LS-SVM with Genetic Algorithm for Forecasting of Runway Incursion Events

Abstract

Forecasting of runway incursion events is very significant to guide the job of civil aviation safety management and it is an important part of the runway incursion early warning management. However, forecasting of runway incursion events is a complicated problem due to its non-linearity and the small quantity of training data. As a novel type of learning machine, support vector machine has some merits, such as dealing with the data of small sample, the high dimension and the excellent generalization ability. Therefore, in this study, least square support vector machine (LS-SVM) with genetic algorithm is proposed to forecast the runway incursion events, among which genetic algorithm is used to determine parameters of LS-SVM. The experimental results indicate that LS-SVM method can achieve greater accuracy than generalized regression neural network (GRNN). Consequently, LS-SVM model is a proper alternative for forecasting of the runway incursion events.