Representative End User Clients
Representative Automation Clients
Representative Software Clients
KEYWORDS: Virtual Twin, Artificial Intelligence, AI Physics, Manufacturing, Industry World Models
Overview
The 3DEXPERIENCE FORUM EUROWEST 2026 highlighted a clear inflection point in industrial transformation, where artificial intelligence, virtual twins, and platform-based collaboration are converging into a unified operational model for industry. Across keynotes from Dassault Systèmes, Valeo, Bel Group, and industrial partners, a consistent message emerged: AI is no longer an experimental layer on top of industrial systems but a core driver of productivity, decision making, and competitiveness.
The forum emphasized that the next phase of industrial performance will depend on the ability to contextualize and operationalize knowledge at scale. Virtual twins are positioned as the central representation layer of industrial reality, enabling organizations to unify fragmented data, simulate complex systems, and embed intelligence directly into operations. Combined with AI companions, physics-based modeling, and synthetic data generation, this approach defines what Dassault Systèmes frames as “Industry World Models.”
Context and Key Takeaways
A New Phase of Industrial Transformation
The forum opened with the recognition that industry is undergoing a structural acceleration driven by artificial intelligence, generative technologies, energy constraints, geopolitical fragmentation, and rising demands for technological sovereignty. In this environment, companies are under increasing pressure to innovate faster while maintaining resilience and efficiency.
The concept of the virtual twin is important - it acts as a shared “map” of industrial operations, ensuring that all stakeholders have a consistent understanding of the situation.
Stéphane Degraeve, Eurowest Managing Director at Dassault Systèmes, positioned this shift as a transition from digital tools to industrial intelligence systems. He emphasized that competitiveness now depends on how effectively organizations can leverage everything they know, including engineering data, operational history, and intellectual property, and apply it consistently across decisions and processes. In this framing, virtual twins become the foundational layer for turning fragmented knowledge into structured, operational intelligence.
Industry World Models: From Data to Industrial Intelligence
Morgan Zimmermann, CEO of the 3DEXPERIENCE, introduced the concept of Industry World Models as the backbone of next-generation Industrial AI. The framework rests on three interconnected pillars.
First, industrial knowledge accumulated over decades, embedded in design systems, production histories, configuration trees, and field incidents, which must be transformed from implicit expertise into explicit, actionable intelligence.
Second, Industrial AI introduces a critical governance dimension. Zimmermann emphasized that AI value creation will increasingly depend on the ability to manage data rights, access, and protection at scale.
Third, knowledge organization must shift away from static repositories toward virtual twins. These twins provide contextual, operational representations of real-world systems.
Physics, Synthetic Data, and the Limits of Pure AI
A key differentiation highlighted during the forum was the need to combine AI with physics-based modeling. While AI is highly effective at pattern recognition, industrial domains such as aerospace, automotive, and energy require deterministic guarantees grounded in physical laws.
Zimmermann further highlighted the importance of synthetic data generation within virtual environments. By simulating systems that do not yet exist, organizations can generate operational data for future products, enabling predictive maintenance, validation, and optimization before physical deployment.
AI Companions and the Operationalization of Industrial Intelligence
Dassault Systèmes introduced AI companions. These systems are designed as role-based collaborators embedded in industrial workflows. Examples include Aura (business analyst), Leo (design engineer), and Marie (scientific assistant). These companions are intended to democratize access to industrial intelligence and ensure that complex analytical capabilities are available across all levels of the organization, not only among experts.
ARC Advisory Group clients can view the complete report at the ARC Client Portal.
Contact Us if you would like to speak with the author.
Obtain more ARC In-depth Research Market Analysis.