Research on Improved Grey Prediction Method of Airport Passenger Throughput

Airport passenger throughput prediction is of great significance to the operation and development of the airport. Based on the grey prediction method, this paper studies the future passenger throughput of the Capital International Airport and the Beijing New Airport. In this paper, the traditional grey prediction model is improved and the passenger throughput of the two airports from 2019 to 2025 is predicted and analyzed. The prediction results show that the improved grey model can improve the accuracy and reduce the prediction error. Introduction The prediction of airport passenger throughput is the basic premise of airport and airline operation, and also an important basis for airport resource allocation. The accuracy of prediction results affects the scale of airport construction and expansion directly. This paper improves the grey model and predicts the passenger throughput of the Capital International Airport and the Beijing New Airport from 2019 to 2025, provides a reference for the future operation and management of the two airports. Prediction methods can be roughly divided into: qualitative analysis [1], trend extrapolation [2], econometric [3], combination prediction method [4-5], etc. Different prediction methods can be used depending on the predicted scene and the predicted data. Grey prediction model (GM) is a time series model based on grey theory, which is used to deal with uncertain and rough data sets. “Grey” reveals an unclear system [6]. Grey theory can deal with incomplete and discrete data. [7], GM model is more robust to noise data and missing data [8]. This model has been proved [9] to be superior to other prediction methods in the processing of short-term prediction. For a long-term prediction, the original GM model needs to be improved. From this point of view, many predictions are for short-term predictions, and there are fewer improvements to the model. Due to the few data of the Capital International Airport and the Beijing New Airport, the grey prediction model can better fit its data development trend, so this paper uses the grey model for prediction and improves the accuracy of the model. Improvement of GM (1, 1) Model The GM (1, 1) Model can be expressed as     (1) ^ (0) 1 , (1) , 0,1,... 1 ak u u X k ce c X k n a a        . (1) For equation (1), when k=0, ... , n-1, the data obtained are fitted values. When k  n, the data obtained are the predicted values. Performing a subtraction on equation (1):       (0) (1) (1) ^ ^ ^ 1 1 , 0,1,... 1 X k X k X k k n       . (2) International Conference on Modeling, Analysis, Simulation Technologies and Applications (MASTA 2019) Copyright © 2019, the Authors. Published by Atlantis Press. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). Advances in Intelligent Systems Research, volume 168