Modelware · Research

Built on a Decade of Research

OML Code did not start as a product idea. It grew out of over a decade of work at Modelware Labs, our research engine, advancing ontological modeling and rigorous systems engineering with corporate, government, and academic partners. That research foundation is why the platform can make claims most tools cannot.

Our Research Engine

Modelware Labs

Modelware Labs is where our work begins. For over a decade, its researchers have tackled the hardest problems in Neurosymbolic Systems Engineering: how to capture engineering knowledge as formal, machine-understandable models, how to reason over those models with confidence, and how to put trustworthy AI to work on them.

Working hand in hand with leading corporate, government, and academic partners, the lab explores these open problems and pressure-tests its ideas on real, high-stakes engineering, not toy examples. What proves out in the lab is hardened and shipped inside OML Code.

What the Lab Pursues

Research Focus

  • Ontological modeling languages and methods
  • Automated reasoning and analysis at scale
  • AI agents grounded in formal models
  • Model DevOps and continuous validation
  • Interoperability across engineering tools
Open by Design

Foundation & Community

Our research lives in the open anchored by an open-source foundation and the community that has grown around it.

Open Source Foundation

openCAESAR

The open source platform for ontological modeling and analysis, proven on flagship space missions. openCAESAR pioneered the rigorous, model-as-code approach that OML Code builds on and productizes today.

Visit openCAESAR
Community

onto:Nexus

A forum and workshop series where practitioners and researchers of ontological modeling share methods, results, and lessons learned, keeping the community connected and the state of the art moving.

Explore onto:Nexus
Proven in Practice

Trusted by Industry Leaders

We have engaged leading aerospace and defense organizations through the open-source openCAESAR foundation.

BAE Systems Leonardo CAE
Academic Partners

University Researchers We Collaborate With

Modelware Labs partners with leading academics to push Neurosymbolic Systems Engineering forward.

Prof. Alejandro Salado
University of Arizona
Prof. Alejandro Salado

Associate Professor of Systems and Industrial Engineering. Collaborating with Modelware on OML CoPilot research since 2024.

Prof. Giancarlo Guizzardi
University of Twente
Prof. Giancarlo Guizzardi

Full Professor and Chair of Semantics, Cybersecurity & Services. Collaborating with Modelware on encoding gUFO in OML so domain ontologies inherit formal analytical patterns.

Prof. Sambit Bhattacharya
Fayetteville State University
Prof. Sambit Bhattacharya

Professor of Computer Science and Director of the Intelligent Systems Lab. Collaborating with Modelware on MCP-based AI tools for OML Code and converting legacy engineering documents into OML knowledge bases.

From the Lab

Recent Publications

Peer-reviewed work by our founder and collaborators. Much of it now ships inside OML Code.

2026
Bridging the Semantic Web and Model-Based Systems Engineering with the Ontological Modeling Language

Elaasar, M., Oakes, B., Kamburjan, E., Hamdaqa, M., Hamou-Lhadj, A.

Proceedings of ISWC 2026, In Use Track, Bari, Italy

Presents OML to the Semantic Web community: the two obstacles that have held back Semantic Web adoption in systems engineering, namely the lack of rigorous semantics for SE models and of an established methodology for using ontologies in SE, and the language, methodology, and tooling that answer them, with adoption at NASA, JAXA, Leonardo, and in academia.

Ontologies Semantic Web MBSE Read the paper
2026
Beyond Syntax: Method-Compliant AI Assistance for Low-Resource Modeling Languages

Amar, S., Elaasar, M., Bhattacharya, S.

Proceedings of MDE Intelligence 2026, co-located with ACM/IEEE MODELS 2026, Málaga, Spain

An evaluation framework and protocol-driven architecture for AI-assisted modeling in low-resource languages: the LLM interprets intent while all edits run through validated MCP tools and a SHACL shape catalog enforces the method. Measured on the FireForce VI case study, tool augmentation helps most on the hardest tasks.

AI Evaluation Modeling Languages
2026
Large Language Models for Systems and Ontological Modeling: Fine-tuning and Evaluation

Alarcia, R., Alhamadah, A., Elaasar, M., Golkar, A., Salado, A., Cornejo, S.

Proceedings of the Conference on Systems Engineering Research (CSER) 2026

Specialized LLMs now serve domain-specific tasks at a fraction of the inference cost of general-purpose ones, but systems engineering has neither a corpus of them nor fine-tuning methods suited to its languages. Contributes a methodology for fine-tuning LLMs for small, underrepresented modeling languages, using OML as the reference, and explores the balance between fine-tuning a model and using a general-purpose one.

AI OML Fine-tuning
Browse all 13 publications

Research You Can Build On

The results of this research ship today inside OML Code. If you want to co-sponsor a Modelware Labs project, propose a new one, or see how a decade of research becomes your engineering advantage, we should talk.

Talk to Us Explore OML Code