Your Digital Twin Already Has the Answers
Madalina Florean, Director of AI, Cadmatic
Fewer than one in four digital twins reach full operational deployment, while only 14% meet expectations. The challenge is often not a lack of information, but data that remains incomplete, unreliable and difficult to use.
Madalina Florean, Director AI at Cadmatic, explains why trusted data matters from the earliest stages of an asset’s lifecycle. Design may account for only 15% of project expenditure, yet decisions made at this stage can commit 75% of lifecycle costs, making early data checks essential for reducing rework and preventing costly problems later.
The presentation identifies four capabilities a digital twin needs before users can rely on its answers: joining equipment records across systems, showing multidisciplinary models in one view, measuring asset completeness and checking data consistently across sources.
Madalina demonstrates how AI can reconcile tags from point clouds and historic documents, convert engineering rules into repeatable tests and use agents to keep information current. These capabilities allow the twin to reveal existing insights, complete defined tasks and explore new possibilities, while human approval and established access rights remain central.
Explore the full presentation below to discover how trusted engineering data and carefully governed AI can turn an underused digital twin into a reliable source of answers across design, construction and operations.
