Run the Model Setup 🚀 ¶

1. Set Up the Model Run Directory 🎂 ¶

a. Open the OpenPaths EMME Shell and a terminal and activate your tm2py virtual environment: 🐍 ¶

OpenPaths EMME Environment is set to:
OpenPaths EMME 25.00.01.06 64-bit,   Copyright 2025 Bentley Systems, Incorporated

Python Path is set to:
C:\Program Files\Bentley\OpenPaths\EMME 25.00.01\Python311\

C:\Users\lzorn\Documents>E:\GitHub\tm2\tm2py_env\Scripts\activate

(tm2py_env) C:\Users\lzorn\Documents>

b. Configure Input and Output Paths ⚙️ ¶

If needed, edit the configuration file located at: tm2py-utils/tm2py-utils/config/develop/setup_config_mtc_2015.toml

c. Run `setup_model.py` 🍰 ¶

This script is a light wrapper for SetupModel and it takes two arguments:

the location of the setup configuration file from the previous step
the model run directory.

Either argument can be relative or absolute paths

(tm2py_env) E:\GitHub\tm2\tm2py>python scripts\setup_model.py E:\GitHub\tm2\tm2py-utils\tm2py-utils\config\develop\setup_config_mtc_2015.toml E:\TM2\2023_TM2_test_20250606

TODO: Fix this Note: You may need to update the emmebanks to the latest version before the model will run.
Also: We used a different WalkTransitDriveSkims.xls file in the CTRAMP folder.

2. Run the Model 🎂 ¶

While still in the activated virtual environment:

Navigate to the model run directory you setup in the previous step.

Run the model:

(tm2py_env) E:\GitHub\tm2>cd E:\TM2\2023_TM2_test_20250606

(tm2py_env) E:\TM2\2023_TM2_test_20250606>python RunModel.py

Understanding Model Outputs

After running the model, comprehensive outputs will be generated. For detailed documentation on output files and data formats, see:

Output Overview - Complete guide to all model outputs
CTRAMP Output File Specifications - Detailed field-level documentation for household travel behavior results

User Configuration 🛠️ ¶

Model Configuration ⚙️ ¶

The model config file allows for customization on the the model run performance settings.

Performance & Processor Configuration 🚀 ¶

For optimal model performance, proper processor configuration is critical. tm2py uses parallel processing in multiple components including EMME assignment procedures, CTRAMP threads, and highway distribution.

Complete Processor Configuration Guide

📊 Processor Configuration Guide - Comprehensive guide covering:

All processor configuration locations (EMME, CTRAMP, highway distribution)
Performance optimization for different system types (desktop vs server)
Format options ("MAX-2", "MAX/3", specific numbers)
Troubleshooting and best practices

Network Acceleration 🚀 ¶

Emme Openpaths provides the network accelerate option, which allows for faster assignment on smaller machines.
WARNING: This has lead to some instability with model runs completing so, especially on large machines, this should remain off 🚧

To enable this under [highway] in toml

    network_acceleration=true

Parallel Highway Assignment 🛣️ ¶

tm2py offers the option to run assignment in parallel to reduce runtime. This can be achieved by including the following configuration under [emme] in the model_config.

    [[emme.highway_distribution]]
        time_periods = ["AM"]
        num_processors = "MAX/3"
    [[emme.highway_distribution]]
        time_periods = ["PM"]
        num_processors = "MAX/3"
    [[emme.highway_distribution]]
        time_periods = ["EA", "MD", "EV"]
        num_processors = "MAX/3"

Otherwise, to turn this feature off, explicitly configure tm2py to use 1 thread:

    [[emme.highway_distribution]]
        time_periods = ["EA", "AM", "MD", "PM", "EV"]
        num_processors = "1"