Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Checking and validating travel distance assumptions #79

Open
2 tasks
sgreenbury opened this issue Dec 13, 2024 · 1 comment · May be fixed by #80
Open
2 tasks

Checking and validating travel distance assumptions #79

sgreenbury opened this issue Dec 13, 2024 · 1 comment · May be fixed by #80

Comments

@sgreenbury
Copy link
Collaborator

From discussion with @Hussein-Mahfouz :

  • Confirm distance units (miles or metres) (specifically affects output plots) (looks like conversion to km here)
  • Explore why the distances for trips appear larger in 2-15km range in AcBM comapred to NTS. Two factors:
    • Interzone short distance trips, the route taken (especially for PT but also for car generally) will not be direct. A factor can be applied (see paper) with Minkowski distance (e.g. apply factor of 1.54 to estimated travel times based on Euclidean distances). Could also include a decay for this factor since longer distances may be expected to be more direct.
    • Intrazone trips have an assumption of distance in proportion to area. Explore if the effect is larger for larger OAs.
@Hussein-Mahfouz Hussein-Mahfouz changed the title Checking and validating assumptions Checking and validating travel distance assumptions Dec 13, 2024
@Hussein-Mahfouz
Copy link
Collaborator

I've made some edits in #80 to improve location assignment. One important thing to note:

  • We should see improvement in assignment of primary locations. These are assigned a zone in 3.1_assign_primary_feasible_zones (see here and here)
  • Assigning feasible zones for secondary locations is based on PAM here. The logic for secondary location in PAM is described in this notebook, and they use a combination of diversion factor, attraction of zone, and leg ratio. This approach does not guarantee finding a solution that matches the reported distance, as this is actually a difficult problem - you are constrained by fixed primary locations on either side. We should be aware of this when validating our results.

Potential future implementations (low priority):

  • consider secondary location assignment when choosing primary zones (i.e we have a set of feasible primary zones, and we choose one that best fits our secondary zone constraints, but that is a very difficult problem and would require a rewrite of the pipeline workflow).
  • Check the approach described in this paper

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
2 participants