OwlEats

Predicted Wait Times

Motivation

We were personally invested in this project as we had walked into crowded servery lines on numerous occasions. As busy students, it would have been great if we knew how long the line would be before making the journey to a servery, which would have helped us optimize our time. These experiences inspired us to help the Rice community by saving people time from having to wait in massively long lines.

Journey

We learned how to develop an Android app with Android Studio, make an interactive user interface in Java, incorporate Python and Java into a single application, use Matplotlib to generate graphs, parse data from a txt file to utilize in Android Studio, and explore multithreading for app development.

One of the main challenges we encountered as a group was debugging in a new IDE, Android Studio, for the first time. All of us are still learning how to program in Java, so some errors were tricky to comprehend. To overcome this challenge, we took advantage of online searching of forums and documentation to identify solutions to our problems.

Next Step

The future wait time predictions currently only rely on averaging past data, but these predictions can be more accurate if we use machine learning models to make predictions and also take into account data from the current day. For future works, we plan to implement a feed-forward convolutional network to improve short term future wait time predictions, such as 30 minutes into the future. We do this by training it to analyze past wait times, new wait time data collected, and factors such as day of the week, special events, or holidays to best estimate the future waiting time. The impact of our project is that it can help students plan out their day better by reducing wait time in line. In the future, we would like to expand this product to student businesses on campus, such as Chauss, the Hoot, EastWest, etc. This could be implemented by using motion sensors instead of card swipes to predict wait time.

Conclusion

Overall, we learned how to identify a problem, formulate different ways to solve the problem, and eventually work together to develop a solution. We are very proud of what we were able to learn and accomplish within such a short period of time, and are proud to better our Rice community by reducing their time spent waiting in lines.