The automotive industry today faces many challenges that include optimizing development costs and shortening time-to-market. In this environment, it is more important than ever to perform tests on the unit under test efficiently and to avoid unnecessary testing. It is critical to anticipate problems early to avoid costly damage to the unit under test and prevent delays in the project timescale.
Since tests are usually executed over several days and weeks, an intelligent mechanism is needed to detect errors and anomalies at an early stage. This enables the detection of a problem hours or even days before an incident happens in order to prevent potential damage to the unit under test and save valuable time.

PUMA 2 Machine Learning can be easily installed on any office computer. Its easy-to-use user interface enables even non-experts to take full advantage of well-known machine learning algorithms and various regression model types. This allows users to use learning patterns from test data without having to program them. PUMA 2 Machine Learning can be used to model and predict any channel values, and works in all test environments.

PUMA 2 Machine Learning allows to detect any problems with the unit under test and the overall testbed equipment at an early stage.
Ease of Use
Easily configure learning patterns without the need for programming skills
Time Saving
Predict values that are not available or require great effort to obtain
Maximize Productivity
Increase productivity through early stop of invalid tests

With PUMA 2 Machine Learning, we want to make the potential of ML algorithms available to our users in a user-friendly way for testbed automation. This can improve the monitoring of the Unit Under Test and thus detect problems at an early stage.
– Stephan Lenhart, Senior Product Manager AVL PUMA 2™ Machine Learning, Test Automation Products, AVL List