Public Transport Model (PTM)
Overview
The Public Transport Model (PTM) is an assignment-only network model owned and maintained by the Transit and Accessibility Modelling team at the Department of Transport and Main Roads (TMR), used to provide detailed forecasts for the South East Queensland public transport network.
PTM inherits many of its inputs from the South East Queensland Strategic Transport Model (SEQSTM), a full 4-step demand model, but analyses public transport in greater detail. The primary function of PTM is to act as an operational level model used to inform service planning initiatives for short to medium term projects (0 to 10 years), as opposed to the SEQSTM which includes full demand response and is generally used for longer-term horizons (10 to 30 years). The key advantage of PTM however is that it takes only a few minutes to run, as opposed to the SEQSTM which takes approximately a whole day, and therefore many service plans can be tested in a short period of time.
The following image provides an overview of where PTM sits in TMR’s suite of models:
PTM represents the consolidation and upgrade of a number of previous models (TIM, Rail Load Model, 2018 Commonwealth Games Transit Model) which have been used by TMR and TransLink for the past decade.
The overall model consists of two core assignment modules; a frequency-based assignment (similar to that used in SEQSTM and other strategic models) commonly used to test alternative service plans, and a schedule-based assignment, used to test alternative timetable options by modelling specific timetable/s and passenger demand down to a minute granularity. An Olympics variant of the PTM timetable assignment (Games PTM or GPTM) has also been developed to model the two-week period covering the 2032 Brisbane Olympics. The user can choose whether the assignment is run with or without crowding. The model does not have fixed parameters such as time periods, modes and user groups – these are defined based on the type of assignment (frequency vs schedule and crowded vs uncrowded).
PTM relies primarily on observed demand however can be configured to import demand from other models, including SEQSTM. For proposed network changes, PTM includes an elastic function which estimates changes in demand based on changes in generalised costs between network scenarios.
Traditionally, network-based models such as SEQSTM are provided with “off the shelf” Reference Cases – a collation of inputs including demographics and transport networks, often based on committed and/or funded projects, which represent a snapshot for a given future year, and provides a starting point from which to build project-specific scenarios. PTM on the other hand is highly customizable; it does not include any Reference Case scenarios – inputs can be sourced from SEQSTM or elsewhere to build scenarios representing any future year.
Purpose
The purpose of this documentation is to:
- Provide an overview of the structure of PTM
- Outline the key steps undertaken for each model process
- Provide a walkthrough for new users on how to setup and use the model
The guide assumes that users are competent with Python programming and have a basic understanding of transport demand and assignment modelling with EMME.
Requirements
PTM is an EMME-based model which uses Python to convert key inputs into EMME compatible formats. At its core, PTM requires only Python and EMME to function, however additional software is required to allow for full functionality within PTM, including visualisation of outputs.
Critical Software Requirements:
| Software | Version Required |
|---|---|
| Python | 3.9+ |
| EMME | 4.7.0.1+ |
Note that PTM v2024 has been tested with Python 3.9 and EMME 4.7.0.1 only. Use of older software versions is not supported. The model, including the deploy_environment.bat file, is currently being updated to work with latest Python 3.13 and OpenPaths EMME 24.
Additional Software Requirements:
| Software | Version Required | Usage |
|---|---|---|
| Visual Studio Code | Source code editor, recommended for running the model. | |
| PostgreSQL | 13.5+ | Used to store PTM data, including key inputs. |
| PostGIS | 3.1.4+ | PostgreSQL extension used to store spatial data within PostgreSQL database. |
| Alteryx | 2021.2+ | Used to process validation and output workflows. |
| Tableau | 2020.4+ | Used for validation and output visualisation. |
A deploy_enviroment.bat file has been provided with this repository, compatible with Python 3.9 only. This batch file creates a virtual Python environment and includes all Python packages required for use within PTM, including the required version.
To use this please ensure that you specify the location of your Python 3.9 installation as well as the preferred directory for the virtual environment. For internal Transport and Main Road users, ensure you update proxy settings before running.
:: Existing Python 3.9 install
set python=C:\Python\Python39\python.exe
:: Virtual environment location
set venv=C:\Python\Python39\venvs\ptm_v2024
:: Proxy settings
::set http_proxy=http://***:***@www-proxy.qdot.qld.gov.au:8080
::set https_proxy=http://***:***@www-proxy.qdot.qld.gov.au:8080
Note that the PTM Assignment Model code used to run PTM through EMME does not use the Python 3.9 environment but instead the EMME interpreter for EMME 4.7.0.1.
v2024 Release
This version has included the following updates:
- Integration of 50c fares
- Inclusion of a process to allow for testing of alternative fare structures
- Updates to the Standard Model Dashboards to be used for PTM including both Frequency and Timetable dashboards
Version History
v2023
This version has included the following updates:
- inclusion of transit lanes
- conversion of network modifications from geojson files to geopackage
- conversion of the SMO process from hosting results within a base schema folder to an analysis folder
- general debugs and improvements identified through model application
v2022
This version included changes to the timetable, including platform based coding, new import functions from SEQSTM and updates to reflect validation changes.
v2021
This version included an upgrade to Python 3 and a major redesign of the PTM structure, including a redesign of the input notebooks used to generate EMME inputs.
Housekeeping Rules
- Remember to clear all outputs prior to pushing to the repo.
- Example schemas have been included within the schemas folders to use as a starting point for a new user.
Contact Information
For more information on PTM, please contact the Transit and Accessibility Modelling team.