✈️ Airplane Boarding Simulation Study

This repository contains a full simulation and statistical analysis of airplane boarding methods.
It was developed as part of an Operations Research / Statistics course project.

---

📂 Contents

Core Simulation

• model_boarding.py — Implements the airplane boarding model (single-aisle, 50 rows, 6 seats per row).
• run_boarding_study.py — Runs repeated simulations and generates main figures (ECDF, boxplots, cumulative means).
• sanity_checks.py — Verifies theoretical expectations (150 / 172.5 minutes targets) and quick validation runs.

Statistical Analysis

• stat_tests.py — Welch-ANOVA, pairwise tests, QQ-plots, ECDF, boxplots, forest plots.
• extra_tests.py — Non-parametric alternatives (Kruskal–Wallis, Mann–Whitney, permutation test).

Sensitivity

• sensitivity_analysis.py, sensitivity_analysis2.py — Boarding time vs. number of rows, both absolute and relative.

Utilities

• export_for_overleaf.py — Collects all generated figures into a for_overleaf/ folder (for report writing).
• Other helpers: fix_images.py, convert_clean.py.

Reports

• Simulation_Boarding_OR (35).pdf — Final submitted academic report (Overleaf).
• סימולציה הוראות.pdf — Assignment instructions.

---

🎯 Project Goal

To empirically evaluate the efficiency of four boarding policies:

• RANDOM
• BACK_TO_FRONT
• FRONT_TO_BACK
• BLOCKS_K2 (our designed method: seat-order + two-row blocks)

---

🔬 Methodology

• Model: event-driven simulation of 300 passengers in a single-aisle cabin.
• Service times: exponential (luggage storage, seat blocking).
• Repetitions: 100 independent runs per method (fixed seeds for reproducibility).
• Analysis: mean boarding time, 95% confidence intervals, effect sizes, robustness checks, and sensitivity to plane size.

---

📊 Key Results

• BLOCKS_K2 consistently outperformed others:
  ~40% faster than RANDOM, ~60–70% faster than directional methods.
• Distribution tails narrower under BLOCKS_K2 (lower risk of extreme delays).
• Robustness confirmed across different plane sizes and alternative tests.

---

🚀 How to Run

# Run main simulation
python run_boarding_study.py

# Statistical analysis
python stat_tests.py

# Sensitivity analysis
python sensitivity_analysis.py
python sensitivity_analysis2.py

# Sanity checks
python sanity_checks.py