Wear detection in machining technology

Wear assessment of components based on high-frequency raw data by combining different process data


Timely recognition of the ageing process of tools.

Prevents the CNC machine from coming to a standstill due to a broken tool.


Quality control is to be carried out during production (inline) so that intervention can take place before quality problems arise. In addition, a connection between process results and process data should be established so that system variables can already be adjusted during processing.  


An AI model is trained with historical data of a healthy machine condition so that the spindle performance can be predicted based on past machine data. If the model is then put into production, this prediction is then compared with the actual measured spindle performance. A high deviation then shows that a different performance would be expected in a normal process and that the probability of a defect in the tool is high.  


The AI model was able to recognise in time that the tool was entering the ageing process due to changed patterns in the process data. In the future, the machine can be stopped at an early stage and the tool can be replaced without the tool breaking, which would result in a longer standstill of the CNC machine.

Are you interested in your own use cases?