Simulation of random tool lives in metal cutting on a flexible machine

This paper describes some numerical experiments related to a tool management model for a flexible machine equipped with a tool magazine, variable cutting speed, and sensors to monitor tool wear, when tool life due to flank wear is stochastic. A computer simulation was performed where decisions about tool loading and cutting speed were based on a deterministic mathematical programming model in which tool setup times are added up to total processing time whenever a tool is required but absent from the tool magazine. Two types of sensor systems are presented: offline sensors and online sensors. It is assumed that the sensor only gives information about whether or not the tool is in good condition to continue processing. The simulation aims at answering six questions: (1) which statistical distributions should be used to simulate the life of a cutting tool? (2) How effective is a deterministic model if tool lives are stochastic? (3) Is the use of a stochastic model still justified when tool life variability decreases (i.e. tool quality and reliability increases)? (4) Does adjusting cutting speed while processing a given part type help improve productivity? (5) If the machine must be stopped to inspect tool conditions, what are the best parameters to use, i.e. inter-inspection times and threshold? (6) How effective are the two sensor systems?