Announcing Alteryx + Snowflake | Alteryx and Snowflake make analytics and data science fundamentally easier. With the new integrated starter kit, you can push down data prep transformations and more into Snowflake for faster data quality and analytics output. Learn More
We are no longer accepting submissions for Use Cases, but please stay tuned for the Excellence Awards submissions that will be coming soon!

Alteryx Use Cases

Read Alteryx customer stories to learn how they transform their organizations into becoming a data-driven business.

Predicting Passenger Flows at Dubai International Airport

StephMills
Alteryx
Alteryx

Overview of Use Case

As Head of the Business Information team, I am responsible for providing performance reporting and analytical support. The majority of our work supports the Strategy & Development and Operational departments, while working in close collaboration with IT.  For this use case, we used Alteryx and Tableau to develop a prototype of a passenger flow prediction tool.

 
Describe the business challenge or problem you needed to solve

Dubai International Airport has a high volume of transfer traffic, with a lot of peaks in passenger volumes throughout the day. An immigration hall or a transfer security checkpoint can be completely empty at times but filled with passengers 30 minutes later.

 

You can plan for the expected passenger load profile to a certain extent, but changes to flight arrival times on the day can have a significant impact on the actual passenger load profile.

 

Actual queue information from sensors and cameras was already available in our Airport Operations Control Center (AOCC), but there was a need for passenger load predictions and resource requirements in advance. This would allow operational teams to collaborate with other stakeholders and proactively open more security or immigration lanes, in advance of the actual passengers arriving. These predictions would allow for resources to be moved from quiet touch-points to locations predicted to be understaffed due to changes in passenger load profiles.

 
 
Describe your working solution

The passenger flow model prototype uses a real-time feed from our flight database, enriched with supporting reference tables. It’s giving us  the latest information on expected flight arrival times and where the aircrafts will be parked. We use Alteryx Designer to create two detailed workflows per terminal: PAX Demand and Counter Supply – which tackle the following:

 

1) Combine the flight-level data with other reference inputs.

2) Cleanse the combined data using basic functions like Select, Formula, Filter, DateParse.

3) Process the data into required PAX demand aggregations using functions such as Generate Rows, Multi-Row, Summarize.

4) Optimize the counter requirement by looping it through a couple of nested iterative loops until it reaches the desired thresholds and service levels.

5) Write the output into a database to be visualized in Tableau.

 

These workflows are scheduled on Alteryx Server to run every 5 mins in order to keep the data as near real-time as possible.

 

The output data from the workflows is used to create visualizations and dashboards in Tableau Desktop that are then hosted on Tableau Server to be shared across the organization and wider airport community. These dashboards are also refreshed frequently to display the latest information.

 

Our deployment strategy was to keep our workflows very simple and streamlined such that any change to the underlying logic or a business process improvement would be a quick and easy tweak. This way, the business can see within a few days if the new process/logic is working as per the plan, and if not, they have the flexibility to change it again.

 

Picture 1.pngPicture 2.png

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Describe the benefits you have achieved

The main goal of the prototype was to find out exactly what our operational teams needed out of a tool like this, and then go to market to purchase it. We didn’t have clear expectations of actual performance improvements.

 

However, days after the tool went live, our operational teams used the information to move resources from one terminal to another, thereby preventing serious overcrowding and queue build-up. This is something that wasn’t possible before because the information simply wasn’t available. After a few months of using the tool, we observed clear decreases in queue lengths and wait times across many different touchpoints.

 

The prototype had done its job and we were ready to replace it for an externally developed tool, fit for a 24/7 airport operation. Then Covid-19 hit and significantly impacted our business and budgets; projects had to be put on hold or cancelled and what was supposed to be a short duration prototype, was now the only option for this functionality.

 

The impact of COVID-19

Covid-19 changed the way airports operate by introducing the need for PCR test stations, segregation of passengers based on origin, temporary closing of terminals, etc. Often changes had to be made overnight which then required updates to our flow model. Because we developed the tool internally, we were able to respond rapidly to any changes and at times released multiple updates per week. Close collaboration between the teams and everything done using Alteryx and Tableau gave us the agility we needed.

 

The tool has been running successfully for over two years. It’s accessed by people from multiple stakeholders through video walls, desktops and tablets. It has given us the flexibility to continue to add value in a very cost-effective way.

 

Picture3.pngPicture4.png

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Related Resources
 
 
Picture5.jpgPicture6.jpg
 
No ratings
Comments
Tanya3
5 - Atom

With so many people logging into the command center, I wonder if that slows the server?