Forecast Analysis Of Parking Demand Based On Time Series Model

The rapid increase of urban population and the steady improvement of residents’ living standards have led to a sharp increase in the number and frequency of urban motor vehicles.The increasing demand for parking makes it necessary to install more parking facilities in urban planning to meet the corresponding parking needs.However,as far as the present situation is concerned,urban land resources available for planning are extremely limited,and the supply of parking Spaces is seriously insufficient.Therefore,under the double constraint of insufficient parking supply and increasing parking demand,parking problem is becoming more and more serious in large and medium-sized cities.The shortage of parking Spaces will make it more difficult to find parking Spaces.Before the vehicle arrives at the parking station,the owners cannot timely grasp the usage of parking Spaces and the shortterm development and change trend,resulting in more congestion on urban roads.At the same time,the site’s parking Spaces are underutilized,resulting in lower economic yields.In the existing research on parking demand prediction,most scholars comprehensively consider population,economy,land and other aspects,and cannot start from the change of parking demand itself.Therefore,in order to alleviate the parking problem in cities,how to accurately and effectively grasp the development trend of parking demand,rationally allocate the existing parking resources and provide reasonable decision support for parking management has become an important research topic.This study starts from the parking demand,taking a single roadside parking station as the research object,intends to analyze the daily parking demand changes,and build a parking demand prediction model.Firstly,this paper analyzes the existing research situation of parking management,parking demand prediction and short-term prediction,and preliminarily determines the research content of predicting the change of parking demand of a single station with the help of time series method.Secondly,relevant theoretical methods are sorted out,which mainly expounds the performance characteristics of parking demand indicators,parking demand analysis and prediction methods and relevant contents of time series analysis methods.Again to stop the raw data for data preprocessing,this part firstly analyzes the parking of the defects in the original data,and the necessity of preprocessing work,this paper analyzes the parking and the basic train of thought,the main task of the data preprocessing and use SQL to data processing software for xi ’an roadside parking monitoring data work,according to the research purpose mentioned above,will stop the raw data into a parking space time-varying monitoring data of time interval for 10 minutes,will stop in the vehicle as the research object of the modeling analysis of time-varying data.Finally,EVIEWS software is used to model and analyze the data sequence obtained after data preprocessing,and ARMA prediction model and smoothing model are established for validation.By comparing the predicted value of the model with the actual value of the fitting diagram,the ARMA model is selected as the optimal model.