Add cost to maximize wait-time and avoid back-travel in optimize transfers - #3654
Conversation
t2gran
added
!Bug
d8d745d to
7ce8fe1
Compare
7ce8fe1 to
35be747
Compare
refactor: Some OTP cleanup to prepare for #3654
binh-dam-ibigroup
left a comment
binh-dam-ibigroup
left a comment
There was a problem hiding this comment.
Approved subject to updating the comment line in change.md.
ee2279c to
fb9e4b5
Compare
t2gran
changed the title
t2gran
changed the title
fb9e4b5 to
4b9b4c2
Compare
hannesj
left a comment
hannesj
left a comment
There was a problem hiding this comment.
Looks good, just one comment on the documentation
optionsome
left a comment
optionsome
left a comment
There was a problem hiding this comment.
Code changes looked good and tested with a case that has back-travel in OTP1 but not with these changes in OTP2.
| return Math.min(MIN_SAFE_TRANSFER_TIME_LIMIT_UPPER_BOUND, minSafeTransitTime); | ||
| return bound( | ||
| minSafeTransitTime, | ||
| MIN_SAFE_TRANSFER_TIME_LIMIT_LOWER_BOUND, |
There was a problem hiding this comment.
This probably needs to be changed once support for defining the GTFS transfers.txt min_transfer_time is added?
There was a problem hiding this comment.
I am not sure if it has anything to do with each other. The minSafeTransitTime is used to add a cost for very short transfers across all possible transfers for a set of trips, while the GTFS min_transfer_time is a restriction for a given pair of stops(& trips). The OptimizeTransferService and the Raptor search must both support the GTFS min_transfer_time - preventing transfers when appropriate.
4 participants
Summary
This is an enhancement of the previous PRs on transfers. It try to balance constrained transfers and avoid back-travel. There is detail documentation on who this works in the PR.
Issue
There is no issue for this.
Unit tests
Unit tests are updated.
Code style
✅
Documentation
✅