✈

AEROSCAN

Airport Security Processing Time Predictor

Model Active
01 — Flight Details

Based on your flight details, all operational parameters are automatically inferred from historical airport patterns.

02 — Auto-Inferred Parameters (14 Features)

These values are pre-filled based on your selected date and time using historical airport operational data.

03 — Prediction Output
-- MINUTES
Calculating...

✓ Linear Regression Model  ·  Test R² = 0.6323
🕐
Recommended Airport Arrival Time
—
04 — Model Performance Comparison

Three supervised learning models were trained on the synthetic dataset (600 samples, 14 features). Linear Regression outperformed ensemble methods due to the near-linear nature of the dataset.

Model Train MAE Test MAE Train R² Test R² Generalization
✓ Linear Regression 1.6944 1.9316 0.7192 0.6323
Random Forest 0.9468 1.9837 0.9029 0.6120
Gradient Boosting 0.1833 2.1358 0.9957 0.5474

⚠ Random Forest & Gradient Boosting overfit on training data — large gap between Train R² and Test R².

05 — Feature Importance (Random Forest)

Feature importance scores from the Random Forest model identify the most influential variables in predicting security wait time.