Hybrid Nested Partitions algorithm for scheduling in job shop problem

Abstract

This paper introduces the main idea of Nested Partitions algorithm, and applied it to solve the job shop scheduling problem. In the algorithm the job shop scheduling problem is considered as a partition tree. The algorithm partitions the feasible region and concentrates the sampling effort in those subsets of feasible regions that are considered the most promising. Genetic algorithm search is incorporated into the sampling procedure, and use the sample points to estimate the promising index of each region. Computation experiments indicated that the hybrid algorithm outperforms the constructive GA search in goodness of searching.