自动化立体仓库的货位分配优化

摘要

To effectively solve the location assignment problem of Automated Storage and Retrieval System ( AS/RS) ,a two-stage multi-objective optimization model based on Polychromatic Sets ( PS) and Particle Swarm Optimization ( PSO) and Simulated An-nealing(SA) is proposed and realized so as to improve the stability of the shelf and efficiency of input /output.The first stage is to zone the rack area and the second stage is to assign goods location in each subarea .In the first stage ,according to the PS theory the problem is solved with considering the efficiency of input /output and the force situation of the shelves ,then use the PSO inte-grated SA to assign goods location in terms of the category and quantity of input goods in the second stage .An example running in the MATLAB shows that compared with Genetic algorithm (GA) and PSO,Hybrid Particle Swarm Optimization(HPSO) has fast speed of convergence and good stability ,and can increase storage efficiency under the premise of guaranteeing the stability of the shelf.%为有效解决自动化立体仓库（ AS／RS），即自动存取系统的货位分配问题，以货架稳定性和出入库效率为目标，结合多色集合、粒子群算法和模拟退火算法三者优势，建立区域划分、货位分配两阶段的多目标货位分配决策模型。区域划分阶段考虑货物出入库效率和货架受力情况，采用多色集合的围道布尔矩阵进行划分。货位分配阶段根据入库货物的类型和数目，采用结合模拟退火算法的混合粒子群算法求解货位分配优化问题。在MATLAB软件中运行实例，结果证明，与遗传算法和粒子群算法比较，混合粒子群算法在求解货位分配优化问题时的收敛速度快、稳定性高，且能在保证货架稳定性的前提下提高出入库效率。