Study On Parking Demand Forecasting Model And Its Application Of City Center

Parking demand is a complex issue,involving many influencing factors,including the social economy, land use, traffic development and so on, thus make theparking demand forecasting becomes difficult.This paper is stressing on the point ofthe parking facilities serve to the retained parking,the work travel parking, the flexibletravel parking,and the shared parking,trying to use principal component analysis toreduce the dimension of several major influencing factors of the parking needs of theurban centers, and taking the results as the input data, the first three types of parkingas the output data, in order to establish the parking demand for the combination ofprincipal component analysis and neural network prediction model to forecasting andplanning the number of parking spaces.This paper is mainly focusing on the followingthree aspects:â‘ Starting from the characteristics of the China urban center parking demand,analysing the factors of parking demand, and combining with the existing parkingdemand forecasting methods involves the main influencing factors,selecting theparking influencing factors based on each type of service of the parking facilities, andto lay the foundation for further study.â‘¡Using principal component analysis to reduce the dimension of theinfluencing factors.Taking the results as the input units of the neural network, andusing retain the retained parking,work travel parking and elastic travel parking as theoutput units of the network, using the improved BP learning algorithm,and use thesample data to train the network simulation and build the parking demand forecastingmodel.â‘¢Taking the Wuzhou central area as example, using the neural networkprediction model to forecast the parking demand,work travel parking demand andflexible travel parking demand, anlysis the prediction,then make some proposal to thecounter measures of the shared parking.