By harnessing artificial intelligence, Data@Hand’s algorithms are able to detect deviations from the norm. “What’s special about this technology is that we can deal with operating states never previously encountered, and we have a system that continuously evolves itself. It learns to recognize normal operation and corresponding deviations,” states Dr. Mario Aehnelt, Head of the Visual Assistance Technologies Competence Center at Fraunhofer IGD in Rostock. He and his team were responsible for developing Data@Hand as an information tool.
The goal is to optimize processes and to assist people with the analysis of large volumes of complex data. Importantly, however, the decision regarding how best to respond to any anomaly remains with the human expert. If there is a change in a machine parameter, such as an increase in the temperature of a compressor, the sensor unit detects it in real time and signals to the operator that they need to intervene. Data@Hand resides in the cloud and works in the background, upstream of existing interfaces or visualization systems such as Fraunhofer IGD’s own Health@Hand or Plant@Hand.
As a result, customers can simply continue to work in their familiar systems environment. The new Data@Hand technology can also add decisive value when it comes to analyzing health data, e.g., by alerting physicians in real time to significant changes in patient data. To achieve this goal, machine-learning solutions are trained to recognize the difference between healthy and unhealthy patient conditions.