Measure updates while appending events on realigned routes

ColinSchut · ‎08-03-2020

Hi all,

I have a question on appending events (in my case changing the to date of an already registered event) with a date range that temporally overlaps with one or more route realignments

Event: From Date: 2010/01/01 To Date: NULL

Route realignment occurred 2015/01/01 for instance.

Appending the event or changing the TODATE to overlap with that realignment will correctly date split the event. But the measures are not updated to the new network locations causing the event to spatially move.

In my case I can locate the updated events to their previous location and update the measure fields manually, but this would be impossible on a more complicated network. The only other thing I can think of is prepping the data to have a full set of event records for every route date range and locating them all. Is there a workflow that will simplify this?

Thanks

RyanKoschatzky · ‎09-09-2020

Colin,

I think you will have to create those timeslices and measures but at least you have the date, most people don't have that and that would make the job even harder.

View solution in original post

RyanKoschatzky · ‎08-04-2020

Colin,

Depending on your event behavior (Stay Put shown below keeps XY, updates measures), internally you should have the following records for a realign in the middle of the route

Route: From Date: 2010/01/01 To Date: 2015/01/01 Measure 1.0

Route: From Date: 2010/01/01 To Date: NULL Measure 1.5

Assuming full coverage event records

Event: From Date: 2010/01/01 To Date: 2015/01/01 Measure 1.0

Event: From Date: 2015/01/01 To Date: NULL Measure 1.5

If the event is not full coverage, you would still have the same timeslices (2) but the measures would depend on where the event began/ended was in relation to the realigned section.

I think your on the right track for an event layer for each route timeslice. I am not sure of workflow to simplify that work effort.

ColinSchut · ‎08-04-2020

To clarify a bit this is referring not to doing a route realignment (the events that were active at the time of the realignment updated properly) but extending an event time range over a realignment that has already been done. Event behaviours don't seem to apply because no route changes have occurred in the edit session.

So when I did the realignment:

Event 1: From Date: 2010/01/01 To Date: 2011/01/01 - Measure 10 to 11

Realignment: 2015/01/01

I now want to change Event 1 to be From Date: 2010/01/01 to To Date: NULL

And I get

Event 1: From Date 2010/01/01 To Date: 2015/01/01 - Measure 10 to 11 (Correct)

Event 1 From Date 2015/01/01 To Date: NULL - Measure 10 to 11 (When it should be 11 to 12 for example).

RyanKoschatzky · ‎08-06-2020

Colin,

I think at least for me what is confusing is the use of the word realignment. That is a route action to me and if using R&H internal events would be subjected to event behavior rules. The use of word realignment when talking about events just doesn't compute to me.

Your realignment happened before the existing event record measures and your expecting the X,Y to stay the same and the measures to update. It appears the data is not being applied the correct event behaviors for the new timeslice of the route measures.

I think you can use the edit log and view the xml data for the shift for individual routes.

Example of one of our realigns

NetworkId="1" RouteId="40001002013" NewRouteId="40001002013" OperationTime="2015-04-29T00:00:00" FirstM="0" LastM="9.552051" IsPerformDownstreamCalibration="true" DoNotApplyEventBehaviors="false" IsFirstPointTouches="true" IsLastPointTouches="true">

You can also use Export Network for a more system changes. Haven’t look at that export data enough but I know some folks use that to update event data in external system.

Just not up on code to help you with that on either method.

ColinSchut · ‎09-08-2020

I believe it's confusing because ideally this is a situation that would never occur at all if an event record has been loaded and route changes created properly over time.

Perhaps a better way to put it would be to say that I have an external excel table or old shapefile containing a bunch of historical information that has never been in an lrs. A record in that table has a "Fromdate" year of 2010 and is still accurate today. Between 2010 and today, that route has had 4 realignments downchain of the event.

If I want to improve my historical data and add it to the database by appending it as is, it will create 5 timesliced records that have the same measures (4 of which will be geographically incorrect). I was wondering if there was any way to calibrate those properly while appending them, or if I need slice that data and locate them separately for ever realignment that has occurred on the route before appending. A big task with a big dataset.

Anyway Ryan, I always greatly appreciate the help you provide on my questions. This only occurred to me because of some improperly modelled data, and I'm unlikely to bother dealing with historical data like this in the future, I will likely close this soon regardless of any answers I get.

RyanKoschatzky · ‎09-09-2020

Colin,

I think you will have to create those timeslices and measures but at least you have the date, most people don't have that and that would make the job even harder.