Traffic travel characteristic data extraction method based on big data

Abstract

The invention discloses a traffic travel characteristic data extraction method based on big data, and belongs to the field of traffic big data analysis application. Mobile phone data act as the core of the big data, and GPS data, coil data and video data act as the auxiliary of the big data. Traffic travel characteristic data include OD data and job-housing ratio data. The main steps are listed as follows: 1) the mobile phone data, the GPS data, the coil data and the video data are acquired and the mobile phone data are preprocessed; 2) traffic cell allocation is performed through combination of the geographic attributes of administrative regions and base stations; 3) traffic travel chain identification is performed through combination of the multi-source data with the mobile phone data acting as the core, and passing points and stop points are identified; 4) inter-cell OD result output is performed through combination of the mobile phone data and demographic data; and 5) all the traffic travel chains within a week are analyzed, residential places and working places are judged through combination of turn-on and turn-off data and conversation data, and job-housing analysis is performed. Wide range of urban traffic travel characteristic data acquisition can be completed in a short period of time by the traffic travel characteristic data extraction method based on the big data.

Description

technical field [0001] The invention relates to a method for extracting traffic travel characteristic data, in particular to a method for extracting traffic travel characteristic data based on big data, which belongs to the field of analysis and application of traffic big data. Background technique [0002] In recent years, with the rapid development of my country's economy and the vigorous development of urban infrastructure construction, changes in the nature of land use have also accelerated. With the application of various advanced transportation tools and various information-based traffic management methods, traffic infrastructure and traffic operation modes are changing rapidly. In this case, the method of obtaining residents' travel characteristics data through traditional residents' travel surveys cannot meet the needs of traffic planning and management in the new era, no matter in terms of economy, accuracy of results, and timeliness. Therefore, there is an urgent ...

AI Technical Summary

Problems solved by technology

[0007] 1) The basic unit of traffic travel analysis is not established in combination with the characteristics of various data, and the traditional traffic area division method is still used. The mobile phone data analysis is used to divide the traffic area, and the demand for traffic analysis is insufficiently considered. Compared with the traditional survey method, in The advantages of high frequency and low cost are obvious, but there are deficiencies in data availability

[0008] 2) Without considering the shortcomings of single-source mobile phone data in terms of low accuracy and large screening errors, the obtained travel chain identification results are at the traffic district level, which cannot meet the needs of road traffic allocation in traffic travel analysis

[0009] 3) The mobile phone data cannot directly correspond to the travel analysis of the actual population, and needs to be processed by a correlation analysis model. However, after simple expansion of the mobile phone retention rate, there are large errors

[0010] 4) According to the judgment method of work place and residence place, there are not many factors to consider the actual work and residence behavior. For example, when judging the place of residence, only the night time is set. During this time period, the number of signaling pieces of a certain base station cell The proportion is greater than 50%, which is the place of residence. Although this method is simple and easy to operate, it does not link the event types in the mobile phone data with people's actual behavior, and cannot further mine the information of the place of work and place of residence.

Images