Classification Rules for the Job Shop Scheduling Problem with Machine Breakdowns

This paper addresses the job shop scheduling problem in the presence of machine breakdowns. In this work, we propose to exploit the advantages of data mining techniques to resolve the problem. We proposed an approach to discover a set of classification rules by using historic scheduling data. Intelligent decisions are then made in real time based on this constructed rules to assign the corresponding dispatching rule in a dynamic job shop scheduling environment. A simulation study is conducted at last with the constructed rules and four other dispatching rules from literature. The experimental results verify the performance of classification rule for minimizing mean tardiness.