摘要
为提高铁路部门服务质量和铁路旅客出行体验,并制定可满足旅客需求多元化的旅客列车开行方案,考虑坐席动态分配对旅客列车开行方案进行研究。从旅客和铁路运营部门2个角度出发,分别以旅客动态乘车广义时间最短和铁路收益最大为目标,以客流守恒、区间通过能力和满足客流需求等为约束建立多目标规划模型。通过Logit模型确定硬座、硬卧、软卧这3类坐席的分担率,确定列车定员数,并在求解过程中依据生成的开行方案不断更新3类坐席的分担率以实现3类坐席的动态分配,直至结果趋于稳定。结合算例采用带精英策略的非支配排序遗传算法(NSGA-Ⅱ)求解,并进行算例分析,计算结果表明:建立考虑坐席类型动态分配的旅客列车开行方案,在实现旅客对坐席选择的同时,铁路旅客服务率提高了3.5%,铁路部门收益增大了1.5%。
Dynamic seat allocation is considered to formulate a plan for passenger train operation meeting the diversified needs of passengers,thus improving the service quality in the railway departments and the travel experience of railway passengers.From passengers and railway operation departments,this paper takes the minimum generalized travel time of passenger dynamics and the maximum railway revenue as the objectives and the conservation laws of passenger flow,interval through ability,and passenger flow volume demand as the constraint conditions.Besides,the multi-objective planning model is established.A Logit model is used to obtain the fixed number of passengers and the share rates in the hard seat,hard berth,and soft berth.In the solving process,the share rates in various seats are constantly updated by the generated operation plan to realize the dynamic allocation of the three categories of seats until the results are obtained.The example is solved using the non-dominated sorting genetic algorithm with elite strategy(NSGA-Ⅱ),and then the analysis of the example is carried out.The results show that the establishment of the plan for passenger train operation considering the dynamic allocation of seat types has increased the passenger-railway service rate by 3.5%and the railway’s revenue by 1.5%,realizing the passenger’s choice of seats.
作者
代万清
杨信丰
DAI Wanqing;YANG Xinfeng(School of Traffic and Transportation,Lanzhou Jiaotong University,Lanzhou 730070,Gansu,China)
出处
《交通信息与安全》
CSCD
北大核心
2021年第4期134-142,共9页
Journal of Transport Information and Safety
基金
国家自然科学基金项目(71761024)资助。
关键词
交通规划
旅客列车
坐席动态分配
多目标规划
开行方案
非支配排序遗传算法
transportation planning
passenger train
dynamic seat allocation
multi-objective programming
train operation plan
non-dominated sorting genetic algorithm