Abstract: The large amount of products and complexity of spatial layout in supermarkets make customers easy to get lost in the room, spending much time searching goods. This paper proposes a genetic algorithm-based shortest shopping guide route recommendation method in supermarkets. The spatial layout structure of supermarket is discretized into a graph, in which nodes represent product zones, and undirected edges represent the walk route between two adjacent zones. The goods on the shopping list which one customer wants to buy are mapped into specific zones based on the shelf-locations they belonged to. The shortest path combining the entrance, the goods zones and the check-out counter is optimized by genetic algorithm. The simulation results in Matlab environment show that the method is simple and effective, and can quickly recommend a shortest path to customers for their shopping decisions.

Key words: supermarket shopping guide, environment modeling, shortest route, genetic algorithm, personalized recommendation

摘要： 大型超市里商品数目的繁多、空间布局的复杂往往容易让消费者迷失在室内，花费大量时间进行商品搜寻。提出基于遗传算法的超市导购路径推荐方法，通过对超市的空间布局结构离散化建模，生成分别用节点和无向边表示商品区域和区域之间可行走路线的平面图；结合消费者的采购清单，根据商品所在的货架位置将商品与具体的区域做出匹配，用遗传算法优化生成一条联结超市入口、要采购的商品区域以及结账柜台的最短路线。Matlab的仿真结果显示，该方法简单、高效，能够快速为消费者推荐出最短路线，供其购物参考。

关键词: 超市导购, 环境建模, 最短路径, 遗传算法, 个性化推荐