TAMUhack26 AA Flight Delay App

FOR AMERICAN AIRLINES CHALLENGE

Inspiration

We were inspired by an American Airlines talk around logistical support in critical holiday seasons and how understaffed zones can have a cascading delay effect that costs immense amounts of money and resources to try to mitigate. We aimed to build an application that could help support logistical and staff movements in peak times.

What it does

We built this application specifically to focus on IAH airport with a mock dataset of flights in and out. The application can be signed into as a worker or as an administrator with proper authentication cookies for each. Once signed in, admins can view a report of all inbound and outbound flights ranked in order from their risk factors. These risk factors were analyzed by an AI integration that determines the relative risk of delay (based on factors like historical data, weather, etc.), as well as the impact of if a delay were to occur (based on how many flights that plane has after a delay, the kinds of airport volumes to deal with, etc.). With these two scores, AI empowers analytics by creating a ranking system of critical flights that admins can view on a map in the center of the screen. They can also send live messages run on a server to any workers signed into the system to redirect them to different stations according to the AI analytics and decisions made by the admin. Workers, once logged in, can see messages, record tasks, see team members, and accept new tasks/assignments with all associated detail.

How we built it

We built it primarily using web based application services running off a node.js background server. The pages are built in html with javascript logic and css styling. Behind the scenes, there are many node.js packages that run cookies, authentication services, file and csv reading and parsing, web http access, and more.

Challenges we ran into

We initially started our project designing based on python with a more application focused project, little to no web integration at all. However, we believed that a web based application would be more realistic, usable, and feasible for this hackathon. So we had to completely reorient our focus to designing in java instead of python.

Accomplishments that we're proud of

We are incredibly proud of the main features that were difficult to debug. These include getting a gemini API key integration to do our data analysis powered by AI. Also, getting a map embed into the web that is interactable and detailed with modern styling. Lastly, we are proud of our authentication, live messaging server, and cookies systems.

What we learned

We learned a lot about working together as a team on separate parts of code to integrate into a bigger project, as well as design principles, proper planning and phase staging (through our use of Trello task board), as well as how to write web and server logic in a multitude of languages and packages.

What's next for TAMUhack26 AA Flight Delay App

The features we would like to implement next are more in depth integration between admin and worker systems, running the messaging system through a database so that all messages are logged, tracked, and persist, and also getting QOL functionality for search bars, information display, and more.