This project investigates the impact of temporal and environmental factors on departure delays for United Airlines (UA) flights using the nycflights13 dataset. By analyzing trends and conducting permutation tests, we aim to provide actionable insights for operational improvements and enhanced customer satisfaction.
- Analyze how time of day, time of year, and weather factors (temperature, wind speed, precipitation, visibility) influence flight delays.
- Apply permutation testing to assess statistical significance of observed differences.
- Use data visualization to communicate patterns clearly and effectively.
- Data Wrangling: Joined flight and weather datasets using R (
tidyverse,nycflights13). - Feature Engineering: Created new variables such as
time_of_day, and categorized weather conditions. - Statistical Analysis: Performed permutation tests to compare delay distributions across multiple factors.
- Visualization: Used
ggplot2for boxplots and point plots to highlight trends and outliers.
- Night-time flights exhibit significantly higher delays compared to mornings.
- Medium-to-high temperatures and low visibility contribute to increased delays.
- Wind speed shows inverse behavior: lower wind speeds often correlate with longer delays.
- Precipitation may reduce delays due to flight cancellations or rescheduling rather than operating during adverse conditions.
- Month of the year has no significant impact on average delays.
- Programming Language: R
- Libraries:
dplyr,ggplot2,tidyverse,nycflights13 - Techniques: Permutation Testing, Feature Engineering, Data Visualization