How do some cyclists manage to avoid busy roads...

[Richard Fairhurst]

roubaixtuesday has given absolutely the right explanation so I don't need to!

I would just add, on this specific topic, that cycle.travel does use real motor traffic data as part of its decision process - i.e. it tries to avoid roads with a lot of cars (subject to available data). That's how it knows, for example, to avoid the B roads between Gloucester and Ross-on-Wye because they're incredibly busy, whereas between Bishop's Castle and Montgomery it'll take the B roads in preference to the NCN route because they're very quiet and less hilly.

[Nearholmer]

I live and learn.

Thanks.

[roubaixtuesday]

roubaixtuesday has given absolutely the right explanation so I don't need to!

I would just add, on this specific topic, that cycle.travel does use real motor traffic data as part of its decision process - i.e. it tries to avoid roads with a lot of cars (subject to available data). That's how it knows, for example, to avoid the B roads between Gloucester and Ross-on-Wye because they're incredibly busy, whereas between Bishop's Castle and Montgomery it'll take the B roads in preference to the NCN route because they're very quiet and less hilly.

It works really well in those sort of situations, but less well in urban or suburban where it will try to avoid busy roads but instead makes you turn on and off them multiple times. Eg here coming out of Marple; the main road is genuinely unpleasant, but turning on and off it multiple times even more so.

I wonder, does the algorithm put any weight on turnings on and off busy streets as opposed to distance on and off them? If it's possible, could that be a good idea?

[Nearholmer]

Whatever works for you of course, I'm just glad we have the choice.

Yes, I often have an entirely back to front approach: I scan OS looking for interesting routes, long sections of byway, interconnected bridleway networks, unfenced roads of the lowest order, that sort of thing, then decide to go to wherever they are, and have a nose round. So the form of route becomes the destination, rather than the destination dictating the route. Even when I am actually “going somewhere”, I often choose what looks like an interesting, rather than direct or quick, way to get there. If I turned my mental process into an algorithm for a routing engine, I’d probably be the only person ever to select it as an option!

[Richard Fairhurst]

I wonder, does the algorithm put any weight on turnings on and off busy streets as opposed to distance on and off them? If it's possible, could that be a good idea?

It absolutely does - there's 300 lines of decision-making code (plus a bunch of supporting C++) to calculate penalties for turns. It's just a very hard problem to solve.

[roubaixtuesday]

I wonder, does the algorithm put any weight on turnings on and off busy streets as opposed to distance on and off them? If it's possible, could that be a good idea?

It absolutely does - there's 300 lines of decision-making code (plus a bunch of supporting C++) to calculate penalties for turns. It's just a very hard problem to solve.

I did think you'd be waaay ahead of me, but thanks for coming back anyway, it's appreciated.

[Bmblbzzz]

roubaixtuesday has given absolutely the right explanation so I don't need to!

I would just add, on this specific topic, that cycle.travel does use real motor traffic data as part of its decision process - i.e. it tries to avoid roads with a lot of cars (subject to available data). That's how it knows, for example, to avoid the B roads between Gloucester and Ross-on-Wye because they're incredibly busy, whereas between Bishop's Castle and Montgomery it'll take the B roads in preference to the NCN route because they're very quiet and less hilly.

In the Gloucester to Ross example, the B roads can be busier than the corresponding section of the A40. Unfortunately an increasingly common situation as car sat nav planners take "quietest" or "shortest" routes rather than main roads.

[mjr]

roubaixtuesday has given absolutely the right explanation so I don't need to!

I would just add, on this specific topic, that cycle.travel does use real motor traffic data as part of its decision process - i.e. it tries to avoid roads with a lot of cars (subject to available data). That's how it knows, for example, to avoid the B roads between Gloucester and Ross-on-Wye because they're incredibly busy, whereas between Bishop's Castle and Montgomery it'll take the B roads in preference to the NCN route because they're very quiet and less hilly.

In the Gloucester to Ross example, the B roads can be busier than the corresponding section of the A40. Unfortunately an increasingly common situation as car sat nav planners take "quietest" or "shortest" routes rather than main roads.

And this problem is becoming worse. When I do use a car sat nav, I now sometimes find myself telling it I'm in a box van to avoid it using more direct but in reality slower (unless you like a high chance of taking your car for a swim!) and some borderline unsuitable-for-motors C/D/U roads across the fens.

[Nearholmer]

Yep, we got caught like this driving to holiday in the summer. Huge tailbacks on the main road, so the satnav decided we’d be better-off taking a scenic route ……. never again!!!

[Bmblbzzz]

C/D/U roads across the fens.

I have a mental image of Angela Merkel on a sit-up-and-beg, like John Major's old ladies to church.

[Jdsk]

C/D/U roads across the fens.

I have a mental image of Angela Merkel on a sit-up-and-beg, like John Major's old ladies to church.

And with clipless pedal systems... ?

Jonathan

[TrevA]

When we did our JOGLE we basically followed the CTC B&B Route. I’m not sure if that is still available. It followed a lot of local cycleways. One thing we did find, is that these routes are often not that direct. We followed parts of the Lancashire Cycleway, but found that it stuck to little roads that went up and down the local hills, whereas we could see the not too busy main road followed the valley bottom, so in the end, we diverted onto that, to get a more direct and easier route.

There’s a lot of good advice on this forum, particularly about how to avoid the steep hills and main roads in Somerset, Devon and Cornwall. We altered our route accordingly.

[pwa]

roubaixtuesday has given absolutely the right explanation so I don't need to!

I would just add, on this specific topic, that cycle.travel does use real motor traffic data as part of its decision process - i.e. it tries to avoid roads with a lot of cars (subject to available data). That's how it knows, for example, to avoid the B roads between Gloucester and Ross-on-Wye because they're incredibly busy, whereas between Bishop's Castle and Montgomery it'll take the B roads in preference to the NCN route because they're very quiet and less hilly.

And having done the B road route between Bishop's Castle and Montgomery I can confirm that it is a mostly quiet and pleasant route.

[Ferrij]

Slightly off piste but I borrowed this from our local library and stuck it on my Xmas wish list. It’s certainly the most comprehensive reference book I’ve come across and works well in conjunction with OS maps when planning routes in the UK.

[Tigerbiten]

I've found after over 60 years and umpteen thousand of miles of cycling on all road, I've a reasonably good idea on how busy a road is likely to be.
As mentioned above, it tends to be the B roads running into a town that are the most variable.
Some are dead quite and others can be silly busy and the busy ones tend to be narrow so are a lot worse vs a wider A class road with the same traffic density.