Near miss at Coltishall Lane User Worked Level Crossing, Norfolk, 21 January 2021

[DerekC]

(As far as I can see there isn't a thread on this already)

RAIB has just published its report on the above incident and it's worth a read. Here's the link and the summary:

https://www.gov.uk/government/news/...6410-4472-ac8b-43c14b8c3aec&utm_content=daily

Summary​

At 17:21 hrs on 21 January 2021, a passenger train narrowly avoided a collision with two cars at Coltishall Lane user worked level crossing, near Hoveton, Norfolk. Road traffic over the crossing was higher than usual as a result of drivers seeking a diversionary route for their journeys, following an earlier road traffic accident on a nearby main road.


The safe use of the crossing relied solely on telephones to warn users in road vehicles of approaching trains, and the near miss occurred because the car drivers did not telephone the signaller before using the crossing. The investigation found that this may have been because the car drivers were unfamiliar with the crossing, because the signs at the crossing were ineffective in prompting users on how to cross safely, and because the level crossing gates had already been opened. It is also possible that factors such as the increased level of road traffic may have affected the behaviour of road users and their decision-making.

The investigation also found that Network Rail and its predecessors had not taken measures to close or upgrade this crossing, despite being aware of the risks it posed. This was possibly because Network Rail’s processes for assessing and controlling risk at the crossing did not take account of some of the factors present, leading to an incomplete understanding of the risks involved in its operation. Additionally, the status of user worked crossings on public roads was not acknowledged within relevant statutory provisions and industry guidance, and this may have affected how safety at these crossings was managed by infrastructure managers. The Office of Rail and Road had previously taken regulatory action, but this had not resulted in action by Network Rail to address the risk at this crossing by the time the near miss occurred.

Recommendations​

RAIB’s report makes three recommendations. The first is that Network Rail should assess, and if necessary reduce, the risks of incidents and accidents at vehicular user worked crossings which may see significant use by unfamiliar users, consistent with current industry best practice. The second is that Network Rail, in consultation with the Office of Rail and Road and the Department for Transport, should improve signage at user worked crossings which may share features identified in this investigation. The third is that the Department for Transport, in consultation with the Office of Rail and Road and Network Rail, should evaluate the extent to which recent guidance on the drafting and making of level crossing orders will better enable the implementation of improved safety measures at level crossings.

There is one learning point that reminds those responsible for the management of safety of user worked crossings to adhere to the principles of level crossing risk assessment published by ORR in June 2021.

There are many issues to consider here. However it seems pretty clear that user-worked crossings should not, except in very unusual circumstances, be used on public roads yet the report seems unwilling to say so directly. I can't help feeling that the issuing of ORR's risk-assessment based guidance in 2021 is not going to help. Should the trend towards standards based entirely on risk assessment continue? It assumes that the necessary expertise and judgement is available in every case on every occasion that risk is assessed and in a large organisation like Network Rail that is very difficult. The tendency is to use a process-based, tick-box approach which can miss key issues. Would it not be better to retain some degree of prescription in standards, whilst providing for risk assessment to be applied in complex or difficult cases? I would be interested to know what others think.

 

[Adoarable]

To start with it’s surprising enough that there are still user worked crossings on public roads.

Even more surprising is the idea that Network Rail seriously expect a member of the public to get out of their car and make a telephone call before opening the gates, especially since the sign that references making the telephone call is so small, wordy, and out of headlight range.

Completely unsurprising is the fact that Railtrack/Network Rail have spent the last 20+ years prevaricating over whether to upgrade the crossing. I imagine that if the ORR told them they had to have a crossing keeper on site 24/7, the crossing would have been upgraded to AHB years ago.

 

[dosxuk]

I imagine that if the ORR told them they had to have a crossing keeper on site 24/7, the crossing would have been upgraded to AHB years ago.

More likely they would have actually pushed for removing the crossing entirely. It's a proper little country lane which goes from nowhere, to nowhere, with, according to the report, an average of 4 users a day. I lived around there for 20 odd years and while I've used pretty much all the other roads around there, I've never once been across that crossing - I honestly thought it had been closed up when they upgraded the line - it really only exists for access to the nearby fields.

 

[Sheridan]

Like others, I was surprised how many user-worked crossings exist on public roads. Quite apart from the problem of individual users failing to telephone, the big problems are surely when more than one vehicle driver (who are not known to each other) wants to cross in quick succession, as happened here. Of course the proper procedure would be for the first to telephone, open the gates, drive across, stop and go back to close the gates, before the next user does the same, but once the gates are opened it’s hard to imagine the average road user doing anything but following the first one across, and so on.

 

[edwin_m]

Miniature stop lights, which would have greatly improved safety compared to expecting every user to phone. In this case the lights changing to red would probably have led to later vehicles in the queue waiting until the train had passed, just as they would at a normal traffic signal, even if the gates had remained open. They were not considered for many years because various guidance and standards suggested they weren't permitted on public roads. As described in the report, this is now allowed if a risk assessment suggests it is the best option.

 

[Class 170101]

Miniature stop lights, which would have greatly improved safety compared to expecting every user to phone. In this case the lights changing to red would probably have led to later vehicles in the queue waiting until the train had passed,

Hmm maybe

just as they would at a normal traffic signal,

But the fatal flaw perhaps - how many car users ignore normal traffic lights?

Frankly either close it or change it to an MCB - OD with full barriers.

 

[edwin_m]

Hmm maybe

But the fatal flaw perhaps - how many car users ignore normal traffic lights?

Frankly either close it or change it to an MCB - OD with full barriers.

I suggest it's rather obvious that providing red and green lights would be safer than telephones only. Traffic lights are relatively well observed, probably because drivers know if they jump them by more than a couple of seconds they are liable to be slammed into by someone on the other arm of the junction. Certainly respected more than speed limits (unless enforced with cameras) or many other road signs. The instruction to phone is not something most drivers would see or expect, and at a "normal" level crossing it only applies to large or slow vehicles, so some might assume the same here. Lights could also be linked to CCTV monitoring.

This is especially so in the situation reported, where there were several vehicles going each way. Second and subsequent drivers would have assumed the first one had done whatever was necessary to get permission to cross, and would have felt under pressure to follow so those behind them and on the other side (probably a single track road) could get through. A green light changing to red part way through that process would be very likely to stop any drivers who hadn't yet passed it, just as it would with a normal traffic light.

What railway enhancement would you cancel to find the several million £ for an MCB-OD?

 

[Taunton]

One is struck in the report by the enormous amount of back-and-forth passing of the issue between different bodies, while in fact many years passed. There seems no ability for someone to just say "Do this". The lack of clarity between the different bodies, who seem only too ready to say (eventually) "don't do this", but quite unprepared to say "do that instead".

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

 

[Dai Corner]

Lets not forget that the crossing has been used for many years without incident. The four users a day were no doubt well aware of the risks and correct procedure and the near miss only occurred because non-locals were trying to avoid congestion elsewhere.

A more proportional response if a similar situation occurs might be to send someone to supervise the crossing, or close the road temporarily.

 

[Railwaysceptic]

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

I've been told that some Sat Nav systems can offer an alternative route when the usual route is unsuitable.

 

[dosxuk]

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

Answered in the report - sat nav's directing people around the road closure on the main Norwich - North Walsham road. If Norfolk Police / Highways had done their jobs and actually put a diversion in place, nobody would have been directed along this road.

 

[Dai Corner]

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

Sat Navs

 

[MotCO]

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked

Probably directed there by satnav.

Would the signalman realise that there was an abnormally high number of users that day? If so could they tell trains to slow to walking pace in that area until clear of the crossing?

 

[Wolfie]

Probably directed there by satnav.

Would the signalman realise that there was an abnormally high number of users that day? If so could they tell trains to slow to walking pace in that area until clear of the crossing?

Probably not surely, unless most of the additional users did actually phone.

 

[edwin_m]

Probably not surely, unless most of the additional users did actually phone.

If one of the first drivers had phoned in and said there were many cars waiting, then I guess the signaller might have cautioned trains.

 

[Class 170101]

What railway enhancement would you cancel to find the several million £ for an MCB-OD?

No scheme cancelled if its too expensive to install MCB OD, just close the crossing then.

 

[edwin_m]

No scheme cancelled if its too expensive to install MCB OD, just close the crossing then.

The report suggests closure would involve building a new section of road, which also costs money.

 

[Class 170101]

The report suggests closure would involve building a new section of road, which also costs money.

Why? Cause inconvenience by a closure or two and perhaps the misuse would stop!

 

[dosxuk]

It doesn't actually sound like the crossing was being misused by its normal users - more that the level of traffic experienced on that day was completely outside of the design standards - and that traffic was by general road users who aren't going to be used to that style of crossing. The report focuses not on the misbehaviour of the motorists, but on the inadequate signage and unsuitability of the setup for that level of traffic.

Closure is a big challenge as you have to get rid of the public right of way, which is far from easy. Building an additional bit of tarmac up to the Tunstead road crossing slightly north would get around that particular issue.

 

[Adoarable]

Probably directed there by satnav.

Would the signalman realise that there was an abnormally high number of users that day? If so could they tell trains to slow to walking pace in that area until clear of the crossing?

According to the report, the last telephone call from that crossing was at 15:54. The accident that caused the road closure happened at about 16:45. The next call the signaller received from the crossing was at 17:24, to report the near miss that had just occurred.

 

[DerekC]

One is struck in the report by the enormous amount of back-and-forth passing of the issue between different bodies, while in fact many years passed. There seems no ability for someone to just say "Do this". The lack of clarity between the different bodies, who seem only too ready to say (eventually) "don't do this", but quite unprepared to say "do that instead".

That's the point I was getting in the OP, really. The change of standards philosophy from prescriptive to process, risk assessment based makes it much easier for the ball to be batted backwards and forwards and nothing to be done.

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

Lets not forget that the crossing has been used for many years without incident. The four users a day were no doubt well aware of the risks and correct procedure and the near miss only occurred because non-locals were trying to avoid congestion elsewhere.

A more proportional response if a similar situation occurs might be to send someone to supervise the crossing, or close the road temporarily.

Isn't the point that any public road that is a link in the network (i.e. not a cul de sac) can suddenly become more highly used when an adjacent road is closed temporarily, and then unfamiliar users will certainly be crossing. The RAIB report notes (although doesn't comment on) the fact that the Highway Authority took no action at all, and it would be unlikely to occur to police that the use of the lane as a diversionary route would be high risk.

An option for this might be to ban vehicular traffic from the road by a Traffic Regulation Order, except for access to adjacent property. That is done sometimes where a road has an unmade section unsuitable for through traffic, and I don't see why that shouldn't apply to this situation.

 

[dk1]

There have been occasional incidents at this crossing. I however have been driving trains for 24 years around that bend where the crossing is, 21 of those years at 75mph & have never had a problem & although it is reasonably well used everybody has been waiting patiently after phoning the signaller for permission. I reckon I could count on one hand the times I’ve ever been cautioned for it & thats only been due to telephone failure.

 

[edwin_m]

That's the point I was getting in the OP, really. The change of standards philosophy from prescriptive to process, risk assessment based makes it much easier for the ball to be batted backwards and forwards and nothing to be done.

The reason it wasn't given miniature stop lights looks to be entirely due to a prescriptive standard.

 

[Taunton]

The reason it wasn't given miniature stop lights looks to be entirely due to a prescriptive standard.

I wonder where this perception that you cannot have miniature lights on a public road arose. They are quite clearly stipulated in the Highway Code, rule 296 in my edition:

296
User-operated gates or barriers. Some crossings have ‘Stop’ signs and small red and green lights. You MUST NOT cross when the red light is showing, only cross if the green light is on. If crossing with a vehicle, you should

• open the gates or barriers on both sides of the crossing
• check that the green light is still on and cross quickly
• close the gates or barriers when you are clear of the crossing.

Laws RTA 1988 sect 36 & TSRGD regs 10 & 52(2)

I would add here that the road users' STOP sign, the quite distinctive octagonal shape, is notably different from the board that the railway uses.

 

[Dr Hoo]

ISTR that when I did my level crossing training on the LM region in about 1975 that Barthomley LC on the Crewe-Kidsgrove line had MSLs and user worked gates on a public road.
I went to the crossing by bike as it happened.

 

[trebor79]

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

Google maps will do that if there is a road closure.
Google once sent me over a user worked crossing. I twigged later it had shaved half a mile off the journey by avoiding a round about and proper level crossing. I was not amused.
I confess I didn't know I was meant to call the signaller before crossing and didn't see the sign telling me to do so until I was shutting the gates.

 

[30907]

Google maps will do that if there is a road closure.

...and even if there isn't (he says, having recently been directed down an extremely narrow back lane in Leeds.).

 

[edwin_m]

I wonder where this perception that you cannot have miniature lights on a public road arose. They are quite clearly stipulated in the Highway Code, rule 296 in my edition:


I would add here that the road users' STOP sign, the quite distinctive octagonal shape, is notably different from the board that the railway uses.

Inspector's report into the Naas crossing collision in 1979: https://www.railwaysarchive.co.uk/documents/MoT_Naas1979.pdf

45. Doubts have been expressed in several quarters about the efficacy of MWL at vehicle crossings. I share those doubts for such crossings on public roads.

Also

47. Following acceptance of all the major recommendations of the Level Crossing Protection Report, revised Construction and Operation Requirements for Level Crossings were issued by the Department of Transport in January 1980. These included the requirements of a new system of protection called Automatic Open Crossing, Remotely Monitored (AOCR). AOCRs will have the standard steady amber and flashing twin-red road traffic light signals which are installed at other modernised level crossings and these will be activated automatically by an approaching train. In this latter respect they are similar to MWL but the lights are much larger and are duplicated on both sides of the road. The standard timecycle will be such that a train will not reach the crossing until the amber light has showed for 3 seconds followed by the flashing red lights for at least 24 seconds. As neither gates nor barriers will be associated with the traffic lights, the message passed to the approaching motorist will be unambiguous

That one worked out well! AOCRs were later phased out due to poor safety record.

I would add here that the road users' STOP sign, the quite distinctive octagonal shape, is notably different from the board that the railway uses.

There was a thread on here a week or so back about new signage for crossings, which get rid of the longstanding railway sign and include the octagonal STOP sign where applicable.

 

[alxndr]

For an "unused and unknown" road, it is surprising how many knew to use it when the main road was blocked.

Having looked at the area on Google maps and Streetview I'd probably take that road if I came across a sudden closure and didn't know the area well enough to know the quirk of that road. Typically I'll try and "re-route" myself first before pulling over and having a look at a map or setting a sat nav.

If you took the first turning either side of the crossing and then the first heading in the general direction of the original direction you'll end up on that road. It's a natural way to try and divert yourself. You might think "oh, this is a bit tight" when coming from the south-west but I certainly wouldn't feel as though there was anything particularly unusual until I got to the crossing.

 

[eldomtom2]

It seems the RAIB are increasingly coming to the conclusion that diagram 103 signs are not fit for purpose.