Analysis of TSA Claims – Less than Expected

As I continue on my journey as an analyst, I keep running across interesting datasets. Over the last few weeks, I have been interested in various datasets around flying. So far I have found TSA claims from the past fifteen years and the number of passengers that have flown in the United States. Starting with the TSA claims, I wanted to see if I could find anything interesting.

First I wanted to acknowledge that it was a bit of challenge to collect and clean this data. Even though the TSA makes it available, I had to extract data from a PDF to get the most recent data and blend it into the existing Excel files. I tried to used the R library tabulizer but ran into issues with Java. I ended up using application Tabula that is built on the same technology. All of my code will be available on my GitHub.

The flight traffic data can be found on the Bureau of Transportation Statistics website.

Let us get a basic understanding of the number of claims files over the last 15 years before we mix it with traffic flight information.

This dataset started in 2002, which had the most claims. I suspect that the TSA inputted previous years because it is hard to imagine they have gotten that much better over the last 15 years, but I guess anything is possible.

The dataset also contains the type of items that were submitted for each claim. A claim can have multiple items per claim, so I had to parse out each item and reaggregate them into a separate dataset. Below is the top items that the TSA seems to continue to damage or loose.

The TSA has gracious also supplied the location the incident happened. Below are the top items broken up by the location of the incident.

It makes me wonder what is happening to all of this stuff at the TSA checkpoints when everything is within 10 feet of you.

Besides the locks, clothes, and luggage computers seem to get damaged pretty often. Either the TSA is building a super computer with all the laptops they are stealing, or traveling with laptops presents a mild risk. Seems you got better odds to bring your laptop on the plane then check it in. Recently I flew without my laptop, where I felt a bit naked, and the LAX TSA agent could not believe that I was not traveling with one. Like I actually had forgotten to take it out after they had been yelling it out 6 times in the past 10 minutes waiting.

Now that we know what the TSA is taking and where they are taking it, I wanted to find out what the odds were of getting money for my broken or stolen laptop. Here is a distribution of the amount of money people claimed for specific items.

Here is the likelihood of getting anything for those items.

Good luck ever getting your watch or MP3 player replaced!

Before we get to some more interesting visuals where I combined claim data with air traffic data, I would like to give a shout out to personal injury claim in JFK on February 12th, 2008 for #3,000,000,000,000. Not surprising the claim was rejected but kudos to you personal injury claim!

Moving past the ridiculous nature of that claim, I was curious to see if I could identify the airline that had the most problems with checked baggage. Below is a visual identifying Continental Airline as the airline with most claims per million passengers. This visual is only looking at checked baggage because I wanted to pinpoint the airlines that have the worst issues with checking bags in.

Taking a step back let us see claims rates by the most popular airports. This is looking at all claims.

It is not a surprise that JFK has the most claims per million passengers. This is where ratios can be a bit deceiving. For every million people that fly there are only about 25 claims. I am sure there is a lot of property that is lost without people reporting, but if someone damaged my MacBook Pro, you bet I would try to get at least something for it.

To put this in more perspective I plotted the total airline traffic next to the number of claims.

For the hundreds of millions of people who fly per year, there are relatively few incidents.

One of the final areas I wanted to explore was how much traffic flew through each state per year.

Even though we all like to hate on the TSA, they do manage to funnel lots of people through their gates. Good Luck TSA!


comments powered by Disqus