Use Case
In the transport and traffic industry, the availability of the vehicle fleet is one of the most important parameters. The ability to react quickly to unforeseen events is the be-all and end-all here. The desired scenario of every company is to detect errors in the system at an early stage in order to be able to act in time and thus prevent unexpected failures.
The precondition for this is the use of data – both data from the train control system and additional sensor data. Instead of collecting as much data as possible, it is important for reliable predictions to collect the right data – i.e., data that provides information on failure risks. Of course, monitoring all components of a vehicle is not practical and only consumes unnecessary resources.
For this reason, it is advisable to monitor safety-critical components on the one hand and components with a high probability of failure on the other. The optimum behavior in normal operation represents the comparative value to the actual state. If deviations are detected here, the recommendation system determines the possible causes and derives recommendations for action. This knowledge is not only of great importance to the operator, but also saves service personnel an enormous amount of time when searching for the cause. The solution offers an extremely high detection probability for deviating system characteristics under realistic load conditions. This is the key advantage of the solution.