Flight Delay Analysis - Atlanta Airport (ATL) 2023

Analysis of 5,000 flight departures to find delay patterns by day of week, hour of day, and passenger volume.

Goal

• Track late departure rate using 15-minute threshold (FAA standard)
• Find which days and times have the most delays
• Compare delay patterns with passenger volume to find correlations
• Understand patterns that could help with scheduling

Data

• flights.csv - ATL outbound flights sample (5,000 records)
• us-daily-passengers.csv - TSA daily passenger counts (2023)

Key Findings

Overall

• Total flights: 5,000
• Overall late rate: 19.8%

By Day of Week

Day	Late Rate
Sunday	22.8% (highest)
Friday	22.5%
Thursday	20.8%
Saturday	19.7%
Monday	19.4%
Wednesday	16.9%
Tuesday	15.9% (lowest)

By Hour of Day

Time Period	Late Rate
Morning (5-12)	~13.8%
Afternoon (12-17)	~18.4%
Evening (17-21)	~27.7%
Night (21+)	~25.8%

Finding: Evening flights are about 2x more likely to be delayed than morning flights. This is because delays build up during the day (cascade effect).

Passenger Volume vs Delays (Extension)

Comparing delay rates with average daily passenger volume:

Day	Delay Rate	Avg Passengers	Pattern
Sunday	22.8% (highest)	2.51M (high)	High volume → High delays
Friday	22.5%	2.55M (highest)	High volume → High delays
Tuesday	15.9% (lowest)	2.10M (lowest)	Low volume → Low delays
Saturday	19.7%	2.17M (low)	Low volume → Lower delays

Pattern: There is a clear correlation between passenger volume and delay rates. Days with more passengers have higher delay rates.

Why this happens:

• More passengers = fuller flights = less buffer time
• Busier airports = more congestion = more delays
• Tighter turnaround times when flights are full

Business Implications

Why morning flights are better:

• Aircraft haven't had delays from earlier flights yet
• Less airport congestion
• Crews are fresh (no timeout issues)

Why Sunday is worst:

• Weekend travel is heavier (high passenger volume)
• Recovery from Saturday operations

Recommendation: If you want to minimize delays, fly on Tuesday morning.

How to Run

1. Install pandas and matplotlib
2. Open flights-delays-project.ipynb
3. Run all cells

Files

• flights-delays-project.ipynb - main analysis notebook
• flights.csv - flight data
• us-daily-passengers.csv - passenger data
• EXECUTIVE_SUMMARY.md - one-page summary
• delay_by_day.png - delay rate by day chart
• delay_by_hour.png - delay rate by hour chart
• comparison_chart.png - side-by-side comparison of delays vs passengers

Context

Completed as part of Python Institute Data Analysis Associate certification.