Root Cause Analysis

Finding the cause of faults step by step


Finding connections

Important information about the status and development of complex machine and IT systems over time can be extracted from the event and measurement data of those systems. For example, by detecting recurring sequences of events or features in sensor data it is possible to identify correlations to provide a better understanding of the system behavior or the cause of faults. 


Verifying the correlations

In many cases, expert knowledge or experiments are required to assess whether the correlations are really relevant. The combination of your domain knowledge with the correlations that we identify produces a selection of relevant relationships with a causal character and, consequently, provides true value for understanding the systems.

Understanding downtimes in the production process better

With self-learning processes, our solutions are able to analyze event data and sensor measurement data in various depths of detail, to structure the results, and to allocate selected events to their most likely causes. The methods that we use for this are extremely varied and depend very much on the type of data. For instance, to prepare the results, they can be presented in the form of an event tree.        

The event tree above shows the links between the fault events of a production plant. Read from left to right, selected faults are matched to their potential triggers. The size of the diameter of a node is a measurement for the strength of the respective correlation.

With the help of root cause analysis, important correlations can be extracted efficiently from historic data and, with interactive visualization, be prepared so that mechanical engineers and service technicians have quick, intuitive access to this information. The knowledge that evolves from this can be used to improve machines and IT systems from a technical aspect, speed up maintenance, or predict faults with software-based solution approaches at an early stage.       

Your benefit from the cluster analysis

Optimize
processes

Improve decision
making processes

Reduce
downtime

Increase
control

Any questions? We are happy to help.

Katana, the data analytics segment of the USU Group, has extensive experience in the application of machine learning methods in the industrial sector. Take advantage of our knowledge and our solutions for building your data-driven business models to reduce your costs and improve your quality and value added.