Urban transportation demand prediction method based on POI

Abstract

The invention discloses an urban transportation demand prediction method based on a point of interest (POI) of a map. According to the method, POI data in an urban space is classified and divided according to a two-stage classification system, then the value of each factor in a traveling generation capacity factor list is generated, and the values are subjected to normalization processing; then the weight of each factor in the capacity factor list is calculated, wherein the weights include occurrence capacity weights, absorption capacity weights and corresponding normalization weights of the factors; finally, the traveling generation capacity index of the POI is calculated according to the normalization values and the corresponding normalization weights of the factors in the factor list. Real-time updated network open source data is used, and the traveling generation capacity index is calculated according to the position and class of the POI data and other attributes, so that the accuracy of urban traffic demand prediction is greatly improved, and the prediction operation process is simplified.

Description

technical field [0001] The invention belongs to urban traffic transportation planning and management technology, in particular to a POI-based urban traffic demand forecasting method. Background technique [0002] The amount of travel generated, also known as travel demand, reflects the travel intensity of residents' social activities in a certain area in the field of urban transportation. The result of travel demand prediction is the key decision-making basis for determining the scale of urban transportation infrastructure construction, and it is an important content that needs to be measured and calculated in the practice of various urban transportation planning projects. [0003] In terms of traffic demand prediction, the existing technology adopts the "four-stage" prediction method, which is to predict the travel volume of local plots based on the urban land use status and the travel generation rate of different types of land, and then further integrate it Calculate the ...

AI Technical Summary

Problems solved by technology

[0004] 1. The accuracy of this aggregation method depends entirely on the accuracy of the land use data used, and the urban land use status is dynamically changing, so it is difficult to obtain data that accurately reflects the land use status in practice;

[0005] 2. In the actual operation process, it is necessary to measure the size of all types of plots in the target area, and the workload is cumbersome and arduous

[0006] In addition, the activity-based prediction method proposed on the basis of the land use data prediction method can greatly improve the prediction accuracy, but this method fails to reflect the individual travel activities from the essence of land use. At the same time, the individual travel The acquisition of activity chain data is also a complex and difficult process, so this method is also difficult to apply in project practice

Images