Parking Spaces Demand Forecasting Based on ARIMA Model

Abstract

The effective planning and management of urban parking resources are crucial for energy conservation and emission reduction. Accurate forecasting of parking demand is essential for determining the number and timing of shared parking spaces. In this study, we analyze the characteristics of commonly used parking demand forecasting models and propose the use of a time series model to forecast parking demand. The occupancy of parking spaces is transformed into a forecasting problem, and the proposed model is used to forecast parking occupancy. The results show that the established ARIMA (Autoregressive Integrated Moving Average) model has high forecasting accuracy and can effectively describe parking occupancy demand changes. This research provides a valuable reference for urban parking management and facility planning.