Aviation Delay Prediction
An end-to-end ML pipeline and decision support system for predicting aviation delays across four major Virginia airports, built with a 3-engineer team processing 10 years of historical data.
What It Is
An end-to-end machine learning pipeline and decision support system for predicting and analysing aviation delays across four major Virginia airports: ORF, RIC, DCA, and IAD. Built collaboratively by a 3-engineer team, the project ingests 10 years of historical flight and weather data across 24+ datasets, trains Random Forest and SARIMAX forecasting models, and surfaces actionable insights through data-driven dashboards for carrier operations and seasonal scheduling teams.
How It Works
Data Pipeline & Feature Engineering
Python (Pandas, NumPy) ingests and normalises 24+ datasets covering flight schedules, weather observations, carrier records, and airport throughput across a 10-year window. A feature engineering step extracts temporal patterns, carrier performance signals, and weather severity scores used as model inputs.
ML Models
Two complementary models power the system: a Random Forest classifier for schedule adherence prediction (on-time / delayed / cancelled) and a SARIMAX time-series model for forecasting delay magnitude across seasonal and weekly cycles. Cross-validation and hyperparameter tuning are automated via scikit-learn pipelines.
Decision Support Dashboards
Seaborn and Matplotlib visualisations translate ML outputs into operational dashboards: delay heatmaps by route and hour, carrier ranking charts, and seasonal forecasting graphs. Outputs are designed for non-technical stakeholders in carrier operations and scheduling.
What I Owned
ML Pipeline Architecture
Designed the end-to-end pipeline architecture: data ingestion, normalisation, feature extraction, model training, and output generation. Established the scikit-learn pipeline conventions used across both models to ensure reproducible runs.
Model Development
Owned the SARIMAX time-series model: stationarity testing (ADF), seasonal decomposition, parameter selection via AIC/BIC, and forecast validation against holdout test periods. Contributed to Random Forest hyperparameter tuning and cross-validation framework.
Visualisation & Reporting
Built all Seaborn and Matplotlib dashboards. Translated ML outputs into interpretable charts and summary statistics formatted for presentation to non-technical aviation operations stakeholders.
Tech Stack
Browse the source code, commits, and technical implementation details for Aviation Delay Prediction.
More Projects
CyberCup.ai
An AI-enhanced cybersecurity training platform combining traditional CTF challenges with LLM-powered learning, actively developed under the Inspire Center organization with a cross-functional team of 8 engineers.
JobNest
A modern, secure job application tracker to manage your entire job search in one place. Features an AI-powered assistant (NESTAi), application analytics, interview scheduling, and email templates. A Techifive product.
SW Habitation
A developer learning platform with tutorials, tools, and interactive resources used by 7K+ users, featuring insight dashboards and usage tracking.