Research On Expressway Travel Time Reliability Based On Trajectory Data Gridding Method

As the pace of modern life continues to accelerate,travelers pay more attention to the value of time,and travel delays and reliability are increasingly valued.Travel time reliability has always been a traffic parameter that researchers are concerned about.As an important indicator of urban road performance evaluation,travel time reliability can not only measure the service level of urban road facilities,but also directly evaluate road users' travel plans.With the comprehensive development of digital information technology,trajectory data has become the most commonly used data input source in travel time related research fields.When using trajectory data as a data source,it is often necessary to use map matching techniques to match trajectories to digital maps.The premise of this is to master map matching and geographic information system(GIS)technology,and to provide a digital map that matches the trajectory.However,the difficulty of digital map acquisition,the difficulty of GIS technology,and the"time-consuming"nature of map matching in the context of big data all restrict the use of trajectory data by researchers who do not focus on GIS technology.Aiming at the problems of the above map matching technology,this paper proposes a method of trajectory data gridding.The method does not require a digital map,and matches the trajectory data to the road to be studied with a grid index,and obtains travel time samples inside each grid as a data source.The gridding method has no complicated algorithm and can process a large amount of trajectory data at a relatively fast speed.In the context of big data,it has very important significance.Based on the gridding method,this paper aims at the reliability evaluation and analysis of expressway travel time,and carries out data processing and data analysis research.The main work and achievements are as follows:(1)A trajectory data gridding method is designed.The meshing method refers to matching a continuous trajectory of a single vehicle to a road network with a mesh index,and at the same time,obtaining a travel time sample formed inside each passing mesh when the trajectory traverses the mesh.This method is simple and efficient,and it can quickly obtain travel time data on adesignated road.(2)From the GPS data of taxis in Beijing in the third quarter of 2015,50.24 million tracks were extracted.From the trajectories,the travel time data on the path of the Beijing South Third Ring Expressway is further extracted by the meshing method.On the condition that it is not rainy,the amount of data on the working day is 73,298,and the non-working day is 22,937.(3)Based on the gridding method,the study of travel time distribution and reliability evaluation of expressway path is carried out.The travel time distribution is fitted by a variety of probability distribution models.After the comparison between the hypothesis test and the fitness,the Burr distribution is an ideal model for fitting the fast travel time distribution.Based on the grid data of trajectory,the reliability evaluation system of urban expressway travel time is constructed.Based on the Burr distribution model,the mathematical expressions of each evaluation index are established,and the appropriate evaluation indicators are selected to evaluate and analyze the reliability of the selected paths.The travel time reliability on working days,non-working days and road sections are respectively three.Level unfolding.Finally,the case analysis proves that the reliability evaluation based on grid method plays an important role in improving the urban expressway traffic travel decision,alleviating traffic congestion and improving the overall traffic service level.Furthermore,the meshing method has a high practical significance for fully and quickly mining the trajectory information.