Research On Association Rule Discovery And Model Of Warning Evaluation About Flight Collaborative Security

The punctuality of flight directly reflects the service quality of civil aviation industry, especially for large and busy hub airport, it is a difficulty and heavy task to ensure the punctuality of flight. As the passenger throughput increasing, many domestic and international large hub airports have tried to introduce flight Collaborative Decision Making(CDM) system to promote the ability of flight security.To some extent, the system optimized flight security process and improved the passenger service quality, but it is short of the real-time intelligent delay warning for key steps of flight security. In view of this problem, the paper effectively mines and analyses the vast flight data from CDM system by using association rule algorithm to find relationship between key security steps and flight delay state, and to provide technical and theoretical support for a new-generation intelligent CDM system that has flight security situational awareness and delay warning function. Therefore, the main research works are as follow:Firstly, the paper research on knowledge discovery of flight security data based on association rule, and extracts the original data generating in flight security process from CDM system, defining flight security attributes which combined with the actual business process of flight security. It is to find the association rules between the completion time deviation value of key security steps and the Given-flight state;Secondly, the paper studies flight security situational awareness and delay warning method based on association rule mining, using Apriori algorithm to mines the rule.And confirming threshold that is between the completion time deviation value of key security steps and flight delay state, the paper proposes a flight delay warning method which by analyzing time deviation of key security steps and taking use of this threshold; Then, a flight security evaluation model for new-generation intelligent CDM system is put forward. The model includes index system that the core are flight's average deviation value of arrival time, average deviation value of departure time, average deviation value of flight service completion time, the accuracy rateabout ‘Target Off-Blocking Time', and the accuracy rate about ‘Calculate Take-Off Time'. The model adopt to association rule mining to confirm the index thresholds that effects flight security and flight delay, and the evaluation magnitude is given by combining with the fuzzy analytic hierarchy process and comprehensive evaluation method, to make an evaluation of collaborative condition for airport, airlines and Air Traffic Management and their work efficiency in flight security; Finally, do simulation experiments by using actual flight operation data of domestic hub airport.The results show that the method to analysis flight security data can get ideal results,and the model also has pretty applicability.