面向高效精准柔性混装作业的智能元胞调度方法研究

柔性混装生产多目标调度系统的动态性、开放性和大规模性导致常规算法失效,因此迫切需要研究新的调度算法。在推理其他算法局限性基础上提出使用Agent形成智能元胞自动机ICA使之具有自学习功能,并基于装配投产序列调度和工位任务分派目标,首次建立智能元胞调度ICAS三维网络模型及构造“弹簧邻居”特殊邻域和“弹跳”和“层跃”特别行为,实现混装作业微观自适应调度。仿真实践证明,该智能元胞调度ICAS方法可靠,计算复杂性降低,为复杂调度问题提供了新手段。

Intelligent cellular scheduling method for high efficiency and punctuality in mixed model assembly line

New algorithm for multi-object production scheduling of flexible mixed model assembly line(FMML) is in dire need,since the conventional algorithm fails in the face of the dynamic,opening and large-scale characteristics of scheduling system.In light of the limitations of the conventional algorithm,intelligent cellular automata and multi-agent are harmonized into a special structure named intelligent-cellular-automata(ICA) with self-learning ability.Oriented towards the detailed goal of concurrent scheduling sequence planning and workstation assignment,3D intelligent cellular automata scheduling(ICAS) network model and special neighborhood of "spring neighbor" and special behavior of "hopping in a certain layer" and "leaping among layers" are constructed for self-adapting scheduling.The simulation proves that ICAS is credible for complex scheduling system and the computation complexity has been reduced.