Deadlock-free Genetic Scheduling Algorithm for Flexible Manufacturing Systems with Key Resources
Huixia Liu,Weimin Wu,Xiaofei Han,Hongyong Yang,Wenhao Ma,Chuang Wang
DOI: https://doi.org/10.23919/chicc.2019.8866617
2019-01-01
Abstract:For flexible manufacturing systems with key resources, a deadlock-free scheduling policy is addressed to minimize the makespan based on Petri net models and genetic algorithms. In the algorithm, a chromosome has corresponded to a part sequence, that is, it is obtained by encoding the operation sequences of all jobs in the systems. The encoding method can guarantee each chromosome to satisfy the operation constraint and resource constraint. By the two-step look-ahead method in the optimal deadlock control policy, the control feasibility of a chromosome is checked and the amending method is proposed. Therefore, the final chromosomes can be feasible ones, that is, they satisfy operation-, resource- and control-constraints simultaneously. Finally, the effectiveness and feasibility of the proposed algorithm are illustrated by experimental results.