Research Of Flight Delay Early Warning Management Based On Data Mining

Flight delay is one of the problems limiting the development of civil aviation. With the age of "big data" is coming, if we can carry out flight delay prediction research based on new technologies such as data mining and the decision support theory to realize early warning, which will be of great help for making airline schedule and the follow-up security work after flight delay.This thesis sets airline flight delay prediction as application background, analysis of a large number of historical flights operating data, mining the intrinsic characteristics of the data to find out the correlation delay spread and the trend between various elements, in order to provide theoretical and methodological support for the implementation of airline controllable factors flight delay early warning management. In this paper, the main research work is as follows:1. Summarized the concepts of flight delay and delay reasons in accordance with the civil aviation regulations and airlines survey data. Researched on data mining model and algorithm technology in classification and prediction, and then proposed the important role of flight delay prediction in the airline operation management.2. The delay prediction analysis for short period flight schedule. According to the study of time series data mining algorithm, two prediction models were established:Flight delay prediction model based on Markov chain and Time series models. Considering of the scheduled flight delay rate, the average time of delay and the delay passengers for the evaluation method are described in the state of delayed flights by employing Analytic Hierarchy Process (AHP). These two models were used to forecast time series status in view of the delay evaluation indicators, and studied pros and cons of the analysis models through simulation results.3. The delay prediction analysis for specific flight schedule. This research employed Bayesian Network and Dynamic Bayesian Network inference as the main modeling method to obtain the probability distribution under different conditions of flight delay. By studying the Dynamic Bayesian Network inference process and simulation, this paper presented a new method for the construction of the flight delay prediction model which is to establish Hidden Markov flight delay prediction model based on the real flight data. Using the Viterbi algorithm of Hidden Markov model decoding problem to predict the flight delay time. This method improved the accuracy of the flight delay prediction object, and gave an example simulation process, which can verify the validity of the model.4. The research on early warning mechanism of flight delay. The early warning management concepts were applied to the flight delay prediction analysis process and used fuzzy mathematical as the main modeling method. The flight delay early warning index system was gained by using fuzzy analytical hierarchy process (FAHP), and the status of flight delay was evaluated based on fuzzy comprehensive evaluation. Finally the actual flitht data is used to simulate the process, the results show that these early warning indexes can reflect the status of flight delay accurately and the outcome is objective, so it can be supported as reference for early warning in airlines.