Research On Parking Planning Method Based On DIDI OD Data

The parking problem is serious in big cities andhas become prominentin small and medium-sized cities.The difficulty in parking and disordered parking is widespread currently,which has seriously affected the development of urban transportation and inhabitants" daily life.How to allocate parking resources scientifically and reasonably,and solve the parking problem effectively has become one of the important tasks of the traffic management department.Previous research of urban parking planning has put a large amount of manpower and resources to obtain the related data,which led to high costs with unsatisfactory results.With the development of information collection and storage technologies,a large number of various traffic data have been accumulated in the urban traffic system,attracting lots of scholars to carry out related research.The study investigated parking planning method based on Origin-destination(OD)data from the DIDI online taxi-hailing service platform used Cangzhou City as the case study.The main research content and conclusions include the following three aspects:(1)The study described of the characteristics of DIDI OD data,extractedand preprocessed the OD data from DIDI online taxi-hailing service platform.By using the processed OD data,the travel distance,travel speed,and the spatial distribution of travel start and end points were analyzed.The analysis results were consistent with the actual travel situation in Cangzhou City,which indicated that DIDI OD data could be used in urban traffic characteristic analysis and parking planning study.To further demonstrate that DIDI OD data can be used in parking planning,the study conducted both quantitative and qualitative analysis on the peak parking volume and OD data by using Cangzhou city as a case study.It was verified that a strong spatial correspondence and correlation existed between the peak parking number and the starting and end points of travel.A parking zone division method based on DIDI OD data was then proposed.(2)The study analyzed the fundamental that DIDI OD data can be used in parking demand prediction and proposed the parking demand prediction method based on DIDI OD data.The method was established through regression analysis using the collected parking data and DIDI OD data,and the overall parking demand prediction could be achieved by the established model.The case study of Cangzhou city showed that the combination of peak parking number and D(end points)number can reach a good regression result,with R2 in simple linear regression model and quadratic polynomial regression model 0.965 and 0.974 respectively.(3)The study identified the basic research idea of parking facilities layout and developed a public parking site selection method based on DIDI OD data.The density clustering method was used to locate high parking demand based on the grid division,so as to determine the construction site of public parking.The study tested the feasibility of the method by using certain parking area(Number 12 and 14)in Cangzhou City for case analysis.Overall,the study proposed the parking planning method based on DIDI OD data,which included the parking demand prediction method and parking facility layout method.Feasibility of the proposed method was verified by using Cangzhou city as the case study.This study made new attempts on the parking planning area,which provided a new direction for solving future parking issues and exploring the potential value of traffic big data.