When moving from preventive to predictive maintenance strategies, many companies adopted vibration condition monitoring technologies. A disadvantage of this technology is that it requires contact accelerometers. Acoustic monitoring solutions do not have this disadvantage, making them easier to deploy. Furthermore, a microphone has the intrinsic capability to monitor multiple components at once, leading to a hardware cost which scales better with machine or production complexity.

However, a microphone can be influenced by background noise often present in an industrial environment which potentially leads to false positive detections. Robustness of the utilized acoustic monitoring solution is therefore very important.

In this talk multiple solutions will be presented which increase the robustness of the acoustic monitoring solution and enable the broader industrial applicability. It will be applied to the use case of ball bearing monitoring, which is crucial to ensure the operation of many machines and vehicles. A comparison with current state-of-practice condition monitoring approaches will be made.