Jump to content
acelafan

ASMAD update - improvements to connecting trains search

Recommended Posts

All,

 

I finally got a chance to update the connecting trains search on ASMAD. Now it will automatically list missed connections in red where the arrival information is known. Sometimes only departure data is known for the first train, and in those cases likely missed connections are displayed in orange.

 

I also include the count and percentage of missed trains at the bottom of the table, like this:

 

20150809_6_50_CHI_connections.png

 

(Link)

 

JB started gathering additional info in the Amtrak Status Maps beginning in late May for this improvement. (Some missed connections may not be automatically flagged before then.)

 

It was a good suggestion by a few people.

 

CJ

Share this post


Link to post
Share on other sites

You guys do a fantastic job! Thanks for sharing!😉

Share this post


Link to post
Share on other sites

Very nice...and very interesting, being able to run checks on various same-day round-trips (e.g. RVR-WAS-RVR in a single day...for example, the NB Star has only misconnected with the SB Meteor twice this summer).

Share this post


Link to post
Share on other sites

Something I've noticed..."likely missed connections" doesn't actually appear even when you have the inbound train arriving at the same time as the outbound train is leaving. 30 seconds may be an acceptable connection time in Germany, but just try that in New York...

Share this post


Link to post
Share on other sites

Something I've noticed..."likely missed connections" doesn't actually appear even when you have the inbound train arriving at the same time as the outbound train is leaving. 30 seconds may be an acceptable connection time in Germany, but just try that in New York...

Good eye! Fixed. I also fixed a bug where a train arriving exactly on time was getting flagged incorrectly.

 

Glad to hear the page is useful.

Share this post


Link to post
Share on other sites

I usually stay out of these sort of threads, but I do have some questions. There are times that connecting trains aren't held if there are no connecting passengers (which happens.) How do you account for that? Additionally, there are times when known connections are going to be missed and passengers are often bussed, cabbed etc to intercept their connections at other stations. Your chart doesn't seem to show that so is this just a "pure" train to train connection you're logging or is this passenger misconnections?

 

While we're at it, does your status page update itself in real time as corrections are made? I bring this up because recently an initial terminal decided to flip flop the consists for two trains without telling anyone. As such, the first train out had the second trains consist and the second train had the first consist.

 

As the first train traveled up the line, it OSd everything an hour early for the arrival and on time for the departure. These are the times that are picked up by your train trackers. When the second train utilizing the first consist departed, every arrival time on the first train was overridden and showed an hour late.

 

It was fixed hours later. Would the adjustment show?

Edited by Thirdrail7

Share this post


Link to post
Share on other sites

I usually stay out of these sort of threads, but I do have some questions. There are times that connecting trains aren't held if there are no connecting passengers (which happens.) How do you account for that? Additionally, there are times when known connections are going to be missed and passengers are often bussed, cabbed etc to intercept their connections at other stations. Your chart doesn't seem to show that so is this just a "pure" train to train connection you're logging or is this passenger misconnections?

Yes to your last question - this is just pure train-to-train connection history information. I have no way of knowing if Amtrak got the through passengers (if any) to their final destination via alternate means in the event of a missed connection. It's a valid point.

Share this post


Link to post
Share on other sites

 

While we're at it, does your status page update itself in real time as corrections are made? I bring this up because recently an initial terminal decided to flip flop the consists for two trains without telling anyone. As such, the first train out had the second trains consist and the second train had the first consist.

 

As the first train traveled up the line, it OSd everything an hour early for the arrival and on time for the departure. These are the times that are picked up by your train trackers. When the second train utilizing the first consist departed, every arrival time on the first train was overridden and showed an hour late.

 

It was fixed hours later. Would the adjustment show?

ASMAD gets its data from Amtrak Status Maps (ASM) once per day. ASM gets its data from Amtrak as a train progresses with its trip. Once the times are recorded by ASM there is no automatic way for it to go back and adjust times that Amtrak has subsequently changed. However, if someone brings the situation to my attention, I can have ASM re-record the times - if it's not too late. Amtrak only keeps about 5 days worth of data on-line. If the times are re-recorded, ASMAD will get the new times when it next gathers data from ASM.

 

jb

Amtrak Status Maps

Share this post


Link to post
Share on other sites

John, thanks for doing the tedious, hands-on work, like forcing a collection of data after the fact. I appreciate the work that went into ASM, and I'm grateful that you are so responsive to comments.

Share this post


Link to post
Share on other sites

I have one more scenario for you if you don't mind. Let's say the notorious 94 is up to its usual tricks and is late operating on the RF&P sub. Amtrak decides to run an on time section of 94 at WAS. Once 94 arrives WAS, it assumes the slot and identity of train 178 to BOS. The Amtrak Status map would show both 94 and 178 as operating on time WAS-BOS, but the actual passengers for late 94 would be entered in the remarks section to show their true arrival times.

 

I suspect something like that wouldn't show up on your page?

Share this post


Link to post
Share on other sites

I have one more scenario for you if you don't mind. Let's say the notorious 94 is up to its usual tricks and is late operating on the RF&P sub. Amtrak decides to run an on time section of 94 at WAS. Once 94 arrives WAS, it assumes the slot and identity of train 178 to BOS. The Amtrak Status map would show both 94 and 178 as operating on time WAS-BOS, but the actual passengers for late 94 would be entered in the remarks section to show their true arrival times.

 

I suspect something like that wouldn't show up on your page?

If Amtrak switches a particular consist to a new identity/number, I believe they re-program the onboard GPS device to that new number. ARROW should then report the AR/DP times for the train that the GPS device is programmed for. That would show up in the ASM files.

Share this post


Link to post
Share on other sites

I think that is his point. Say I board 94 in Richmond. We arrive in Washington late. Meanwhile, Amtrak runs an on-time section of 94 leaving WAS on time, and arriving at NYP on time.

 

Meanwhile, 94 becomes 178 in WAS and also runs on time to NYP.

 

Our hypothetical traveler from Richmond gets off in New York later than planned, but that isn't reflected in the database since both trains are "on time".

 

With some data analysis, you could find it, look for a 94 that departs WAS before it arrives in ALX. :D

Share this post


Link to post
Share on other sites

With some data analysis, you could find it, look for a 94 that departs WAS before it arrives in ALX. :D

Ah, yes that's possible, maybe on a rainy day. ;)

 

Also, a note may appear at the top of the status maps file if John knows it's a split train. In fact, this happened for Train #79 originating July 22 but my code doesn't look for that.

Share this post


Link to post
Share on other sites

I had to look.

 

Two good examples of #94 departing WAS before departing ALX are 03/03/2014 and 08/20/2014.

 

In the first case the train was 7+ hours late at ALX; in the second it was 3+ hours late.

But they both got to BOS less than 30 minutes late. :giggle:

 

Share this post


Link to post
Share on other sites

By the way, when they start out an extra train set to cover the regular train schedule of a late train, we get the following four scenarios because of the way Amtrak Status Maps (ASM) works. In order to keep the network traffic to a minimum, ASM only looks for a train at a station when it's possible. For example, on the 3/3/2014 version of 94 in the last post, being 7H 49M late in ALX, it wouldn't even start looking for the train in WAS until 9.24 p.m. (regular arrival time 1.35 p.m.) because it thinks that 94 couldn't possibly get to Washington before that. This leads to four scenarios:

 

A) Amtrak identifies the extra train as the original train, and unidentifies the late train when it gets to the origin station for the extra train. On the maps this will show original 94 up through WAS and then stalled until new-94's time gets late enough for ASM to start looking for it again. Then suddenly it will jump way up the line.

 

B) Amtrak does not identify the extra train as the original train. In this case the new train will be invisible and never appear on the maps at all.

 

C) Amtrak identifies BOTH trains as the original train. In this case the maps will show the progress of the original train. The extra train will be invisible.

 

D) Amtrak unidentifies BOTH trains. In this case the extra train will be invisible and the original train will seem to stall.

 

That's enough of these thought experiments. I've got to get ready to go to work.

 

jb

Edited by John Bobinyec

Share this post


Link to post
Share on other sites

Another place where this kind of thing happens is Albany when an eastbound Empire Service train is late from Buffalo.

 

Potentially, this can also happen on Mondays at IND when 51 is late and IP's equipment is to deadhead to CHI.

 

For these "extra" movements to be shown on ASM and accessible in ASMAD, Amtrak needs to record them in ARROW, and I have to find out about it and set up the ASM system to follow it - before Amtrak's 5-day window of on-line data expires.

 

jb

Share this post


Link to post
Share on other sites

Another place where this kind of thing happens is Albany when an eastbound Empire Service train is late from Buffalo.

 

Potentially, this can also happen on Mondays at IND when 51 is late and IP's equipment is to deadhead to CHI.

 

For these "extra" movements to be shown on ASM and accessible in ASMAD, Amtrak needs to record them in ARROW, and I have to find out about it and set up the ASM system to follow it - before Amtrak's 5-day window of on-line data expires.

 

jb

 

For the record, your task should be easier these days. There are a slew of alerts that are now generated if the tracking systems think something isn't tubed correctly. It isn't foolproof and you still have to wait for confirmation from an offending terminal, but at least CNOC can initiate a call to find out what is happening.

Share this post


Link to post
Share on other sites
I had some requests for minor updates to the "Missed Connections" checker on ASMAD.


With the usual caveats that Amtrak data can sometimes have errors, I color-coded the good connections as green and missed as red. Ambiguous data suggesting a likely miss is orange. Otherwise unknown is gray.


You can also set the threshold for the amount of time between trains for a good connection. For example, the default is 5 minutes but you can set it to 10, 20, etc should you want more "cushion" between arriving and departing trains.


Also, sometimes the resulting table has rows of incomplete data. There is now a choice to hide those rows if desired.


Attached is a screenshot of the Lakeshore Limited (49) to the Texas Eagle (21) at CHI with the link.


Hope it proves helpful. I am grateful to both Amtrak Status Maps and ASM transitdocs for the source data.




post-2944-0-89163800-1515777577_thumb.png

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×