Flight Delay Prediction Using Random Forest with Enhanced Feature Engineering

Mary Dufie Afrane; Yao Xu; Lixin Li; Kai Wang

doi:10.1109/southeastcon56624.2025.10971443

Flight Delay Prediction Using Random Forest with Enhanced Feature Engineering

Mary Dufie Afrane, Yao Xu, Lixin Li, Kai Wang

Computer Science

Georgia Southern University

Research output: Contribution to book or proceeding › Conference article › peer-review

Abstract

Delays in air travel continue to disrupt aviation operations, causing passenger inconveniences and economic losses. This research proposes a flight delay prediction framework using a random forest model enhanced with carefully engineered features from flight and weather data. Key features introduced include historical delay metrics and network centrality attributes designed to capture flight network dependencies. The proposed model achieves remarkable predictive performance, with 92.36% accuracy, 98.29% precision, 86.22% recall, 91.86% F1 score, and 96.78% AVC-ROC. These results demonstrate the potential of combining Random Forest with domain-specific features to predict flight delays effectively.

Original language	English
Title of host publication	IEEE SoutheastCon 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1055-1056
Number of pages	2
ISBN (Electronic)	9798331504847
ISBN (Print)	9798331504847
DOIs	https://doi.org/10.1109/southeastcon56624.2025.10971443
State	Published - Mar 22 2025
Event	2025 IEEE SoutheastCon, SoutheastCon 2025 - Concord, United States Duration: Mar 22 2025 → Mar 30 2025

Publication series

Name	SoutheastCon 2025

Conference

Conference	2025 IEEE SoutheastCon, SoutheastCon 2025
Country/Territory	United States
City	Concord
Period	03/22/25 → 03/30/25

Scopus Subject Areas

Computer Networks and Communications
Software
Electrical and Electronic Engineering
Control and Systems Engineering
Signal Processing

Keywords

feature engineering
flight delay prediction
network centrality
random forest

Access to Document

10.1109/southeastcon56624.2025.10971443

Cite this

@inproceedings{436b081793f941d082e98a6433b5a31d,

title = "Flight Delay Prediction Using Random Forest with Enhanced Feature Engineering",

abstract = "Delays in air travel continue to disrupt aviation operations, causing passenger inconveniences and economic losses. This research proposes a flight delay prediction framework using a random forest model enhanced with carefully engineered features from flight and weather data. Key features introduced include historical delay metrics and network centrality attributes designed to capture flight network dependencies. The proposed model achieves remarkable predictive performance, with 92.36\% accuracy, 98.29\% precision, 86.22\% recall, 91.86\% F1 score, and 96.78\% AVC-ROC. These results demonstrate the potential of combining Random Forest with domain-specific features to predict flight delays effectively.",

keywords = "feature engineering, flight delay prediction, network centrality, random forest",

author = "Afrane, \{Mary Dufie\} and Yao Xu and Lixin Li and Kai Wang",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE SoutheastCon, SoutheastCon 2025 ; Conference date: 22-03-2025 Through 30-03-2025",

year = "2025",

month = mar,

day = "22",

doi = "10.1109/southeastcon56624.2025.10971443",

language = "English",

isbn = "9798331504847",

series = "SoutheastCon 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1055--1056",

booktitle = "IEEE SoutheastCon 2025",

address = "United States",

}

TY - GEN

T1 - Flight Delay Prediction Using Random Forest with Enhanced Feature Engineering

AU - Afrane, Mary Dufie

AU - Xu, Yao

AU - Li, Lixin

AU - Wang, Kai

PY - 2025/3/22

Y1 - 2025/3/22

N2 - Delays in air travel continue to disrupt aviation operations, causing passenger inconveniences and economic losses. This research proposes a flight delay prediction framework using a random forest model enhanced with carefully engineered features from flight and weather data. Key features introduced include historical delay metrics and network centrality attributes designed to capture flight network dependencies. The proposed model achieves remarkable predictive performance, with 92.36% accuracy, 98.29% precision, 86.22% recall, 91.86% F1 score, and 96.78% AVC-ROC. These results demonstrate the potential of combining Random Forest with domain-specific features to predict flight delays effectively.

AB - Delays in air travel continue to disrupt aviation operations, causing passenger inconveniences and economic losses. This research proposes a flight delay prediction framework using a random forest model enhanced with carefully engineered features from flight and weather data. Key features introduced include historical delay metrics and network centrality attributes designed to capture flight network dependencies. The proposed model achieves remarkable predictive performance, with 92.36% accuracy, 98.29% precision, 86.22% recall, 91.86% F1 score, and 96.78% AVC-ROC. These results demonstrate the potential of combining Random Forest with domain-specific features to predict flight delays effectively.

KW - feature engineering

KW - flight delay prediction

KW - network centrality

KW - random forest

UR - http://www.scopus.com/inward/record.url?scp=105004572745&partnerID=8YFLogxK

U2 - 10.1109/southeastcon56624.2025.10971443

DO - 10.1109/southeastcon56624.2025.10971443

M3 - Conference article

AN - SCOPUS:105004572745

SN - 9798331504847

T3 - SoutheastCon 2025

SP - 1055

EP - 1056

BT - IEEE SoutheastCon 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2025 IEEE SoutheastCon, SoutheastCon 2025

Y2 - 22 March 2025 through 30 March 2025

ER -

Flight Delay Prediction Using Random Forest with Enhanced Feature Engineering

Abstract

Publication series

Conference

Scopus Subject Areas

Keywords

Access to Document

Other files and links

Fingerprint

Cite this