Building on our final project for the course Intro to Optimization and Mathematical Programming, we extended our problem settings to be closer to real-life scenarios. We used Genetic Algorithm to further optimized the NYU Shanghai shuttle bus schedules.

Special thanks to Professor Zhibin Chen, the supervisor of this research project, for offering us guidance along the way and Dean's Undergraduate Research Fund (DURF) for generously providing us research funding.

The research outcome is presented in the form of a report. Please see DURF report.pdf.

I have also turned the report into a series of blogs to explain things in a less formal way. Hopefully, they are easier to understand.

• Genetic Algorithm & Spatio-Temporal Networks: Optimization of Campus Shuttle Bus Schedule Part 1: Spatio-Temporal Networks for problem formulation

• Genetic Algorithm & Spatio-Temporal Networks: Optimization of Campus Shuttle Bus Schedule Part 2: Genetic Algorithm for Baseline Problem (with Python Code)

• Genetic Algorithm & Spatio-Temporal Networks: Optimization of Campus Shuttle Bus Schedule Part 3: Python Genetic Algorithm for Extended Problem & Web App Demo

webapp directory is a Python Flask app where you can play with the application in your local host. Make sure you have Python and all the packages in requirements.txt installed.

1. Clone this GitHub repo to your preferred directory

git clone https://github.com/AlisonYao/DURF-Bus-Schedule-Optimization.git

2. Go to webapp

cd webapp

3. Run app

FLASK_APP=init.py flask run

4. Click open the URL below

* Debug mode: off
* Running on http://127.0.0.1:5000/ (Press CTRL C to quit)

5. You should see a window like this:

6. Scroll down and make your own changes. You can solve either the baseline problem or the extended problem.

7. Click the Submit button at the bottom, then you should see the results!

Here is a video that showcases what you should be seeing:

Try for yourself and see you can get a better result than mine.

Baseline problem solution: please see baseline_solution.py code

Extended problem solution: please see extension_solution.py code

The file Genetic Algorithm has all of my codes in progress.

• toy_GA_example.py contains the YouTube tutorials I followed to learn GA.
• baseline.py is the first attempt at solving the baseline problem. I enforced the demand constraint as a requirement, rather than a penalty. This version is significantly slower to run because its one iteration could take the time of 50 iterations, only because the first 49 times failed to meet the demand constraint. Although the solutions are guaranteed to met the demand, it performs poorly.
• baseline_v2.py is the successful second attempt at the baseline problem where I converted the demand constraint into penalty. Although meeting the demand is not guaranteed at first, the solution will eventually satify demand over time. Same with the rush hour constraint and the max working hour constraint. baseline_solution.py is the duplicate of this file.
• extension.py is a failed first attempt at the extended problem.
• extension_v2.py is a successful solution to the extended problem. extended_solution.py is the duplicate of this file.
• hpc_shuttle_v2.py just outputs results into txt files.
• temp.py is just a file for testing that doesn't have anything important.