CNC milling is widely used for many different applications. The milling tools wear out over time and must be replaced, which is cost intensive. Naturally, using a milling tool for as long as possible without breaking it, is the goal for any manufacturer. Today, manufacturers rely on reactive maintenance, tool changes after breakage or a preventive maintenance approach. Any of those maintenance systems do not guarantee an optimal point of tool change. To find the optimal point of tool change, it is necessary to monitor the CNC process parameters during machining and use advanced data analytics to predict the tool abrasion. This approach is called predictive maintenance.
In our partnership AI laboratory in Hong Kong, predictive maintenance research has been conducted with the goal of developing a cost-efficient upgrade kit for legacy CNC milling machines and using AI to predict the optimal point of tool change. That way even SMEs (Small and Medium Enterprises) can afford advanced analytics tools without investing in modern CNC machines. In the published paper, a practical solution is presented with a holistic hardware/ software setup, including an edge device and multiple sensors. In addition to that, the user interface visualizes the machine condition for the maintenance personnel on the shop floor for easy access to the data.
Read the full paper here