Method for Constructing the Model of Parking Demand Forecasting Based on Time Interval Statistical Analysis

Abstract

Urban parking lot planning and management are the main measures to solve the problem of parking and road traffic congestion, and accurate understanding of parking demand is a prerequisite for parking lot planning and design. In the existing research on forecasting methods of urban parking demand, there arefew researches on forecasting methods of dynamic changes of daily parking demand. Therefore, this paper proposes a parking demand forecasting method based on time interval statistical analysis. The basic idea of the method is as follows: First, according to the travel characteristics of urban residents, the parking demand within a day is divided into several stages. Secondly, within a predetermined period, the parking demand in different stages of a single piece of land is counted. Then, the statistical average value of each stage during the period is calculated and used as the initial parking demand forecasting data of a certain piece of land at a certain stage. Finally, according to the coefficient of trip attraction of a certain piece of land in a period of time, the initial parking demand forecasting data is revised, and the revised data can be used as the parking demand forecasting result of a certain stage of a certain piece of land. This method simplifies the complexity of the existing forecasting model, with simple calculation and clear thinking, which can be a reliable basis for the planning and management of urban parking lot.