Project Meeting 2017.01.26

Model run problem

• On @Hu's machine, the Python scripts can't find other scripts in the current working directory
• This use to work for him and still works for Nagendra and PSRC
• @RSG to update the scripts by appending the current working directory to all scripts so it works

Networks

• Bellevue updated the networks but can't test them since the model run is broken (above)
• @Bellevue will test once model run works and then send to RSG
• @Bellevue will attach highway and transit counts as well

Time-of-day

• @RSG to change AM time period to 6 to 9 and PM from 3:30 to 6:30
• @RSG to make sure this is done correctly everywhere - transit, truck, supplemental, link capacities, skims, etc.

Documentation

• GitHub wiki site coming along and conversion to Word docx using pandoc looks good
• Still a couple days of work to do
• @RSG add Daysim data dictionary specifics for BKR
• @RSG integrate the content from Mark's recent memo of inputs

Employment

• LEHD/LODES-based data is our alternative to the ESD-based data
• PSRC can create the Daysim parcel input file with either - the format is the same
• Maybe RSG uses LEHD and BKR uses ESD?
• @ESD - we are still waiting on their decision about RSG getting access to their data

Calibration

• We compared PSRC's latest results to BKR's and the survey
• There are no major differences and results generally look pretty good
• Remember, BKRCast is using the latest SoundCast calibration coefficients
• BKRCast has 15% more trips
• PSRC's model run is with ESD data; BKR uses LEHD
• BKRCast is partially calibrated at this point