摘要
本文根据有序样本聚类的Fisher算法 ,给出一种峰值曲线的优化方法 ,通过该方法我们得出了上行客流峰值为 5个 ,其峰值区间为 :5 :0 0 6 :0 0 ,6 :0 0 9:0 0 ,9:0 0 16 :0 0 ,16 :0 0 18:0 0 ,18:0 0 2 3:0 0 ;下行客流峰值为 5个 ,其峰值区间为 :5 :0 0 7:0 0 ,7:0 0 9:0 0 ,9:0 0 16 :0 0 ,16 :0 0 19:0 0 ,19:0 0 2 3:0 0。 然后 ,依据峰值区间建立确定发车间隔的算法Ⅰ模型和算法Ⅱ模型 ,对两种算法模型计算结果进行比较分析 ,得出结论 :两个间隔高峰类时间段用算法Ⅰ进行求解 ,其余 3个类时间段用算法Ⅱ进行求解。在各个时间段结合处用光滑法进行优化处理 ,并以处理后的数据为基础制定出两个起点站的发车时刻表 ,并求出全线共需要 47辆车 ,乘客对方案的满意程度为 98 2 % ,公交公司的满意程度为 76 2 3%。 最后 ,运用随机服务系统的相关理论建立随机规划模型 ,给出概率灵敏度和误差分析 ,进而得出采集运营数据的较好方案。
This paper presents an optimal method of peak curve according to the Fisher algorithm of serial specimen clustering. We conclude 5 uphill passenger flow peak ranges: 5:00 6:00, 6:00 9:00, 9:00 16:00, 16:00 18:00, 18:00 23:00, and 5 downhill passenger flow peak ranges: 5:00 7:00, 7:00 9:00, 9:00 16:00, 16:00 19:00, 19:00 23:00. Then, under the peak ranges, two algorithm models, Ⅰ and Ⅱ, are established. Comparing the calculated results of the foregoing models, we conclude: model Ⅰ is applied to two interval high peaks, and model Ⅱ is applied to three others. With the smooth method between every two time sections, we make the bus time schedule of two starting stations, and get 47 needed buses at least. In this scheme, passengers' satisfaction rate is 98 2%, and the bus ccompany's is 76 23%. By the end, we set a random optimum model by the theory of random service system, and give the probability sensitivity and error analysis. Further, we get a better scheme for collecting operation data.
出处
《工程数学学报》
CSCD
北大核心
2002年第F02期67-74,共8页
Chinese Journal of Engineering Mathematics
