Modeling of Flight Delay State-Space Model - Advanced Research on Computer Education, Simulation and Modeling - page 26

Modeling of Flight Delay State-Space Model
HaiYan Chen, JianDong Wang, and Hao Yan
College of Computer Science and Technology, Nanjing University of Aeronautics and
Astronautics, 210016, Nanjing, P.R. China
[e-mail]
Abstract. Flight delay prediction remains an important research topic due to its
dynamic nature. Dynamic data-driven approach might provide a solution to this
problem. To apply the approach, a flight delay state-space model is required to
represent relationship among system states, as well as relationship between sys-
tem states and input/output variables. Based on the analysis of delay event se-
quence, a state-space model was established and the input variable was studied.
A genetic EM algorithm was applied to obtain global optimal estimates of pa-
rameters used in the mode. Validation based on probability interval tests shows
that: the model has reasonable goodness of fit to the historical flight data, and
the search performance of traditional EM algorithm can be improved by ideals
of Genetic Algorithm.
Keywords: Flight Delay Prediction, Dynamic Data-driven Approach,
State-space Model, Genetic Algorithm.
1 Introduction
As a result of excessive demand for air transportation, flight delay becomes an urgent
problem that exacerbates national transportation bandwidth limitations. Over the past
decade, researches were focused on analyzing flight delay factors, predicting delay
and propagation, and decreasing delays and other issues [1-3]. Real-time prediction of
flight delay is essentially state estimation of dynamic system. Flight operation process
is monitored in real time, which provides an opportunity to apply dynamic data-driven
approach [4] to achieve more accurate and more reliable prediction. The challenge
remains in establishment of the delay state-space model, which is the foundation in
applying the dynamic data-driven approach.
In this paper, a flight delay state-space model was proposed with a statistic way to
calculate delay caused by random factors. In order to search for maximum likelihood
estimates of parameters in the model, Genetic Algorithm (GA) was combined with the
traditional EM Algorithm to avoid the local maximum problem. Model validation and
performance comparison of the two EM algorithms were given as well.
2 State-Space Model of Flight Delay
2.1 State-Space Model of Flight Delay
From departure at an airport to arrival at the destination, an aircraft accomplishes a
flight task. For efficiency and cost considerations, an aircraft should perform multiple