Operational Optimization Strategies And Simulation Of Flex-route Transit Service

With accelerated urbanization and economic growth over the past few decades,residential communities with low population densities in suburban and rural areas have increasingly developed.The low and unstable travel demand in these areas has led to the low efficiency of traditional fixed-route transit systems,which are largely dependent on a concentrated demand,with a lack of flexibility in terms of route scheduling.As a result,an increasing portion of people must rely on private automobiles as their main travel mode,which increases congestion and pollution in urban areas.To encourage more people to utilize public transit,new forms of transit systems are being developed to meet the new travel patterns in these areas.Flex-route transit,which combines the low-cost operability of fixed-route transit with the flexibility of demand responsive transit,is one of the most widely used flexible transit services and can be used to meet the diversified travel needs in these low-demand suburban areas.However,although flex-route transit services have many years of operation experience in foreign developed countries,there are still many problems that needed to be solved in both theoretical research and practical operation.That is why the development of this innovative transit service is far below expectation.On the other hand,due to the different socioeconomic contexts,travel demand characteristics and transportation networks structures,the existing successfully flex-route transit operation experience in other developed countries can hardly be applied directly in China.In view of the above problems,this paper intends to study the operational optimization strategies and simulation of flex-route transit service with the hope of improving its level of service and reliability.The main research contents of this paper are as follows:Firstly,by analyzing the operating characteristics of flex-route transit systems,we discuss the feasibility,necessity and challenges before its implementation in China.Then,we conduct a stated preference survey on the residents in the suburban area of Nanjing city.The questionnaire consists of four sections,including stated choice experiments,socioeconomic characteristics,attitudes towards the current transit system,and the service design preferences of flex-route transit.Three discrete choice models,namely,multinomial logit model,nested logit model and panel mixed logit model,are applied to describe the mode choice process using the data collected from the survey.Based the estimation results of the models,the main factors that influencing the choice of flex-route transit are analyzed and the potential users are identified.Besides,the values of walking time,waiting time and in-vehicle time for different travel modes are estimated.We also predict the sharing rate of different travel modes after changing some key parameters through elastic analysis.The above analysis results can be used to assess the potential demand and passengers’ service design preferences prior to the deployment of flex-route transit service in China.Secondly,to reduce the number of rejected passengers and idle time at checkpoints resulting from uncertain travel demand and the limitation of predetermined slack time in operation,a slack arrival strategy and a meeting point strategy are proposed,respectively.The two strategies can inject more flexibility into the system by relaxing the departure time constraints of the checkpoints and space constraints of the pick-up and drop-off locations.By constructing the theoretical model,simulation model and vehicle routing model of flex-route transit,the benefits of introducing the two strategies are investigated.Experiments over a real-life flex-route transit service show that the proposed two strategies could greatly improve the system reliability and level of service,the rejection rate and system cost can be reduced by up to 63% and 40% respectively with no additional operating cost.Thirdly,to improve the reliability,responsiveness,and productivity of the flex-route transit service,the vehicle scheduling and routing problem under a dynamic operating environment is investigated.First,we discuss the new operating polices after the introduction of intelligent transportation systems(ITS),including automatic vehicle location(AVL)system,mobile data terminals(MDTs)and a computer-aided dispatch(CAD)system.Then,corresponding on-line scheduling schemes are presented to tackle different dynamic events,including dynamic requests,cancellations of requests,customer no-shows and travel time fluctuations.Simulation experiments are conducted to assess the influence of different dynamic events.The results demonstrate that the proposed operating polices and scheduling schemes based on ITS technologies are reliable for coping with various dynamic events,which can greatly improve the system reliability during the real-time operation of flex-route transit service and reduce the operation cost for the operators.Finally,based on the agent modelling theory,the basic framework of the multi-agent based flex-route transit simulation system is constructed,including the detailed design of passenger agent,vehicle agent,dispatching center agent and their communication interaction process.On this basis,a prototype simulation system is developed using JADE platform and Java language,which can be used to provide decision support for transit operators in the system design process before the service implementation and evaluation process after the service implementation.This paper studies the operation optimization strategies of flex-route transit from five perspectives,namely,feedback of the stated-preference survey,slack time redistribution between checkpoints,dynamic allocation of pick-up and drop-off locations,dynamic routing and scheduling of operating vehicles,and simulation system construction.We believe that our work have some important practical values for improving the service level and reliability of flex-route transit and can help guide the service design and decision-making for operators before or after the actual operation.Furthermore,it could also promote the development of flex-route transit service,reinforce the public transport system,and alleviate the growing traffic congestion problem in China.