Airport flight large-area delay risk prediction method

Abstract

The invention discloses an airport flight large-area delay risk prediction method, and the method specifically comprises the following steps: collecting historical data, and carrying out the preprocessing; screening key factors by using a random forest algorithm; carrying out category feature numeralization and normalization processing on the screened data, and randomly sampling to divide the datainto a training set and a test set; for the data, carrying out rough clustering by using canopy clustering to find out an initial cluster; taking he cluster found by the cannopy as a k value, and adopting a K-means algorithm to perform multi-feature fine clustering; calculating the occurrence probability of large-area delay in each cluster; carrying out classification by inputting airport weatherand flight information of a certain hour, and achieving risk prediction of large-area delay of airport flights of the hour. According to the invention, the risk of large-area delay of an airport in ashort time can be accurately predicted; the defect that only single flight delay prediction is carried out is overcome, a plurality of influence factors can be comprehensively considered, and the prediction accuracy degree is improved; the risk prediction problem of large-area delay of airport flights at present is solved.

Description

technical field [0001] The invention belongs to the technical field of civil aviation, and in particular relates to a method for predicting the large-area delay risk of airport flights. Background technique [0002] With the rapid development of the civil aviation industry, the problem of delays has become increasingly prominent. There are many reasons for delays, which can be roughly divided into airline reasons, weather reasons and traffic flow control. Flight delays not only have a great negative impact on airlines and passengers, but also have a negative impact on the normal operation of airports. Accurately predicting the risk of large-scale delays at airports is of great practical significance for timely adoption of targeted strategies. The current delay prediction is mainly divided into: delay time prediction, delay quantity prediction and delay level prediction according to the prediction content; according to the prediction object, it is mainly divided into: single...

AI Technical Summary

Problems solved by technology

[0003] However, most of the current forecasts are regression forecasts based on historical data; and most of them are forecasted for a single flight, lacking an overall risk forecast for airport delays within a certain period of time; most of the forecasts are specific forecasts for a single factor of delay, such as Time, quantity, grade, without a risk forecast for the probability of delay

However, in actual operation, large-scale delays at airports are caused by the combined effects of many flights and many influencing factors, and the situation is complex and changeable.

Images