Digital twin
A digital twin is a context-dependent digital representation connected to a real-world entity or process through an ongoing exchange of data, often supporting observation, analysis, prediction, or control.
Direct definitions with security implications and source context. Terminology varies; local architecture and authority remain decisive.
A digital twin is a context-dependent digital representation connected to a real-world entity or process through an ongoing exchange of data, often supporting observation, analysis, prediction, or control.
A digital shadow is a digital representation updated primarily through one-way data flow from the physical system to the digital representation.
A digital model is a digital representation that does not necessarily maintain automated, ongoing synchronization with a physical counterpart.
A digital thread is the connected flow of information across a product, system, or asset lifecycle and its engineering and operational contexts.
A cyber-physical system integrates computation, communication, and physical processes so digital behavior and physical behavior affect one another.
Telemetry is measurement or status data transmitted from a device, process, or system for remote observation or use.
Synchronization is the process of reconciling physical and digital state with defined timing, ordering, conflict, and freshness rules.
Twin fidelity is the fitness of a twin’s representation and behavior for a stated purpose, operating range, and decision context.
Twin identity is the set of attributes and credentials used to distinguish and authenticate a digital twin instance and bind it to its intended counterpart and owner.
Model provenance is the traceable history of a model’s origin, data, code, parameters, validation, approvals, versions, and transformations.
Data provenance is evidence describing where data originated, how it was collected, and what transformations, custody, quality, and timing apply.
Sensor provenance connects a measurement to an identified device, calibration context, configuration, time source, location, and processing history.
A trust boundary is a transition where data, identity, authority, or assumptions move between components or domains with different control or assurance conditions.
A control loop measures a process, compares it with a desired state, and changes inputs to influence physical behavior.
An actuator converts a command or control signal into a physical action.
Edge computing places processing near devices or physical processes, often to reduce latency, bandwidth, or dependency on remote services.
Industrial IoT connects industrial devices, sensors, machines, and software to support monitoring, coordination, analytics, and operations.
Operational technology is programmable hardware and software that detects or causes changes in physical processes, devices, and events.
A PLC is an industrial computer designed to execute deterministic control logic for machines and processes.
Supervisory control and data acquisition systems monitor and supervise geographically or functionally distributed industrial processes.
A simulation executes a model to explore how a system may behave under selected initial conditions, assumptions, and inputs.
Model integrity is confidence that a model, its parameters, dependencies, configuration, and execution have not changed outside authorized and verified processes.
Model poisoning is the intentional corruption of training data, model artifacts, feedback, or update processes to alter model behavior.
State desynchronization occurs when the digital representation and relevant physical or authoritative state diverge beyond accepted limits.
Digital twin federation is a relationship in which independently governed twins exchange state, models, identity, or decisions.
A software bill of materials is a formal inventory of components and supply-chain relationships in software.
Machine identity is the verifiable identity assigned to a device, workload, service, application, or other non-human actor.