Inspiration
We came across this idea while investigating common issues airline passengers presented as inhibitory to their flying experience. The most common complaint was passengers being unable to comfortably perform tasks during their flight duration simply because of their seating arrangement relative to other passengers with different flight routines.
What it does
Our website takes in two major quantifiers of passenger data: passenger preferences on a scale of 1 to 5 from most important to least important (eg. sleeping at 1 for most important, and working at 5 for least important) and a binary preference for 5 major passenger criterion (sleeping,socializing, working, light on, fan on) outputting only a 'yes' or 'no' for each category.
How we built it
First, we theorized different clustering algorithms geared towards arranging passengers on a plane based on their preferences. We ultimately chose Djikstra's shortest path algorithm because we wanted an ordered sequence of passengers based on the weights of their preferences. After building the algorithm, we set up a Firebase back end that stored randomly generated passenger data and user inputs. We also set up a Flask server on Google App-Engine that we used to host our algorithm. Finally, we developed a user interface for our website that would visualize passenger arrangements before and after the implementation of our algorithm.
Challenges we ran into
It was difficult to decide on which algorithm would best fit our intended goal of optimizing passenger arrangement based on weighted individual preferences. The implementation of the algorithm/back end into Google Cloud Platform also proved difficult because of the long deployment time. Finally, optimizing our website for different web browsers presented a significant challenge that we are still troubleshooting
Accomplishments that we're proud of
We were able to develop a working platform by which to arrange airline passengers based on their in-flight preferences relative to other passengers with conflicting or like-minded orientations
What we learned
We learned how to apply mathematical concepts such as Dijkstra's algorithm towards real-life applications that can improve customer acquisition and customer loyalty for airline companies.
What's next for Fly Savvy
We hope to present our novel algorithm to airline providers, and potentially aid them in their customer acquisition and retention due to the improved flying experiences we provide for their airline passengers.
Log in or sign up for Devpost to join the conversation.