What Are Digital Twins?
A digital twin is a digital representation of a physical object. Simple cases involve sensor readings such as temperature or RPMs, while complex cases cover expensive capital assets including diesel engines, turbines, and heavy-duty mining and construction equipment. These complex twins can comprise finite state machines with potentially tens of thousands of discrete states, requiring machine learning algorithms to build accurately.
In these cases, the digital representations will be incredibly complex, comprising finite state machines with potentially tens of thousands of discrete states. Due to their complexity, these digital twins can only be built accurately using machine learning algorithms . Once created, digital twins and analytics facilitate more accurate diagnostic, predictive and optimization processes.
Do Digital Twins Improve Operations?
Digital twins and analytics facilitate more accurate diagnostic, predictive, and optimization processes. However, a common misconception holds that digital twins should visually replicate the physical object, perhaps using CAD data. According to Kevin Walsh of Bsquare, only a minority of IoT use cases, such as augmented reality for repair technicians, benefit from digital twins that incorporate spatial or structural information.
Where Does Digital Twin Data Come From?
Digital twins must be built entirely on characteristics of the device that machine learning algorithms can learn and maintain over time. Any manual addition of information, such as CAD files, impairs scalability and makes it difficult to preserve accuracy. Visual representations of physical objects may serve as presentation aids, but they do not contribute to the machine learning models that power real analytics.
More ways manufacturers can capitalize on data analytics can be found in our previous article.
This article was written by Kevin Walsh, Vice President of Marketing at Bsquare and originally was published here. In this capacity, Kevin drives all aspects of marketing including product and partner strategy as well as marketing communications and public relations.
