"No two machines are alike at our company." "We don’t have any idea what our machine is producing." "What we produce on the machine varies so much we’d be comparing apples to oranges; that wouldn’t work at all." All these assertions and more are things we hear from our customers. But even in heterogeneous production environments, evaluating data with the help of AI is possible and useful. Our example of machine fingerprinting demonstrates this.
We collect data on the state of a machine under “normal conditions” — the fingerprint. This can, for example, be a special measuring run that is recorded during commissioning as the normal condition. The condition assessment can be run on this basis regardless of the specific production involved. The prevailing status is determined by restarting this measuring run between two production cycles and analyzing the data. This makes it possible to detect deviations from the fingerprint. Unlike classic, continuous condition monitoring, the state of a machine is “only” determined at certain intervals and under defined conditions.
Machine fingerprinting provides USU customers with a simple, reliable and clever solution for effectively assessing a machine’s condition even in a heterogeneous production environment.
Lorem ipsum dolor sit amet, consetetur sadipscing elitr.
Our experts will be happy to answer your questions and support your projects.