Modal filtering

Posted on
  • @dicki posting about filtering in his area here has just brought to mind that we don't have a general modal filtering thread, only an old one I started about London Fields some time ago. I'll write a longer post here later.

    Here's an article about communication:­bike-blog/2020/sep/03/road-closed-open-s­igns-low-traffic-neighbourhoods

  • I'll post something long and boring here later.

  • I'll post something even long and even more boring here later.

  • I'm replying here to some of the posts in the other thread mentioned above.


    Thanks Oliver, very comprehensive reply. I think that option number 1 looks really sensible and is v helpful in understanding how you can achieve the same impacts by having filters located away from the edge of the cell.

    Great to hear that comes across! I've been repeating this point ad nauseam, as it's one of the main points that people constantly get wrong. It's really to achieve a number of purposes, and the impacts aren't quite the same (but I know that's not what you mean). It does away with injudicious turning movements off main streets while maintaining the presence of side street junctions with main streets so that you don't get them to be completely de-prioritised; the purpose of a transport network is ideally not to prioritise one direction of travel over any other, but to enable interchange by means of people travelling in many directions. That may sound a little abstract, but to make it more concrete, if you take most junctions that have been closed off at the main street, you'll usually (not always) find that crossing the street there on foot or by bike is inconvenient or even hard, e.g. there may be a signalised pedestrian crossing twenty yards away and not near the (former) junction, or no dropped kerb for people cycling, and so forth. The upshot of this is always a suppression of travel across the main street, which is often what traffic engineers wish to achieve (see also dual carriageways with railings in the centre). You want to ensure that does not happen so much, and a proper junction mouth does help with that. Needless to say, the junction mouth should still be built in a modern way to enable people with mobility difficulties to cross easily, but not de-prioritised. If it's at the edge of a filtered area, there will be less through motor traffic coming out of it, giving transport planners less to worry about, but people on foot or cyclists will actually have an easier time if they need to cross the main street, as will the relatively small number of drivers coming out of there.

    I'm also pretty local, and am guilty of previously using Chobham Road as a convenient run through to Stratford.

    Well, who could blame you? The only other alignment is Cann Hall Road/Crownfield Road, and that's actually a little narrower, and obviously comprised of residential streets, too.

    It forms (or did form) part of a larger rat run that span across to the east of Leytonstone Road. I think it's more sensible to route that traffic down Major Road and Crownfield Road. It's a minor detour and the roads are much more suitable to the level of traffic. It also frees up Chobham Road as a more direct cycle route - up to this point it was a pretty horrible road to ride down.

    It's very interesting how that perception arises--in terms of alignment, (I would argue) Cann Hall Road and Crownfield Road are actually much more direct and continuous, and not a detour for most drivers. Unlike Chobham Road, it's not possible to argue that Crownfield Road should not be a cell boundary street, though.

    Although you stole the joke I was going to make about Major Road, so now I'm not happy.

    Pffft, I own all rubbish jokes, and if anyone is stealing anything, it's you. :)

    The best bit is where Major Road turns into quite a minor street, of course.

  • There's a modal filter near me in E10, part of the Francis Road project, which is right outside a secondary school.

    Every day at school pickup and drop off time, it is rammed with parents dropping their sons off or picking them up.

    There is less through traffic, but it has turned the road into a mini car park.

    I'll try and get some photos tomorrow if anyone is interested.

    There are loads of scooters and mopeds who bomb it through, this is a personal pet peeve of mine and I'm not too sure why

  • aggi:

    In terms of rat-running I do wonder what difference google maps, Waze, etc has made to it.

    As others have said, it has definitely made a difference, although I feel that is now a long time ago (not quite when massive lorries were getting stuck between drystone walls on tiny country lanes, but certainly when people started to use satnav systems heavily). I doubt that the effects will have worsened that much after the initial shock of the new, but it would definitely be interesting to see data, which should be easy to map. I've never looked at it as I think filtering should be done as a matter of principle,

    The danger is that with the additional pressure to filter, some of which has undoubtedly come on because of on-line/app/whatever routing information and drivers rat-running/driving at inappropriate speeds down inappropriate speeds/causing more crashes, you then end up with an additional impetus to increase motor traffic capacity on the perimeter as a way to compensate--i.e., something very similar to what has been happening to town centres/high streets up and down the country for decades, e.g. in Waltham Cross--high street filtered and a massive by-pass/ring road built.

    I see this in the context of the absurd situation that we supposedly have a £27bn road-building budget (which I hope now won't get spent, but I suspect it will be regardless), which would mainly increase capacity around the perimeter of centres, and at the same time a drive to filter local areas (in places like London and Brighton (I think) at least. Obviously, in London it won't easily be possible, short of changes in signalling methods (on-demand signal phases etc.) to increase perimeter capacity, and apart from the Silvertown Tunnel and the constant agenda to increase capacity on the North Circular, no really massive road-building projects are planned imminently, but the mistake of 'compensating' for filtering by making it easier to drive around centres must be avoided. Most European countries have made that mistake--the upshot is that larger centres benefit disproportionately, as people avoid the smaller centres by driving around them (using those convenient ring roads and the like) and then shop/go out/etc. in larger cities. This in turn causes the economy of smaller centres to suffer, leading to business closures and job losses there, in turn causing people who previously had worked quite locally to have to commute further afield, and so on.

    The majority of the time I'll get it directing me off an A road to a B road to avoid traffic (routes are almost identical length or the diversion can be longer). For those who know it the most obvious one is that you get directed off Green Lanes/Seven Sisters onto Wightman Road/Tollington Park almost every journey. You can see that a lot of traffic is also following the same diversion as people will all turn at the same points, etc

    I think where you get the effect of satnaving isn't so much on streets like Wightman Road, which has been a problematic rat-run for decades, but on the more minor streets. One of the things I do when I look at local areas is to follow the rat-runners around. They're always easy to spot because of aspects of their driving style, and as they typically don't get through local areas more quickly than me on a bike, I usually get a very clear idea of what they're doing. It's obviously never very clever stuff, as a bit of work with a map will usually show very easily where the likely rat-running alignments are (following decades of haphazard pseudo-filtering and alternating one-way operation schemes, it's usually the loopholes that people use, and those were always known to locals long before satnavs came along), but always interesting to see it being enacted live, as you get to see the conflict points, where queues develop first, etc.

  • oat44:

    Oliver, your local knowledge of my area astounds me. Very impressive!

    Well, it may appear that way, but it's really just a bit of work with on-line maps and StreetView. As I said, I know the area reasonably well, but certainly not as well as most of Hackney. Filtering is hardly rocket science and the principles are always the same. What's important is to apply a good methodology, and what you often find is that people use very bad method, often implementing the first thing that seems most obvious, which is rarely a good solution.

    Lots of cars are already using Colegrave/Edith as a way of bypassing the long queue heading W on Crownfield. They did before this, and now more so. It is moderately entertaining as they are often met by people heading the opposite way trying to skip the queue from Major Road to Crownfield. As Colegrave is a single lane road but still 2-way it can get quite heated at times, but there's usually just enough empty spaces that everyone gets through.

    Yes, and I forgot earlier to suggest that a filter could also be on the western arm of Colegrave Road, so I've edited the post accordingly. That would still lead to rat-running along that short stretch of Colegrave Road (east), but would probably cause other problems.

    I wondered about a filter both on Edith and eastern arm of Colegrave so at least it could be used as a turning space for bin lorries and delivery trucks.

    I thought about that, but the junction is definitely not wide enough. With public buildings, you need to talk to the people running them about what their requirements are. For instance, does the Community Centre run daycare for elderly people who are collected and brought every day by minibus/mediumbus? There's a height restriction on their car park, so that such buses would generally not enter that but require that they can simply approach via Edith Road or Colegrave Road and leave via the other.

    Paths for the Waste Service are also important, but generally operators would prefer having keys for gates that they could open and close. They don't like turning or reversing (although they often have to reverse when met by queues of drivers from the opposite direction). As it would be easy to install an openable gate here, that's not so much of a problem, and the Community Centre might also not worry about any of the above, but you first have to ask them to find out.

    I am happy they are having a go, but I was surprised that Chobham was closed off and miffed that the net result of that is an increase in angry traffic on Colegrave.


    I hadn't paid much attention to the implementation of the western section of the LTN (I live over on the Eastern side of area 2 and am fiercely parochial in my interests, down with the Marylanders etc etc) , but there does appear to be a glaring omission when it comes to the northern edge of area 1. Can't understand the lack of intervention there.

    Yes, what often happens is that engineers are given the brief who have no experience of filtering and actually don't see how gradual displacement of through motor traffic happens. You really have to do a whole cell in one go, there is no other way. And yes, that includes small, unconnected parts of the cell like the streets between Chandos Road and Crownfield Road.

  • Backstop:

    Yeah, but, upfront consultation would mean they wouldn't get off the ground, the only way to get anything implemented of this form is to do it and let people see the benefits, all else is just wishful thinking that's got nowhere for 20 years.

    Well, what used to happen was obviously not good, either, but in different ways. Usually, a group of residents got some wind under their wings to go through the motions with the council, took a deputation to a full council meeting, got articles in the local newspaper, etc., got their councillors to support them, with the upshot that something (that usually wasn't very good) was then designed by some highway engineers and put out, top-down, to 'consultation' (which many people think is a vote, but which is actually just an information-gathering exercise and the wrong vehicle for 'voting').

    That then led to the predictable counter-campaigning group to be formed ('we were never informed') who often managed to get schemes watered down or defeated. Again, obviously, that's not good. If there were no alternative to this happening, sure, one could just do it top-down, but there are also examples where things were not imposed top-down, but where communications staff facilitated genuine engagement exercises and where neighbours actually talked to each other.

    It needs to come up from the streets if you want the local knowledge. Top-down generally brings very poor outcomes--I think I've seen one scheme of the current batch that's even halfway acceptable in terms of where the filters are, although I haven't looked at them all. Sure, areas are then filtered, but badly, and they then don't get touched again for 20 years, or get disfigured when the inevitable crash clusters happen that are easily explicable by the poor filtering scheme in operation, but which engineers generally fail to treat adequately, only seeing on-the-spot solutions where area-wide measures, and a complete review of the bad filtering, is needed.

    (That may seem too abstract, but one example is the junction of Southgate Road and Northchurch Road, a very problematic junction, despite looking quite innocuous, where there have been a large number of horrible crashes. The reason why these problems exist at this junction is because the eastern arm of Northchurch Road is filtered at the junction. This causes drivers on Southgate Road, the cell boundary street, to drive like idiots. I can't think of another junction in Hackney where during junction observation I've seen such bad driving. If this junction had a junction mouth on both sides for drivers approaching along Southgate Road, their behaviour in driving through it would be very different, and the junction layout could also be improved significantly. The junction in question has been tinkered with since forever, and I originally didn't understand what the problem with it was, either. However, having thought about it for a long time, I now know that the filter at the junction must be removed and pushed deeper inside the cell. That in turn requires other changes to the filtering of that cell, so it becomes an area-wide problem that people often find it much harder to get their heads around and that, especially in view of the long history of, and the hard-won fight for, the De Beauvoir filtering scheme, is political. There are hundreds and hundreds of examples like this with edge filtering in London, and lots for the other ways of getting filtering wrong.)

  • There's a modal filter near me in E10, part of the Francis Road project, which is right outside a secondary school.

    Every day at school pickup and drop off time, it is rammed with parents dropping their sons off or picking them up.

    There is less through traffic, but it has turned the road into a mini car park.

    I'll try and get some photos tomorrow if anyone is interested.

    I take it you're referring to Norlington Road and and Norlington School? That's definitely not a very good scheme. It seems rather isolated. Many streets around there, e.g. Claude Road, Murchison Road, and Albert Road, are still one-way (one of the main things that proper filtering must enable is to return local streets to two-way and to get rid of those (mostly one-way) loops that former generations of engineers created), and on StreetView you can see a traffic island with its plastic illuminated bollards shaved off in Norlington Road (south), which indicates speeding and.or badly-managed conflict. I haven't been in that specific area for a long time, but again, filtering needs to be done by area, not by trying to address problem spots.

    I actually saw a similar problem just recently outside Uphall Primary School in South Ilford/Loxford. There's a kind of turning hammerhead at the end of Wingate Road created by very poor filtering that as usual isn't area-wide. I rode past just before chucking-out time and watched in fascination as six or seven drivers turned up to pick up their kids from school and tried to somehow share this area in a really elaborate elephants' ballet.

    With schools, what people have found is that you have to define a kind of wider exclusion zone. It can be time-limited to school run times. I think this works quite well at Gayhurst School in London Fields, which is one example I know. 'School streets' are no substitute for good filtering, but I think they're better than an isolated filter like the one at Norlington School. There are still one-way loops that people can drive along to access the school and those need to be got rid of in the same go.

    There are loads of scooters and mopeds who bomb it through, this is a personal pet peeve of mine and I'm not too sure why

    Quite understandably, as it undermines the purpose of filtering. I don't know if it's now possible to include powered two-wheelers in ANPR, but famously this was still impossible when Congestion Charging was introduced in 2003, leading to some guff about how two-wheelers generally helped with congestion. Until it is possible to fine people for driving PTWs through filters, this will continue, unfortunately--the only thing that could stop them would be the towpath-style gates designed to exclude them. These we don't want to see, as they're a pain to navigate for cyclists, and unless the footways were fitted with them, too, which isn't possible, all two-wheelers would just go along the footway, anyway. This is an unresolved problem, unfortunately.

  • The one way streets were created when Francis Road was closed to through traffic between the hours of 10 to 8.­-your-area/leyton-town-centre/constructi­on-updates/

    The works prevented anyone with a car in the northern part above Albert Road travelling to anywhere south, without going into the High Road, and vice versa.

    A lot of traffic was heading down Scarborough Rd, the other side of the train line, which has led to that being closed to left hand turns from Grove Green Road.

    If I had a fiver for every time I have given confused visitors in cars directions of how to get one road south of the closures, or one road north, I'd be a few hundred pounds richer.

    Come around 8.30 or 3pm (1 on a Friday) to hear the Norlington school chorus of car horns.

  • Thanks, that sounds like a poorly-thought out scheme. One does hope that they'll revisit it soon.

  • Hammersmith and Fulham created a massive LTN in Sands End East with 5 ANPR cameras. We
    now have a huge discussion raging in the local community.

  • What is getting my goat is:

    • people arguing that traffic has increased on their road despite traffic counts showing a reduction (e.g. on streets outside the LTN, which were formerly a rat run, which are no longer a rat run because a filter somewhere stopping the route as a whole being suggested by Google Maps)

    • people who live in nearby areas filtered decades ago, arguing against new filtering projects while passionatly defending their own traffic filters

    • people complaining that cutting off a 4000 vehicle a day rat run is unacceptable because 20 vehicles a day are now using their street

    • people

  • It's certainly annoying a lot of people down here. It's pushed me to be quite a lot less accepting of parts of the car driving community.

    I do drive but I've been cutting down over the last 15 years. I'm involved in the local community and a big fan of push bikes and bikes in general so I guess I was destined to come down on the side of believers in the LTN.

    There's plenty to complain about if you don't like them though. Ours has near invisible signage and I'm guessing a lot of fines have been handed out.

    We've had some incidents with speeding where it would have been impossible before, at times when children are around.

  • Sorry, I managed to miss this. I've had a quick look at this. Obviously, as usual I support the general intention behind it, but there are a few problems.­ads/sw6-traffic-reduction-scheme

    First, you need to define cells properly. This may not seem important, but it is. The graphic suggests that this is one cell, but Townmead Road is actually a boundary street. As it's the only connection to Chelsea Harbour that crosses under the railway line all the way up to the King's Road, it won't get filtered. The small cell south and east of Townmead Road next to the river is also unfiltered and may not get filtered. It probably attracts a small amount of rat-running if Townmead Road is stuffed up, but probably not enough to warrant filtering.

    So, you have a cell bounded by the A217 Wandsworth Bridge Road, Townmead Road, Harbour Avenue, Chelsea Harbour Drive, Lots Road, and the A308 King's Road/New King's Road. This is, in effect, an expansion of a smaller, pre-existing cell in the centre with the boundaries of Broughton Road, Townmead Road, Imperial Road, Harwood Terrace, Bagley's Lane, and Pearscroft Road--effectively a cell within a cell, as it left streets parallel to main streets as boundary streets and therefore through routes for motor traffic--plus the existing filter in Michael Road. This doesn't make much sense; I mean, it may have done at the time, but it was only a matter of time before it had to be reconsidered. The new filtering is now an attempt to complete this work while leaving the pre-existing filters in place by filtering some streets (Imperial Road, etc.) that were previously boundary streets.

    The main problem with this methodology is that it leaves the cell unbalanced. You can see how large the green area is compared to the others. This is caused mainly by where the existing filter is in Bagley's Lane and where the new filters are in Broughton Road. The ones in Elswick Street and Stephendale Road are also in the wrong place, although obviously the attractive Stephendale Road streetscene scheme won't be removed (or would cost a lot to change in a sensible way). With this sort of imbalance, you have a lot more motor traffic coming into the green area from Wandsworth Bridge Road than into the two other colour codes. It also omits to mention that motor traffic will come into the green area from the Chelsea Harbour direction via Townmead Road.

    In short, the position of the existing filters causes some of the new filters to be in the wrong places--the old ones were either at the centre of, or too close to the edge of the old cell. All existing filter locations should therefore have been reviewed and similar filters doing similar jobs moved closer to the centre of the new cell where possible. This means fewer detours for residents when driving out of the cell, so less local anger, and potentially fewer loops, and there are a lot of loops left in there (these can be a problem close to main streets, as lost drivers often use loops through side streets to turn around, etc.).

    Now, obviously, I've seen local anger before. I don't generally agree with it, but if you listen carefully through the noise, people often do have valid points (alongside not-so-good points), and they'll be related in a concrete way to what I say above in a rather abstract way. Well-conceived filtering schemes still have their opponents, but they'll be much easier to use for answering concerns.

    And again, I've always said these schemes need to come up from the streets. The 'old' filtering schemes generally did. The problem with them was usually as above, that they were partial and had some logical mistakes in them, such as filtering at the edges. Often, they were watered down in poor compromises and then persisted for decades. When the campaigners of the time retired, there was often a long lag before anyone took up the cudgels again. The same things will happen to the current batch of schemes. Some will be watered down and some will be removed altogether, while some will be changed for the better--it'll all depend on the quality and understanding of local campaigners.

  • There's plenty to complain about if you don't like them though. Ours has near invisible signage and I'm guessing a lot of fines have been handed out.

    Yes, I have to say that I'm still not convinced that filtering by ANPR is a good idea. I'd much prefer well-conceived schemes that can do with only proper physical filters, maybe the odd gate that emergency services and the waste service can open.

    We've had some incidents with speeding where it would have been impossible before, at times when children are around.

    Was this because of the above-mentioned imbalance in the cell, i.e. long straight stretches left open (and then possibly only filtered by ANPR (that may not have been put in yet))?

  • It's a very complicated balance. At the moment the west side of Wandsworth bridge road struggling and Wandsworth bridge itself has restrictions due to works. There's talk of making Imperial southbound only. WBR has no right turn into it from New Kings Road and now none for non residents into Imperial. Hammersmith Bridge being closed is adding to the traffic.

    I'm guessing you have some professional experience in the subject. This system would make an interesting case study I'm sure, there's very little room for manoeuvre.

  • Yes, I forgot to mention the issue of banned turns on cell boundary streets. Cell boundary streets must be two-way, with all turns permitted, for filtering to work.

    No professional experience, no, just 20 years' volunteering experience (which may be better than professional in some ways). I researched filtering and realised that there are no flawless filtering schemes anywhere, so I wanted to find out why and came up with a list of dos and don'ts. These are hard to achieve but worth it.

  • Certainly in my own experience of community volunteering you can get a view of things that the professionally interested parties don't have. How to get the information to the decision makers is usually the problem.

    There are currently discussions about the west side having the same scheme, main downside to that is our neighbours on Wandsworth Bridge Road who will be stuck with a huge amount pollution.

  • Post a reply
    • Bold
    • Italics
    • Link
    • Image
    • List
    • Quote
    • code
    • Preview

Modal filtering

Posted by Avatar for Oliver Schick @Oliver Schick