Analyzing My Flying Patterns: Creating a Fictitious Airline and Determining Where and How Often I Fly

Ajay Jain
10 min readMar 31, 2020

--

With COVID-19, I’ve had more time on my hand to pursue some technological projects. Over the course of my stay at home as opposed to university, I have ended up creating graphs and determining findings from my Facebook messaging data (since 2012), created a new cooking website for my recipes, and determined how many times I have traveled through different airports since I was born.

For the four main airports I traveled through going up, Chicago O’Hare (ORD), Chicago Midway (MDW), Philadelphia (PHL), and Raleigh (RDU), the numbers are slight estimates from 1998–2015 based on my normal travel pattern of flying roundtrip through PHL and RDU at least once a year (with a couple exceptions). However, the data for the remaining airports is correct.

A Map of Every Airport I Have Traveled Through

Since I was born in 1998, I have traveled through a total of fifty four airports ranging across four continents (Africa, Asia, Europe, and North America).

In the United States, these airports are spread across sixteen states dotted throughout the country and through every time zone in the continental United States.

Ajay Airlines: My Flying Patterns As An Airline

I thought that one of the interesting takeaways from my data would be to convert my airport visits into a theoretical airline’s hub network. The big five airlines in the United States of America (Alaska Airlines, United Airlines, Delta Airlines, American Airlines, and Southwest Airlines) have a total average of nine hubs per airline. As such, I have decided to allocate Ajay Airlines nine hubs as well. In order to do this, I am going to designate my most frequented airports as hubs:

My top visited airports in the United States

Given a sneak peak at the data, there is some allocation that needs to be done. Obviously, an airline would not have a hub at both Chicago O’Hare and Chicago Midway. Furthermore, an airline would also not likely have a hub at Willard Airport near Champaign, Illinois (although a puny five gate airport would make for an entertaining hub). Therefore, I decided to do some combinations based on metropolitan area and airport size.

The hub allocations end up being the following:

  • Chicago O’Hare (118 total visits: ORD (96 visits) + MDW (15 visits) + CMI (7 visits))
  • Philadelphia (54 visits)
  • Raleigh (38)
  • San Francisco (15: 12 (SFO) + 3 (SJC))
  • Los Angeles (7)
  • Baltimore/Washington (7: 4 (BWI) + 3 (DCA))
  • New York-Kennedy (6: JFK (5) + EWR (1))
  • Seattle (6: SEA (5) + PAE (1))
  • Boston (4)

I ended up merging Champaign’s airport with the Chicago metropolitan area just because Champaign is not that far outside of Chicago and I felt that it should not be left out of the calculation given that I had flown through Champaign more than I had Seattle, New York City, and Boston. In terms of tiebreakers for determining which hub would be ranked over another if there were the same amount of visits, I decided to place whichever airport I most recently visited first. This is why LAX has a bigger use in my network than BWI, JFK is over SEA, and why MIA is left out of the map altogether (even though I have traveled through a lot of places, I have not flown through Miami since 2002).

Ajay Airlines’s hub network map ends up looking like this:

The airline network has mostly coastal hubs (three on the West Coast and five on the East Coast), with the exception of its main hub at Chicago O’Hare International Airport. I would argue that Ajay Airlines’ is a direct competitor to American Airlines and its partner Alaska Airlines. With the exception of Boston Logan, all of Ajay Airlines’ hubs are either within a market that American/Alaska has a hub in or lies within 150 miles of a competitor hub. On the West Coast, Seattle, San Francisco, and Los Angeles are both Alaska and Ajay Airlines’ hubs (with LAX also being an American Airlines hub). Ajay Airlines’ busiest hub, ORD, is American’s main Midwestern hub. On the East Coast, Ajay Airlines’ shares its Philadelphia and New York-Kennedy hubs with American, its Baltimore hub lies near American’s Washington National hub, and Ajay Airlines’ Raleigh hub (a former American Airlines hub) lies just 130 miles away from American’s Charlotte hub.

Based on how many times I have gone through these airports, however, my airline’s strategy actually wouldn’t play out exactly like American Airlines’ strategy. Rather, it would look something like this:

  • Chicago O’Hare (ORD): would be utilized similarly to DFW in American’s network and ATL in Delta’s network. As the airline’s overwhelming hub with extensive connections, Ajay Airlines would utilize ORD as a gateway for heavy demand flights in Africa, Asia, Australasia, Europe, North America, and South America that would be supplemented by connections flying to and from O’Hare. ORD would also be used as a gateway to regional Midwestern and Great Plains destinations.
  • Philadelphia (PHL): would be utilized similar to PHL in American’s network and would provide extensive service up and down the East Coast and to heavy demand and seasonal European destinations, thus acting as the primary European hub.
  • Raleigh (RDU): would be utilized similar to CLT in American’s network. RDU would act as a southeastern and Caribbean gateway (similar to Houston in United’s network) as well as a secondary European hub.
  • San Francisco (SFO): would be utilized similar to SFO and DEN in United’s network. SFO acts as the main Western state hub as well as the primary Pacific hub.
  • Los Angeles (LAX): would be utilized similar to LAX in Delta’s network, essentially acting as a small hub for heavy demand destinations in Europe and Asia.
  • Baltimore/Washington (BWI): would be utilized similar to DCA and PHL in American’s network, acting as a hub for Northeastern destinations and operating shuttles to and from the nation’s capital.
  • New York-Kennedy (JFK): would be utilized like JFK in American’s network, acting as a hub for high demand transatlantic destinations and a connecting point to airline partners.
  • Seattle (SEA): would be utilized similar to SEA in Alaska and Delta’s network, operating as the main Pacific Northwest hub.
  • Boston (BOS): would be utilized similar to BOS in Delta’s network, acting as a small reliever hub for demand that cannot be fit at other surrounding hubs (in our case, ORD, PHL, and JFK).

Realistically speaking, the airline’s hub network is way too coast-centric. If Ajay Airlines was actually real, shifting the Seattle hub and Boston hubs to more centrally located cities (such as Denver, Dallas, or Austin), would be a much better strategy.

Shifting Back to the Real World: How Many Airports Have I Visited?

Since I was born, my estimate is that I have flown through Chicago O’Hare International Airport ninety six times. Because I have lived within fifteen minutes from ORD my entire life, this makes a lot of sense, and as such, many of my flights originate or end at O’Hare. This can be seen quite well from my first graph, where the number of visits at O’Hare skyrockets the other fifty three airports:

A vast chunk of my flying has come over the past five years. My flying habits remained somewhat steady during the first ten years of my life until the recession in 2008, when train travel became much cheaper for my family and I than plane travel. After 2015, plane travel became frequent in my life, and I had my most airport visits in 2019 when I traveled through an airport fifty times. 2020 was on pace to break that mark, with 18 visits in a span of just two months. However, with COVID-19 disrupting the airline industry, this unfortunately might not be the case. That being said, assuming my current travel plans can remain intact (flying to and from Washington DC for an internship and to India for a wedding), I might still be able to reach my 2018 numbers.

Another data visualization that puts into perspective how much I travel is this stacked bar chart. On my y axis is the percentile of airports I have visited, with 1.00 being all fifty four airports that I have traveled through. As you can see from the graph, I visited my first twenty seven airports between 1998 and 2016. My remaining twenty seven airports that I have visited came within the last four years.

What Are My Flying Patterns? Where Do I Fly?

In this graph, a linked destination is a destination I have flown to or from. For example, RIC is a linked destination for ORD because I flew from ORD to RIC and SEA is a linked destination for ORD because I flew SEA-ORD. Destinations where I have flown both to and from (e.g. I have flown both ORD-PHL and PHL-ORD) are only counted once.

Similar to our first graph, we can see that O’Hare has the most linked destinations with 27 destinations (spread across three continents). Likewise, Philadelphia also comes in second with 9 destinations (spread across two continents). The amount of times I have visited Raleigh does not correlate with a greater number of linked destinations, as it is tied with a bunch of airports (Chicago Midway, London Heathrow, Mumbai, New Delhi) that I have visited far less frequently.

O’Hare vast number of destinations is quite evident given its importance in the world’s aviation system as well as its distance to my house. From a map, we can see the diversity of ORD’s linked destinations:

Honestly, not bad for twenty two years of flying.

I also wanted to examine how diverse my trips were when going through different airports. Therefore, I decided to add an additional statistic called a diversity ratio, which is the ratio between number of destinations and number of visits. The higher the ratio, the more likely I would be on a unique airline route.

At the top of the list are airports I have not frequented often and when I have, they typically been hubs utilized for connecting to another flight. As such, their diversity ratios are higher than the other airports and are seen as outliers in my data.

Abu Dhabi’s ratio makes a lot of sense, as each linked destination is a unique continent (flying from Los Angeles, flying to Cairo, flying from New Delhi, flying to London). It is the only airport in my data that is connected to a linked destination on each of the continents I have traveled to.

Half of these airports are outside of North America, and five out of the eight of these outlier airports were used to connect from the United States to either Europe or India. The airports within the United States happened to be connecting points during odd routes when I would have to connect instead of flying directly from O’Hare to my actual destination.

On the other hand, airports with lower ratios are typically airports that I frequent a lot more. VOBG (the old airport in Bangalore) is the exception here. All three of my flights at the old BLR were to or from Mumbai.

Despite flying in and out of RDU since I was just two months old, I have only traveled to or from six unique destinations (Chicago O’Hare, Chicago Midway, Philadelphia, Detroit, St. Louis, and Dallas Fort Worth). Similarly, Philadelphia is also the same way, with its linked destinations being Charlotte, Chicago O’Hare, Chicago Midway, Detroit, New York-Kennedy, London Heathrow, St. Louis, Raleigh, and Rome.

As a result, Chicago Midway, Mumbai, New Delhi, and London have much higher diversity ratios than Raleigh, because they have a lot less airport visits given the same amount of destinations. LAX, which I visited seven times, had trips to and from five unique destinations (one destination less than RDU), giving a ratio of 0.7142857.

Conclusions

Given that a vast majority of my airport visits are in Chicago, Philadelphia, and Raleigh, it can be easily inferred that most of my traveling is to see my relatives in Delaware and North Carolina. I suspect, however, that a much bigger spike in airport visits will occur in the Washington, DC area as I eventually make my way towards moving over there. As a result, my visits to Philadelphia’s airport will decrease because I won’t fly from Washington to Philadelphia. This was actually seen last year, when I would take the Amtrak from DC to Delaware instead of flying through PHL.

Furthermore, I travel way more often than the average person my age. If we look at data from Statista, the average American between the ages of 18–24 took 2.3 airplane trips in 2017. In 2019, I took fifty flights (essentially one flight a week). In terms of total trips (counting round trip flights), I took sixteen during 2019.

A lot of my traveling was also done over the past five years. Prior to my senior year of high school, I traveled about as much as the average American. However, with cheaper airline tickets, more family trips, job opportunities, aircraft retirement flights, and airline inaugural routes, my plane trips skyrocketed.

The number of linked destinations for Mumbai and Delhi actually rivaled and exceeded most airports I have traveled through in the United States. This makes sense considering I have made three trips to India in the span of fifteen months. Furthermore, because a lot of my travel is from a massive hub for American and United, I typically do not take connecting flights. When I do, I typically haven’t gone through the same hub more than once (Etihad’s hub in Abu Dhabi and Delta’s hubs in Detroit and New York-Kennedy are exceptions).

Hope you all enjoyed this!

Ajay

--

--

Ajay Jain

Working in civic tech as a software engineer. Previously worked in politics, interned in government. Freelance travel writer.