CAP 3321C Final Project · Airports case study

Lost, Damaged, Denied

Insurance claims against U.S. airports, 2002–2009

Data Wrangling · Fall 2026

Introduction · The data

Three files and nine questions

TSA claims
tsa_claims2.csv, one row per claim, columns
Global airport database
IATA code with latitude and longitude, used to place airports on the map
State outlines
Shapefile of U.S. states, read with GeoPandas
Q1Most common claim type
Q6Share of the claim that gets paid
Q2Most common claim site
Q7Five airports with the most claims
Q3Most common type at each site
Q8Total close amount over time
Q4Typical claim amount
Q9Map of claims by airport
Q5Overall approval rate
Part 1 · Data cleaning

The data and how we cleaned it

Part 1 · First look

Every column arrives as text, and the values say why

Reducing every value to its shape exposes the formats a column actually holds: amounts read $9;999.99␣, and cells hold the text -.

Part 1 · Missing values

The file holds two recording periods

claims received 2002–2009, with close amounts and dispositions
claims received 2010–2015: Close Amount and Disposition never recorded

Part 1 · Recover before removing

Outcome labels and broken dates are restored before dropna()

Status values use the disposition vocabulary. Status and Disposition match one to one, so these labels are mapped and missing dispositions are filled.
Claim number2005051985108 Incident date2005-04-30 Date received17-May-55 → 2005-05-17
receipt dates repaired: in impossible years and contradicted by the incident and claim-number dates
of
incident dates written 17-MAR-0201 recovered: from the year field where it fits, otherwise from the receipt date

Part 1 · Cleaning pipeline

Twelve small functions, chained with pipe()

Part 1 · Result

The required dropna() keeps about  claims, all received 2002–2009

complete claims with float amounts and true dates
incomplete rows removed, including every claim received after 2009
Part 2 · Section 1 questions

What the claims say

Q1 · Claim type

Two in three claims are for lost property

Passenger Property Loss
Property Damage
the two together; theft, injury, and the rest are rare

Q2 · Claim site

of claims start in checked baggage

claims at Checked Baggage
at the security Checkpoint ( claims)

Q3 · Type at each site

Loss leads in checked baggage, damage at the checkpoint

of checked-baggage claims are Passenger Property Loss
of checkpoint claims are Property Damage (loss: )

Q4 · Claim amount

The typical claim is , not the average

median claim: half of all claims are below it
95% of claims are below this amount
largest claim, a personal-injury demand that was denied

Q5 · Approval rate

Fewer than half of decided claims are approved or settled

approved or settled: the claimant received a payment
approved in full

The rate counts decided claims only (): approved, settled, or denied.

Part 3 · Section 2 questions

Payouts, airports, trends, and the map

Q6 · Share of the claim paid

Approved means paid in full; settled usually means half

of approved claims are paid exactly 100%
median payout on a settled claim; are paid exactly half
median for settled personal-injury claims

Q7 · Airports

Five large hubs account for one in five claims

of all claims at
claims at , the most of any airport

Q8 · Close amount over time

Total payouts fell from the peak

paid on claims received in
paid on claims received in 2009
→
share of claims paid anything, vs 2008: fewer payments, not smaller ones

Q9 · Building the map

Claims meet coordinates: merge on the code, then find each airport's state

rows in the airport database sit at latitude 0, longitude 0, a placeholder. Removing them makes every U.S. code unique, so the merge is one to one.
The nearest-state join keeps coastal airports such as Key West that fall just outside a simplified state outline, where a latitude/longitude box would drop them.

Q9 · The map (seaborn + GeoPandas)

About claims at continental U.S. airports

Q9 · Interactive

The same data, live

Full screen: map.html

Conclusion

What goes wrong, where, and what gets paid

  • 1Belongings, mostly in checked bags. Lost property () and damage () are nearly every claim; start in checked baggage, where property goes missing, while the checkpoint sees more damage.
  • 2Small claims, and payment is the exception. The median claim is ; of decided claims receive any payment.
  • 3Payment is all or half. of approved claims are paid exactly in full; settled claims receive a median of .
  • 4Volume follows the hubs; payouts fell. Five hubs hold of claims, and payouts fell after as fewer claims were paid.
  • 5The data needed repair first. Two recording periods, two vocabularies, and broken dates; the analysis covers the complete claims received 2002–2009.

Thank you. Questions?