Resetting the numbers

[FosseWay]

Why is a hub for road numbering an advantage?

AFAICS, the zone idea for road numbering came about as a matter of administrative convenience at a time when classification and numbering were purely a way for the MoT to categorise roads internally. It wasn't an integral part of the public-facing aspect of road classification that came about later.

As I said, there is clearly a point to giving the most major routes numbers with fewer digits. But beyond that, I don't see the advantage to collecting all roads starting with a given digit in one place, and particularly not when that place is the shape of the current zones, which bear little relation to any aspect of how the roads are used (AFAICS). On the other hand, there can be a disadvantage to having very similarly numbered routes that intersect or are close by each other, especially with less memorable four-digit numbers. People not familiar with the area may misread signs being read in haste while driving, and anyone - local or stranger - may mishear instructions given orally.

It would therefore strike me as ideal to "randomize" the allocation of route numbers. In reality it wouldn't be random, but it would look random to the casual observer. First you classify the roads you want to number into four groups (with 1, 2, 3 or 4 digits) depending on their significance. Then you allocate the available numbers in each group at random, but you create a number of rules to ensure, for example, that consecutive numbers or numbers containing the same digits in different orders do not intersect. This rule may perhaps be irrelevant for A1-A9 or A1-A99 - you may want to allocate the first nine to specific routes for other reasons, rather than randomly, and that's fine.

[Ruperts Trooper]

Why is a hub for road numbering an advantage?

AFAICS, the zone idea for road numbering came about as a matter of administrative convenience at a time when classification and numbering were purely a way for the MoT to categorise roads internally. It wasn't an integral part of the public-facing aspect of road classification that came about later.

As I said, there is clearly a point to giving the most major routes numbers with fewer digits. But beyond that, I don't see the advantage to collecting all roads starting with a given digit in one place, and particularly not when that place is the shape of the current zones, which bear little relation to any aspect of how the roads are used (AFAICS). On the other hand, there can be a disadvantage to having very similarly numbered routes that intersect or are close by each other, especially with less memorable four-digit numbers. People not familiar with the area may misread signs being read in haste while driving, and anyone - local or stranger - may mishear instructions given orally.

It would therefore strike me as ideal to "randomize" the allocation of route numbers. In reality it wouldn't be random, but it would look random to the casual observer. First you classify the roads you want to number into four groups (with 1, 2, 3 or 4 digits) depending on their significance. Then you allocate the available numbers in each group at random, but you create a number of rules to ensure, for example, that consecutive numbers or numbers containing the same digits in different orders do not intersect. This rule may perhaps be irrelevant for A1-A9 or A1-A99 - you may want to allocate the first nine to specific routes for other reasons, rather than randomly, and that's fine.

I'd agree that the 9 most important routes in the UK should be allocated digits 1-9 with the next 90 important routes allocated 10-99 regardless of prefix, how that looks on a map or whether it creates zones from its segments.

[KeithW]

I think this should be on the Fantasy Road Forum.

That said I would have far fewer A Roads -- major inter-regional roads, How many A roads can you think of that you would realistically do at least 60 miles and cross between regions or nations. Regional B roads, road where reasonably to do at least 30 miles, and local C roads, any other road worth numbering. B and C numbers can be reused in different parts of the country.

Quality wise A roads would normally be D2, but there maybe exceptions, C roads would be at least S2.

You are proposing something rather akin to the US system where you have 4 tiers of road numbers.
Interstate Highways e.g. I 71
National roads such as US 40
State Roads such as OH 315
County Roads which in some areas are numbered but in others just have names or any variation thereof

Examples
Clark County Nevada 312
Wisconsin County Z
Scioto County 28

This is where things can get really complicated as in many areas the numbers at County level are really only used by the highways authorities for maintenance. County Road 104 in New York State is not signed as such even though its shown on many maps, on the ground the signs say West Chenango Road

In the midwest you have another complication in that the homestead acts of the 19th century resulted in a grid of minor roads being built with numbers such as N 600 W

God help you if you need to navigate them and dont have a really detailed map, usually the best you can do is keep driving in straight line until you meet a real road.
https://www.google.co.uk/maps/@40.41110 ... 1664?hl=en

The reality is the largest single group of roads in the UK are those that are unclassified or C roads. Signs displaying a C number are rather rare. The importance of these roads is often rather higher than this implies. One of the major north south roads across the North Yorkshire Moors (and most beautiful) is the route from the A170 east of Kirkby Mills through Hutton-le-Hole over Blakey Ridge to Castleton.

Personally I think the current road numbering system generally works well but what I would like to see is the addition of a formal Scenic Byway numbering system for such routes, another in North Yorkshire is the B1257. Farther west one of my favourites is the B6255 from Ingleton to Hawes and then take the A684 to the A1. The A66 is faster but in summer this is a lovely alternative pennine crossing. As a bonus you can take a peek at the Ribblehead viaduct on the Settle and Carlisle railway.

https://www.google.co.uk/maps/@54.21051 ... 6656?hl=en

At a different end of the scale is my old friend the B1040 winding its way through the flat lands of Lincolnshire and Cambridgeshire from NE of Peterborough to Biggleswade. If all else failed I knew I could get off the A1(M) at Peterborough and take it through Whittlesey, Ramsey and St Ives to my home in Gamlingay. Back road Britain at its best.

[B1040]

my old friend the B1040

[KeithW]

It would therefore strike me as ideal to "randomize" the allocation of route numbers. In reality it wouldn't be random, but it would look random to the casual observer. First you classify the roads you want to number into four groups (with 1, 2, 3 or 4 digits) depending on their significance. Then you allocate the available numbers in each group at random, but you create a number of rules to ensure, for example, that consecutive numbers or numbers containing the same digits in different orders do not intersect. This rule may perhaps be irrelevant for A1-A9 or A1-A99 - you may want to allocate the first nine to specific routes for other reasons, rather than randomly, and that's fine.

I see no good reason for such a complex system. The fact that the A66 and A67 intersect is scarcely a problem nor is what happens within a few miles of my house where the A174 intersects with both the A171 and A172.

As for significance the issue is that the relative importance of roads alters with time. In the 1960's the A19 was not an especially busy or important road, today between Thirsk and Seaton Burn its a more important and busier road than the A1(M) Darlington bypass and is now being gradually widened to D3. The A66 is another example that has changed out of all recognition from a minor link road between the A1 and Penrith into a major cross country strategic road. At the same time many of the older routes such as the A5, A6, A30 etc have become less important and the A20/M20 is now the major route to the channel ports not the A2/M2.

The reality is that only a small number of human beings consider the road number of any significance. As far as the maintenance of highways or satnavs are concerned you can simply allocate each road a GUID that the database uses to provide a Globally Unique Identifier. 700bd15f-556d-49e1-ab3a-fada44b46593 may not mean anything to you but to a computer is can signify a section of the A19 between Thirsk and Osmotherley. Going back to the A171 and A172 the reality is that the local humans call them them Marton Road and Cargo Fleet Lane while the A1032 is Acklam Road. These are the identifiers you will hear on local radio traffic announcements. Some of the major new routes around the southern periphery of the Teesside Conurbation are actually unclassified. The Stainton Way, Ingleby Way and Myton Way are now routes of major importance so much so that traffic now being diverted off the A19 and A172 onto those roads.

This is the A172
https://www.google.co.uk/maps/@54.53018 ... 6656?hl=en

This is Stainton Way
https://www.google.co.uk/maps/@54.52611 ... 6656?hl=en

Last but not least the unclassified Queen Elizabeth Way
https://www.google.co.uk/maps/@54.53144 ... 8192?hl=en

[FosseWay]

It would therefore strike me as ideal to "randomize" the allocation of route numbers. In reality it wouldn't be random, but it would look random to the casual observer. First you classify the roads you want to number into four groups (with 1, 2, 3 or 4 digits) depending on their significance. Then you allocate the available numbers in each group at random, but you create a number of rules to ensure, for example, that consecutive numbers or numbers containing the same digits in different orders do not intersect. This rule may perhaps be irrelevant for A1-A9 or A1-A99 - you may want to allocate the first nine to specific routes for other reasons, rather than randomly, and that's fine.

I see no good reason for such a complex system. The fact that the A66 and A67 intersect is scarcely a problem nor is what happens within a few miles of my house where the A174 intersects with both the A171 and A172.

I take your point about road numbers not being of huge interest to the majority, and about changes in significance of routes. But my suggested system is nothing like as complex as the current one, with two different zoning systems (A/B vs M) and various arcane requirements about what number a route can have depending on whether it crosses boundaries clockwise or anticlockwise and what happens if you reroute one of the Ax routes so the boundary moves, and so on. Random is basically as uncomplex as it's possible to be, and my suggestion is essentially random unless there are good specific reasons to depart from it, as in the allocation of the lowest 100 numbers.

I would also suggest that (a) you (generic) are unlikely to get confused by similar numbers in an area you know well, and (b) you (personally, and all the rest of us on here) are not a good test subject to see whether confusion can arise. We are interested in the whole question and tend to know off the top of our heads where a given road starts and finishes even if we've never been there. And there's still the problem of crackly phone reception when giving a location: "Hello. Ambulance please. There's been an RTA on the A17<crackle> near Ormesby Hall." If only one A17x number was anywhere near that location, there would be no ambiguity.

[Arcuarius]

If we have a hub, it should be around somewhere like Birmingham or Manchester. All major ring roads (such as the M25) should be renumbered to x0 in a shield. I think the zone system itself works quite well but probably should be sub divided, so where you're in the 1-zone, the next number should generally (as far as possible) indicate where in that zone it is. For this reason any non-strategic 3-digit A and B roads would be changed to a 4-digit number better reflecting their location.

Finally, all non-motorway roads would have a 4-digit number allocated, whether for not they are deserving of a better number. That way they can occasionally be swapped in and out of these "strategic" numbers as varying traffic flows allow, without any additional administrative stress.

[Bryn666]

Remember that Germany and the USA have numbering systems that don't rely on the capital city.

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

[KeithW]

I would also suggest that (a) you (generic) are unlikely to get confused by similar numbers in an area you know well, and (b) you (personally, and all the rest of us on here) are not a good test subject to see whether confusion can arise. We are interested in the whole question and tend to know off the top of our heads where a given road starts and finishes even if we've never been there. And there's still the problem of crackly phone reception when giving a location: "Hello. Ambulance please. There's been an RTA on the A17<crackle> near Ormesby Hall." If only one A17x number was anywhere near that location, there would be no ambiguity.

To which the response is 'say again', in reality of course any local would either say 'on the Parkway' or on 'Ormesby bank'. Many would probably not even know what the road number was. An accident on either would generate more than one call and would show up on the traffic sensors. There is no compelling reason here for the cost and disruption of a mass change of road numbers and all that accompanies it in the form of changed maps, paper and electronic and road signs.

[FosseWay]

There is no compelling reason here for the cost and disruption of a mass change of road numbers and all that accompanies it in the form of changed maps, paper and electronic and road signs.

No, of course not - right with you there! But the whole point of this thread was a hypothetical from-scratch renumbering.

[Steven]

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

We could just number by historic county...

[Bryn666]

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

We could just number by historic county...

The Blackburn Hundred becomes the new hub. Sorted.

[Steven]

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

We could just number by historic county...

The Blackburn Hundred becomes the new hub. Sorted.

Except nowadays it's called Blackburn Nil.

[c2R]

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

We could just number by historic county...

....how about a Spanish style numbering system based on historical county rather than administrative body, with historical routes given preference in terms of important numbers, and arbitary numbering changes at the county boundary. That would work even better, of course, for counties with detached parts....

[M5Lenzar]

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

We could just number by historic county...

I think Yorkshire would have a few more problems than Rutland!

[KeithW]

If we were to start from scratch I would advocate a two tier system.
1) Major strategic roads. In the past we kind of did this by showing trunk roads on maps with the suffix T.

2) Major regions. Highways England effectively already do this splitting the country into the following major regions.
South West, South East, Greater London, East, Midlands, North East and North West.

The first digit would indicate the origin of the road and subsequent digits the number and any suffix for quality.

In my own region the A19 between Thisk and Seaton Burn might become the T19 while the Thirsk to Doncaster section would be the NER 19A

Had we had this in the past renaming the old A14, now A1198 would have been the ER14 and the new A14 could have been the T14. You could even handle the case of seriously sub standard roads that are classed as strategic routes such as the A259. It could be the T259A or even T259B.

[RichardA626]

Remember that Germany and the USA have numbering systems that don't rely on the capital city.

The shape of the UK makes a grid difficult to work, but I think a loose sector system for all numbers above x100 (the first 100 routes being the most important in the country) rather than rigid zones would be better. At least it would stop people being upset about the A14 and A42

Maybe if the numbering was like the American system, & starting point was in Eastern Kent then it would be reasonably centred on London.

[Altnabreac]

I did a grid style system renumbering for Scotland a few years ago.

Would be trickier to make it work UK wide.

Here's my system.

N1-N9 define the boundaries of zones (roughly).

N1 Lamberton - Greenock A1 - A720 - M8 - A8 - M8 - A8
N2 Tradeston - Gretna M74 - A74 (M)
N3 Tradeston - Cairnryan M77 - A77
N4 Erskine - Inverness M898 - A898 - A82 - Inverness West Link
N5 Provan - Broxden M80 - M9 - A9
N6 Kirkliston - Bannockburn M9
N7 Claylands - Fraserburgh M9 Spur - A90 - M90 - A90 - AWPR - A90
N8 Dyce - Inverness A96
N9 Perth - Kirkwall M90 - A9 - Ferry - A965

Zone 1 is SE Scotland - Firth of Forth - Borders - M74
Zone 2 is SW Scotland - M74 - A77
Zone 3 is Ayrshire west of the A77
Zone 4 is the west coast west of the A82
Zone 5 is between the A82 and the A9
Zone 6 is Central Scotland within triangle of M80-A9, M8, M90
Zone 7 is East Coast east of A90
Zone 8 is the North east between the A9 and A90
Zone 9 is the North west from the A890 - A832 northwards

The F99 are then defined on a grid system. The first digit corresponds to Zone and the second digit is even for north south roads and odd for east west roads. The F99 change number when crossing into a new zone but can keep the same suffix digit.

Thus the A71 becomes Route 15 Edinburgh - Larkhall, Route 25 Larkhall - Kilmarnock and Route 35 Kilmarnock - Irvine.

The F99 are also ordered within zone so the numbers run 0-2-4-6-8 east to west and 1-3-5-7-9 north to south.

[marconaf]

Setting aside clearly this never will, or due to the cost and confusion, should, happen, in a context where locals know the roads by name or destination and they plus longer distance traffic all use satnav/phones anyway so the numbers are less and less relevant to people.

The big thing for me when using a map, digital or paper - is to understand what the “going” will be like on a route. I’d like to know what are grade separated dual carriageway sections on which I can expect to progress well, and which are S2 that I’m likely to find in villages with low limits, junctions and slower traffic.

So ideally different prefixes to identify the road standard:
M for full motorway
A for grade separated dual carriageway
B for no GS or S2 type roads but still main roads
C for less important Bs
D for even less important ones
U for unclassified (oxymoron!)
(There is an argument M to A but the now abandoned? Expressway gradings would allow a division).

Map colours would reflect these, say M blue, A green, B orange?!, C white/black, D ?

Numbers would be used for routes, much as they are now, the single digits being the main routes and thus lifting the existing Mways which would effectively be merged with the rest of the road network.

Where these roads drop Mway standard, eg. M5 @ Exeter to Plymouth, they’d become Ax.

10-99 and 100-999 etc. would be allocated zonally using the Mways as boundaries, Most current A numbers would thus be retained although as a mixture of A and B prefixes depending on standard.

I think London still works as a centre for the zones as so many of the UK’s principal routes do radiate from it. For cross country routes as with say current A14, the “clockwise from origin” feels a but desperate - create their own series of numbers?

I’d suggest E instead of A and thus keep most As as is instead of them becoming B on this system which also reflects Expressway terminology, except it confuses with the Euro routes. [I hate the Ax(M) designation intensely as confusing - is it a motorway or an A road? and overlong with the brackets.]

Problems would be renumbering A1-6 and any other M duplications - especially as they do still have some high quality sections, plus if a road is a bit schizophrenic in quality it might get a bit confusing on signs and colours varying over a route (A6!).

The key thing I would like to see however is an easy way to tell apart a good road like the A120, from a terrible one like much of the A6.

[KeithW]

Setting aside clearly this never will, or due to the cost and confusion, should, happen, in a context where locals know the roads by name or destination and they plus longer distance traffic all use satnav/phones anyway so the numbers are less and less relevant to people.

The big thing for me when using a map, digital or paper - is to understand what the “going” will be like on a route. I’d like to know what are grade separated dual carriageway sections on which I can expect to progress well, and which are S2 that I’m likely to find in villages with low limits, junctions and slower traffic.

So ideally different prefixes to identify the road standard:
M for full motorway
A for grade separated dual carriageway
B for no GS or S2 type roads but still main roads
C for less important Bs
D for even less important ones
U for unclassified (oxymoron!)
(There is an argument M to A but the now abandoned? Expressway gradings would allow a division).

The trouble is that apart from the Motorways there are very few roads that you can fit nicely into one of those categories. Lets start with the most obvious - the A1.

It starts in the city as an urban road continues as such to Finchley up to the A406 which it crosses at grade. From there it heads NW up the Great North Way, now this is dual carriageway but has at grade junctions and it continues in this vein to Apex Corner and Stirling Corner before it final loses the at grade junctions and you climb the dizzy heights to motorway standard until you get to Baldock where you are back on the old school A1 with at grade junctions all the way to Peterborough and this is barely the start of the entire road. Once you get north of Morpeth and Alnwick there are a lot of S2 sections with very dodgy junctions and of course the last leg into Edinburgh is back to city streets.

Practically every major road including many strategic routes has the same story to tell. They are clearly identifiable routes which have sections that are of different quality. There would be no advantage at all in splitting roads such as the A1, A14, A2, A3, A5, A38, A66 etc into separate chunks especially when the possibility exist that they may in future be improved as the A1, A14, A19 and A66 have been in recent years. Your system means that there would be a continual changing of numbers as roads were upgraded, or downgraded for that matter. Great for bureaucrats but terrible for road users and mapmakers.

The current method of assigning a route number such as A1 and assiging a suffix such as A1(M) to sections of higher quality is better and is why I would be supportive of having an A1(X) option for Expressways. Much of the A19 north of Thirsk is already close to that as will be the A66 between Scotch Corner and the M6 once the RIS 2 improvements are completed.

In my opinion at a national level we need to differentiate between routes of national or major regional importance and local roads which is what the A and B classes do now. The prefix says nothing about the quality of the road its about the importance of the route. The A19 between Thirsk and York is in its own way just as important as the section between Thirsk and Seaton Burn and is part of a contiguous route between Doncaster and Tyneside. Few people are likely to drive the route in full on a single journey but I have driven almost its full length at different times which means if come across it I know where it will take me without having to look at a map.