Introduction
My friend James Dunkerley is a cool dude and he does many great things in Alteryx. You can find his website by clicking here. James is also really smart and does a great job sharing his knowledge.
Tonight I borrowed a technique from James to animate airline flight paths using a great circle approach. There are websites available to do these types of calculations but James wanted to do the calculations for usage in Tableau. Although I didn’t have time to ask James if I could borrow his work, I’m sure he won’t mind!
Once I used his approach in Alteryx to do the calculations, I decided to modify the approach to add a degree of realism to the animations I was able to produce in Tableau.
I decided to use the distance between the origin and destination of the flight segments to proportionately speed up shorter flights. This technique required a bit of work on the Alteryx workflow so I decided to write this article.
The Original Alteryx Workflow Developed by James
You can click this link to find the work of James. Figure 1 shows his workflow.
Here is the Tableau animation created using the data created by this workflow.
The Revised Alteryx Workflow
I made a few modifications to the workflow to allow a variable number of segments along each flight path. This allows the animation to appear more realistic, with shorter flights arriving before longer flights. Figure 2 is the revised Alterxy workflow.
Here is the Tableau animation created using the data created by this workflow.
Going Professional Grade
The following two animations are real-world data. As shown in Figure 3, an expanded Alteryx workflow was built to convert actual travel data to the format needed for the Great Circle calculations. That output file was sent to the workflow shown in Figure 2.
The first animation represents only 1 month of domestic travel. The only flight segments shown here are for the origin to the first destination. Connecting flights and return flights are not shown for simplicity.
In the following animation, all departing flight from Detroit are shown for the first 4 months of 2016. The animation is now a function of time rather than the movement along the flight paths.
Previous Work
This article is a part of a continuing series that I am writing for processing travel data. Previous articles can be accessed by clicking on the word “Travel” in the word cloud that exists on the right hand side of this blog, as shown in Figure 3. The airport lat/long values used in the animations in this article are available in the custom geocoding article shown in Figure 3.
Future Work
I’m going to expand my Alteryx workflow to process all flight segments for domestic and international flights. Once I have that, I’ll add additional Tableau controls to be able to view travel for an individual, for a division, or for the entire company. After that, more analytics, more visualizations, and more fun!
Don’t be surprised to see a part 2 to this article as I strive to make a general-usage tool for Alteryx and Tableau. Once I make further progress, I’ll publish my workflows.
Also, work like this has been done previously in Tableau by Alan Walker (and possibly others). I don’t remember what other tools Alan might have used to process the data. However, if you are interested in learning more about this approach, be sure to search for Alan’s work because he is one of the leaders in doing this type of geographical analysis in Tableau. Thanks again, James, and Alan!
Awesome work Ken! The animation really makes these types of visualizations so much more impactful. I love how it just requires a handful of steps in Alteryx to generate the requisite data.
Hi Mike,
Thanks for writing! I’ve taken about 6 months off to recover after the end of the blogging experiment. I hope to write the article that describe what I learned in the blogging experiment. Hopefully, that will occur soon, although work and family are taking a lot of my time.
Great to hear from you,
Ken