基于贝叶斯网络的航班过站时间分析与延误预测

Flight turnaround time analysis and delay prediction based on Bayesian network

从航班延误链式波及的角度出发,分析了影响航班过站时间的多种因素,建立了贝叶斯网络模型,模型能够清晰地反映多种因素对下游航班过站时间的影响。提出了基于贝叶斯网络参数估计的航班延误预测算法,当航班发生起飞延误时能够预测下游航班的起飞时间和延误状况。对算法进行了实现,并利用实际航班数据进行仿真,结果表明了该算法有比较高的预测准确率。

From the standpoint of flight delay propagation prediction, multi-factors that influenced flight turnaround time are analyzed, and a Bayesian network model is established, which could clearly reflected the influence of various factors on the downstream flight tur- naround time. A flight delay prediction algorithm based on Bayesian network parameter estimation is proposed, and when flight delay occurred during take-offstage the downstream flight departure time and delay conditions could be predicted. The algorithm is programmed and the actual flight data is used to simulate the process. The results show that the algorithm has higher prediction accuracy.