-
Notifications
You must be signed in to change notification settings - Fork 100
Project Meeting 2021.05.18
Ben Stabler edited this page May 25, 2021
·
2 revisions
- Issues / PR review
- Maybe just use conda forge for the test system moving forward
- Discuss updated telecommute model design
- Essentially using the SANDAG model design
- Will approximate the SANDAG multimodal accessibility measure as best we can with the existing activitysim accessibility calculator
- SANDAG accessibility calculator has mode constants to get at the weighted average multimodal logsum measure
- The survey tells us trips by mode and a simple model was estimated with auto as the base
- In phase 7 lets do the more complete accessibilities calculator
- We learned about the MAG model at the peer review
- SEMCOG plans to use the adopted approach
- Maybe for SEMCOG we tag workers with their work zone employment mix and then sample from a distribution to pick an employment type for them
- Then use for the telecommute freq model for industry specific telecommute freq testing
- This is important for equity analysis as well so persons by income also important
- Let's finish the WFH model as is and the SEMCOG research into the employment mix idea can be done in parallel and contributed later if desired
- I'm now testing the chunker improvements