Project Meeting 2021.04.08

Task 1 Project Management (milestone)

• Working on documentation/examples/tidying-up for upcoming release v0.99
  • Examples all downloadable, including SANDAG
  • Getting started, examples, notebooks, submodels sections updated, still working on core, test system, estimation
  • PopulationSim documentation updated

Task 2 Complete Estimation Mode (milestone)

• See task 1

Task 3 Telecommute Model in Cooperation with SEMCOG (milestone)

• Discussed Telecommute google doc here
• I'll address the comments, which are all straightforward except for the what-if analysis need
• We discussed the what-if analysis in detail and agreed to build an iterative procedure into the model that compares the work from home share to a target and iterates by adjusting a coefficient until the target is met (or max iterations is met)
• This doesn't need to be a systematic generic auto-calibration routine for everything activitysim; rather it can be just for this model
• It's similar to shadow pricing but it works at the global level instead of zone level
• Let's not let the perfect get in the way of good enough
• I'll work on a little mockup/design for further discussion and update the memo
• Need to discuss (but not implement) the DaySim in/out commuters feature here as well since it relates to telecommuters

Task 4 Transit Pass Ownership Model in Cooperation with SEMCOG (milestone)

• Joel to report back on data analysis on Tuesday

Task 5 Performance Enhancements (milestone)

• See task 1

Task 6 Prototype multiple models test system (milestone)

• See task 1

Task 7 Input checking and error handling and pipeline improvements (milestone)

• Discussed potential improvements list
• Votes below
• Since the chunking improvement is a big lift and we've heard a lot of feedback related to it, we're going to do that
• We can also do the most popular one - print info on no alt found - since it's small
• We're tentatively going to do 407, 406, 405, and 411
• Let's think it through a bit more and discuss next week
• Let's create an issue to improve logging when users get high trip fail rate for trip destination, like PSRC is experiencing. This is new issue #416.

feature	issue	votes	LOE estimate
when no alt found print expressions turned off	#404	7	easy
chunking	#406	6	modest to involved
input checking	#401	5	wide open
parsing configs	#402	4	wide open
turn off restart	#405	2	easy
expression timing	#412	2
long trace file names	#409	1
better inheritance file documentation	#411		I'll do as part of the documentation updates

Task 8 Maintenance and Support (milestone)

• See task 1