Digital Twin

A digital twin is a computer model of a specific physical machine, production line, or system that stays continuously updated with real data measured from its physical counterpart, rather than remaining a fixed design drawing or a one-time simulation, so that the model's behavior can be checked against the real machine's actual condition at any moment and used to predict how it will perform or fail before that happens in the physical world. The concept grew out of decades of engineering simulation practice, including NASA's use of physical and computational models to diagnose problems on Apollo missions from the ground, but the specific term and its present definition were formalized around 2010 in work associated with NASA's own technology roadmaps. In industrial and manufacturing engineering the idea has extended from a single product to whole production processes, letting an engineer test a proposed change to a factory floor, a maintenance schedule, or a production run against the digital twin first, catching a problem in the model rather than on the real line. Because a digital twin is only useful if it is actually kept in sync with its physical counterpart, the approach depends on the same networked sensors and real-time data feeds that spread through manufacturing under the broader banner of industrial automation.