Duplicate files in Journey Planner datastore - is there a way to choose the right one?

There are several services that have more than one file in the Journey Planner Datastore, all with the same operating period: 19, 22, 260, 314, 349, 683 (3 files).
Examining the case of 19 manually, I think what is going on is that one file has the normal journey pattern, and the other has a diversion due to long term road works in Sloan Street. The diverted service seems to be the one actually running on the ground.
Is there some systematic recommended way to choose the current operational file in these situations?

Thank you!

1 Like

Very good question! Journey Planner itself must be able to distinguish, so whatever it uses, can we have it please?

If you had asked the question a year ago, I would have said “look at the Service Change Number” and compare with the current Working Timetable on the TfL site. Unfortunately that will not work any more because the five digit Service Change Number used in the Journey Planner file names now appears to have no link. It seems to be “add 1 to a previous SCN used for a route” and in some cases it is just a single digit. This abandonment of the true SCN in the Journey Planner files seems to have come in with Adiona but I have no idea why. The goddess of the return journey seems to have a lot to answer for.

To be fair, even a year ago my workaround would not have, er, worked around fully, because seasonal variations and diversion timetables were often numbered on the “+1” basis. Except at Christmas, at least 90% of JP files aligned directly with the WTTs.

Thank you @mjcarchive , unfortunately my time travel skills are limited so last year’s answer is less useful :rofl:

Having said that, I would still like to understand your solution. What is the Service Change Number?
And are the Working Timetables the PDFs in http://bus.data.tfl.gov.uk/schedules/ ?

1 Like

The TfL bus working timetables can be found at Bus schedules - Transport for London. You enter the route, get a list of current WTTs and open one up. The SCN (Service Change Number) is visible within the first line of the file.

Take the 247. The WTT contains the SCN 45645. Sure enough the JP Datastore zip file contains a file tfl_60-247-_-y05-45645.xml. But that only works (now) because the 247 has not been revised (not even driver duties) since 2018. It won’t work on any route which has been revised since last summer (though where there is only one JP file for the route you don’t need it to work).

For the 19, the WTT on the TfL site contains the SCN 63089 for the normal daily schedules (Fr, MT etc). In tonight’s JP zip file the two files available for the 19 include the numbers 61932 and 61935. The latter has the diversion and is current.

Now Working Timetable (Time Schedule) Graveyard - Route 19 lists year’s worth of WTTs for the route. Near the bottom you will find 61930 and 61935. However that 61935 is not current. What’s going on, you might well ask?

The answer to that, I think, is that all the JP files since 61930 have been given phoney SCNs by adding 1 to the previous one; I can see that the full range from 61931 to 61935 has been created since last summer. Usually these phoney SCNs do not correspond to a real SCN for the route. Just to add confusion, 61935 is an exception but phoney 61935 must actually equate to real 63089!

The loss of this link since last summer makes it difficult for an outsider to tie the two sources together (and is an outrage to anyone who favours tidiness!) but presumably it has zero operational impact, or someone would have done something about it (wouldn’t they?).

You will notice from the tab19 page (or equivalents for other routes) that quite trivial changes can generate a new SCN. I should also add that there have always been files in the JP zip file which have no equivalent in the WTTs. The most obvious examples are the Christmas period specials, which for some routes (but not all - what do you expect - consistency?) have their WTTs brigaded under the existing SCN but get given phoney SCNs in the JP set.

Right, that lot should either fry the brain or send you to sleep.

Thank you for that, that makes things a bit clearer; no brain frying smoke alarm sounding yet.