摘要
为克服传统客流分配方法在旅游轨道交通规划应用中的局限性,提出MCA(基于马尔可夫链的旅游轨道交通客流分配)模型。旅游出行满足马尔可夫链“无后效性”,通过站点转移矩阵和站点选择矩阵运算求得路径选择概率,从而实现OD(起讫点)量在区间上的客流分配。以五台山风景名胜区实际案例为研究背景,对旅游阻抗函数进行多元线性回归分析,剔除1个非显著变量后计算其他5个显著变量参数取值,并对分散度系数进行敏感性分析;分别计算本模型以及AON(全有全无)、UE(用户均衡)和SUE(随机用户均衡)客流分配模型在各区间的客流分配量;对比4种模型的计算精度和效率。平均相对误差计算结果的优劣排序依次为MCA,UE,SUE,AON,均方根误差计算结果的优劣同上。MCA的计算效率最高。研究结果表明,所提模型更适用于旅游交通场景。
To overcome the limitations of applying conventional passenger flow assignment methods in tourist rail transit planning,a MCA model(tourist rail transit passenger flow assignment based on Markov Chain)is proposed.Tourist travel process satisfy the non-aftereffect property of Markov Chain,station transition matrix and selection matrix are built for the calculation of routes selection probability,so as to accomplish OD(origin-destination)flow assignment to each interval.Taking the practical case of Mount Wutai as research background,multiple linear regression analysis of tourist impedance function is carried out.1 non-significant variable is eliminated and parameters of other 5 significant variables are estimated,and sensitivity analysis of disperse coefficient are conducted.Passenger flow assigned to each interval are calculated by using MCA,AON(all-or-nothing),UE(user equilibrium)and SUE(stochastic user equilibrium)respectively.By comparing computational accuracy and efficiency of the 4 models,the mean relative error calculation results are ranked from best to worst as MCA,UE,SUE,AON,and the root-mean-square error calculated results are ranked the same as above.MCA has the highest calculation efficiency.Research results show that the proposed model is more suitable for tourist rail transit scenario.
作者
董皓
王何斐
雷佳祺
DONG Hao;WANG Hefei;LEI Jiaqi(Beijing Public Transit Tram Co.,Ltd.,100080,Beijing,China)
出处
《城市轨道交通研究》
北大核心
2022年第9期38-44,共7页
Urban Mass Transit
关键词
轨道交通
旅游交通
客流分配模型
rail transit
tourism transportation
passenger flow assignment model
