Closed Hussein-Mahfouz closed 6 months ago
Which method suits our purpose better?
Without much context, method 2 seems to work well, what's the downside?
This is my current (incomplete) thinking. If we are looking at which bus routes to remove, then is it better to look at:
(a) could give us a ceiling for the demand on a route, and we could combine it with another step - identifying the total demand that is uniquely* served by each route - to inform which routes are more important and which are redundant
(*) demand between A-B is uniquely served by a route X if it is the only route that goes directly from A to B
(**) Approach from chapter 10 of Modelling Transport
The multi-path approach in the last comment can be calculated using Aequilibria, see relevant documentation here
:+1:
Demand at the OD level is processed using issue #4 and PR #14
How do we project this demand from the OD level to the route level? Currently I am using two different methods. Take an OD pair with demand = 20 that is served by 3 routes: Route A (freq = 6 buses/hour), Route B (freq = 3/hr, Route C (freq = 3/hr):