🚀 Feature: Replace MAZ-assignment Distance Threshold with Intrazonal (TAZ) threshold

[DavidOry]

User Story

The TM2 modeling system currently assigns short trips via an all-or-nothing assignment at the MAZ scale. The hypothesis is that very short trips have more of an impact on local roads than regional roads, so assigning at the MAZ scale would be better than the TAZ scale, where a large portion of the trip may be assigned to TAZ centroids.

The modeling system has not yet been used enough to know if this is a good or bad idea. But the distance threshold is potentially problematic. There are short trips, say less than 1 mile, that may not take the shortest path due to congestion. To avoid this, would it be better just to assign intrazonal TAZ trips (that are not intrazonal MAZ trips) at the MAZ scale? This would more directly leverage the multi-resolution network without the need for an arbitrary distance threshold.

Progress:

• Sufficiently defined
• Approach determined
• Tests developed
• User story satisfied
• Doc strings
• General documentation
• Passing tests

Priority

Low

Level of Effort

Low

Resolution Ideas

As part of the resolution of #117, the change could be made to the logic that separates the MAZ-scale assignment trips and the TAZ-scale assignment trips.

Project

Not right now

Who should be involved?

Users: @lmz, @FlaviaTsang, @gregerhardt
Reviewers: @i-am-sijia, @AshishKuls

Risk

Very low

Tests

• Test that all of the trips from the trip list end up in either the TAZ or MAZ scale databank matrices