|
As China's \Since traffic forecasts for the entire project design, construction, operation, driving organizations, and economic evaluation and other aspects have a huge impact, and Passenger Transport Corridor as a new mode of transport, it currently lacks specific prediction theory and methods. MD model is Japan's Transport Bureau of Investigation experts through long-term studies in the 1970s proposed a new mode of transport for traffic forecasting methods. Papers in the current domestic and foreign scholars have published research results, based on the MD model exploration and research, and further improve its theories and methods, and to establish a model of the traffic on MD forecasting process. Firstly, the current domestic and international rail passenger traffic forecasting theoretical methods were analyzed and summarized, such as time series, forecasting models random time trend forecasting methods, factors to consider passenger volume regression analysis, econometric models and other theories, as well as widely used at home and abroad of the four-stage method; discusses neural networks, logit model, gray theory, support vector machines and other rail traffic in recent theoretical prediction specialized applications, and analyzes the scope of application of various methods and advantages and disadvantages. Secondly, we discussed the railway passenger traffic trends and influencing factors. Changes in the structure according to the current transport, passenger dedicated lines studied structural characteristics of the train vehicle characteristics, economic and technical characteristics and the traffic characteristics of the formation mechanism and structure, and according to the nature of its transportation services, analyzes Passenger transport channels in the whole and other the relationship between transport modes and functions for the passenger line of the traffic forecast has laid a theoretical foundation. Then MD model by studying the basic principles of the process of its implementation to identify the difficulties and shortcomings, and its improvement and refinement: passenger traffic by analyzing the relationship with potential passenger demand, passenger demand to establish the potential for solving OD Methods: The model is presented in the MD generalized travel time and generalized travel cost concept, time-consuming travel, travel costs, travel comfort, the probability of accidents, transportation and many other factors into punctuality travel expense calculation; introduced the GDP-based method for determining the time value of resources; analyze trends in traffic, induced traffic growth and metastasis volume relationship in the MD model to further refine and improve the MD model theory and methods, to establish a complete set of MD model forecasting process. MD with unmodified model and logit model predictions comparative analysis proved that the improved MD Passenger traffic forecasting model reliable and rationality. Finally, the model of MD thesis on a number of points and limitations made a note, and directions for further work are briefly discussed.
|