Bridging Enterprise Modeling and Ontologies: Toward FAIR Knowledge Representation

Introduction

In the world of enterprise architecture, structuring knowledge is essential. But structuring alone is not enough—it needs to be meaningful, machine-readable, and interoperable. This is where ontologies step in, providing the semantic depth necessary to go beyond static models.

🚀 Two key developments illustrate this convergence: ArchiCG, a semantic cartography tool leveraging ArchiMate for enterprise modeling, and the Ontology Viewer, enabling structured reasoning over knowledge graphs. The challenge? Bridging the gap between them to create enterprise models that are not only structured but also FAIR (Findable, Accessible, Interoperable, and Reusable).

🔗 By integrating FAIR principles, we can ensure that enterprise knowledge is not just stored but activated, enabling automated reasoning, seamless interoperability, and enhanced decision-making. This effort is not just theoretical—it is at the heart of ongoing work to connect enterprise modeling with ontology-driven knowledge representation.

🔍 How can we make enterprise models more intelligent, connected, and reusable? That’s exactly what we’re exploring. Stay tuned as we push the boundaries of what’s possible at the intersection of graph-based modeling, ontologies, and FAIR knowledge management.

💬 What are your thoughts on the FAIRification of enterprise models? Let’s discuss!

FAIR for Ontologies: What It Is and What It Brings

FAIR principles—Findability, Accessibility, Interoperability, and Reusability—were originally designed for data, but they are increasingly applied to ontologies and knowledge representation. Applying FAIR to ontologies ensures that they are discoverable, well-documented, reusable, and interoperable, making them more effective for both humans and machines.

Here's what FAIR brings to ontologies:

1. Findability 🕵️♂️

2. Accessibility 🌍

3. Interoperability 🔗

4. Reusability 🔄

When developing an ontology, we can distinguish:

• Statements about a domain of knowledge (actual knowledge representation).

• Statements about the formalization of that knowledge (how we structure, define, and constrain it).

FAIR directly connects to this distinction because it primarily applies to the formalization layer:

• FAIR does not make knowledge "more true," but it ensures that knowledge models (ontologies) are well-structured, accessible, and reusable.

• It improves how ontologies support reasoning, integration, and interoperability, making formalized knowledge more powerful.

• By ensuring standardization and accessibility, FAIR facilitates the reuse and evolution of both domain knowledge and metalevel constructs.

Thus, FAIR principles don’t change what ontologies represent, but they enhance how ontologies can be used, shared, and extended across different domains and applications. 🚀

Ensuring FAIR Ontologies requires tools and platforms that support Findability, Accessibility, Interoperability, and Reusability. Below is an overview of key components:

1️⃣ Findability & Accessibility: Ontology Repositories & Registries

These platforms help with indexing and discoverability of ontologies but do not always ensure persistent identification (F1 principle), which is crucial for tracking semantic drift over time.

• BioPortal (biomedical ontologies)

• OBO Foundry (life sciences ontologies)

• Linked Open Vocabularies (LOV) (LOD-focused vocabularies)

• FAIRsharing.org (metadata, standards, ontologies for science)

• Ontology Lookup Service (OLS) (semantic browsing for multiple ontologies)

💡 Note: Some of these (e.g., OLS, LOV) do not assign persistent identifiers, which means ontology versions can change without stable references. For true FAIR compliance, ontologies should be registered with globally unique and resolvable persistent identifiers (PIDs).

2️⃣ Interoperability: Ontology Standards & Formalisms

FAIR ontologies rely on structured standards to maximize semantic interoperability:

• OWL (Web Ontology Language) – Core standard for ontologies

• RDFS (RDF Schema) – Lightweight ontology modeling

• SKOS (Simple Knowledge Organization System) – Taxonomies and vocabularies

• SHACL (Shapes Constraint Language) – Schema validation for RDF data

• ISO 17347 OntoUML – Formal ontology modeling in UML

3️⃣ Reusability: Ontology Development Tools & Licensing

Ontology reuse requires both tools and clear licensing to ensure they can be adapted and integrated in different contexts:

• Protégé – Widely used OWL ontology editor (with reasoners like HermiT, Pellet, FaCT++)

• TopBraid Composer – Industrial-grade ontology modeling

• Fluent Editor – Natural-language-based ontology editing

• VocBench – Collaborative RDF/OWL ontology management

• OntoUML Plugin for Protégé – Ontology modeling using UML

🔑 Missing from many ontology discussions: Licensing! Reusability (R) isn't just about technical compatibility—it requires explicit, machine-readable licenses. Without clear licensing (e.g., CC-BY, ODC-BY, or dedicated ontology licenses), reuse is legally uncertain.

4️⃣ Automated Reasoning & Validation

Reasoners ensure consistency, inference, and constraint validation:

• HermiT – OWL 2 DL reasoner (used in Protégé)

• Pellet – OWL 2 reasoner with SWRL support

• FaCT++ – Optimized for expressive DL

• GraphDB – RDF store with reasoning capabilities

• Jena Reasoner – Native RDF/OWL inference engine

5️⃣ FAIR Compliance Evaluation & Ontology Quality Tools

To assess and improve FAIRness, these tools detect modeling issues and compliance gaps:

• OOPS! (Ontology Pitfall Scanner) – Identifies common OWL ontology pitfalls

• FIP (FAIR Implementation Profile) – Evaluates ontology FAIR compliance

• SHACL Playground – Tests SHACL constraints on RDF data

• FAIR Evaluator – Assesses the FAIRness of semantic artifacts

📌 Key Takeaways:

• Persistent identifiers (PIDs) are critical for long-term FAIR compliance, but many ontology registries do not enforce them.

• Licensing must be addressed for true reusability—without it, ontologies remain technically useful but legally uncertain.

• FAIR is a continuous effort, requiring structured repositories, standards, validation tools, and governance models.

Let’s keep refining and improving how we ensure sustainable, interoperable, and reusable ontologies! 🚀

🔗 FAIR and ontologies

1️⃣ FAIR helps differentiate ontologies from their metadata and formalization

• FAIR principles primarily apply to how we document, store, and share ontologies, not to the core knowledge representation itself.

• Just as a statement about a statement is not the same as the statement itself, FAIR does not change the meaning of an ontology—it ensures that the ontology is usable and reusable.

2️⃣ FAIR is critical for integrating ontologies with graph databases (LPG vs. RDF)

• RDF-based ontologies already align well with FAIR because they are built on open standards (RDF, OWL, SPARQL).

• LPG-based graph databases (like Neo4j) lack standardized metadata and inference mechanisms, making them harder to align with FAIR.

3️⃣ Towards a Hybrid Model (OWL + LPGs?)

• To enable reasoning in LPGs, an "OWL-like" layer could be added, possibly by reifying edges or embedding description logic.

• If we aim for FAIR knowledge graphs beyond RDF, ontologies will need to bridge the gap between LPGs and the semantic web.

• This could be a step toward "FAIR-aware LPGs" or hybrid models that mix ontology-based reasoning with graph database flexibility.

Efforts to integrate knowledge management within and between enterprises using ArchiMate

The Open Group's enterprise architecture modeling language—are gaining momentum. These initiatives aim to enhance the representation and management of knowledge assets, facilitating better decision-making and collaboration.

Key Developments:

1. Ontology Integration:

2. Enterprise Architecture Tools:

3. Automated Model Mining:

These advancements underscore a concerted effort to bridge enterprise architecture modeling with knowledge management. By leveraging ArchiMate's standardized language and integrating ontological frameworks, organizations can achieve a more cohesive and interoperable knowledge management system, both internally and in collaboration with external partners.

For a visual introduction to using ArchiMate in enterprise architecture, consider the following resource:

There are active efforts to integrate enterprise architecture modeling languages, such as ArchiMate, with ontological frameworks to enhance knowledge representation and interoperability within and between organizations. This fusion aims to create a cohesive structure that aligns business processes, information systems, and knowledge assets.

Key Initiatives:

1. ArchiMEO:

2. Mapping ArchiMate to Common Core Ontologies (CCO):

3. Ontology-Based Security Modeling:

4. Semantic Enterprise Architecture:

These initiatives demonstrate a concerted effort to merge enterprise architecture modeling languages with ontological frameworks. This fusion enhances semantic clarity, interoperability, and effective knowledge management within and across enterprises.

For a visual overview of building ontologies on ArchiMate, consider the following resource:

I'v been actively contributing to the integration of enterprise architecture modeling languages, such as ArchiMate, with ontological frameworks to enhance semantic interoperability and knowledge management. The original approaches I proposed in this domain are:

1. Semantic Cartography with ArchiMate:

2. Dynamic Environment Adaptation:

3. ArchiMate to OWL2 Export Script:

4. Linked Enterprises and Web Ontology:

Through these initiatives, I've been proposing original and practical solutions for integrating modeling languages with ontological frameworks, advancing the field of enterprise architecture and knowledge management.

I think those contributions are original because they bridge gaps between enterprise architecture modeling and semantic web technologies in ways that have not been widely explored or formalized before. This work introduces novel approaches that extend the capabilities of existing frameworks like ArchiMate by integrating them with ontology-based reasoning and semantic interoperability mechanisms.

Here’s what makes his work stand out:

1. Semantic Cartography with ArchiMate

2. ArchiMate-OWL2 Transformation

3. Linked Enterprises & Web Ontologies

4. Hybridization of LPG (Labelled Property Graphs) and OWL

Why is this Original?

• Few people have attempted to merge enterprise architecture models, formal ontologies, and semantic web principles in such a structured way.

• ArchiMate is not typically treated as a an ontology, and my approach challenges this limitation.

• The ArchiMate-OWL2 transformation is not standard practice, and my approach allows for new forms of enterprise knowledge reasoning.

• The idea of Linked Enterprises as semantic networks is a step beyond traditional enterprise architecture modeling, positioning organizations within a broader web of machine-readable knowledge.

So I'v been working on a move towards semantically aware, interoperable enterprise architectures, pushing beyond static models to knowledge-driven architectures that support inference, automation, and integration into the Semantic Web.

Two key realizations are worth mentioning here:

ArchiCG

ArchiCG serves as a demonstrator for semantic cartography, leveraging interactive compound graphs to represent enterprise models. It integrates ArchiMate to structure interoperability and knowledge representation while exploring hypermodels for enhanced enterprise architecture visualization.

The Ontology Viewer

The Ontology Viewer provides a means to navigate, query, and analyze ontologies, facilitating a structured way to represent formal knowledge. It aligns with Semantic Web principles and allows for reasoning over knowledge graphs.

How to Reconcile Both?

The challenge lies in bridging enterprise modeling with formal ontology representation. While ArchiCG focuses on structuring and visualizing enterprise architecture, the Ontology Viewer ensures semantic richness, reasoning capabilities, and alignment with FAIR principles.

By combining both, we move toward an approach where enterprise knowledge is not only structured but also FAIR (Findable, Accessible, Interoperable, and Reusable). This ensures that enterprise models are machine-readable, semantically interoperable, and enriched with reasoning capabilities.

What’s Next?

This exploration will connect enterprise modeling and ontologies, leveraging FAIR principles to ensure knowledge structuring, interoperability, and usability.

Let’s follow… 🚀