Shortest route query - Acton Bridge to Manchester

[Jack Hay]

Hi, yesterday I travelled Acton Bridge (ACB) to Manchester return (not easy owing to disruption at Cheadle Hulme) and this caused a 'permitted route' puzzle which I've not met before. I had an Anytime Return to Manchester Stations.

The shortest route from ACB to Manchester (any station) is via Warrington BQ, but this is unusable because there are no trains in that direction from ACB. So the usual route, which I took, is via Crewe.

The alternative route is via Liverpool South Parkway (LPY), but the Anytime Return is not valid via Parkway, according to the NRES journey planner.

Now if I was travelling to Piccadilly, this is understandable because via Crewe is marginally the shortest route (that has a train service).

But I was travelling to Oxford Road, and that is marginally shorter via Parkway, but NRES does not allow it. I have not come across this situation where the shortest route is apparently not permitted. Please can someone explain it for me?

 

[Mcr Warrior]

Although it's not an answer to the question as asked, is travel from Acton Bridge to Manchester permitted by travelling the one stop to Hartford and then walking the mile or so along the A559 to catch an onward train from Greenbank into Manchester?

 

[setdown]

Probably not, but when I've done this in the past I've had no issues. One guard even said "No problem, the ticket says Any Permitted, so you can get there however you want!"

Could never quite work out if he was having a laugh or not.

 

[yorkie]

Probably not, but when I've done this in the past I've had no issues. One guard even said "No problem, the ticket says Any Permitted, so you can get there however you want!"

Could never quite work out if he was having a laugh or not.

Difficult to say; they were probably exaggerating and merely indicating that they would accept anything that seemed reasonable. I doubt they meant it literally!

Hi, yesterday I travelled Acton Bridge to Manchester return (not easy owing to disruption at Cheadle Hulme) and this caused a 'permitted route' puzzle which I've not met before. I had an Anytime Return to Manchester Stations.

The shortest route from ACB to Manchester (any station) is via Warrington BQ, but this is unusable because there are no trains in that direction from ACB. So the usual route, which I took, is via Crewe.

The alternative route is via Liverpool South Parkway, but the Anytime Return is not valid via Parkway, according to the NRES journey planner.

Now if I was travelling to Piccadilly, this is understandable because via Crewe is marginally the shortest route (that has a train service).

But I was travelling to Oxford Road, and that is marginally shorter via Parkway, but NRES does not allow it. I have not come across this situation where the shortest route is apparently not permitted. Please can someone explain it for me?

The shortest route data - which booking engines have to adhere to - is only 32.02 miles, therefore it's way over, at 43.05 miles via LPY.#

The fare is higher from LPY so you are not likely to see any sort of easement; a pragmatic solution would be for a route: LPY fare to be introduced, at a value nominally higher than the LPY to MAN fare.

 

[Mcr Warrior]

The shortest route data - which booking engines have to adhere to - is only 32.02 miles, therefore it's way over, at 43.05 miles via LPY.#

Acton Bridge to Hartford is 2 miles 54 chains and then Greenbank to Manchester Piccadilly (via Stockport) is a further 29 miles 11 chains, so, assuming the walk along the A559 is deemed to be zero mileage, there's not much in it.

 

[yorkie]

Acton Bridge to Hartford is 2 miles 54 chains and then Greenbank to Manchester Piccadilly (via Stockport) is a further 29 miles 11 chains, so, assuming the walk along the A559 is deemed to be zero mileage, there's not much in it.

I am not quite sure what you are saying here.

You can't randomly come up with such an itinerary; it has to be supported by the data. There is no fixed link between Greenbank and Hartford stations.

 

[Mcr Warrior]

You can't randomly come up with such an itinerary; it has to be supported by the data. There is no fixed link between Greenbank and Hartford stations.

Didn't say it was necessarily valid, but it does, nevertheless, appear to result in the shortest mileage.

 

[alistairlees]

Didn't say it was necessarily valid, but it does, nevertheless, appear to result in the shortest mileage.

Yes, but this is meaningless I’m afraid. The shortest route between Carlisle and Dover is probably by train to dalston then to walk to Martin mill then train to Dover. Just a few miles! But also, obviously, ridiculous (and not supported by data). In determining the shortest route it’s the data (in the Routeing guide) that matters; nothing else.

 

[yorkie]

Didn't say it was necessarily valid, but it does, nevertheless, appear to result in the shortest mileage.

It's not the shortest mileage (and cannot be); it's not even a valid interchange.

The shortest route is as per the data, which is based on the shortest route by rail.

Other routes, which use valid interchanges, may be valid under the shortest route rule.

But any route which involves a walk between stations and is not a valid/recognised interchange (as defined in the data) is irrelevant.

 

[Watershed]

The shortest route is directly permitted under the wording of the NRCoT - i.e. without needing to refer to the Routeing Guide. The NRCoT don't say how the shortest route is to be determined; therefore, as long as you are confident of accuracy, you could use your own sources rather than the (often inaccurate) industry data. The NRCoT do, however, say that the shortest route has to be "by passenger services", which suggests it cannot include any walk or other non-rail links between stations.

You are only valid to travel on routes within 3 miles of the above by relying on provisions of the Routeing Guide - and this specifies that mileage data from the eNRT must be used.I think fixed links remain in somewhat of a grey zone, seeing as there is nothing in the Routeing Guide which explicitly mentions them.

Coming back to the original questions raised by the OP, travelling via Liverpool South Parkway - even to Piccadilly - is the shortest route by any definition, at 44.75 miles in the eNRT vs 45.5 miles via Crewe, or 44mi47ch vs 44mi79ch via Crewe according to the Sectional Appendix (the latter assuming the use of the rarely served Manchester Airport avoiding line, which is marginally shorter than the route via Stockport).

The eNRT obviously doesn't show any mileage between Acton Bridge and Warrington Bank Quay, as there are no services which call at both stations, so this route cannot be considered.

 

[Jack Hay]

Thank you Watershed, that answers my question but creates a new one! I took the mileages I quoted from the RailMiles tool. You obviously have access to the official tool which gives slightly different figures - and makes the route via Liverpool South Parkway the shortest. So why does the NRES journey planner not allow it? NRES says two tickets are necessary to travel that route, yet it is the shortest! That is something I've not met before.

 

[yorkie]

Thank you Watershed, that answers my question but creates a new one! I took the mileages I quoted from the RailMiles tool. You obviously have access to the official tool which gives slightly different figures - and makes the route via Liverpool South Parkway the shortest. So why does the NRES journey planner not allow it? NRES says two tickets are necessary to travel that route, yet it is the shortest! That is something I've not met before.

@Watershed was stating the de facto contractual position. Journey planners must adhere to the data.

I think you missed my earlier post:

The shortest route data - which booking engines have to adhere to - is only 32.02 miles, therefore it's way over, at 43.05 miles via LPY.

This is why NRE does not allow it.

 

[Jack Hay]

Thanks Yorkie, but I'm sorry to say I don't understand. Your mileage of 32.02 is via Warrington, isn't it? The two routes with train services are both around 44 miles, and Watershed is saying that via LPY is shorter than via CRE, though as those two are within 3 miles of each other, both should be allowed. I don't see how the 32 mile route with no train service is relevant here.

 

[yorkie]

Thanks Yorkie, but I'm sorry to say I don't understand. Your mileage of 32.02 is via Warrington, isn't it? The two routes with train services are both around 44 miles, and Watershed is saying that via LPY is shorter than via CRE, though as those two are within 3 miles of each other, both should be allowed. I don't see how the 32 mile route with no train service is relevant here.

You asked why NRE was not able to validate the route via LPY and I explained it is because the data states the shortest route is 32 miles long.

Journey planners have to adhere to the shortest route data, rather than the de facto contractual position.

A similar issue exists in many other cases, such as Newton to Carluke, Tutbury & Hatton to Rugeley Trent Valley and many, many more...

 

[alistairlees]

Thanks Yorkie, but I'm sorry to say I don't understand. Your mileage of 32.02 is via Warrington, isn't it? The two routes with train services are both around 44 miles, and Watershed is saying that via LPY is shorter than via CRE, though as those two are within 3 miles of each other, both should be allowed. I don't see how the 32 mile route with no train service is relevant here.

There are two sets of distance data

1. the data you see in the published timetable - the eNRT data
2. the data used by journey planners - from the routeing guide

They are not the same.

Journey planners must use #2. This shows that the shortest distance (wholly by rail) from Acton Bridge to Manchester Oxford Road is 32.02 miles, and is via Warrington Bank Quay and Earlestown.

 

[Jack Hay]

Right, thanks to you both. Let me check I've understood. The shortest distance ACB-MCO is via WBQ, but there is no train service that way. We all agree on that. So I gather NRES must offer the shortest route which has a service, and no alternative routes because any alternative is too long. What still puzzles me is why the only route offered is via CRE, when via LPY is shorter. Can you explain this please?

 

[sonic2009]

It's going to be something to do with the routeing guide and failing fares check?

 

[JB_B]

Right, thanks to you both. Let me check I've understood. The shortest distance ACB-MCO is via WBQ, but there is no train service that way. We all agree on that. So I gather NRES must offer the shortest route which has a service, and no alternative routes because any alternative is too long.

NRE will be calculating the shortest distance working from the routeing guide distance dataset - it definitely won't be checking if a passenger service actually exists over each link when doing that.

The RG distance dataset goes against the spirit of the public-facing routeing guide in as much as it includes some links between stations pairs where no practical passenger service actually exists.

The Acton Bridge to Warrington Bank Quay link is an example of this.

( Some of these are probably simply errors or a failure to update to reflect changing service patterns - there's plenty of other inaccuracies in there - and some are deliberately introduced fictions intended to shut down certain classes of exploit. )

What still puzzles me is why the only route offered is via CRE, when via LPY is shorter. Can you explain this please?

Crewe is valid as it's a mapped route ( via Liverpool Parkway isn't.)

 

[Merseysider]

It's always worth looking at split tickets.

Acton Bridge to Glazebrook is £8.90 via LPY
Glazebrook to Manchester is £4.90

Total £13.80, allows travel via LPY, is cheaper than the through single, but does restrict you to the infrequent service via GLZ. Haven't checked return fares; other splitting points may offer better value / journey opportunities.

 

[Jack Hay]

NRE will be calculating the shortest distance working from the routeing guide distance dataset - it definitely won't be checking if a passenger service actually exists over each link when doing that.

The RG distance dataset goes against the spirit of the public-facing routeing guide in as much as it includes some links between stations pairs where no practical passenger service actually exists.

The Acton Bridge to Warrington Bank Quay link is an example of this.

( Some of these are probably simply errors or a failure to update to reflect changing service patterns - there's plenty of other inaccuracies in there - and some are deliberately introduced fictions intended to shut down certain classes of exploit. )






Crewe is valid as it's a mapped route ( via Liverpool Parkway isn't.)

Thank you, I shall read up about mapped routes. I think that is the answer to my original question.