Task

You work for American Airlines as a Data Analyst. The big bosses want analysis on recovery from the COVID downturn and have sent specific R FIs. Your supervisor gave you the following questions to answer and a dump of data to go along with it. She wants a concise, defensible report with clear conclusions to respond to the board with general technical information in support.

On-time Analysis

• What proportion of our flights arrive on time (on or before scheduled arrival)?

– What is our overall on-time arrival rate?

– How has it changed between Aug20 and Feb21?

– Report overall rate for AA-operated flights and a breakout by contractors.

– How do our rates compare to the other major carriers?

– What proportion of flights depart late but arrive on time?

• Compare on-time rates of the unique flight routes.

– Are there habitually late routes?

– How do our rates compare to the other major carriers?

– Is our performance improving compared to the other major carriers?

• Which route had the most cancellations? Diversions?

• What is the trend in our performance over time?

Flight and Route Analysis

• Has our flight volume increased since Aug20?

• Which route has the most traffic over-all?

• Which route has the most American Airlines flights?

– What is our performance on that route?

– Are any competitors trying to take our market share?

Partner Analysis

• Compare our contracted airlines’ performance to their performance when working for our competitors.

• Should we stop working with any contractors on specific routes?

Excursion

Over the course of your analysis you typically find some insight or piece of interesting analysis that was not specifically asked but you feel the sponsor may care about. Find one in these data and explain why the conclusion is important.

Examples: on-time departure analysis, analysis by time block, compare flight times between carriers, etc. . .

Data Overview

The data are Department of Transportation flight database pulls. Each onTime data file contains the same information covering the flight operator, major carrier, scheduled times, actual times, and other information.

The data dictionary provides an explanation of the different data fields. Additionally, you have data on the airports (code, location, time zone, ect) and decoding files. The some fields in the onTime data contain shorthand codes, the decoding files provide full name details (e.g. AA = American Airlines). All time are local times, you must account for time zone changes.

Deliverables

1. Summary report, including

a. Brief problem description

b. Analytic methods discussion

c. Analysis answering the given questions

d. Supporting tables and visualizations as necessary

e. Excursion

2. Code

a. Clearly commented

b. Logically ordered

c. Syntax error free

d. Concise (unnecessary items removed)

Criteria:

• Does the report explain what you did, are the conclusions supported by the analysis in the report?

• Is your analysis correct?

• Are the figures and tables meaningful, attractive, correctly used, and correctly referenced?

• Is your excursion interesting and the conclusions sound?

• Does your code support the analysis and conclusions? (Can I replicate the info in the report)

• Is the code well commented/documented? (i.e. can I follow your logic and understand what the steps are doing?)

• General grammar and writing.

• Use an applicable data analysis technique (like you learned in Stats) to support your conclusions. Provide the code.

• Submit your report as an RMarkdown pdf report.

Data