Research On Flight Recovery Decision Making Based On Data Mining

Now a days we have entered the information technology cloud era,the civil aviation industry has begun to seek integration points with big data.How to use big data mining to improve flight delays and help airlines improve operation efficiency and operation cost is currently a hot research issue that airlines pay attention to.This article progressive discourse the delay prediction and flight recovery problem.Through processing and analyzing historical flight operation data,it explores flight delay factors and characteristics,analyzes the flight delay situation,predicts the delay level,adjusts the recovery flight,and improves it.Research on the operational benefits of airline flights.The main research work of this paper is as follows:First,use the collected flight operation data and meteorological data,through a series of pre-process data,and analyze characteristics of flight delays in multiple dimensions from the aspects of overall flight delay characteristics,different airlines,different time periods,and different flight sectors.Aim at the problem of flight clusters and delay propagation,establishing a Bayesian network model to analyze the spread of flight delays in the flight chain,verify the accuracy of examples,and provide a basis for flight adjustment and recovery.Secondly,based on the K-means clustering algorithm to divide the flight delay time,and according to the previous flight delay time.Construct the C4.5 decision tree model of the arrival flight prediction under severe weather,combined with the flight operation case,generate the results,and verify the correctness of the model from multiple angles.Simplify the flight adjustment decision-making process,predict the delay level of arrival flights,and provide decision support for the subsequent flight recovery.Finally,construct a spatiotemporal network diagram based on the arrival flight delay prediction,construct a spatiotemporal network recovery model based on the delay prediction,and add airline preference weights to the model to optimize the objective function.Finally,solve an example to verify the effectiveness of the flight recovery model under severe weather.According to the actual operation of the airline,the proposed measures are proposed to dynamically adjust the flight in order to achieve the purpose of data-based and intelligent flight recovery.The completion of the research work in this article will help reduce airline flight operating costs with the help of data science,meet the airline’s different flight recovery decision-making needs,improve flight delay rates and operational quality.Provide scientific predictive evaluation models and flight recovery models.Guarantee the operating efficiency of airlines.