ARC ActivitySim Implementation

Atlanta Regional Commission (ARC) prototype of the ActivitySim Framework

Folder and File Setup

1. Clone the repository to a directory (e.g., C:\Projects)
2. Create a folder named 'Data' in the above directory
3. Download the data (e.g., households.csv, persons.csv and ZoneData.csv) and place those in the 'Data' folder
4. Create a folder named 'Skims' under 'Data' and place the skim matrix (e.g., skims_complete) in that folder

Installation

1. Install Anaconda 64bit Python 3
2. Create an ActivitySim-oriented virtual python environment. Recommend using a virtual env with the environment.yml included in this repository.

   conda env create -f environment.yml

3. Switch the Virtual Env

   conda activate arc_activitysim

Data Processing (If necessary)

1. Update the data path in build_h5.py (e.g., C:\Projects\Data) located here: C:\Projects\client_arc_activitysim\scripts
2. Navigate into the scripts folder

   cd scripts

3. Run 'build_h5.py' to create the necessary inputs for ARC ActivitySim.

   python build_h5.py complete

   Available options

   • micro (25 sample zones)
   • small (500 sample zones)
   • complete (all zones)

Simulation

1. Run the simulation

activitysim run