September 2024 Meeting

Bill Hereth · September 9, 2024

Meetings

Date/Time: Thursday September 12, 2024 10:00-11:30am
Location: MAG Offices - 586 E 800 N, Orem, Utah 84097

Agenda

Welcome & Introductions

  • Anyone attending for the first time?
  • User Spotlights

Discussion Topics

  • Model Status Update (link)
    • Utah Statewide Travel Model (Hayden Atchley, UDOT)
    • Wasatch Front TDM (Chad Worthen, WFRC)
    • Cache TDM (Jeff Gilbert, Cache MPO)
    • Dixie TDM (Nicholas Gayer, Dixie MPO)
    • Summit-Wasatch TDM (Hayden Atchley, UDOT)
    • Iron TDM (Hayden Atchley, UDOT)
  • Git Training (Greg Macfarlane, Brigham Young University)
    • Date: Tuesday, October 15, 2024
    • Time: 2:00 - 4:00 pm
    • Location: Redwood B Conference Room @ UDOT Complex
    • Summary: Greg Macfarlane will provide an introduction to software version control for transportation analysts. This training is aimed at those who have heard of Git, but have never used it, or who have used it but didn’t really know what they were doing.
  • Utah MUG Website (Chris Day, WFRC)
  • Statewide Resources
  • Others?

Presentations

  • Visualizing WF TDM Outputs via the New vizTool (link)
    Bill Hereth – Data Scientist, Wasatch Front Regional Council
    Summary: The upcoming v9.1 release of the Wasatch Front Travel Demand Model will feature a new visualization tool, vizTool, designed to enhance the way users interact with and analyze model outputs. This tool simplifies the process of exploring complex data sets by providing an intuitive and user-friendly interface, allowing users to easily interpret and present key insights from scenario runs. The vizTool offers dynamic, accessible visualizations tailored to travel demand model users. We will solicit feedback from those present at the meeting to see how we can best support users to make this a successful tool.

  • Half Time Break
    Ivan Hooper, Avenue Consultants

  • Illustrating the use of Activity-based models in Utah (link)
    Greg Macfarlane – Assistant Professor, Brigham Young University
    Tim Baird – Associate, Fehr & Peers
    Hayden Atchley – Travel Demand Modeler, Utah Department of Transportation
    Summary: In the past 20 years, many MPO’s in the United States have moved to an activity-based model for their transportation planning efforts, and lessons learned from this transition have affected the decisions of other agencies for almost that time period. A recent UDOT and WFRC- funded research project examined attitudes towards activity-based models from current practitioners, and illustrated how scenario analysis works in both models. The results of the research suggest that modern activity-based models might not be as difficult to configure and use as commonly believed.

Next meeting

  • Date/Time: Thursday, January 16, 10:00-11:30 am @ UDOT

Suggestion Box

  • Please enter any suggestions here.

🍽 lunchiMUG

  • Gather with us at Culvers (1211 W 800 N Orem) for lunch.

Notes

Model Status Update

  • Utah Statewide Travel Model (Hayden Atchley, UDOT)
    • Working on updates to update employment data, SE data, and the newest model version
    • Working to stitch in the sub area models into the Statewide model (end of the month)
    • 2027 long range plan (end of the month)
    • Working on recreational factors for the areas with state parks
  • Wasatch Front TDM (Chris Day, WFRC)
    • 9.1 getting close
    • Working towards a release date of October 31st
    • Python updates
      • Python environment included in the model
      • Instead of ArcPi we are switching to Geo Pandas?
        • Rearranging of some of the scripts
        • Changes to HOT speed
    • VizTool updates
      • Bill presenting on this
      • Awesome visualization tool
        • Enhanced 6-7 scripts
        • VHT, VMT, V/C ratios, boarding reports, mode breakdown, access to opportunities, TAZ data
      • Tool is in the model and should open up at the end of the model run
    • Transit and Road network updates
      • Fairly minor
      • Direction updates
    • Detailed documentation will be released with the model update
  • Cache TDM (Jeff Gilbert, Cache MPO)
    • Working with Fehr and Peers
    • Traffic count data base
    • Working for more model calibration
    • Attempting to get a statewide database of count data
      • Nicholas Black
      • Had a meeting to discuss how to expand this. Nothing official yet
    • November/December release based on new calibration
  • Dixie TDM (Nicholas Gayer, Dixie MPO)
    • Working with Horrocks on updates
  • Summit-Wasatch TDM (Hayden Atchley, UDOT)
    • Updating transit ridership
    • Focusing on the winter season
  • Iron TDM (Hayden Atchley, UDOT)
    • Model was released
    • Updating python scripts
    • Updating network processor and switching to GeoPandas

Git Training (Greg Macfarlane, BYU)

  • Date: Tuesday, October 15, 2024
  • Time: 2:00 - 4:00 pm
  • Location: Redwood B Conference Room @ UDOT Complex
  • Summary: Greg Macfarlane will provide an introduction to software version control for transportation analysts. This training is aimed at those who have heard of Git, but have never used it, or who have used it but didn’t really know what they were doing.
  • Link to sign up here

Utah MUG Website (Chris Day, WFRC)

  • New tools and resources have been added to the website
  • SE forecasts
  • Traffic Volumes
  • Household Travel Surveys

Visualizing WF TDM Outputs via the New vizTool Bill Hereth

  • This came directly from the polls that we did at the last meeting
  • What TDM outputs do you use routinely? How can I use the VizTool?
  • BETA
    • Documentation and user guides will be released
    • If something isn’t working reach out to Bill or the VizTool Wiki
  • Main Components
    • Groups
      • Roads and Transit
      • Trips
      • Communities
    • Modules
      • Road Segments
      • Road Segments Trends
      • Transit Segments
      • Transit Stops
      • Transit Stop Trends
    • Attributes
    • Filters
    • Scenarios
    • Legends
    • What filters are currently applied
  • VizTool is for data visualization ONLY
    • If you need to make edits, make them in the TDM and re-run it
  • Intro Demo
    • Scenarios Folder - .VizTool folder
    • All self contained
    • Runs a local host
    • Options to divide by HH, Pop, EMP, etc.
    • VMT by district, City, TAZ, etc
    • Transit Ridership
    • Lots of different filters
    • Lots of new features still under development
    • Scenarios Configuration File
      • Define which outputs you want to be working with
    • Trips
    • Zone Trips Ends by Mode
    • Production, Attractions
    • Communities
      • SE data
      • Population, HH, owned vehicles
      • Employment by category
      • Zone Metrics
        • VMT produced by various variables
      • Access to Opportunities
        • Access to jobs, HH, etc.
  • Why the VizTool
    • Nearly instantaneous visualization of outputs
    • Broad range of data
  • What’s it made of
    • ArcGIS javascript ACI, calcite, chart.js
    • Data is stored in text files called JSON
    • Jacascript classes that make it scalable as more data and functionality is added
    • Self contained and transferable
  • Where is the VizTool going from here?
    • BETA will be released with TDM v9.1
    • Updates will be flexible and not always tied to the newest TDM release
    • Intent is to extend to other models (TDMs in the state, bikeway models, etc)
  • Best way to learn and be supported on the VizTool
    • GitHub repository with issues and wiki documentation
    • One on one training
    • Small group training
    • Webinars
  • Are there plans to host base runs on the WFRC site
    • Planning on hosting, but not advertising
  • Feedback meeting
    • Maybe a couple of months after the release
  • Have to make sure that all of your links that you want to show up need to have a SEGID otherwise it’s not going to show up in VizTool
  • Network Comparisons
    • Scenario Selector
    • Functions as a “this scenario” minus “this scenario.”
    • Once you collapse the “Compare To” it stops the comparison
    • It’s easy to misinterpret data, so make sure you know the network and the why well.
    • Truck percentages not currently included
  • Last slide appears to say which variables are currently included and what is planned

Illustrating the use of Activity-based models in Utah Greg Macfarlane

  • Project got finished yesterday, should be published soon
  • Trip vs Activity Based Models
    • Set of choices differ
    • Trying to get to the same point, but how they get there is going to vary. Example is people going to work before lunch vs people who go straight from home to a restaurant
  • Practitioner Survey
    • When and why did your agency transition to an activy based model
    • What did development and adoption look like?
    • Benefits
      • Peak spreading, pricing, demographic change, equity analysis
    • Drawbacks
      • Longer run times, few benefits for basic highway capacity analysis, more complex scenario analysis, small consultant pool, model sensitivity vs random elements
  • Comparative Illustration
    • WFRC TDM used the ActivitySim as part of the Highway Skims section
    • Opens up a lot of choice
    • ActivitySim not a new concept, but it’s a good methodology to follow
    • Focusing on tours rather than single trips
    • Breaking down the different trips based on the mode
    • Non-home based trips are very hard to explain in a trip-based model
    • Mode switching from scenario to scenario
      • Driving alone, carpooling (2), carpooling (3+)
      • You can see what modes people use throughout the tour where in the trip-based model you can’t
      • When you double the commuter rail in a day, how many people are switching and where?
  • Remote Work Considerations
    • Rebound effect
      • Fewer work trips may cause more non-work trips
    • Trip Length
      • Longer commute more likely to work remotely
    • Household structure
      • Single vs dual parent
      • Age of children
  • Conclusions
    • Cited difficulties
      • Computational complexity
        • Similar runtime on same hardware
      • Complicated design
        • Easier interpretation with activity based
      • Lack of interoperability
        • ActivitySim remote work sub-models taken from SEMCOG
        • Simple to add/edit parameters
      • Staff Training
        • Similar time/effort between models