialized in diverse organizational forms, either institutional or non-institutional [33]. The overall agreed body of knowledge, accreditation principles, standard curriculum templates, multi-level education programs, approved credentials, and professional licensing determine the industrial occupations and professional development [34].

This research addresses the institutionalized forms of higher education, teaching, training [35], and learning [36] activities sublimated in an arbitrary higher education institution (HEI). The HEI is a complex enterprise SoS with an embedded enterprise information system (EIS) whose digitalization status determines the overall organizational digital maturity level [37]. The information system is an intrinsic component of any organizational system with the mission to supply (feed) the rest of the system’s components with the information resources needed to fulfill the individual or emerging goal/goals. In general, the existence of a digitalized form of an information system is not obligatory. Digital transformation in contemporary higher education is firmly associated with the specification, development, and operational support of HEI’s enterprise information system (EIS).

In an institutional sense, teaching, assessing, and learning reside in the higher education institution’s internal context frame and are the consequence of time and modal characteristics respecting the freedom of autonomously associated actors. On the other hand, HEIs exist in a broader SoS external context governing the overall structure and behavior.

We claim that the external contexts frame education in the addressed domain and level and depend on the main structuring and organizing principles that guide the education process. This is the main reason for selecting the education context management component as a representative among the large set of higher education institutional enterprise information system building blocks.

The individual, country/regional-level-based contexts of an HEI habitat represent a reach referent source of valuable information related to the specificity of geo-political aspects of higher education. Generally, it is possible to isolate two main groups of contexts: one that is decentralized and highly diverse and legally independent from the federal government higher education system (does not enforce centralized roles, standards, or metrics, where the accreditation is not mandatory), as is the case with the US, and the more often encountered counterparts relying on centrally directed (hybrid or centralized) approaches worldwide.

Based on the analyzed sources and our experience with enterprise information systems design in a higher education domain, we believe it is a representative candidate for domain-specific illustration of STSCDT meta-concepts and meta-models’ reusability potentials.

3. Results

The elaboration supporting the rationalities related to the significance evaluation of the individual hypothesis (Section 2, Materials and Methods) justifies the meta-model-based approach to STSCDT representation. According to hypothesis RH1, the research results section follows the nested meta-modeling approach, starting with the hyper-document envelope formulation (Section 3.1), followed by detailed specification of domain-independent meta-concepts and meta-models (Section 3.2).

3.1. STSCDT Representation (RH1)—A Virtual Twin Hyper-Document and Project Package Tree

For the specification and development of a virtual twin hyper-document, we used the SAP Sybase Power Designer Data Architect Version 16.1 modeling tool. Because this research article focuses solely on the meta-modeling MOF level, we specified individual meta-models with the ER-Merise conceptual modeling formalism augmenting standard ER modeling with the inheritance extension.

The current structure of the proposed virtual twin hyper-document (Figure 4) follows the multilayered research methodology framework specified by Figure 1.

The inheritance extension enables the reduction of meta-model diagram density as a highly desirable feature aiding the meta-modeling abstraction layer comprehension and readability. At the meta-modeling level, it is composed of the following packages representing hyper-linked meta-model containers:

• 01_GenericContextManagementPackage—aligns with the research hypothesis RH2 and contains four generic packages and associated meta-models specifying the collection of reusable meta-concepts, commonly referenced in other packages (refer to Section 3.2 for detailed elaboration);
• 02_EducationDomainGenericPackage—aligned with the research hypothesis RH3, containing three education-domain-related generic packages and associated meta-models specifying the collection of reusable meta-concepts specific for the education domain and commonly referenced in other packages (refer to Section 4.1 for detailed elaboration);
• 03_ExternalEducationContextPackage, aligned with the research hypothesis RH3, containing the additional five packages and associated meta-models specifying the collection of external context meta-concepts (refer to Section 4.2 for detailed elaboration).

3.2. STSCDT Representation—Generic Context Management Packages and Meta-Concepts (RH2)

Based on the concept inheritance property, the 01_GenericContextManagementPackage meta-concepts are reusable and support model-density balancing over the inheritance mechanism.

Due to the meta-model fidelity feature, the package is further split into four generic sub-packages (See Figure 4):

• 01_01_GenericConceptPackage, containing GenericConceptMetaModel, specifying a reusable StructuredConcept meta-concept (Section 3.2.1);
• 01_02_GenericAssessmentAndRakingPackage, containing GenericAssessmentAndRankingMetaModel, specifying reusable accessing and ranking meta-concepts (Section 3.2.2);
• 01_03_GenericTechnologyPackage, containing GenericTechnologyMetaModel, specifying a reusableGenericTechnologyConcept meta-concept (Section 3.2.3);
• 01_04_GenericPerformerPackage, containing GenericPerformerMetaModel, specifying a reusable GenericPerformer meta-concept (Section 3.2.4).

In all meta-models, attribute-related marker <pi> designates the primary identifier, while <M> designates the mandatory existence.

3.2.1. Generic Concept Package and Meta-Concepts

The GeneriConceptPackage contains the Generic Concept Meta-Model specifying the generic meta-concepts used in the specification and modeling of other meta-concepts (Figure 5).

AttributeConcept is a universal meta-concept enabling the specification and meta-modeling of the semantic dimension of an arbitrary meta-concept. It is potentially composite over the Cartesian product StructuredAttribute meta-concept supporting the specification and meta-modeling of structured attributes over ContainerAttrib and ContainedAttrib strong dependency relations. The AttributeConcept meta-concept may participate in the identification (IdentifyByAttrib strong dependency), extension (ExtByAttrib weak dependency), and classification (ScExtencedAttribures with Classifier? set true) of a StructuredConcept and all its specializations.

The StructuredConcept is an identity-opened, semantically opened, classification-opened, and connectivity-enabling generic meta-concept, enabling specification and modeling of other specialized meta-concepts. The inheritance mechanism aids the simplicity of derived meta-concepts and related meta-models. Any meta-concept that inherits a StructuredConcept inherits its structure and behavior. Specialized meta-concepts expand through meta-model-to-model transformation only at a model level.

The identity-opened feature relies on the AttributeConcept meta-concept and enables data-driven clustering and identification of a StructuredConcept meta-concept through a strong dependency relation (IdentifyByAttrib). A strong dependency relation enables the formation of the composite concept’s primary identifier and the cascading proliferation of the type-identifier to the arbitrary model level. The cascading composite primary identifier proliferation enables meta-model to platform-dependent model-instance transformations with either SQL or NoSQL data-repository types and fosters the elimination of explicit joins in business logic specifications.

The semantically opened feature relies on the ScExtendeAttributes meta-concept and enables data-driven configurations of arbitrary complex attribute collections extending the StructuredConcept over strong dependency ExtAttributes relation. The ScExtendedAttributes meta-concept is a container for data or meta-data attributes used to specify the semantics of an arbitrary specialization.

The classification-opened feature relies on the ScExtendeAttributes meta-concept resident attribute marked as Classifier? and enables data-driven configurations of arbitrary complex classification mechanisms extending the StructuredConcept over strong dependency ExtAttributes relation and serving for indexing mechanism creation.

The connectivity-enabling feature assumes the ability to form different tracing paths and support the configuration building of an arbitrary StructuredConcept. It supports the following:

• successor–predecessor connectivity (ScSuccessor-ScPredecessor relations among StructuredConcepts);
• mash-supportive connectivity over Cartesian product MashRelated meta-concept enabling specification and modeling of an arbitrary mash structure over SCSource and SCDestination strong dependency relations with PayBackAttrib-defined cost or benefit value when traversing from source (SCSource) to destination (SCDestination) StructuredConcept;
• hierarchy connectivity (HierarchyRelated multilevel composition of StructuredConcept, enabling the formation of arbitrary tree topologies following the hierarchy-related leveling of a StructuredConcept by InHierarchy-related StructureConcepts).

In any connectivity, it is essential to guarantee referential integrity by implicit, structure-based, or explicit, behavior-based mechanisms.

3.2.2. Generic Assessment and Ranking Package and Meta-Concepts

01_02_GenericAssessmentAndRankingPackage contains the related Generic Assessment and Ranking Meta-Model and enables the assessment and ranking of arbitrary meta-concepts that inherit either AssessedConcept or RankedConcept meta-concepts (Figure 6).

The meta-model specifies seven universal meta-concepts:

• AssessedConcept—represents a core assessment and ranking meta-concept enabling specification and meta-modeling of the assessment mechanisms for an assessor, representing a GenericPerformer (AssessOrRanks strong dependency relation—Figure 6), used assessment technology, and serves as a placeholder for derived assessments or rankings. Consequently, the AssessedConcept configures three additional meta-concepts: the AssessmentMechanism meta-concept over HasMechanisms strong dependency relation, the AssessmentTechnology meta-concept over AppliesTechnology strong dependency relation, and the Assessments meta-concept over HasAssessments strong dependency relation. Additionally, it serves as a foundation for derived rankings.
• AssessmentMechanism meta concept is a specialization of StructuredConcept (AssessMechSpec specialization relation) enabling specification and meta-modeling of arbitrary assessment mechanisms that may be used by the GenericPerformer through related AssessedConcept to assess or rank arbitrary meta-concepts that inherit either the AssessedConcept or RankedConcept meta-concepts. It configures the MechanismIndicators meta-concept over the DefinedIndicators strong dependency relation and relates with the applied AssessmentTechnologies (many-to-many relation MechTech).
• MechanismIndicators meta-concept is a specialization of StructuredConcept (MechIndSpec specialization relation) enabling specification and meta-modeling of indicators related to the AssessmentMechanism meta-concept. It configures the IndicatorParameters meta-concept over the IndicParams strong dependency relation.
• IndicatorParameters meta-concept is a specialization of StructuredConcept (IndParamSpec specialization relation), enabling specification and meta-modeling of parameters related to the MechanismIndicator meta-concept.
• AssessmentTechnology meta-concept is a specialization of the GenericTechnologyConcept (AssessTechSpec specialization relation). It supports meta-modeling of the assessment technology versions related to the AssessedConcept over AppliesTechnology strong dependency relation and relates with the applied AssessmentMechanism meta-concept (many-to-many relation MechTech).
• Assessments meta-concept is a specialization of the AssessmentMechanism meta-concept (AssessmentsSpec specialization relation) and enables the derivation of the assessments related to the AssessmentConcept according to the referenced AssessmentMechanism.
• RankedConcept is a specialization of AssessmentConcept (RankedSpec specialization relation), enabling specification, meta-modeling, and derivation of ranking meta-concepts according to the ranking meta-model in compliance with the inherited AssessmentMechanism.

3.2.3. Generic Technology Package and Meta-Concepts

The 01_02_GenericTechnologyPackage contains the GenericTechnologyMetaModel specifying the GenericTechnologyConcept reusable meta-concept used in the specification and modeling of other meta-concepts. Figure 7 represents the meta-model of a GenericTechnologyConcept meta-concept.

The GenericTechnologyConcept meta-concept represents the specialization of three generic meta-concepts:

• StructuredConcept (GenTechStructSpec specialization relation) inheritance defines the GenericTechnologyConcept as an identity-opened, semantically opened, classification-opened, and connectivity enabling meta-concept.
• AssessedConcept (GenTechAssessedSpec specialization relation) inheritance defines GenericTechnologyConcept as an assessable meta-concept according to the generic assessment and ranking meta-model (Section 3.2.3).
• RankedConcept (GenTechRankSpec specialization relation) inheritance defines GenericTechnologyConcept as a rank-able meta-concept according to the generic assessment and ranking meta-model (Section 3.2.3).

Additionally, the GenericTechnology meta-concept configures TechnologyDocumentation (specialization TechDocSpec of the GenericDocumentation meta concept specified in Section 3.2.4) over DocumentedTechnology strong dependency relation.

3.2.4. Generic Performer Package and Meta-Concepts

The 01_04_GenericPerformerPackage contains the GenericPerformerMetaModel, which specifies the GenericPerformer reusable meta-concept and enables the specification and modeling of other specialized meta-concepts. Figure 8 represents the meta-model of GenericPerformer-related meta-concepts.

GenericPerformer meta-concept is a course-grained specialization (GenPerfSpec) of a RankedConcept (Section 3.2.3) representing either a real or virtual meta-concept (attribute Virtual?). A RankedConceptGenericPerformer enables ranking according to the assessment mechanism specified by an arbitrary GenericPerformer acting as a ranker. It performs documented activities (PerformerDocuments strong dependency relation with the GenericDocumentation meta-concept) in a generic language and alphabet environment (PerformerLangAlph many-to-many relation with LanguageAlphabet meta concept) and generic process configuration context (PerformerProcess strong dependency relation with the GenericProcess meta-concept).

It uses generic resources (PerformerResources strong dependency relation with the GenericResources meta-concept) and the related technologies (RelaysOn strong dependency relation with PerformercTechnology meta-concept) to deliver tangible or intangible generic documented products (PerformerProduct strong dependency relation with the GenericProduct meta-concept).

GenericPerformer is further recursively specialized by two composite configuration meta-concepts:

• InternalConfiguration is a specialization of the GenericPerformer meta-concept (IntConfigSpec specialization relation) and configured over the InternalPerformers strong dependency relation.
• ExternalConfiguration is a specialization of the GenericPerformer meta-concept (ExtConfigSpec specialization relation) and configured over the ExternalPerformers strong dependency relation.

The GenericPerformer configures four additional specializations of an AssessedConcept (Section 3.2.3) (enables the assessment of its specializations over the assessment mechanism specified by a GenericPerformer acting as the assessor) and one additional specialization of the GenericTechnologyConcept meta-concept (RelaysOn specialization relation):

• GenericProcess is a specialization of AssessedConcept (GenProcSpec relation) configured by the PerformerProcess strong dependency relation. It enables specification and meta-modeling of a GenericPerformer’s process dimension.
• GenericProduct is a specialization of AssessedConcept (GenProdSpec relation) configured by the PerformerProduct strong dependency relation. It enables meta-modeling of a GenericPerformer’s product dimension.
• GenericResources is either a specialization of AssessedConcept (GenRessAssessSpec relation) or RankedConcept (GenResRankSpec) depending on the generic resource nature. It is configured by the PerformerResources strong dependency relation enabling specification and meta-modeling of a GenericPerformer’s resources dimension.
• GenericDocumentation is a specialization of AssessedConcept (GenDocSpec relation) configured by the PerformerDocuments strong dependency relation) and enables meta-modeling of the GenericPerformer’s language and alphabet clustered documentation (DocLangAlph strong dependency relation).
• PerformersTechnology is a specialization of GenericTechnologyConcept (PerfTechSpec relation) configured by the RelaysOn strong dependency relation. It enables the meta-modeling of a GenericPerformer’s technology dimension.

The GenericPerformer meta-concept specifies the following cross-configuring many-to-many relations:

• ProcessRelated enables direct coupling between InternalConfiguration and the GenericProcess meta-concepts (who internally runs the process);
• ExternalProducts enables direct coupling between ExternalConfiguration and the GenericProduct meta-concepts (who uses the products either in a supplier or consumer role);
• ProcessProduct enables direct coupling between the GenericProcess and the GenericProduct meta-concepts (which process delivers/uses which product);
• ProcessResources enables direct coupling between the GenericProcess and the GenericResources meta-concepts (how the process consumes resources);
• ProductResources enables direct coupling between the GenericProduct and the GenericResources meta-concepts (how the product consumes resources);
• ProductTechnology enables direct coupling between the GenericProduct and the PerformersTechnology meta-concepts (which technology supports the generic product);
• ProcessTechnology enables direct coupling between the GenericProcess and the PerformersTechnology meta-concepts (which technology supports the generic process);
• ProcesDoc enables direct coupling between the GenericProcess and the GenericDocumentation meta-concepts (how is the process documented);
• ProductDoc enables direct coupling between the GenericProduct and the GenericDocumentation meta-concepts (how is the product documented).

While the previous four packages are globally generic and universally reusable in an arbitrary domain, the following paragraphs elaborate on a Higher Education domain-specific meta-models and reusable meta-concepts derived from the generic meta-concepts in the particular meta-modeling context.

4. Domain-Specific Reusability Verification of STSCDT Meta-Concepts

The proposed universal meta-concepts and meta-models reusability potentials are verified through the derivation of the Higher Education External Context Digital Twin Model (HEECDTM). The Higher Education Sociotechnical System (HESTS) was selected due to its extensive impact on the overall scientific, technical, and operational development of contemporary STSs. The specific descriptions of related meta-concepts and their roles, inheritance mechanisms, configuring features, and relations appear in a tabular form representation. While analyzing the descriptions systematized in tables, it is necessary to relate them with previously elaborated generic meta-concepts and the generic

Table 1. The generic table specification.

Meta-Concept	Role	Inherits	Configures	Related with
Contains the name of the specified meta-concept.	Describes the generic role of the specified meta-concept in the meta-model.	Lists all of the meta-concepts inherited by the specified meta-concept. If multiple inheritance is specified, the resulting structure and behavior represent the union of the inherited concepts.	This column specifies all strongly related meta-concepts with the proliferated meta-concept’s ID down the tree hierarchy.	This column specifies all weakly related meta-concepts forming the referential proliferation of the specified meta-concept’s ID as a non-mandatory or mandatory foreign key.

, describing the table’s structure and the semantics associated with the individual columns.

4.1. Domain-Specific Generic Package and Meta-Concepts (Support for RH3)

• 02_EducationDomainGenericPackage addresses higher education domain reusable course-grained meta-concepts (Figure 4—Tree View).
• Due to the inherent complexity, the package splits into three sub-packages:
• 02_01_GenericEducationDomainPackage containing GenericEducationDomainMetaModel specifying the collection of domain-related reusable meta-concepts referenced in other packages (Section 4.1.1);
• 02_02_GenericCurriculumPackage containing GenericCurriculumMetaModel specifying the collection of reusable curriculum-related meta-concepts referenced in other packages (Section 4.1.2);
• 02_03_GenericAccreditationStandardsAndNormsPackage containing GenericAccreditationStandardsAndNormsMetaModel specifying the collection of accreditation-related reusable meta-concepts referenced in other packages (Section 4.1.3).

4.1.1. Generic Education Domain Package and Meta-Concepts

02_01_GenericEducationDomainPackage addresses higher-education-specific reusable meta-concepts. It contains the Generic Education Domain Meta-Model (Figure 9), specifying a hyper-structured, course-grained meta-concept containing the generic higher education domain meta-concepts. It is cross-related with the GenericCurriculumMetaModel (Section 4.1.2) and the corresponding generic meta-concepts.

EducationDomain meta-concept supports two interrelated meta-modeling branches:

• domain-directed (EduDomDesct configuration of the EducationDomainQualification meta-concept);
• domain body-of-knowledge-directed (HasABoK configuration of the BodyOfKnowledge meta-concept).

Both branches enable cross-coupling over QualLevEduDomQual many-to-many relation between the EdicationDomainQualification and EducationTypeAndLevelRelation meta-concepts (See Figure 9).

4.1.2. Generic Curriculum Package and Meta-Concepts

02_02_GenericCuriculumPackage addresses reusable curriculum-related meta-concepts. It contains the Generic Curriculum Meta-Model (Figure 10) and specifies a hyper-structured, course-grained meta-concept containing the generic curriculum and generic course-related meta-concepts.

4.1.3. Generic Accreditation Standards and Norms Package and Meta-Model

02_03_GenericAccreditationStandardsAndNormsPackage contains the related Generic Accreditation Standards And Norms Meta-Model enabling the specification and modeling of accreditation documents, standards, rules, and norms framing the contemporary accreditation environment, if applicable (Figure 11).

It is cross-related with the ExternalEducationContextPackage reusing the LegalSubject meta-concept used for the AccreditationAgency meta-concept specialization.

4.2. External Higher Education Context Management Packages and Meta-Models (Support for the Hypothesis RH3)

According to the HEI Context Profile responsibilities, the corresponding Meta-Model splits into five packages (Figure 4—Tree View):

• 03_01_GeoPoliticalContextPackage (containing GeoPoliticalContextMetaModel specifying the external context meta-concept) (Section 4.2.1);
• 03_02_ContactPackege (containing ContactMetaModel specifying course-grained Contact meta-concepts) (Section 4.2.2).
• 03_03_SocioEconomyContextPackage (containing SocioEconomyContextMetaModel specifying course-grained LegalSubject meta-concept) (Section 4.2.3);
• 03_04_CulturalContextPackage (containing CulturalContextMetaModel specifying course-grained cultural context-related meta-concepts) (Section 4.2.4) and
• 03_05_LegislationContextPackage (containing LegislationContextMetaModel specifying course-grained legislation meta-concepts (Section 4.2.5).

4.2.1. Geopolitical Context Meta-Model

03_01_GeoPoliticalContextPackage contains a related GeoPoliticalContextMetaModel, enabling the specification and modeling of context-related concepts regarding geo-topology and political aspects (Figure 12).

4.2.2. Contact Meta-Model

03_02_ContactPackage contains a generic context-related ContactMetaModel supporting modeling and specification of contact and locating information for an arbitrary addressable concept (Figure 13).

4.2.3. SocioEconomyContextMetaModel

03_03_SocioEconomyContextPackage contains a related SocioEconomyContextMetaModel enabling the specification and modeling of context-related legal subjects forming the socio-economy context (Figure 14).

4.2.4. CulturalContextMetaModel

03_04_CulturalContextPackage contains a related CulturalContextMetaModel enabling the specification and modeling of cultural aspects of context-related concepts (Figure 15).

4.2.5. LegislationContextMetaModel

03_05_LegislationContextPackage contains a related LegislationContextMetaModel enabling the specification and modeling of context-related legislation (Figure 16).

We believe that the formulation of Higher Education External Context domain-specific meta-models verifies the reusability potentials of domain-independent meta-concepts and meta-models at the focused meta-modeling level.

5. Discussion

The process of building DT for a physical asset assumes the existence of a corresponding configurable and reusable digital template with a rich collection of tunable attributes, enabling its ad libitum convergence to the desired twin. An attempt to apply the same methodology while creating a digital twin of an organizational system (DTOS) fails because of the unavailability of a universal digital templates that overcome the underlining uniqueness and the emergent nature of individual organizational systems.

This research article’s motivation emerges from two main foundations:

• results of the comparative analysis of the contemporary research publications addressing digital twinning templates existing in general sociotechnical systems (STSs) and the particular higher education context representations (elaborated on in the Section 1 and Section 2);
• practical experiences gained throughout model-based enterprise information systems engineering and reengineering in different domains (business, banking, healthcare, government, education, and national accreditation agency), partially elaborated upon in our previous publication [38] (pp. 25–27).

To mitigate the inherent STSDT complexity, this research article proposes one general (RH1), one universal (RH2), and one domain-specific (RH3) research hypothesis, establishing the framework of the proposed meta-model hyper-document supporting the STS context digital twin representation.

By abstracting the practical meta-modeling examples in different domains and contemporary pinpointed research challenges concerning the contemporary STS theoretical foundation and digital twinning methodology application in digital transformation of complex STSs, we first established a multilayer research methodology framework compliant with the MOF architecture model (Figure 1). Secondly, the proposed observer-based object model, extending the adopted ontology model (Figure 2), frames the behavioral aspects of complex sociotechnical system-of-systems (STSoS) with the explicit conceptual segregation of the context STSoS, corresponding to all related individual STSs excluding one under the consideration, as well as the particular addressed STSoS (Figure 3).

The primary mission of this research article follows the dominating claims that the sociotechnical system’s context represents a virtual-twin hyper-document large data object enabling the verification and validation of different domain-specific processes in life-long assessment and data-driven constant upgrading while fulfilling the stated mission, wherein we have proposed multilayered meta-concept-based templates with high reusable potentials to fill the apparent gap in digital twinning of STSs. The detailed elaboration of the proposed hyper-document content appears in the Section 3 of this article.

To cope with the complexity of the proposed meta-models, in compliance with the research hypothesis RH2, we introduced five novel universal, domain-independent meta-concepts encapsulating the core features used to build other meta-concepts over the inheritance mechanism (StructuredConcept, GenericTechnologyConcept, AssessedConcept, RankedConcept, and GenericPerformer) (see Section 3.2).

Consequently, the secondary mission of this research article is to pinpoint higher education external context influencers throughout a comprehensive novel higher education external context meta-models. We focused on higher education due to its domain-related diversity. On the other hand, it is apparent that a complex higher education system-of-systems mission highly outweighs a simple knowledge transfer. It plays a crucial role in maintaining social order, transmitting values, promoting social mobility, and fostering societal change [39]. The higher education system symbiotically interacts with the higher education external context and evolves within it. By comprehending the role of external context in this symbiosis, it is possible to perceive its incubation capacity to profile future higher education. Consequently, with a better understanding of a higher education within the external context, it is possible to comprehend its ability and potential to shape the future of the entire higher education ecosystem.

According to the research hypotheses RH3, we propose four education-domain-related generic course-grained meta-concepts: EducationDomain, GenericCurriculum, GenericCourse, and GenericAccreditationStandardsAndNorms, representing domain-dependent specializations of SDSC digital twin representations, followed by the external context layer meta-concepts specifying five packed meta-models addressing GeopoliticalContextMetaModel, SocioEconomyMetaModel, CulturalContextMetaModel, LegislationContextMetaModel, and ContactMetaModel. The specified meta-concepts, defined in the previously discussed meta-models, enable meta-modeling of the arbitrary external context influencers affecting the higher education institution as a legislator of the internal education context.

The proposed meta-models and specified meta-concepts enable the derivation of model, model-instances, and digital twin dataset layers (Figure 1), which are beyond the scope of this research article and represent the most challenging steps in future research endeavors due to the potential application of generative artificial intelligence algorithms for the automatic transformations of meta-models to models; meta-models to model instances; models to model instances [38,40]; and fabrication of the corresponding digital twin repository, hereby recognized as a STSCDT’s virtual twin.

Also, one of the additional challenges concerning the observer-based synchronization of STSCDT’s virtual twin and its physical counterpart, hereby recognized as an immutable dataset instance, remains a promising future research direction.

The comparative analysis as the verification and validation mechanism is challenging due to the lack of previously published research articles covering the higher education external context representation over the proposed meta-concepts. The rest of the Section 5 cross-compares the selected research articles’ over the proposed meta-modeling framework and ranks their relative coverage of 17 domain-related features.

Table 2. The related reference assessment of the proposed meta-concepts.

Comparison over Features		Assessed Research Articles (Figure 17)
	Ranked from 0 (Does Not Cover) to 5 (Exactly Covers) in	R1	R2	R3	R4	R5	R6	R7	R8	This Research
		References
Feature No. (Figure 17)	Feature	[41]	[42]	[43]	[44]	[45]	[46]	[47]	[48]	-
1	Teaching-oriented	0	0	0	0	5	0	3	0	0
2	Learning-oriented	0	0	0	5	5	5	5	0	0
3	StructuringConcept	0	0	0	0	0	0	0	1	5
4	GenericTechnologyConcept	0	0	0	0	0	5	0	1	5
5	GenericPerformer	0	0	0	0	0	0	0	3	5
6	Assessment	5	0	5	0	5	0	3	1	5
7	Ranking	5	0	0	0	0	0	3	0	5
8	EducationDomain	2	5	3	0	2	2	0	0	5
9	Curriculum	1	5	5	0	5	0	0	0	5
10	Course	1	5	5	0	5	0	3	0	5
11	Accreditation	0	0	0	0	0	0	0	0	5
12	Geo-political and economy	0	3	0	0	0	3	0	0	5
13	CulturalContext	0	0	0	0	0	2	0	0	5
14	LegalSubject	0	0	0	0	0	0	0	5	5
15	Legislation	0	3	0	0	0	0	0	2	5
16	Contact	0	0	0	0	0	0	0	0	5
17	Internal HEI	4	3	3	0	2	2	2	0	1

contains the related work rankings with cross-comparative visualization of ranked research represented by Figure 17.

The proposed meta-concepts represent the foundations of multi-dimensional hyper-document refinements and are tailorable regarding the significance of individual context-related meta-models for the particular modeled higher education external context.

Limitations of the Proposed Meta-Modeling Approach

The inherent limitations of the proposed approach, specified meta-concepts, and derived meta-models are threefold: the general limitations that emerge from adopting the meta-modeling abstraction layer, specific restrictions concerning the context-dependent digital twining of complex sociotechnical systems, and domain-dependent limitations of the proposed templates.

Firstly, among the generally addressed meta-modeling and meta-concept foundation obstacles, the most common are more profound domain expertise affecting stakeholders’ comprehensibility; the risk of unnecessarily high generalization affecting applicability; and the inherent limitations concerning change-tracking, scalability, and effective handling of a large number of parameters. The significant mitigating approaches, elaborated on in the limited set of analyzed related references, address higher computational demands [49], representational complexity of highly parameterized systems [50], model consistency management [51], and data-driven modeling methods [52].

The flexibility and data-driven configurability of the proposed StructuredConcept and the proposed hyper-document integration mitigate the mentioned obstacles, excluding domain expertise, meta-modeling, and model transformation services skills [53], which are highly dependent on the significant lack of disciplined, systematic, and methodical dimensions in contemporary engineering education [34], as well as the integrated digital twinning software tools supporting all of the discussed challenges.

Secondly and thirdly, although the proposed meta-concepts and meta-models have high reusability potential, we claim that a crucial limitation is a potentially demanding expertise in the meta-modeling methodology as a prerequisite for effective and efficient use of the proposed meta-concepts and meta-models. Although we believe that this is a challenge, not a limitation, the concluding claims of our previous publication [22] (“…the expertise is not free of prejudices and reluctance to the approaches that go beyond the current comprehension.”) suggest precaution.

6. Conclusions

Contemporary digital twinning of complex sociotechnical systems faces two main obstacles: universal templates of different sociotechnical systems do not exist, and a mutually agreed perception is lacking.

We believe that the proposed meta-models with an open set of universal constructs, domain-independent and domain-dependent derived course-grained meta-concept specification, significantly contribute to the sociotechnical digital twin representation related to the context representation endeavors. They serve as a foundation in mitigating the particular challenges existing in arbitrary STS external context management project phases from vision, requirements specification, repository design, and service specification and development through a model-based software development methodology.

This research result verification focuses on a higher education external context digital twin (HEECDT) motivated by higher education diversity potentials, varying fidelity levels, lower inertia to novel technology adoption and incorporation into the education core, and significantly higher impacts on the HEECDT representation. We emphasize that it is essential to strictly differentiate between incorporating novel technology achievements into the educational process as a supportive means and educating students on the core foundation of these technologies and their utilization in real-life problem-solving.

The industrial context favors novel technologies due to the higher entrepreneurship potentials of successful endeavors despite the inherent risks that usually have short-term effects on the overall organizational system. Immature technologies in mission-critical projects are rarely welcomed or justified. The question is, is higher education a mission-critical endeavor or not? If the answer is yes, and we think it is, then the immanent inertia is understandable and justifiable in supporting the longevity of stated aims.

The proposed meta-models and meta-concepts open several challenging potential future research directions.

The first challenging future research direction is transforming the specified meta-models into models and model instances and creating the foundations for automatic meta-model-to-model, meta-model-to-model-instance, and model-to-model-instance transformations. It is the essential phase in implementing generic AI methods and parametric, data-driven, context-dependent transformation roles in mitigating the most serious limitation of the operational usability of the proposed meta-modeling-based templates.

The second challenging future research direction wraps the proposed three layers of the MOF hyper-document by the meta-meta-model, a domain-specific language, thereby completing the full spectrum of OMG MOF specification abstractions. The DSL development would further raise the abstraction level and enable a language-based approach to the STSCDT building.

The third challenging future research direction is the specification and development of a software tool that, based on the specified virtual twin model, enables automatic service generation of the corresponding real-twin, enabling create, update, delete, and search operations on arbitrary large objects derived from the proposed hyper-document MOF model and on-fly generation of the tool’s functional and visual characteristics according to the externally specified and run-time-alterable specification-to-code transformation standards.

The fourth challenge addresses the generally reusable nomenclature aspects of complex sociotechnical systems, particularly in the higher education domain. It directs the research toward the systematized nomenclature of STS (SNoSTS) and the systematized nomenclature of higher education (SNoHEd), enabling the transcription of meta-models, models, and model-instance concepts to different languages. The potential existence of a variety of distinct configurations in a constantly growing dataset enables the application of a large language model methodology to gain overall transparency of the language-independent nomenclature core.

The fifth challenge addresses further specification of the internal higher education context representation resulting from the proliferation of specified external context-related benefits to more specific internal context management decision-making support, empowered by digital twinning methods and techniques.