What needed to be understood.
The dataset contained passenger and flight details but no direct delay-cause field. The analysis therefore needed to identify patterns without turning correlation into an unsupported causal explanation.
How the analysis was built.
- Created a month field and summarized delays by month.
- Compared delays across continents, pilots, airports, and total flight volume.
- Used pivots, conditional formatting, and charts to surface concentrations.
- Built a storyboard that connected questions, analysis, findings, recommendations, and limitations.
What the evidence showed.
- North America recorded the most delays in the submitted analysis.
- Monthly delays were comparatively spread out, providing little evidence of one dominant seasonal pattern.
- Pilot-specific differences were limited within the one-year dataset.
- California airports appeared repeatedly among high-delay locations, while also handling high flight volumes.
What should happen next.
- Investigate airport congestion and operational processes in the California locations highlighted by the analysis.
- Add weather, maintenance, staffing, and reason-code data before attributing causes.
- Normalize delay counts by flight volume and compare delay duration, not only frequency.
What this project does—and does not—prove.
The dataset covered a limited period and omitted explicit delay causes, so the findings are exploratory rather than causal.