EVOLUTION BASED LEARNING IN A JOB-SHOP SCHEDULING ENVIRONMENT

A class of approximation algorithms is described for solving the minimum makespan problem of job shop scheduling. A common basis of these algorithms is the underlying genetic algorithm that serves as a meta-strategy to guide an optimal design of local decision rule sequences. We consider sequences of dispatching rules for job assignment as well as sequences of one machine solutions in the sense of the shifting bottleneck procedure of Adams et al. Computational experiments show that our algorithm can find shorter makespans than the shifting bottleneck heuristic or a simulated annealing approach with the same running time.