How Agentic AI and TEAMS® Are Transforming System Health Digital Twin Development
At the 2026 AHFE (Applied Human Factors and Ergonomics) Conference, Christopher M. Norton of Qualtech Systems, Inc. presented a new approach to one of the most persistent challenges in System Health Management (SHM): the knowledge acquisition bottleneck. In this talk, we shared how agentic AI, combined with TEAMS® and a neuro-symbolic modeling framework, is fundamentally changing that equation.
While advances in AI, sensing, and diagnostic algorithms have accelerated dramatically over the past three decades, one critical limitation has remained unchanged: The process of converting engineering knowledge into executable models is still largely manual. This bottleneck—not inference capability—has been the primary barrier to scaling digital twins across complex systems.
Breaking the Knowledge Acquisition Bottleneck in Digital Twins
TEAMS® has demonstrated scalable diagnostic reasoning across complex systems for decades. The limiting factor is not the ability to reason over failure modes; it is the time and effort required to convert manuals, schematics, FMECA tables, logs, and engineering expertise into structured, maintainable models.
- Manual modeling slows digital twin deployment.
- Model maintenance becomes harder as systems evolve.
- Scaling requires a faster, more repeatable way to capture engineering knowledge.
Agentic AI changes the modeling workflow
The approach uses AI agents to ingest engineering artifacts, extract components, functions, failures, tests, and relationships, and assemble candidate model structures. Instead of replacing engineers, the agents accelerate the early modeling work so subject matter experts can focus on review, correction, and approval.
Formal models keep AI output executable and trustworthy
A key theme of the presentation was the need to pair AI with formal causal models. The AI helps extract and structure information, while the model framework preserves the rigor required for diagnostics, failure propagation, observability, modes, uncertainty, and certification-oriented review.
- Ingest: Normalize engineering artifacts.
- Extract: Identify components, failures, tests, and relationships.
- Structure: Build candidate causal models.
- Verify: Keep engineers in the loop through review, approval, and correction.
The results show practical acceleration
The AHFE presentation reported significant productivity gains from this workflow, including:
- Approximately 80% reduction in engineering effort.
- 4× to 17× acceleration in modeling tasks.
- Greater than 90% correctness and completeness in extracted model elements.
Validated models become reusable enterprise assets
Once validated, a causal digital twin becomes more than a diagnostic model. It can support real-time fault isolation, prognostics, guided troubleshooting, training, FTA/FMECA, sensor optimization, lifecycle cost analysis, and future model updates as field data becomes available.
The long-term goal is a living digital twin
The long-term vision is a closed-loop environment where field data, new failure modes, AI-generated updates, and SME validation continuously improve the model. In that environment, the digital twin connects design, operations, maintenance, and training through a single evolving knowledge asset.
Human expertise remains essential
The message is not that AI replaces engineers. AI provides scale, speed, and consistency; engineers provide context, judgment, and accountability. That human-AI partnership is especially important for safety-critical systems, certification needs, and long-term trust in digital twin outputs.
Looking Ahead
As digital systems become more complex, the ability to maintain accurate, up-to-date models will become even more critical. Agentic AI, when combined with formal model structures and human oversight, provides a practical path forward. At Qualtech Systems, we are excited to be leading this transition—helping organizations move from Manual modeling → Scalable automation → Adaptive digital twins.
Learn More
This blog post highlights the main takeaways from the AHFE 2026 presentation. For the full technical discussion, including the modeling architecture, validation workflow, results, and application examples, download the accompanying white paper and watch the full presentation video.
Qualtech Systems, Inc. (QSI) delivers advanced solutions for System Health Management, Digital Twins, and Model-Based Engineering, integrating physics-based modeling, AI-driven automation, and decades of domain expertise to support complex systems across their full lifecycle.
