PREVISÃO DE ATRASOS DE VOOS POR MACHINE LEARNING: REVISÃO BIBLIOMÉTRICA E METANALISE

HÉLIO DA SILVA QUEIROZ JÚNIOR - UNIVERSIDADE FEDERAL DE PERNAMBUCO (UFPE) -
VIVIANE ADRIANO FALCÃO - UNIVERSIDADE FEDERAL DE PERNAMBUCO (UFPE) -

Abstract

A COMMON PROBLEM AT AIRPORTS AROUND THE WORLD IS DELAYS IN COMMERCIAL FLIGHTS. THE INCREASING DEMAND FOR AIR MODAL CAUSES THESE DELAYS TO BE INCREASINGLY RECURRENT, CAUSING ADDITIONAL COSTS AND REQUIRING CONSTANT ADJUSTMENTS IN FLIGHT MANAGEMENT. DEFINING THE MOST EFFECTIVE METHOD IN PREDICTING THE OCCURRENCE OF THESE DELAYS IS A RECURRING THEME IN TRAFFIC OPERATION SEARCHES. IN THESE STUDIES, THE SPECIFICITY OF THE ANALYZED AREA (BE IT AN AIRLINE, AN AIRPORT OR THE ENTIRE OPERATION OF A COUNTRY), THE COMPLEXITY OF THE OUTPUT SOUGHT (PREDICTION BY REGRESSION OR CLASSIFICATION) OR THE SIZE OF THE DATABASE USED REQUIRE MORE ROBUST METHODS OF ANALYSIS, AND MACHINE LEARNING IS A COMMON ALTERNATIVE TO THE USE OF CLASSICAL STATISTICAL METHODS. HOWEVER, THE NUMBER OF PREDICTION METHODS THAT USE THE MACHINE LEARNING PRINCIPLE IS VAST. THE DIVERGENCE BETWEEN THE SCENARIOS STUDIED ATTRIBUTES DIFFERENT ACCURACY SCANS BETWEEN THE RESPONSES OBTAINED IN THE STUDIES PERFORMED. THEREFORE, THIS PROJECT AIMS TO DEFINE THE MOST APPROPRIATE METHODS OF MACHINE LEARNING TO ESTIMATE THE OCCURRENCE OF DELAYS AT AIRPORTS. BASED ON THE METHODS OF LITERATURE SYSTEMATIC REVIEW, GROUPS OF PERIODIC PUBLICATIONS ARE RAISED IN THE MAIN INTERNATIONAL DATABASES, FROM GROUPS OF DEFINED KEYWORDS. THUS, THE ANALYSES INDICATE THAT THE DEEP BELIEF NETWORK, RANDOM FOREST, GRADIENT BOOSTING TREES AND RECURRENT AND CONVOLUTIONAL NEURAL NETWORKS HAVE A BETTER EFFICACY IN THE PREDICTION RESPONSES, AND THE IDEAL SCENARIO OF ANALYSIS IS THOSE SPECIFIC TO AIRPORTS OR AIRLINES. THEREFORE, THE DATA SIZE, THE COMPLEXITY OF THE SCENARIOS AND THE FORECASTS SOUGHT TEND TO OBTAIN MORE APPROPRIATE RESULTS WHEN MORE COMPLEX METHODS OF PREDICTION ARE APPLIED.

Keywords: FLIGHT DELAY - PREDICTION - MACHINE LEARNING

Download

Full text available here

Citing

@inproceedings{SIT125,
    author = {HÉLIO DA SILVA QUEIROZ JÚNIOR; VIVIANE ADRIANO FALCÃO},
    title = {PREVISÃO DE ATRASOS DE VOOS POR MACHINE LEARNING: REVISÃO BIBLIOMÉTRICA E METANALISE},
    booktitle = {Proceedings of the 2022 Air Transportation Symposium},
    series    = {SITRAER 2022},
    year = {2022},
    pages = {273-283},
    publisher = {SBTA - Brazilian Air Transportation Research Society},
   address = {São José dos Campos, Brazil,}

}

Save link as to get the .Bib file

License

The contents of this article can be used given that the authors are cited. The reproduction of the whole article in any medium is forbidden, unless authorized by the authors.