I would love it to be entirely in-house, being a retired IT manager who detested the enormous folly of outsourcing. I ask because it is now well over a year since 222 and U5 changed contract and hence the database needed a REVERSE_DEPOT_END function. Does it exist? The countdown predictions near both routes’ former depots (222=Hounslow, U5=Uxbridge) are still clearly expecting buses to be despatched regularly on schedule, while near the present depots (222=Uxbridge, U5=Hayes) the predictions are expected as dependant on buses coming in the other direction, clearly no longer the reality.
This must be happening on many other routes where contract changes have changed the depot location. It must be causing much distress and mistrust of the system. If the above function does not exist, it seriously needs to be developed.
Previous thread on this issue was “Countdown data on 222 seriously bad” Feb 15 to Sep 7.
This really is a bit of a mystery. Lots of routes change depot without any issues, though it is probably rare to move from a depot precisely at one end of the route to precisely the other. The U5 is slightly different, the depot being several minutes away from the first point where it takes up service, near Hayes Station.
I’m told that the 222 buses are actually kept somewhere quite different. Perhaps they only get on the system once driven to Uxbridge to take up service? But that would not explain some of the oddities my analysis showed up. Nor would it explain the U5, unless something similar is happening to other routes based at Hayes.
No I can vouch for the 222s being garaged and parked right in Uxbridge bus station, rare overflow go around the corner to York Rd but next departure never brought from there. I have seen drivers remember to log-in after leaving and when I look at the stops ahead, that service suddenly appears.
OK, I thought I might be clutching at straws. I think they are parked elsewhere though. On the working timetable the first few buses starting service at Uxbridge are allowed 10 minutes from UX, which is the code for the garage right next door (similarly for the last few buses finishing in service at Uxbridge). Then there is one allowed 1 minute. After that there are few such journeys starting from UX but there is one late morning (MF) that is back to 10. Some simplifications of reality going on, perhaps, but that should not matter as long as it affects neither the service nor (as it clearly does) information about the service.
It would be nice if someone took ownership of this issue within TfL though!
@MikeYates Does my reply to your previous post about Countdown data on the 222 explain the issues you refer to here too? Let me know and if not I’ll go back to the Buses team for more info.
Thanks
Theo
No, Theo, I’m afraid it doesn’t.
Mine is a “new” issue, relating to the “predictions” given from the timetable before a bus leaves the “depot”, being the incorrect depot in these contract-changing cases and those given near the turnaround, also incorrect, from incoming buses.
That means that the events surrounding the “first stop” you mention incorrect. There is a big difference between the “first stop” after the depot and the “first stop” after the turnaround.
I hope that makes it clear at last.
I base my theoretical diagnosis on 15 months of personal experience using the 222 and on long experience of database systems.
I doubt that it is easy to “change depot” for a route. That eventuality may have been unknown during development of the system. Perhaps the administrators may be reluctant to delete and re-enter a route to avoid downtime.