Agenda

• Security: eMudhra CA key compromise activitysim#1135
• Deprecate python 3.10, run CI for python 3.11, 12, 13  activitysim#1127: decide whether we test regularly against the latest, the oldest, or both.
• Faster random numbers activitysim#1103: Existing evidence does not support merging. An EET test could tell a different story, but had not been done (yet). Who will and with what timeline?
• exposing network_los to annotator and preprocessor activitysim#1134: New PR needs reviewer.
• Reduce peak memory use in destination choice and large trip models activitysim#1110: Reviewed by @janzill and approved with a few notes. Needs more rigorous benchmarking before merge? And if so who will do that?
• Addition of Application and Analysis Guide activitysim#1086: Still marked as draft. Plan to close out?

Ready to Review

• Calibration activitysim#1107 -> @janzill
• Skip failed sampling choices activitysim#1130 -> @jpn--
• Explicit Telecommute: new telework_arrangement and telework_duration components activitysim#1126 -> @janzill & needs agency reviewer
• Reduce memory of summarize step by dropping unused variables before wide merge activitysim#1125 -> @jpn & needs abench
• Fix for simulation-based shadow pricing segmentation activitysim#1083 -> @bhargavasana + @jpn--

Merged 🥳

• Allow zero probs option in pnr lot choice activitysim#1123
• exposing network_los to annotator and preprocessor activitysim#1134