Hastus 'Non-Public' Stops and GT solutions.

[El McG.]

Hello all,

Does anyone use so-called 'non-public' stops in Hastus? When the
stops are exported from the GEO module for the GTFS, the non-public
stops are exported as well but unfortunately are included in the trip
solutions for customers. If the stops are not exported, this will
cause a disconnect between the stops that are available in the table
and the stops that are used on the trips.

There are a couple of reasons they are used in Hastus:

1. In the GEO module, the algorithms calculate shortest
distances between stops when creating itineraries. These generated
itineraries don't always represent the actual path taken by the
route. "Non-public" stops are used to 'force' the itineraries to
respect the actual path taken by the route. NOTE: This can be
avoided if the street segments of the actual path are selected and
used to create the itineraries, which is the method employed now.
2. When a single route serves multiple districts, the system
uses the stop-to-stop distances to generate statistics such as hours
and miles that will be attributed to each district based on the
location of the stop. As such we have placed "non-public" stops at
the city boundaries so that the reports will show the correct figures
for each district. This is important since the reports are used to
bill the cities for their respective service.

If anyone has some experience with this please reply.

[Devin Braun]

Are you using the automatic export function of HASTUS, or are you using
OIG files for your export?

In San Diego, we use our own OIG files that we wrote in-house. The
export specifically ignores the stops that aren't marked as "Boarding"
or "Debarking" in the stp02 window (the "Public Information" option in
the stop properties window).

If you are using OIG scripts, you can change them to ignore stops that
aren't marked as public in HASTUS. Or, if you're using the built-in
function, this might be a fix that GIRO could make for you.

Devin Braun
San Diego MTS

[El McG.]

Hi Devin, thanks for the response. We are using in-house OIG
scripts. As you mentioned, the OIG scripts can be changed and I've
done it both ways, considering and ignoring the stops that are not
marked "Public Information" in the stop details window. When I
consider them, they turn up in solutions. When I change the script to
ignore them, I get validation errors since those same stop id's are
included in the trips file. Every stop that's in the trips file must
be in the stops file. (See stop_id field in both trips.txt and
stops.txt).

> If anyone has some experience with this please reply.- Hide quoted text -
>
> - Show quoted text -

[Tom Hixson]

These stops are also handy for timepoints to control drivers (layover
points). Can you set pickup_type and drop_off_type in stop_times.txt
to 1?
Tom
Sacramento

[El McG.]

I will certainly try. These fields are listed as optional in the feed
specifications. Does the designation determine if they'll be returned
in trip plans or not? If not then it just may be the solution to this
dilemma. If so, does it include a note or comment stating that a
particular stop is unavailable for pickup or dropoff?

[Tom Hixson]

Yes, trip plans obey. We have buses wait for the train by using 2
timepoints (arrive and depart) at the same stop. The first is drop-
off only; the second is pickup only. You can't see it because it
applies to the trip, not the stop. But you could add a warning to
stop_name in stops.txt; this shows when you click a stop icon.
-Tom

> > - Show quoted text -- Hide quoted text -

[daniel.p...@giro.ca]

Note that recent versions of HASTUS now include an interface to Google
Transit that was done by GIRO. That interface skips these non-public
stops for the stops.txt and the stop_times.txt files so this problem
is avoided altogether. You could do the same in your OIG script and
skip these stops in both files.

> > - Show quoted text -- Hide quoted text -

[El McG.]

It's great to finally see that we're not the only agency that uses the
arrive and departure concept (reference places, too?).

[El McG.]

Hi Daniel,
I got an OIG line from Devin that got rid of the non-public stops from
the stop_times file which was the missing link. Thanks for the
update, I look forward to getting the latest version later this year.