Research On Application Of Data Mining Technology In Urban Roadside Parking Management

In recent years,with the continuous development of Chinese economy,more and more Chinese family began to possess their own family cars.As a result,the parking problems in the process of urban development appear more and more severe.As for HangZhou,since the HangZhou Municipal Government began to pay attention to the parking problems much earlier,the parking pressure of HangZhou at present is much lower than other big cities.In the process of parking management in Hangzhou,facing with a huge parking space gap,the Hangzhou Municipal Government has an initiative to respond to the growing parking demand by developing parking spaces along with urban roads.And in the process of continuous development of road parking spaces and model management,a wealth of parking data on urban roads has also been accumulated which is the research basic of this article.This paper briefly explains the current situation of parking management in big cities and analyzes the research results of urban parking management in various aspects in recent years at first.Then this paper introduces some theories and techniques of data mining which are related to the research in this article.Based on the in-depth study of parking data in Hangzhou,this paper analyzes the various influencing factors that affect the parking demand on urban roads.And then,a MLR-ARMA model which combines a multiple linear regression model with an autoregressive moving average model is proposed to consider the characteristics of the parking lots and external environmental facors in the meantime.This model is mainly used to fit the parking data and predict the future parking demand of a single parking lot and the experimental results show that it has better fitting and prediction capabilities than simple multiple linear regression models or ARIMA models.At last,this paper proposes a solution on how to conduct an overall analysis of the parking lots on the entire city roads and uses the aforementioned MLR-ARMA model for experimental verification.This paper mainly uses the relevant technology of data mining to research and analyze the historical parking data of Hangzhou and hopes that it can provide help and reference for future work of parking management,including evaluating the effectiveness of parking management policies,the assessment of the size of parking lot,the construction and integration of parking resources,etc.,thereby facilitating the solution of parking problems.