2020 Volume 48 Pages 42-46
In our training factory, there are manufacturing education programs for engineering students. Though the supporting staff teaches manufacturing methods and machine tool operation, it is getting more difficult to maintain the educational quality because there are not enough staff to complete all tasks. Since it is hard to keep making the rounds in the factory, a system which detects when a student needs educational support while working, would be useful. In this paper, to address this problem we have developed a system which automatically detects when students need advice about machining operations and/or machining processes and sends the notification to staff's smart watch. We show that the presented neural network technique determines suitable timing for providing targeted assistance from technical staff members. The timing is based on machining proficiency data derived from measurement of electric power consumption of the target machine. When the neural network estimates low proficiency in machining operation in the experiment, coaching and training education of the student is necessary which demonstrates that the presented system can detect suitable timing to give students educational advice.
