Predictive maintenance using artificial intelligence


Article: Predictive maintenance using artificial intelligence

Is it realistic to assume that people got to the doctor every Friday in order to have him diagnose potential illnesses? Of course, it is not! People learned to ask for doctor’s advice only in case of discomfort. In some cases, we are even in the position to classify yesterday’s irregularities as tomorrow’s commonplace. Humans keep learning, and so can machines, if we empower them to do so. 

DB Systel drives forward this empowerment in the context of Predictive Maintenance for technical plants of Deutsche Bahn. Deviations in the “state of health” of technical plants can be recognized by means of so-called machine learning algorithms, which can also help to provide information as to when the next service must be carried out. The prediction is based on special methods of artificial intelligence. 

Mechanical Plants “Speak”, AIM “Listens”

An example is the DB Systel startup "Acoustic Infrastructure Monitoring" (AIM). There are around 1,000 escalators at German railway stations. Again and again, systems grind to a halt. Sometimes they are out of service for a long time – because the failure is not immediately noticed, and the repair is often very complex.

For some years, the team has been dealing with the set-up of a service which detects early malfunction and the development of which is based on operating noises of mechanical plants. The AIM Service consists of an AIM box and two microphones, which are installed at the plant as well as a software, which analyzes operating noises. In case a malfunction is looming, an early message will be sent to the operator or service provider and technical staff can examine the systems where there are problems in a targeted way, thus ensuring that they do not fail in the first place. “This saves operating and maintenance costs and ensures, for example, that fewer passengers are stranded with their suitcases in front of a broken escalator”, says Jens Glöckner, founder of the AIM team.

The service is a universally useable retrofit solution which can be fitted to existing plants without having to interfere with the system control. We already tested this service in different mechanical plants e.g. for the use of cranes in terminals or for the monitoring of rotating components such as motors or gearboxes. In this regard, the use of this service in logistic warehouses to surveil critical components is imaginable.

In addition to the AIM products and solutions, the team also offers comprehensive consulting on predictive maintenance and AI topics.

AIM has a wide range of possible applications - also on the DB-external market

Mechanical Plants “Speak”, “Listens”

DB is already using AIM in escalators since it was proven during the testing phase that the maintenance costs could be reduced by approximately 25%. In addition, the availability of the moving staircases was considerably improved. In future, the predictive maintenance approaches by Deutsche Bahn will contribute to an improved punctuality of the trains, since information received from sensors in almost real-time can be analyzed and evaluated.” says Glöckner.

Do you have other use cases in mind, or you would like to try AIM?

We would appreciate if you get in touch with us.


More information:

AIM on

AIM on Twitter