## Inspiration

Whenever we wanted to board planes, we thought that it takes too long, and there must be a more efficient way to do this. We also thought about how air hostesses do not get paid until everyone is on the plane, so we thought reducing the time that the boarding process takes would improve the working conditions

## What it does

It’s takes in a description of a plane and turns it into an array that’s meant to represent it. And after taking in all the details about the seating arrangements, and different classes of the plane, it decides the most efficient way for the passengers to board in. We had already predetermined an algorithm we believed would work most efficiently, so the challenge was making that algorithm work on our plane and in the most general form we could muster.

## How we built it

We used Python and it’s Numpy function, to create an arrays of arrays which would represent the different aspects of the planes/seating options as numbers. For instance the first class seats are given the number ‘8’, window seats on Odd rows the number ‘9’, on even rows the number ‘2’, etc.

## Challenges we ran into

In the initial stage, we had hoped to be able to use an API to turn a photo of a plane schematic into a platform we could adjust. But we never were able to figure it out.

Upon deciding to run Python, we had challenges on making our algorithm to differentiate the odd rows and the even rows for certain key points such as the middle rows, window seats, etc.

We also had issues on further improving our algorithm to consider things like families, groups, people with disabilities, etc.

## Accomplishments that we're proud of

For starters, we have a finished product.

Secondly, we were able to come up with something rather General, where we could somewhat show our idea, and help the customers visualize what the end results will look like.

We were able to make ways to identify between the numbers that are actually passenger seats and those which are empty spaces where people can walk through.

## What we learned

In general, we learned ideas we tried to incorporate from the Workshops such as how API can make photos into interactive platforms from the Shuttlestock presentation.

We also learned how to try and make generalized maths based codes from the talk from Ubisoft.

And we also learned a little about deep learning from the Matrox présentations.

And from completing our work, we learned more about what we could do with Python and Numpy, and what other things we could learn to improve this work. And also, what skills we should try to acquire from now do that the next works will be more sophisticated.

And personally I experienced how things work, since this is my first hackathon.

## What's next for Bombi

If we could add functions to make it even more general, and more efficient and to make it accommodate for families, people of disabilities, etc.

Also to improve the methods how we represent the final answers, and maybe add a little animation of how people enter the planes and show in real time how our model would look compared to how current airplane boarding systems look like.