Open AccessAirport Passenger Throughput Forecast Based on PSO-SVR Model
Author(s) -
Yuanxun Li,
Xiushan Jiang
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/780/6/062006
Subject(s) - exponential smoothing , particle swarm optimization , throughput , support vector machine , autoregressive integrated moving average , computer science , beijing , genetic algorithm , data mining , operations research , time series , engineering , machine learning , geography , telecommunications , archaeology , china , wireless , computer vision
Accurate prediction of airport passenger throughput can provide a scientific basis for airport construction, aircraft procurement, and route planning. Based on the characteristics of airport passenger throughput, the support vector regression machine prediction model based on particle swarm optimization (PSO-SVR) is proposed. First, the conditional attributes and decision attributes are selected in the historical data of airport passenger throughput. Secondly, the particle swarm optimization algorithm is embedded to optimize the kernel function parameters and penalty factors. Finally, the support vector regression machine is used to predict the airport passenger throughput in the next year. Beijing Capital, Shanghai Pudong, Guangzhou Baiyun and Chengdu Shuangliu were selected as examples to verify the validity of the model. And compared with the moving average method, exponential smoothing method, ARIMA precision. The results show that the PSO-SVR model improves the prediction accuracy, and the average absolute percentage errors for the four airport throughput forecasts are 3.21%, 4.37%, 2.38, and 4.43%, respectively.