Study On Urban Mass Transit Network Planning Methodology Based On Rail Station

Recently, especially in the swollen cities, the amount of motor vehicles is increasing rapidly. while the construction of transportation facilities is relatively lagging.so traffic demands is bigger than supply。 As a result, various traffic problems relevant come out. such as aggravating traffic jam, frequently happened traffic accident, and traffic environment deterioration. Now A consensus between political circles and academia has emerged that the priority of the development of urban public transportation so as to ease traffic problems.Limited by the capacity of bus system, the construction of the rapid transit has become a development tendency in large cities, such as BRT. light rail, and subway metro, etc.The spatial form is the key point of network planning. The main research method in this area is to combine qualitative analysis and quantitative analysis, and qualitative analysis is used as the major method. Passenger volume forecast of rail net is an important part of quantitative analysis method. The four-step method is one of the widest applications in forecasting technology, has its own integrated theory system and technique。It has four stages namely:trip generation,trip distribution,model split and traffic assignment。This type of models has comparatively perfect theories, but suffers from a few drawbacks in practice. For the planning methods of spatial form, this article explores the key techniques of spatial form of rail network based on rail transit station so as to optimize the rail net using genetic algorithms and predict the passenger volume of rail transit station. Thus we can provide more quantitative analysis methods for the planning of network.The decomposition of spatial data based on the microcosmic unit is good for build the model related to accessibility. The actual data indicate that passengers at the rail station mainly come from walkers around the site. Based on the principle about accessibility, we can predict the overall passenger volume of rail transit station by forecasting the pedestrian volume around the sites。Based on this principle, this paper studied the methods of passenger flow prediction at the rail station by analyzing the accessibility of the rail station. In the case of Wuhan, the results showed that the forecast data is accurate comparatively.The optimization of network based on the rail station is favorable to ensuring the coverage percentage of the public transport station. Based on the volume of station, this paper studied the optimization of rail network using the genetic algorithm. Firstly, based on the principle of random process, we assessed the level of each station by computing the weight of each station, and the optimized nodes of higher level are seen as the candidate sites of the rail network.Secondly, under certain constraints, find out the K-shortest-paths between the terminal stops, and these paths are regarded as the set of candidate lines.Then, an integrated network is gained from the set of candidate lines by the application of genetic algorithms. In the case of Wuhan, we put this theory into practice using a program developed by our project group, the results showed that the forecast data of rail net is analogy to the real indicators of Wuhan urban rail transit network plan.