-
Notifications
You must be signed in to change notification settings - Fork 180
faq 110269439
by Gerald Richter on 2017-06-27 13:27:42
I do have a larger network which had some isolated (unroutable) areas,
so I ran the netcleaning procedure on the graph
(</ac:plain-text-link-body></ac:link>)
But trying to run the simulation on the resulting network still causes the mentioned errors.
- Why could that be?
- what to do or check?
I checked the vicinity of the nodes for unidirectional linkage,
but everything seems OK and linked in both directions...
I tried with an old nightly and a more recent one, with same results
Update:
I checked the shortest path between the indicated nodes of the errors externally
with a directed-graph Dijkstra implementation.
And got back a distance for all of them...
Remark:
The error message does not indicate, which person / plan (id) it is, that causes the problem,
so I have no means to start looking for the detail of the problem.
by Gerald Richter on 2017-06-29 10:11:18
answering my own question now... hoping
the explanation of the solved problem will save some time for someone else
I had some stray links traversable by "car", which led away from a riverside promenade which was itself only traversable in mode "walk". these stray car-links were not connected to the rest of the network otherwise.
A person taking the car from a location close to the stray link was snapped to one of the nodes, which left it stranded in the mode-routing sense.
I wonder how one could streamline such necessary manual fixes...
Since "car" is mostly the main mode considered,
- the network cleaning process could put out some info on mode-restricted islands
- or run with an additional paramter indicating mode(s) to check the network for
You are viewing an archive of the previous MATSim Q&A site. The real site is now at https://matsim.org/faq