Traffic Demand Forecasting Model Of Economic Region Base On Density-based Clustering Analysis

The development of China’s economic region along with the improvement of the level ofurbanization and improvement in the level of regional economic integration, thus thedevelopment of the regional transport also put forward higher requirements, the study ofregional transportation needs are still many deficiencies, it is difficult to meet the travel needsof the growing inter—regional, regional traffic demand forecast is different from land—usedata and residents travel survey of urban transport demand forecasting method can not followthe traditional use of socio—economic data regression analysis to determine the regionaltransportation needs.Paper by external traffic within the economic region on central city district division, usingcluster analysis on how to to merge external traffic area division and regional transportationmodel and the model for urban traffic data are separated by the community and the use of howto dock is discussed, so as to adapt to economic zone traffic model is modeled as a goal,probes into the method is suitable for the data to construct the economic.Paper established a regional traffic demand database, the use of density—based clusteringanalysis, the center of the economic zone outside the city traffic division of the cell, select theindicators of external transport outbound traffic as the main basis of cell division: theoccurrence of passenger and cargo flows amount to attract volume, passenger and cargo flowsEnds, hub area, transportation and so on. Application of this algorithm can be divided inforeign traffic zone to take full account of the temporal and spatial distribution characteristicsof external traffic.Paper—based feedback mechanism of the "window" docking method to establish a modelof urban transport can be embedded regional models. By this method the external point of thewindow determined by an external point of implementing the exchange of data betweenmodels, while the use of the feedback mechanism, the formation of the combination oftop—down and bottom—up cell division correction method to achieve for external trafficarea was further optimized, application of this method to make traffic zone division of datainto account both the inheritance and the realization of the regional traffic forecastingtargeted.In this paper, the above methods are Guanzhong—Tianshui Economic RegionTrafficDemand Forecast Modeling analysis, Xi’an, using the existing model data on cluster analysisto optimize the cell, while the Guanzhong—Tianshui Economic Region travel demand model using optimized based on feedback mechanisms window docking method, through theregional travel demand model for road network and community feedback correction features,so as to achieve economic zone and the central urban transport traffic model consistent withthe model.