- Source code to accompany Defunctland's video "FASTPASS: A Complicated Legacy"
- Download the video at https://www.youtube.com/watch?v=9yjZpBq1XBE
- This source code is licensed under the Creative Commons 4.0 International License
- See the file named LICENSE for details
The simulation is written in Python and has been tested with python 3.6.9. Download the latest version of python here: https://www.python.org/downloads/
The code also uses Jupyter Notebooks, available here: https://jupyter.org/install
An Anaconda environment has been provided (shapeland.yaml) that automatically installs all necessary dependencies to a new interpreter. To create the environment (also named shapeland), execute the following from the Anaconda Prompt application:
conda env create -f shapeland.yaml
After that, you can show the associated Jupyter Notebook from the Anaconda Prompt, by first activating the shapeland environment and then starting the server in your browser:
conda activate shapeland
jupyter notebook aps.ipynb
If you're using an IDE, choose this newly created Conda environment as the interpreter when running the Notebook. Below we have provided instructions for the two most common IDEs among students.
For Visual Studio Code:
- Ensure that you have the Jupyter extension by Microsoft installed. If you don't have that yet, look it up on the Marketplace, install it and reboot Visual Studio Code.
- Open the .ipynb file. You may already be prompted to choose an interpreter; if so, select the shapeland Conda environment.
- If you aren't prompted automatically, there's a button in the top-right of the Notebook window that lists the version of Python currently selected. Click that, then "Select another kernel..." and "Python Environments".
- This should give you a list of all available interpreters, both within and outside Anaconda. If you can't see the shapeland environment yet, click the Refresh icon in the top right of the search bar.
For PyCharm: apparently the Community version only supports reading Jupyter Notebooks, and to run them, you need a Professional trial or license. So regarding PyCharm, my advice would be: use Visual Studio Code.
There are 5 main classes in this simulation:
-
activity.py: An activity is something an agent can do inside the park. Activities include going on rides, eating, and so on.
-
agent.py: Simulates one guest making decisions in the park.
-
attraction.py: Encapsulates all of the calculations to simulate an attraction, including whether it has FASTPASS, its hourly capacity, how that capacity is split among different lines, and so on.
-
behavior_reference.py: Each Agent has a behavioral archetype. -- Ride Enthusiast: wants to stay for a long time, go on as many attractions as possible, doesn't want to visit activites, doesn't mind waiting -- Ride Favorer: wants to go on a lot of attractions, but will vists activites occasionally, will wait for a while in a queue -- Park Tourer: wants to stay for a long time and wants to see attractions and activities equally, reasonable about wait times -- Park Visitor: doesn't want to stay long and wants to see attractions and activities equally, inpatient about wait times -- Activity Favorer: doesn't want to stay long and prefers activities, reasonable about wait times -- Activity Enthusiast: wants to visit a lot of activities, reasonable about wait times -- Archetypes can be tweaked and new archetypes can be added in behavior_reference.py.
-
park.py: The park contains Agents, Attractions and Activities. -- Total Daily Agents: dictates how many agents visit the park within a day -- Hourly Percent: dictates what percentage of Total Daily Agents visits the park at each hour -- Perfect Arrivals: enforces that the exact amount of Total Daily Agents arrives during the day -- Expedited Pass Ability Percent: percent of agents aware of expeditied passes -- Expedited Threshold: acceptable queue wait time length before searching for an expedited pass -- Expedited Limit: total number of expedited pass an agent can hold at any given time