On the relevance of data science for flight delay research: a systematic review

ABSTRACT

Flight delays are a significant problem for society as they evenly impair airlines, transport companies, air traffic controllers, facility managers, and passengers. Studying prior flight data is an essential activity for every player involved in the air transportation system. Besides, developing accurate prediction models for flight delays is a crucial component of the decision-making process. Prescribing actions to solve on-going delays is an even challenging task due to the air transportation system complexity. In this regard, this paper presents a thorough literature review of data science techniques used for investigating flight delays. This work proposes a taxonomy and compiles the initiatives used to address the flight delay studies. It also offers a systematic literature review that describes the trends of the field and methods to analyse the applicability of newly proposed methods.

KEYWORDS:

• Flight delay
• data science
• data management
• data analytics
• systematic review

Acknowledgments

The authors thank CNPq, CAPES, FAPERJ, and CEFET/RJ for partially funding this research. We would also like to thank a subset of the authors for the initial effort on developing a preliminary preprint version of this paper (Sternberg et al., 2017), which was not published in any place before. It was a basis from which we were able to conduct a systematic review presented here.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Search string used: (“airport” or “airline” or “flight”) and (“root delay” or “delay propagation” or “arrival delay” or “departure delay” or “airline delay” or “airport delay” or “flight delay”) and (“classification” or “predict” or “forecast” or “cluster” or “pattern” or “statistic” or “data science” or “data mining” or “data pre-processing” or “big data” or “data cube” or “data warehouse” or “data analysis” or “data analytics” or “visualization” or “network analysis” or “network science” or “complex network” or “outlier” or “anomaly”).

2 Disclosed at DAL (2020).

3 Disclosed at DAL (2020).

4 Disclosed at DAL (2020).

Additional information

Funding

This work was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq); Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES): [Grant Number 001]; Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ).