Machine Health-Driven Dynamic Scheduling of Hybrid Jobs for Flexible Manufacturing Shop.

In the multi-type & small batch production mode, jobs of small quantity (i.e., hybrid jobs) are implemented in the job shop. The production scheduling is a significant activity and it is necessary to predict the disturbance in advance. Some methods and tools to tackle the production scheduling have been provided. From the perspective of duration, our work extends the method of dynamic scheduling with incorporation of machine health prediction. The primary result of this work is the efficient generation of feasible scheduling solution when the machine health is warned. The machining quality data-based method is proposed to predict the machine health status and the relation between machine and quality characteristic is established. Combination of K-means, data equalization algorithm and particle swarm optimization (PSO) is designed to predict the machine health. Mathematical model in terms of duration is proposed and improved genetic algorithm (GA) is applied to generate the feasible scheduling solution. A prototype system is developed and a case study of a metalworking workshop is implemented. The results show that the work can reduce the risk of machine health to production. Using the prototype system, the engineers can filter out infeasible scheduling solutions automatically and acquire a successful one efficiently. [ABSTRACT FROM AUTHOR]