AI Glossary
The complete dictionary of Artificial Intelligence
Digital Twin
Dynamic virtual replica of a physical asset, process, or system, using real-time data to simulate, predict, and optimize its behavior throughout its lifecycle.
Model Fidelity
Degree of accuracy with which the digital twin reproduces the characteristics, behavior, and performance of its physical counterpart, often measured by error and correlation metrics.
Real-Time Synchronization
Continuous process of updating the digital twin with operational data (sensors, IoT) from the physical asset to ensure the simulation reflects the current and real state of the equipment.
What-If Scenario Simulation
Use of the digital twin to virtually test the impact of various conditions, failures, or maintenance strategies on equipment performance and lifespan, without risk to the actual asset.
Data Contextualization
Process of enriching raw sensor data with contextual information (history, environment, operating mode) to give meaning to simulated states in the digital twin.
Shadow Mode
Operation of the digital twin in parallel with the physical asset, receiving the same inputs and performing the same calculations to validate new strategies or software updates before deployment in production.
Persistence Twin
Type of digital twin that maintains the complete history of an asset's states, events, and maintenance, enabling analysis of degradation trajectories and validation of lifespan models.
Time Warping
Simulation technique that accelerates or slows down the time scale in the digital twin to quickly assess the long-term consequences of a maintenance decision or observe a slow phenomenon in detail.
Digital Thread
Communicational data flow that connects the digital twin to all of the company's information systems (PLM, ERP, MES), ensuring data traceability and consistency throughout the asset's lifecycle.
State Twin
Instantaneous digital representation of a physical asset, focusing on its current state (temperature, pressure, vibration) for real-time monitoring and diagnosis, without necessarily modeling its future dynamics.
Variable Geometry
Ability of a digital twin to integrate and simulate geometric changes of equipment due to wear, deformation, or breakage, to predict their impact on functional performance.
Uncertainty Model
Component of the digital twin that quantifies and propagates uncertainties from sensor measurements and model approximations, in order to provide probabilistic predictions and confidence intervals.
Twin Aggregation
Combination of multiple digital twins of individual assets to form a larger system twin, enabling simulation of interactions and cascade effects within a production line or factory.