Project Meeting 2019.05.17

Verification

• See the verification table for the current summaries, which keep looking better and better
• Issues we fixed since our last call, which are in pull request PR#261
  • Add at least one non-preschooler to filter for HHs that participate in the joint tour freq
  • Calculate time overlaps for all persons in HH for joint tour freq
  • Correct ptype, cdap expressions for calculating HH time overlaps
  • Correct use of emp status (full or part time) instead of person type in atwork tour freq
  • Correct non-worker indicators in atwork tour freq
  • Add constants for preschoolers for non-man tour freq
  • Correct expressions offset error for school and preschool ptypes for non-man tour freq
  • Add constants for joint tours for non-man tour freq
  • Turn on stop freq num stops by purpose expressions
  • Correct income expression for stop freq atwork
  • Correct max distance by purpose for stop freq
  • Correct tour mode choice IVT coefficients by transit submode
  • Correct auto ownership filter for transit for tour mode choice
  • Drive alone not available for escort for tour mode choice
  • Fix transit origin and dest walk times for tour mode choice
  • Use offpeak parking costs for non-man for tour mode choice
  • Divide transit skims by 100 (because they are in Cube format) in tour mode choice
  • Correct trip mode choice IVT coefficients by transit submode
  • Set trip mode choice max walk to 1 mi instead of 1.5 like tour mode choice
  • Fix transit origin and dest walk times for trip mode choice
• Re-ran TM1 with comparable setup, which includes these additional UEC fixes, see PR#22
  • IndividualMandatoryTourFrequency.xls - not divide auto skims by 100
  • JointTours.xls - moreCarsThanWorkers calculated incorrectly
  • ModeChoice.xls - Drive transit coefficients on IVT incorrect
  • TripModeChoice.xls - correct walk mode walk time max
• Still working on reviewing tour stop frequency
• Plan to publish (release) an updated version after that
• Likely do one more pass through to correct additional differences
• Updated verification script to run for either ASIM (or ASIM and TM1)

Admin

• Welcome Dennis Farmer from MetCouncil
• @Joe follow-up with agency partners about implementation plans for workshop
• @Jeff working on updating our multiprocessing runtimes for our TRB Planning Applications conference paper
• @Ben to draft SOW for next phase of work. I'm planning to scope the following tasks for consideration:
  • Features
    • Support for multiple zones
    • Support for AVs/TNCs
    • Incorporation of PopulationSim into the consortium
    • Parking location choice
    • OD demand matrix builder
  • Usability
    • Improved input validation / checking for easier use
    • Output reporting / summarization functionality for easier use
    • Additional user documentation / tutorial with exercises in the wiki
  • Additional performance tuning (i.e. optimization)
  • DaySim like estimation mode to output alternatives, logsums, etc
• Project wiki
  • @Ben to add the Phase 1 (Synthicity) materials to the wiki
  • @Ben to restructure wiki for new users - move current stuff up, move historic stuff down
  • @Ben add a possible future features page with ideas, ranking, and target release date for @Alex and others to contribute