Service Data Analyze for the Available Parking Spaces in Different Car parks and Their Forecast Problem

Abstract

In order to save people's travel time in searching parking spaces, this paper firstly analyzes the changes data of available parking spaces in five parking lots with different capacities, which are located in Xi'an, China. Then this paper uses two methods to predict the data of available parking spaces. The results show that: (1) Available parking spaces in the urban parking lots are fluctuating according the survey data, parking spaces with large capacity of the parking lot will have a larger data fluctuation range, and parking saturation of parking lot will also be affected by the parking lot size and location. (2) Back propagation (BP) neural network model can be used to predict the available parking space of parking lots, no matter the capacity and different geographical location of those carparks is different. Moreover, a model of combination of genetic algorithm (GA) and BP neural network is more reliable than a simple BP neural network model according the results of data contrast are used.