The RDR 2024.2 public release includes updates across the entire tool suite, primarily focused on use of publicly available data to generate an RDR analysis. The documentation has been expanded to include public data workflows as well as a new Reference Scenario based on the 2022 Ferndale earthquake. Updated visualizations, particularly in the Tableau workbook, provide better insight into where and under what scenarios benefits are found. Several highlights of the release are detailed below.
RDR Metamodel
- Reformatted the Model Parameters input file and eliminated the User Inputs input file to improve usability and clarity of how to define the scenario space.
- Added two new user parameters to control convergence criteria for the core model (AequilibraE user equilibrium routing): maximum gap and maximum number of iterations.
- Fixed a bug in the regression module that was assigning core model run outputs to the incorrect resilience projects.
RDR ROI Analysis Tool
- Updated methodology for the resilience return on investment (ROI) module for monetization of trips lost/gained and calculation of safety, noise, and emissions benefits.
- Corrected a bug in the benefits calculation related to associating network performance metric savings with project benefits.
- Developed a new Map dashboard in the Tableau workbook for visualizing the user's resilience projects geospatially. The map allows users to dynamically explore their project outcomes and symbolize network links by average exposure, project cost, and regret ranking.
- Refactored ROI module for readability and efficiency.
- Refreshed Tableau workbook with clearer text and resolved visualization bugs.
RDR Benefits Analysis Tool
- Renamed tool to emphasize its focus on understanding the distribution of benefits across users of the transportation network.
- Improved usability of the TAZ metrics HTML report by adding toggle on charts so users can switch between showing data for (a) all TAZ/TAZ pairs versus (b) only the subset of TAZ/TAZ pairs that would be impacted by the disruption.
- Refactored the equity overlay tool to rely on geospatial processing via geopandas instead of arcpy, meaning that an ArcGIS license is no longer needed to run the tool.
- Updated default layer for the equity overlay component to use data from the Council on Environmental Quality (CEQ) Climate and Economic Justice Screening Tool (CEJST) in lieu of the prior default data from the Department of Transportation (DOT) Equitable Transportation Community Explorer.
Public Data
- Documented publicly available datasets and general workflows in the User Guide and Reference Scenario Library for developing:
- Road and transit networks,
- Trip tables,
- Hazard impacts.
- Added helper tools demonstrating various public data workflows, including (a) creation of a road network from OpenStreetMap data, (b) implementation of a gravity model to translate production-attractions to origin-destination trips, (c) overlay of historical earthquake data on a real-world network to calculate link-level capacity and damage impacts.
- Created a new Reference Scenario to demonstrate the above public data workflows based on the 2022 Ferndale earthquake.
Helper Tools
- Added new functionality to the RDR User Interface:
- Updated the order that parameters are set to prioritize those essential to running RDR.
- Added more input validation to improve usability.
- Updated scenario input validation helper tool to produce a CSV output that allows the user to review summary statistics on certain fields in the node and link files to check their reasonableness.
- Built in more user flexibility for the suite of Format Demand helper tools.
- Deprecated the GMNS_link_conversion Excel helper tool.
Other Updates
- Rebuilt the RDR conda environment, which now includes geopandas and osm2gmns.
- Updated default values in the ROI module to align with the latest USDOT BCA guidance from November 2024.
- Updated default repair cost values to align with the latest FHWA data.
- Converted all dollar values to 2023 dollars.
See documentation files for additional details.