Data mining application for real-time distributed shop floor control

The classification model based on a decision tree representation offers a solution to the real-time scheduling, sequencing, and routing of jobs on various machines at a manufacturing shop floor. In the paper, a decision tree classifier is induced from data generated from simulation runs of a real-time distributed shop floor control system. The simulation evaluates the various possible states of the shop floor, using information about the existing machines, the expected waiting time in the queues, the number of jobs waiting in each queue, the throughput rate, cycle time, and the average earliness and tardiness. The decision tree classifier develops a description or model for each type of machine behavior, which can be used to select the best machine behavior with respect to the machine´s goal