Project Meeting 2021.03.04

Task 1 Project Management (milestone)

Some revisions to the bench contract will be shared with the consultants
Partners please review the phase 6b scope Joe shared
AMPO and Joe discussed the various technical resources needed and will discuss some more soon
SEMCOG university model
- How transferrable is the model to other regions/universities?
- Could be useful to GATech and UW
- Like a micro ABM within a regional ABM
- Would probably be of value to other places
- Size terms using building capacity instead of enrollment and where students live in the synpop is critical
- Should look at the ASU survey as well
ARC presentation
- Does the use of np.log versus np.log1p have a material difference in the system? Do we need to make and revisions to the software as a result?
PSRC/SANDAG add runtimes to multi-region timing runtimes by submodel
- Let's have folks add to it over time
- SANDAG runtimes without presampling....and will get faster with representative logsums
- ActivitySim should always continue to work on performance. Something great about ActivitySim is that it is worth investing in (...worth making faster)
PopulationSim multiprocessing PR getting close to pullable (is that a word?)
FYI, PSRC using latest PopSim and results are better, see conversation here

Reviewed ODOT's input checker (including the list of checks here)
The ODOT checker reminds me of the task 7 work below
I need to work on the plan some more
Like the new agency examples, added a simple regression test to check overall results to our core example example_mtc

Continued to help PSRC get their example working, which it now does
Due to example_mtc regression test above, found and fixed a bug in multiprocessing repeatability related to zone sequencing