Published Online:July 2025
Product Name:The IUP Journal of Computer Sciences
Product Type:Article
Product Code:IJCS030725
DOI:10.71329/IUPJCS/2025.19.3.34-46
Author Name:Abdulrahman Gamji Aminu, Mohammed Baba Hammawa and Fatimah Binta Abdullahi
Availability:YES
Subject/Domain:Engineering
Download Format:PDF
Pages:34-46
The paper focuses on the Nigerian Airspace Management Agency (NAMA) and presents an Enhanced Electronic Flight Plan Management System (EEFPMS), augmented with machine learning (ML) predictive analytics capabilities. The system is capable of using historical flight data records to preprocesses and carry out data engineering features and train random forest (RF), linear regression and gradient boosting ML models, experimented using NAMA flight arrival times (ETA) data, in order to make accurate predictions. RF model forecasts from experimental investigations showed impressive results of 20.25, 2466.64, 49.67, 0.717 for MAE, MSE, RMSE, R², respectively, for initial prediction values and also a post-tuning result of 19.9, 2049.14, 45.27, 0.7649 MAE, MSE, RMSE, R², respectively. Using python jupyter note, RF demonstrated improved operational efficiency compared to other ML models and traditional error-prone manual processes. In compliance with international standards, this study provides actionable insights into aviation air traffic controllers’ flight management data for a transformative process from a manual flight management process to an automated electronic process. EEFPMS presents a substantial improvement in usability, error management, and compliance adherence in airspace management. This study contributes to modernizing the Nigerian aviation systems using intelligent and adaptive ML solutions.
The increasing need for air transportation in different parts of the world has encouraged governments and international aviation agencies to come up with different aviation solutions.