Flight Delays R Shiny

[This article was first published on R – NYC Data Science Academy Blog, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Shiny Dashboard 2018 flight delays:

https://huiminou.shinyapps.io/Flight_Delay/

 

Background

There are thousands of flight delays in the United States on a daily basis. Direct Aircraft Operating Cost is $74.2/min. Flight delays cost airlines billions of dollars every year. It is important for airlines to minimize flight delays. I have created an interactive visualization app to help airline operations better visualize fight delays.

 

Data

I used the 2018 flight delay data from Kaggle.com. The original dataset size for 2018 flights is more than 7 million rows. Because the free version of Shiny App has a capacity limitation, I created a subset of the data to include only flights from American airline and Delta Airlines. After the subsetting, the new dataset size includes 310,983 observations.

 

Definitions:

Carrier Delay: Delay caused by a carrier such as an aircraft cleaning, aircraft damage, and baggage loading.

Late Aircraft Delay: Delay caused by previous aircraft arriving late.

NAS Delay: Delay caused by National Aviation System.

Security Delay: Delay due security such as the evacuation of terminal and re-boarding due to security breach.

Weather Delay: Delay due to weather conditions.

 

Interactive Data Visualization:

  1. Airline Operations can use “Map” to visualize the average delay hours per flight for airports in different months.

Interesting findings:

In 2018, for Delta Airline, the highest average delay hours per flight is 3.69 hours at Bismarck Municipal Airport in Bismarck, ND. Direct Aircraft Operating Cost is $74.2/min.  On average, each delayed flight at Bismarck Municipal Airport cost Delta Airlines $16,437 in 2018.

For American Airlines, the average delay hours per flight is 2.8 hours at Ted Stevens Anchorage International Airport in Anchorage, AK. Each delayed flight cost American Airlines on average $12,448 at Ted Stevens Anchorage International Airport in 2018.

 

  1. Airline Operations can use a dodge bar chart to compare reasons for the delay

 

 

Interesting finding:

For American Airlines, average delay hours for late aircraft is often longer than average delay hours for carrier delay. For Delta Airlines, delay hours due to late aircraft are often shorter than delay hours due to carrier delay. In other words, in most months of the year, late aircraft delay costs American Airlines more than carrier delay. It is the opposite of Delta Airlines.

In addition, American Airlines experienced seven months of weather delays in 2018 while Delta Airlines has weather delays almost every month.

The delay caused by NAS is similar between the two airlines.

For both Delta and American Airlines, there is almost no delay caused by security.

 

  1. Airline Operations can use a dodge bar chart to compare the reason for delay by airport

 

In major airports such as BOS, LGA, ORD, passengers spend more time waiting for flight due to late aircraft delay compared to spending time for other reasons.

 

It is interesting that at JFK airport, flights are delayed longer due to National Aviation System compare to other flight delay reasons.

 

  1. Airline Operations can compare delay frequency by hours and weekdays

For American Airlines, flights are often delayed from 5PM to 8PM on Monday, 4PM to 9PM on Thursday, 3PM to 9PM on Friday, and 6PM to 8PM on Sunday.

AA flights are delayed more often on Monday and Friday compared to any other days of the week.

There are more flight delays occurring from 6AM to 12PM compared to flight delays from 12PM to 6AM. It makes sense because airlines only schedule late-night flights in a few popular traveling destinations.

For Delta Airlines, flights are often delayed from 3PM to 9PM on Thursday and Friday and 5PM and 8PM on Monday.

 

Compare two airlines:

American Airlines has more flight delays than Delta Airlines.

For both airlines, Monday and Friday are the two days with most flight delays. Saturday is the day with the least flight delays compared to the other days of the week.

 

  1. Compare Flight Delay Frequency by Hours and Weekdays in a different airport

For American and Delta flights, the flight delay situation in different airports varies. Below is an example of the flight delay for American and Delta Airlines at JFK. Feel free to play with the Shiny app if you are curious about American and Delta flight delays at other airports.

 

For JFK airport, American and Delta flights are frequently delayed at 5-8 PM on Friday,  4-6PM and 7-9PM on Thursday, 5-8PM on Wednesday, 7PM on Tuesday, 7PM on Monday, and 8PM on Sunday.

JFK Airport is not busy from 12AM to 6AM for American and Delta Airlines.

 

  1. Airline operations can visualize the peaks of the frequency of delay

 

For American Airline, there are not many flight delays from January to April. This may be due to the fact that people tend to travel less in winter.

Flight delay frequency increases from April to August and peaks on Aug 13. There are a lot of flight delays from May to August. This is possibly due to the fact that people start to travel more when the weather gets nicer, and families may plan more travel when students are off school. The flight delays occur less after students are back to school in September.

However, delays increase around holidays such as Columbus Day in October, Thanksgiving in November, and Christmas in December.

The second peak of flight delays is on Dec 28. This is probably due to people traveling for the New Year.

  1. Use boxplot for visualizing delay hours by month

 

For American Airlines and Delta Airlines, the median of flight delay for all months are similar. It is around 0.5 to 0.8 hours.

For extreme flight delays, a lot of American Airlines flights are delayed over 20 hours in multiple months, and three of them are delayed more than 30 hours. However, only  one Delta flight appears to have been delayed over 20 hours.

 

  1. Use boxplot for visualizing delay hour by weekday

 

In each day of the week, American and Delta Airlines often have flight delays for around .65 hours on average.

  1. Use boxplot for visualizing delay by the hour of the day for each airline

 

 

Interesting finding:

For American Airlines and Delta Airlines, from 12PM and 5AM, the median flight delay hours are 1 to 2.5 hours. It is significantly longer than the median of the average flight delay hours between 5AM and 12PM.

 

Resources:

2018 Flight Data: Data Source: https://www.kaggle.com/yuanyuwendymu/airline-delay-and-cancellation-data-2009-2018

Direct Aircraft Operating Cost per Minute: Information Source: https://www.airlines.org/dataset/per-minute-cost-of-delays-to-u-s-airlines/

 

If you are interested to see the code, please view my Github:

https://github.com/hmo99/Flight_Delay

 

 

 

To leave a comment for the author, please follow the link and comment on their blog: R – NYC Data Science Academy Blog.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)