提出一种能在两类“货到人”订单拣选系统(即自动存储及取货系统(Automated storage and retrieval system, AS/RS)和Carousel系统)之间依据不同订单类型进行系统适用性选择的系统优化方法。订单拣选系统最常用的两种作业策略分区策略和分批策略本质上都是对订单的聚类，因此订单可以被分为很多一维单元化网格；根据各系统的逻辑运动关系，得到单元化网格的拣选时间表达式；利用作业时间模型和空间填充线，对单元化网格进行聚类，得到二维系统的总订单拣选时间，采用遗传算法计算各系统的最优值；比较各系统的时间选择适用性，即选择总订单拣选时间更小的系统。通过对四类不同类型的订单进行试验研究，初步得出两类系统对四类订单的适用性分析和系统参数敏感性分析。

A systemic method is presented to choose the suitable order picking system between automated storage and retrieval system (AS/RS) and Carousel system for different types of customer orders. As both batching and zoning, the two frequently-used operational policies, are essentially order clustering, the customer order sheet can be divided into many unit grids. After the time formulation for each system in one-dimensional unit is defined according to the logical movements, a time sequence models for two-dimensional systems is presented. The genetic algorithm is adopted to search the optimal value for each system. In the experimental study, not only the suitability of each system for given orders with the hypothetic value of parameters by comparing the order picking time is established, but also the effect of some key factors on suitability is discussed such as zone number and batch size.