01589902

Component

Short term remaining parking spaces forecasting is an important function of urban parking guidance system (PGS) which is an effective approach for improving urban parking management and relieving the parking problem. Remaining parking spaces of public parking lots can be provided to drivers by Variable Message Signs, internet, broadcast or even on-board equipment. The reliability of PGD will be affected if the number of remaining parking spaces cannot be forecasted accurately. Previous methods are difficult to be applied in practice because of low accuracy or long computing time. In this study, time series method which originates from economics field is applied to establish an approach for forecasting remaining parking spaces accurately and improving the efficiency of PGS. Based on the order and periodic characteristics of the real world data collected in Nanjing, China, a short term forecasting model, Auto Regressive Integrated Moving Average (ARIMA) model, is established and a real-time forecast process is presented. The performance of this forecasting model is analyzed by comparing   forecast data with collected data. Results indicate that time series method can be applied to forecast the number of remaining parking spaces accurately. The real-time forecast process is appropriate for the established model and can improve the forecast accuracy. In addition, the proposed ARIMA model is compared with neural network method. Results show that time series method is better both in individual error analysis and collective error analysis. This method is appropriate for applying in real world and can improve reliability of PGS.

This paper was sponsored by TRB committee ABJ30 Standing Committee on Urban Transportation Data and Information Systems.

01584066

16-5060

English

18p

2016

Transportation Research Board 95th Annual Meeting

Digital/other

Figures; Maps; References; Tables

Forecasting; Neural networks; Parking facilities; Parking guidance systems; Real time information; Time series; Urban areas

Nanjing (China)

Highways; Operations and Traffic Management; Planning and Forecasting; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2016 Paper #16-5060

TRIS, TRB, ATRI

Jan 12 2016 6:12PM