Likhitha Veganti

Likhitha Veganti

Data Science Student at Seattle University

Flight Gain Analysis for United Airlines (R Project)

Link to my GitHub Repository Github Code.

Introduction:

Suppose you’re part of United Airlines, code-named UA. After a previous exploration of departure delays, the current investigation centers on gain per flight. The net gain is a crucial metric, representing the disparity between departure and arrival delays. By introducing a new variable to measure this net gain, this analysis aims to provide insights using confidence intervals, hypothesis tests, and exploratory data analysis.

Project Objectives:

  1. Late Departures Analysis:
    • Explore if the average gain differs for flights that departed late versus those that did not.
    • Examine the average gain for flights departing more than 30 minutes late.
  2. Common Destination Airports:
    • Identify and describe the five most common destination airports for UA flights from New York City.
    • Analyze the distribution and average gain for each of these top airports.
  3. Gain per Hour Analysis:
    • Calculate the gain per hour by dividing the total gain by the flight duration.
    • Investigate if the average gain per hour differs for late versus non-late flights and for flights departing more than 30 minutes late.
  4. Flight Duration Impact:
    • Assess whether the average gain per hour varies for longer versus shorter flights.

Executive Summary:

In this analysis, we filter the nycflights data to focus on United Airlines flights and introduce new variables, including late, very_late, gain, and gain per hour. The exploratory data analysis involves boxplots, histograms, and statistical tests to answer the project’s questions. Here’s a summary of key findings:

1. Average Gain for Delayed Flights

  • Late vs. Not Late: Permutation tests and t-tests suggest a significant difference in net gain between flights that are delayed and those that are not, with non-delayed flights having a higher average gain.

  • Very Late vs. Not Very Late: Similar findings indicate that flights not delayed by 30 minutes or more have a higher average gain than very late flights.

2. Most Common Destination Airports

  • The five most common destination airports for UA flights from NYC are IAH, ORD, SFO, LAX, and DEN.

  • Detailed distributions and average gains for each airport are presented, showing normal distributions with various skewness.

3. Average Gain per Hour for Delayed Flights

  • Late vs. Not Late: Permutation tests and t-tests reveal a significant difference in gain per hour between late and non-late flights, with non-late flights having a higher average gain per hour.

  • Very Late vs. Not Very Late: Similar results indicate that flights not delayed by 30 minutes or more have a higher average gain per hour than very late flights.

4. Gain per Hour for Longer and Shorter Flights

  • Permutation tests and t-tests show a significant difference in gain per hour between shorter and longer flights, with shorter flights having a higher average gain per hour.