Abstract: To avoid the decrease of solution set of cellular layout problem by solving intra-cell and inter-cell layout problem separately, a multi-objective integrated optimization model was constructed. The problems in layout such as intra-cell facility layout, inter-cell facility layout and orientation of facility were described simultaneously, and the research space of problem solution was exptended. Aiming at the complexity of proposed model and the weakness of classical Particle Swarm Optimization(PSO)to fall into local optimal, a structured real-code Cooperative Particle Swarm Optimization(CPSO)was proposed. The gene bank of each sub-group was constructed with CPSO, and genetic operation was used to deal with particles in gene bank. Thus the search ability of proposed algorithm was improved. The example of cellular layout solution was applied to demonstrated the effectiveness of model and algorithm.

Key words: cellular layout, multi-objective integrated, particle swarm optimization, co-evolution

摘要： 为避免单元间布局和单元内设施布局分开孤立研究所导致的问题解空间损失，建立了多目标集成优化模型，对布局中涉及的单元间布局、单元内设施布置、设施摆放方向三个问题同时进行描述，扩大问题解的搜索空间。针对模型的复杂性和传统粒子群算法容易陷入局部最优的弱点，设计了基于结构化编码的多粒子群协同进化算法，在多个粒子群协同进化的同时，对各子群构建基因库，并对基因库粒子进行遗传操作，提高了算法的寻优能力。通过求解单元布局问题实例，验证了模型和算法的有效性。

关键词: 单元布局, 多目标集成, 粒子群优化