You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In moving away from transit access points and to a zone-based transit assignment, the Emme path builder now generates a generalized cost estimate for each interchange (by access mode, by time of day). Skims from these paths are then returned to the mode choice model. The mode choice model applies a series of coefficients to the skim values to compute a second, different generalized cost used in mode choice. This is inconsistent and likely inefficient. It would be interesting to replace the mode choice expressions with the single generalized cost expression and see how the model performed. Other skims could still be brought in to make targeted calibration adjustments.
Progress:
Sufficiently defined
Approach determined
Tests developed
User story satisfied
Doc strings
General documentation
Passing tests
Priority
Low
Level of Effort
Modest
Resolution Ideas
The mode choice UECs would be modified to replace the utility expressions with a reference to the generalized cost. The generalized cost from the transit skimming and assignment would need to be stored and skimmed. The mode choice models would then need to be recalibrated.
User Story
In moving away from transit access points and to a zone-based transit assignment, the Emme path builder now generates a generalized cost estimate for each interchange (by access mode, by time of day). Skims from these paths are then returned to the mode choice model. The mode choice model applies a series of coefficients to the skim values to compute a second, different generalized cost used in mode choice. This is inconsistent and likely inefficient. It would be interesting to replace the mode choice expressions with the single generalized cost expression and see how the model performed. Other skims could still be brought in to make targeted calibration adjustments.
Progress:
Priority
Low
Level of Effort
Modest
Resolution Ideas
The mode choice UECs would be modified to replace the utility expressions with a reference to the generalized cost. The generalized cost from the transit skimming and assignment would need to be stored and skimmed. The mode choice models would then need to be recalibrated.
Project
No project in place yet
Who should be involved?
Users: @gregerhardt, @lmz, @FlaviaTsang
Reviewers: @DavidOry
Risk
Aside from moving the model out of calibration, the risks are low.
Tests
This is primarily a change to the UECs, so the tests would involve debugging the UECs to make sure they are being implemented properly.
The text was updated successfully, but these errors were encountered: