Study On Optimization Method Of Public Transport Network In Small And Medium-sized Cities Based On Web Open Source Data

The rationality of public transport network plays a decisive role in the development of public transport,and the theme of studying urban traffic problems using a large amount of non-aggregated web open source data is unfolding.The paper based on that data to build a public transportation network optimization method for small and medium sized cities.Not only can it provide the basis for operations,but it can also open up new ideas for traffic applications in the new era.Firstly,through the analysis of the process of network open source data generation and development,the significance of traffic data in network data is explained.Further extend the POIs,reviews,and Weibo check-in data for this study,and analyze the Python-based web crawler framework,that provides a solution to acquire the network open source data.Secondly,preliminary methods such as culling,cleaning,transformation and unification of coordinate systems were adopted for data processing.At the same time,to solve the contradiction between POIs classification and traffic applications,combining the purpose of travel and the theory of urban land classification,construct POIs Reclassification System Based on traffic applications.Based on multiple sources such as reviews and city statistics,subjective AHP and objective entropy method are combined to determine the weights of various types of POIs.Thirdly,put forward the “standardized density” indicator of POIs in traffic areas to determine the strength of their passenger flow,and combined with layout conditions to determine the optional bus stop.calculate the proportion of combinations of weighted POIs and identify urban functional areas with the reclassification system.The DBSCAN method is used to analyze the resident sites,in combination with the characteristics of land use and POIs,the resident sites are identified as work sites,residences,and other sites,and build three kinds of travel models for small and medium-sized cities based on Weibo check-in data.estimate origin-destination matrix by Python.Furtherly consider the limitations of network data such as Weibo,direct expansion the OD matrix based on scale and age structure,and adjust OD matrix based on OD inverse model.Fourthly,build a strong passenger flow contact layer with the goal of reachability,and build a non-strong passenger flow contact layer with the goal of user optimization.In order to meet the tiered optimization goals,respectively adopt all-nothing and user-balanced methods to achieve passenger flow distribution.The initial public transport network to be evaluated is laid out in three levels: macroscopic,meso-,and microscopic,and consider optimization goals again,accessibility evaluation using technical specifications for network density,site coverage,and average transfer coefficient,user opportunity evaluation by using the ratio of various types of POIs,iterative optimization of public transit network with seven optimization techniques,and then build the final optimized network.Finally,take Xingyang in Henan Province as an example,the validity of the research method was verified,and explained the implementation of research methods detailedly,get a better result for public transportation network optimization application.