Project Meeting 2024.12.19

Agenda

• Admin
  • Admin Home
• Technical
  • Explicit Telecommuting Discussion

Administrative

• Admin Home
  • Zephyr board has no objections to hosting ActivitySim
  • Joe has discussed w/ lawyer what Zephyr needs to do to host ActivitySim
  • Do not foresee much movement on this, due to holidays; will revisit post-TRB

Technical

• Explicit Telecommuting Discussion
  • Discussed extensively on consortium-only call
  • Want to understand absolute magnitude of how much these behavioral differences matter
    • What does it matter in big picture; is it worth making these changes?
  • Is it the only way and is it the best way to handle telecommuting?
    • What additional design thinking do we need to do to get everyone on board with the approach?
  • According to WSP, people agree on Task 3, but not Task 4 (how to tweak the SANDAG model)
    • What they are proposing is not a big software change, and it is almost completed
    • Would like to showcase it w/ SANDAG model; though what SANDAG officially plans to implement it is up to them
    • Hope calibration won't have to be extensive, but because believe most work-from-home is similar to work-from-office patterns
  • Joe has issues with idea of a "fake tour"/"special snowflake tour"; What is the implication of that? Do we ignore it?
    • Ex: If it is fake tour, and it has intermediate stops, how does user/survey processor know how to treat it/process it differently?
    • Want to lay out potential implications more explicit
  • David is concerned that our terminology changes with this software change
  • Jeff suggests that software changes suggests that HTS will have to change to accommodate these changes and produce data to be able to use functionality well
    • Anyway we build the model is based on assumptions with how people behave and not observed data
    • Alex - too much survey burden to be able to collect in-home activity information; Joe agrees
  • Sijia argues that the software changes allow for all these different type of options with how people behave
    • Software changes can accommodate the tracking w/o assigning as tours, as defined by travel
  • Stefan - Why do we have to block off time for those who work from home?
  • Joe - we need to work out how to figure out best approach, because enough questions have been raised about the proposed WSP approach
    • Only want to implement a software change if people are going to use it
  • Alex - likes idea of tracking tours
  • Greenlight has been given to software change related to flagging folks who work from home
    • Jeff - A few options after that
      • Do nothing and choice of modeler of whether to use it as a covariate
      • Do larger software changes that give additional options for how they are represented in the model
        • Flexibilty may result in added system complexity that in practice makes it more cumbersome
  • Do we need to do Task 4 to know whether or not it works; or can we think it through without spending all the money
    • Do not want to dive into tasks and later regret the investment if it can be
  • Will find alternative time to have regular technical discussions