UK Journey planners' weirdness for Oxford -> Combe

[Nicholas43]

Any theories on why, asked for a 0630 depature from Oxford to Combe, DB Hafas gets it right (double back from Charlbury); NRE says double back from Kingham, and First Great Western says double back from Moreton in Marsh ?

NRE and FGW also wrongly say "buy two tickets", although I believe this trip is covered by an Anytime Return Oxford<>Charlbury, stopping short at Combe on the return leg. I guess there's no hope of getting any algorithm to suggest that.

Agreed, there can't be many takers for this journey, so I assume FGW is not just trying to maximise revenue. Maybe they are charitably advising that the better chance of seat on the up Oxfordshire halts train is to go out to Moreton??

 

[dzug2]

Quicker to walk from Hanborough?

Maybe what they are really saying is there is no service at this time of day

 

[yorkie]

NRE and FGW also wrongly say "buy two tickets", although I believe this trip is covered by an Anytime Return Oxford<>Charlbury, stopping short at Combe on the return leg. I guess there's no hope of getting any algorithm to suggest that.

I agree, and I can't see booking engines suggesting travelling further in order to come back and break the return journey, although it would technically be possible.

 

[RJ]

NRE and FGW also wrongly say "buy two tickets", although I believe this trip is covered by an Anytime Return Oxford<>Charlbury, stopping short at Combe on the return leg. I guess there's no hope of getting any algorithm to suggest that.

I've known the NRE journey planner to suggest buying one ticket and stopping short for certain journeys - by no means for all journeys though!

 

[yorkie]

Interesting. Got any point to point examples?

(Not rovers/rangers as I'd argue that's a different matter and there us no real concept of finishing short on them)

 

[Paul Kelly]

Charlbury is not designated as an interchange station in ATOC data, whereas Kingham is designated as a small interchange, and since you can get to Combe at the same time by changing there, journey planners should, if correctly following ATOC data and rules, suggest the change at Kingham.

No idea why WebTIS (FGW site) is suggesting taking an earlier train from Oxford and having 1 3/4 hour layover at Moreton-in-Marsh. Perhaps it is a bug that comes to light only because it couldn't find any valid fares for the itinerary with the change at Kingham, so abandoned that completely, tried a different itinerary, couldn't get a valid fare for that either and just gave up and dumped that second best itinerary out?

 

[Nicholas43]

Charlbury is not designated as an interchange station in ATOC data, whereas Kingham is designated as a small interchange ...

Ah, thanks. Presumably because Kingham has the bus link to Chipping Norton?

No idea why WebTIS (FGW site) is suggesting taking an earlier train from Oxford and having 1 3/4 hour layover at Moreton-in-Marsh. Perhaps it is a bug that comes to light only because it couldn't find any valid fares for the itinerary with the change at Kingham, so abandoned that completely, tried a different itinerary, couldn't get a valid fare for that either and just gave up and dumped that second best itinerary out?

FWIW, if WebTIS decides to offer the 0514, it has to say change at Moreton, because it's non-stop Oxford>Moreton.

Quicker to walk from Hanborough?

Yes. Traveline indeed suggests walking from Hanborough - but also offers the weird 0514 via Moreton-in-Marsh.

 

[RJ]

Interesting. Got any point to point examples?

(Not rovers/rangers as I'd argue that's a different matter and there us no real concept of finishing short on them)

It was a couple of years ago, from Loughborough to somewhere in the Warrington area. A text box appeared saying something to the effect of it being cheaper to buy a ticket to Chester and stop short.