Notebooks

Four notebooks have been retained for the processing of demand:

  • 01_build_base_year_demand.ipynb – designed to build the base demand for PTM.
  • 02_import_future_year_demand.ipynb – import future year demand from a STM and other approach.
  • 03_grow_future_year_demand.ipynb - designed to grow base demand by growth seen in STM demand matrices.
  • 03_modify_centroids_connectors_w_demand.ipynb - creates a weighted centroid location to improve validation outcomes.

Files Created

These notebooks create the following inputs in ..\schemas\{SCHEMA_NAME}\_emme_inputs\_matrices:

  • imported_matrices.omx
  • imported_matrices_list.csv

Details

01_build_base_year_demand.ipynb

The Jupyter notebook is designed to build a base year demand model for a PTM, including:

  1. User Defined Variables: The notebook starts by defining user-specific variables such as schema names and paths for factored transactions. The notebook also sets up the schema and initialises model parameters, paths, and the PTM engine.
  2. Import Factored Transactions: Factored transactions are imported from a specified path and displayed. Reference tables are also imported for route types, ticket status mapping, and user groups are imported and processed.
  3. Filter and Join Data: Transactions are filtered to valid route types and joined with user group information. Lost transactions are identified and handled.
  4. Aggregate Data: The valid transactions are aggregated by date to calculate the total factored journeys.
  5. Map to Centroids and Zones: Transactions are mapped to centroids and zones based on the model type (timetable or frequency).
  6. Create Journeys Table: A journeys table is created by sorting and aggregating the mapped transactions.
  7. Factor Intra Zonal Demand: Intra-zonal demand is factored into the model.
  8. Match to Time Periods: The journeys data is matched to specific time periods using imported time period data.
  9. Aggregate to Avg Daily: The journeys data is aggregated to average daily journeys, with adjustments based on the model type.
  10. Output to OMX / Journeys File: The final journeys data is exported to OMX files or journey files based on the model type.

02_import_future_year_demand.ipynb

The notebook is designed to import future demand data from an STM (Strategic Transport Model) source and save the results in a format suitable for modelling within PTM (Public Transport Model). It includes the following key sections:

  1. User Defined Variables: Defines the schema name for the PTM model. Initialises the PTM schema and sets up the necessary parameters and paths.
  2. Frequency Models with Direct Demand Import: Imports demand data directly from STM for frequency models, renames matrices to PTM format, and exports the data to PTM.
  3. Timetable Models with Direct Demand Import: Similar to the frequency models, but includes additional steps for binning demand into 15-minute intervals and creating timetable matrices.
  4. Import Events Demand: Handles the import of demand data for event-based models, with specific assumptions for events including the Olympics.

The notebook uses various functions to import, process, and export the demand data, ensuring it is correctly formatted for use in PTM.

03_grow_future_year_demand.ipynb

This notebook outlines a process to project future demand based on current demand using growth factors derived from STM (Strategic Transport Model) runs. It includes the following key sections:

  1. User Defined Variables: Defines schema names, base and future demand schemas, extraction folder paths, and time period matrices. Initialises the schema and log the setup process.
  2. Data Ingestion: Ingests base, future, and current centroids from specified paths and ingests base, future, and optionally greenfield matrices.
  3. Match Centroids to SA2: Matches the centroids to SA2 (Statistical Area Level 2) regions for base, future, and current datasets.
  4. Sectorise Matrices by SA2: Sectorises base and future demand matrices by SA2 and calculates the demand for different time periods.
  5. Obtain Growth Rates: Calculates growth factors by comparing base and future demand.
  6. Apply Growth and Add Greenfield Demand: Applies the growth factors to the current matrices to project future demand. This notebook also optionally add greenfield demand if available. Saves the updated matrices and log the changes in demand.

The final output includes updated demand matrices reflecting the projected future demand.

03_modify_centroids_connectors_w_demand.ipynb

This notebook aims to adjust the existing centroids and connectors to better reflect demand patterns for base scenario runs. The process retains the node end of the connectors and involves two main steps:

  1. Modification of Existing Centroids: The centroids are adjusted to be demand-weighted, ensuring they better represent the actual demand.
  2. Modification of Connector Lengths: The lengths of the connectors are modified accordingly.Key steps in the notebook include:
    • Setting up the schema and importing necessary data such as zones, stops, and demand.
    • Calculating weighted centroids for regular zones based on demand.
    • Adjusting connectors to align with the new weighted centroids.
    • Exporting the modified centroids and connectors for further use.