Information system - Rome to Rio

[whole diary]

[grahame]

Come across this a couple of times in recent weeks ... posting under "Smoke and Mirrors" because I don't know much about it, and I question the accuracy of the information provided and indeed (on other tests made, rather than on this sample) of the practicality of its suggestions. 



http://www.rome2rio.com/about

Rome2rio, based in Melbourne, Australia, is organising the world's transport information. We offer a multi-modal, door-to-door travel search engine that returns itineraries for air, train, coach, ferry, mass transit and driving options to and from any location. All in one place.

and

Users of Rome2rio don't need to search multiple websites to research and compare their travel options ...

Yeah, right!   

[bobm]

I'd query those train fares too.  It is cheaper NOT to split at Swindon for a start!

[TaplowGreen]

Interesting that driving is so much quicker & cheaper.

[IndustryInsider]

I don't think anyone would find it surprising that there are many journeys that are much quicker and cheaper in the car than by rail.  Melksham to Kemble via road is virtually a straight line via the A350 and A429, whereas by rail you have to do a pretty severe dogleg via Swindon.  That was the case even before Beeching came along.

Though the 32 minute road journey was estimated to take 43 minutes when I just checked on Google Earth, so is always vulnerable to the vagaries of traffic, especially given the vast majority is on single carriageway roads with plenty of roundabouts to negotiate.

[Bob_Blakey]

A quick check on t'interweb gives 41 minutes / ^11-^12 (using running costs from the AA) by car against a possible 50 minutes / ^12.90 by train at certain times of the day. Pretty marginal really.

[stuving]

A quick check on t'interweb gives 41 minutes / ^11-^12 (using running costs from the AA) by car against a possible 50 minutes / ^12.90 by train at certain times of the day. Pretty marginal really.

The AA's figures always include fixed (per-year) costs, but this site's ones look like they don't. If you want the A vs B cost difference that applies to the choice of mode, then only the variable (per-mile) costs are relevant.

[grahame]

There is, I'm afraid, so much I would query in this example that I didn't know where to start - just put the example up.

However ... you folks have started for me  so I can carry on ...
a) Frequency of services - wrong
b) Implication of ticket split being need on pricing - wrong
c) Overall journey time - wrong (well - not allowing for changes / waits)
d) Are there really 25 places to stay ("Accommodations") in Kemble?
e) Drive cost looks very low
f) Taxi cost looks rather high
g) Bus routing rather interesting ... and again I think connection time missed
h) All pricing for single journeys only?
i) Drive times - you just might manage 32 minutes in an emergency in the middle of the night. Peak I would allow an hour.

Train time IS accurate for twice-daily each way through services; pricing RETURN is ^18.60 or ^13.00 off peak.  Average journey time (each way) is around 75 minutes taking services offered by online booking / timetable engines.

We do have passengers who regularly commute Melksham to Kemble - although it's a dog-leg, it's not really silly if the services connect. And the through service that goes on to Cheltenham Spa is hugely useful for the Midlands and North. Indeed one of the TransWilts' looked at cases was (past tense) linking at Swindon so that the 2-hourly Cheltenham to Swindon unit would carry on south via Westbury, using up much of its wait time at Swindon and providing a good route to the north from Chippenham, Melksham, and (in one scenario) Frome, Yeovil, Dorchester and Weymouth. (Trowbridge and Westbury too, but that would have abstracted 'via Bristol' traffic)

[LiskeardRich]

The driving cost looks petrol price only- 24 miles melksham to Kemble. Gallon of fuel is around ^5-6 I think these days. My car does 25mpg roughly so the car cost shown is certainly petrol only in my case at least.

[Kate_Rome2rio]

Hi everyone,

Friendly Rome2rio representative here.

Thanks for all the feedback and input! If you ever find something on Rome2rio that's not quite accurate we encourage you to let us know! You can send any comments to: feedback@rome2rio.com

The team here have seen all your comments and we are now looking into correcting these problems, primarily the frequency estimates, definitely something going wrong there.

We actually get our data for UK^▸ (United Kingdom) train routes from Network Rail and then we do some calculations on our end to try and get an idea of the frequency.

Hopefully we'll be able to make improvements on all of the above points you have made.

As I said above should you find any further problems please let us know so we can look into them.

Thanks again,

Kate
Rome2rio, Content Manager.

[grahame]

Welcome to the forum, Kate ... thanks for taking this on board. 

As sampled the other day, I fear the data temps people to use the services as a "turn up and travel" one, on the understanding that it's frequent ... only to face a wait of up to 2 hours for a 9 minute journey.  The figure you're quoting looks more like the minimum interval between services - happens just once a day, and not necessarily at peak times (most frequent service on Swindon - Kemble is mid-evening) - please let us know once you've fixed it or removed the erroneous data.

Good to read that you're taking the Network Rail data feed - with algorithms sorted that should help keep you up to date with any changes.    And - hopefully in just a few years time - those frequencies you've suggested may become more of a reality and once again shown on your site