Research On Rail Transit Passenger Route Choice Question Based On MNL

With the development of our country'economy, the urban traffic question is serious day by day. Urban rail transit is becoming the best way to solve the question. Along with the country to the investment of infrastructure enlarging, our country' railway transportation enters the high speed development time. But the research on the aspect of rail transit passenger flow is quite deficient in our country. This thesis has conducted the correlation research on the rail transit passenger path choice, also has obtained the corresponding research technique and research conclusion. This thesis has the vital significance of grasping passengers' rules of the whole urban transit network, organizing passenger flows and allocating the transportation capacity reasonably for the urban transit managers and operators. At the same time, theoretical basis is provided for the forecast of the urban spatial distribution of traffic after Large-scale construction of urban rail transit.On the basis of reading large of literatures at home and abroad and studing on the urban traffic planning theory in depth, the disaggregate model based on individual travel behavior in traffic planning is introduced into the rail transit path of choice .This thesis obtained the factors influencing passengers' route choice of urban rail transit through the traffic survey as well as the experience and the general knowledge, and has carried on the quantification and the analysis to each influence factor. In order to obtain accurate urban rail transit passenger flow assignment models that have a goodness of fit index, this thesis divides the influence factors into the fundamental factors, transfer factors and other factors taking each influence factor as a foundation. On the basis of fundamental factors, the initial model has been established. Then transfer factors and other factors were considered gradually. Through adding and subtracting the influence factors unceasingly, I obtained two models which have a goodness of fit index using the simulation process of establishing model - examining model - establishing the model again repeatedly.Finally, taking the urban rail transit network of one city as the example, I obtained all the effective ways between all the OD of the network using the improved Dijkstra algorithm and the chart traversal algorithm. Then the utility of each effective way was calculated, and passengers were assigned into each effective way through the two obtained model C and model D. In the end, the calculation result and the investigation result were compared so that we can confirm the two model's accuracy and feasibility. Also, we draw the conclusion that the path choice of urban rail transit has the relation of in-vehicle teavel time, numbers of transfer, transfer walk time, length of the path and stops. At the same time, passengers have different sensitivity of the influencing factors.