基于模糊聚类和识别的出租车合乘算法研究

发布时间：2018-04-27 02:06

本文选题：交通工程 + 出租车合乘　；参考：《交通运输系统工程与信息》2014年05期

【摘要】：针对目前部分大城市出租车合乘效果差,合乘效率低等现状,本文采用模糊聚类和模糊识别方法,研究出租车行驶路线模糊聚类,并利用行驶路线、行驶时间和合乘人数创建隶属函数,实现合乘乘客与出租车的合乘模糊识别.通过随机生成的多组出租车出行和合乘乘客样本数据,发现在假定的出租车合乘条件下,出租车样本数量决定了合乘的成功率,但同时也发现,在合乘人数固定的情况下,无限制的增加出租车样本数量会增加合乘乘客的搭载成功率,平均每辆合乘出租车的收入并不会随着样本数量的增大而增大,而是趋于稳定值.仿真结果说明,该算法适合于大样本的出租车合乘问题,是一种可以提高出租车合乘成功率的有效方法.
[Abstract]:In view of the bad effect and low efficiency of taxi ride in some big cities at present, this paper adopts fuzzy clustering and fuzzy recognition method to study the fuzzy clustering of taxi driving route, and uses the driving route. The driving time and the number of passengers create a membership function to realize the fuzzy recognition of the combined passenger and taxi. By randomly generating multiple sets of taxi travel and passenger sample data, it is found that the number of taxi samples determines the success rate under the assumption of taxi co-rider, but at the same time, it is also found that, under the condition of fixed number of co-passengers, the number of taxi samples determines the success rate of co-rider. Unlimited increase in the number of taxi samples will increase the passenger ride success rate, the average income per taxi will not increase with the sample size, but tend to a stable value. The simulation results show that the algorithm is suitable for large sample taxi ride-sharing problems and is an effective method to improve the success rate of taxi ride-sharing.
【作者单位】：兰州交通大学交通运输学院;兰州交通大学经济管理学院;
【基金】：国家自然科学基金(61364026,61164003) 兰州交通大学青年基金项目(2011044)
【分类号】：U491

【参考文献】