Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!
Counting fatalities is always going to produce statistical outliers. I recall that the solitary fatality at Grayrigg was a heart attack, not a physical injury.
For injuries and fatalities to train crew, is is in one sense immaterial whether the train was in passenger service or not.
Whilst safety measures have undoubtedly imnproved, both in signalling and design of rail vehicles, trains run faster and more frequently than they did 100 years ago, and this should be taken into account in comparing statistics. It is no coincidence that most collisions happen on the busiest lines.
Although there you have the problem of a fairly small population of people that ever flew Concorde, so that one crash with the loss of all passengers and crew really distorts the statistics.
But I wonder if we start to reach a point whether the law of diminishing returns kicks in, and the tough question of how much it costs and is it worth it kicks in?
As said somewhere further up the topic, before any figures approach being useful, you need a whole bunch of definitions.
Statistics can only show you the past, they say nothing about the current or the future. It's unwise to make forward projections IMHO. There are too many factors at play including an element of randomness. Hence be careful saying that railways are safer today.
Counting only fatalities is also not helpful. Ideally injuries and incidents should also be included, but the trouble is, data is not available for all of these.
For example, an allegation may be made against the signalling system or a fault on the signalling system may be reported. The railway then conducts an internal investigation. The outcome is not published publicly, but a serious issue may have been found. That in slightly different circumstances may have resulted in a train crash. I once found a serious error in a circuit for a signal on a passenger line. And I saw an alteration to a signal control circuit that did result in an incorrect signal aspect sequence. Neither caused any accident but if the circumstances were different, who knows what may have happened.
My point is, statistics are only as good as the underlying data, and we don't have enough good enough data.
Some of the posts that I have quoted make important points.
Are you including or excluding railway employees of contractors working on the railway? What about emergency services operating on the railway? E.g.the fire service being lineside when fighting an lineside embankment fire? Or the police chasing someone?
For example, I worked on the railway for over 35 years. No one got hit by a train, but there were some "near misses" ( I hate that phrase). One employee in the department where I worked did die on duty, but he was in the depot at the time and died of natural causes.
In one of the near misses, on a "slow speed" branch line, I myself came close to being hit by a DMU. Another was finding a 360 ( "JCB" type machine) getting far too close to me, which I was unaware of until my team member alerted me. This was in an engineering worksite.
And although safety systems, the design and build of equipment, training and procedures and safety culture has in general improved, we may be going backwards in some areas (e.g. due to budget limits, maintenance is occurring less often than traditionally and in parts of the maintenance organisation, there are staffing issues including insufficient skills and experience).
The way the railway struggles with vegetation is one maintenance area that can be seen.
And that is what could in theory happen on the railway tomorrow, unlikely to kill anywhere near that number of people, but certainly a big enough crash to seriously affect the statistics. Imagine if a heavy fully laden freight train went into the back of say a BR vintage multiple unit at line speed...
I think Talerddig is unnerving because it shows that just focusing on signalling and driver assistance is not enough to prevent fatalities. Hopefully one silver lining is that network rail might now specify the additional feature in future ETCS rollouts that is missing from Cambrian and may have helped the situation. We also need to be investing far more in vegetation management too as ultimately aside from level crossings I think weather events are going to be the biggest issue going forwards.
The only things that stop a moving train are it's brakes working (including the wheel/rail interface) or it hitting something else... In this case, the brakes were working but the automatic sander wasn't.
Counting fatalities is always going to produce statistical outliers. I recall that the solitary fatality at Grayrigg was a heart attack, not a physical injury.
For injuries and fatalities to train crew, is is in one sense immaterial whether the train was in passenger service or not.
Whilst safety measures have undoubtedly imnproved, both in signalling and design of rail vehicles, trains run faster and more frequently than they did 100 years ago, and this should be taken into account in comparing statistics. It is no coincidence that most collisions happen on the busiest lines.
Today's trains are different from those of 50 years ago. Many lines are more busy. Lots of things have changed. We think they have got better, but at some point in the future, something may happen that finds a weakness and If the conditions allow, an incident or accident may occur.
== Doublepost prevention - post automatically merged: ==
For which aspect of safety? Does Network Rails "Everyone Home Safe Every Day" only apply to railway employees?
Is the national railway system in this country (the U.K.) a safe system to use when compared to other modes of transport, yes. Is it as safe as it can reasonably be? That depends on how you define reasonably safe.
Edited to correct typos and words changed by autocorrect.
There is a well-known concept of the "tyranny of small numbers" - that small sample sizes can throw up erratic and largely meaningless outcomes, evidenced already in this thread by the Concorde or Monarch examples.
In the context of railway accidents LTC Rolt picked up something similar in pointing out that some locations recorded multiple accidents, but purely by chance rather than by something inherent to the location.
If you are including Salisbury in your list of fatal train crashes, then you need to add the 150/HST collision at Plymouth in April 2016 and the 800/HST collision at Neville Hill in November 2019 - both of which also had 0 fatalities.
Having chewed over my earlier attempt to answer one of the OP's questions I need to correct my opening sentence which stated "such small samples". This is wrong! I should have stated "such small data". The samples are huge, not simply 10 years but 3652 days or ....minutes or ... seconds - the time is irrelevant and if we converted 6.5 years to the same days/minutes/seconds we would end up using the same Po(1.95) distribution. I'm rusty!
I continue to agree that there are better measures to use over the periods rather than their relative lengths: train journeys, passenger journeys, train kilometres, passenger kilometres, .... But I stand by my conclusion - the increase in fatalities wrt the time periods alone is not significant (as there would need to be 5 or more fatalities in the 6.5 years of the 20s, at the 95% confidence level).
Finally, Disraeli's quote does not mean that responsibly calculated statistics should be automatically dismissed. What he highlighted is the need to be clear and responsible with respect to what you claim you are determining, and readers to be alert to abuses. And many people are not.
The only things that stop a moving train are it's brakes working (including the wheel/rail interface) or it hitting something else... In this case, the brakes were working but the automatic sander wasn't.
In this case (Talerddig) there was also a one-shot emergency sander in working order that was not used IIRC. Which makes it ultimately a human factors accident not an engineering-related one.
In this case (Talerddig) there was also a one-shot emergency sander in working order that was not used IIRC. Which makes it ultimately a human factors accident not an engineering-related one.
In this case (Talerddig) there was also a one-shot emergency sander in working order that was not used IIRC. Which makes it ultimately a human factors accident not an engineering-related one.
Or it could be asked why, on a train with a system (ETCS) that had already determined the train wasn’t meeting the required braking curve and so was going to exceed its movement authority, there was no automatic discharge of the one shot sander nor alert to deploy it?
I assume there was also no indication to the driver that the automatic sander was not operating? Something which might also have alerted the driver to deploy the one shot sander?
Human factors indeed, but from an engineering shortfall.
For which aspect of safety? Does Network Rails "Everyone Home Safe Every Day" only apply to railway employees?
Is the national railway system in this country (the U.K.) a safe system to use when compared to other modes of transport, yes. Is it as safe as it can reasonably be? That depends on how you define reasonably safe.
1. I would suggest nearly all, when considered in context rather than isolation (ie the safest railway is one with no trains, but that means diverting people onto more dangerous modes of transport) As no-one’s life is spent 100% on the railway, we may have reached the point where some of the money spent chasing that final 1% on the railway would be better spent elsewhere.
2. No but see 1 - with the caveat that as employees are at greater risk money and time spent on their safety is more cost effective (sorry to make it sound so heartless)
3. If, as has been mentioned above the most dangerous part of a rail journey is getting to and from the station, would it not be better to funnel money into that rather than lowering everyone’s chances of being involved in a railway accident from 1 in 200 years to 1 in 200 years and 25 minutes? Yes - we should continue to invest in safety on the railway but we need to push back against the populist “Something must be done. This is something so we will do it and the more noughts on the end of what it costs the better it must be”.
I wonder what % of serious incidents on the railway (and elsewhere) have as their root cause someone not doing what they’re told. Solution costs no money at all - just (self-)discipline.
In this case (Talerddig) there was also a one-shot emergency sander in working order that was not used IIRC. Which makes it ultimately a human factors accident not an engineering-related one.
It was actually both - if the automatic sander had worked (or various other things had happened differently) the train would have stopped without needing to resort to the one-shot sander.
1. I would suggest nearly all, when considered in context rather than isolation (ie the safest railway is one with no trains, but that means diverting people onto more dangerous modes of transport) As no-one’s life is spent 100% on the railway, we may have reached the point where some of the money spent chasing that final 1% on the railway would be better spent elsewhere.
2. No but see 1 - with the caveat that as employees are at greater risk money and time spent on their safety is more cost effective (sorry to make it sound so heartless)
3. If, as has been mentioned above the most dangerous part of a rail journey is getting to and from the station, would it not be better to funnel money into that rather than lowering everyone’s chances of being involved in a railway accident from 1 in 200 years to 1 in 200 years and 25 minutes? Yes - we should continue to invest in safety on the railway but we need to push back against the populist “Something must be done. This is something so we will do it and the more noughts on the end of what it costs the better it must be”.
I wonder what % of serious incidents on the railway (and elsewhere) have as their root cause someone not doing what they’re told. Solution costs no money at all - just (self-)discipline.
With the above as an example, most rail accidents are the result of a combination of events, any one of which if different would have prevented it. The best known analogy is probably Dr Reason's Swiss Cheese model, where each slice represents an event and the holes have to line up for an accident. RAIB also shows this by taking the causes separately in its reports. Hence, it's rarely possible to say an accident has a single root cause. This may also be one reason railways are much safer than roads, where the sequence of events leading to an accident is often much simpler and may just be a driver error.
There's also a suggested hierarchy of protection measures known by acronyms such as ERIC - Eliminate Reduce Isolate Control. Basically it says that when reasonably practicable to do so it's best to eliminate the hazard by designing it out in engineered systems, rather than just to control it by setting operating rules.
I wonder what % of serious incidents on the railway (and elsewhere) have as their root cause someone not doing what they’re told. Solution costs no money at all - just (self-)discipline.
I'm not an expert but I thought that the current view on human factors is that it isn't always as simple as that.
As for "Everyone Home Safe Every Day", it's a nice slogan but it can't mean that money is no object when improving safety for staff any more than it is for passengers.
Absolutely. But isn't that why nowadays we consider things like this to be human factor issues rather than just blaming the driver or pilot for making an error?
I'm sure there are cleverer statisticians on this forum than me. Can the increase in fatal train crashes in recent years be considered statically significant?
Your data shows 4 fatal accidents in 24 years 2003-26.
Suppose the hypothesis that fatal accidents are random events, with a 1 in 6 chance in any given year. That's like throwing a 6 with a 6 sided die. This is a hypothesis for statistical demonstration, I'm not suggesting it as a way of modelling fatal accidents.
Now imagine that you are throwing a die, and get this sequence, where * is "not 6".
****6************6***6*6
Anyone who has ever played any dice game, where throwing a 6 matters, will know that sometimes it takes ages to get a 6, and sometimes they seem to just keep on coming.
There's nothing to suggest that this sequence is not random.
The latter point is one for debate. Clearly the railway is safer than it was in the 80s or 90s. But is it safer than it was in the 2010s? One could argue that it isn’t simply because there’s been several accidents since 2020 that could reasonably be defined as serious, whereas there weren’t for a near 15-year period prior to that, although including Croydon alters that.
From a perception point of view, I suspect a lot will depend on what arises from the report into the Bedford collision. If it transpires it was a relatively inexplicable error from a driver with an unblemished record then it will likely just be a case of looking at whether the TPWS risk criteria could benefit from being altered. If however it turns up a pile of mess like we saw at Southall or Ladbroke Grove, especially in relation to driver training/supervision/assurance, it’s reasonable to suggest that things could have started to slip.
I will be interested to know what the report says about that last sentence. In my opinion training quality has certainly diminished in the last decade.
While we can say things such as ‘the railway is on an upward safety trend’ or ‘constant improvements are being made’ and we can see statistics seemingly backing that up.
I think there may be small things lurking in the background that suggest all is not well and perhaps things aren’t constantly improving but starting to degrade but not at a rate that we’ve noticed yet.
I have no knowledge of statistics so my post will have nothing scientific to back it.
But as an example, something I was once told by a manager. TOCs view minor station overruns as safety incidents despite no-one being hurt because they could be a precursor to more serious incidents.
The idea being that many platforms have signals or buffers at the end so misjudging a stop could be a warning of something serious to come.
My point is that it is possible that the number of small overruns (eg one door not platformed) could rise significantly over a period while SPADs could lower.
Would that mean the railway is getting safer or not? SPADs are the more serious incident and more likely to cause death. And the overruns would probably not injure anyone. So we could say SPADs have lowered and the railway is therefore safer.
But if the accepted wisdom is overruns are a precursor is it actually safer? And while the overrun statistics would be clear and available. Would the industry actually have any urgency to act? I’m sure there would be briefings issued and human factors research undertaken but I doubt an acceptance the railway was on a downward trend would come unless the bigger incidents down the road from these ‘precursors’ began to rise.
While high profile accidents may be a statistical anomaly and could be coincidence, there are things in the background that I’m seeing currently that concern me personally. And I mean multiple things, and multiple things I’ve not seen before in my 20+ yr career.
But my concerns would be neither scientific nor am I qualified to officially make them.
While we can say things such as ‘the railway is on an upward safety trend’ or ‘constant improvements are being made’ and we can see statistics seemingly backing that up.
I think there may be small things lurking in the background that suggest all is not well and perhaps things aren’t constantly improving but starting to degrade but not at a rate that we’ve noticed yet.
I have no knowledge of statistics so my post will have nothing scientific to back it.
But as an example, something I was once told by a manager. TOCs view minor station overruns as safety incidents despite no-one being hurt because they could be a precursor to more serious incidents.
The idea being that many platforms have signals or buffers at the end so misjudging a stop could be a warning of something serious to come.
My point is that it is possible that the number of small overruns (eg one door not platformed) could rise significantly over a period while SPADs could lower.
Would that mean the railway is getting safer or not? SPADs are the more serious incident and more likely to cause death. And the overruns would probably not injure anyone. So we could say SPADs have lowered and the railway is therefore safer.
But if the accepted wisdom is overruns are a precursor is it actually safer? And while the overrun statistics would be clear and available. Would the industry actually have any urgency to act? I’m sure there would be briefings issued and human factors research undertaken but I doubt an acceptance the railway was on a downward trend would come unless the bigger incidents down the road from these ‘precursors’ began to rise.
While high profile accidents may be a statistical anomaly and could be coincidence, there are things in the background that I’m seeing currently that concern me personally. And I mean multiple things, and multiple things I’ve not seen before in my 20+ yr career.
But my concerns would be neither scientific nor am I qualified to officially make them.
I'd say there is a good argument to be made on human factors grounds (as well as more efficient operations) that station overruns should not be in the same way as SPADs. Training drivers to use a different and more conservative braking technique when approaching a platform on a single yellow might make it less likely that the driver forgets that is what they are doing.
The thing is that approaching on a yellow means that TPWS grids (where provided) are armed so some degree of caution over normal is required but I see your point in that a driver could then in their mindset be approaching a “red” rather than “a station” and if that red steps up they could power up whereas if they were just concentrating on hitting the stop marker regardless of the aspect they’d make the stop and all would be well.
You also make a good point about human factors and mandating certain things in certain situations can lead to drivers doing them because they could be picked up for not doing so rather than consciously thinking about what they are doing and why.
One of my bugbears is that post incident mitigation sometimes brings in rules that we are then bound by despite potentially as an individual finding them off putting after years of doing things another way.
So you start following the new process but miss something minor because you used to have a completely different system of your own but are now focussed on doing things the new mandated way.
Since fatal accidents are the tip of a large iceberg, there's a strong case for analysing all the reportable incidents to see whether there are any categories where there are adverse trends.
As an observer of the scene, by no means an expert, two sub categories strike me as worrying.
Misinterpretation of signals---- Grantham, Peterboro, Bedford, Ilkeston, Doncaster Decoy etc--- are these on the increase and is there any commonality of underlying cause?
Infrastructure failures related to weather--- Carmont, Salisbury, Talerddig--- is environmental risk increasing and if so, then what?
So you start following the new process but miss something minor because you used to have a completely different system of your own but are now focussed on doing things the new mandated way.
That is the kind of thing that concerns me as well.
Some of the things I’m seeing that concern me…..
-Local Multi Spad list is the highest it’s been in my career.
-Most Spad bulletins seem to be “first recorded spad at this signal.
-A lot of other signals that do have previous spad history returning to being spaded after many years or even decades without.
(It appears records go back to c1985. I’m seeing things “this is the 4th recorded spad at this signal. The others were 1989/1991 and 1993” Then a 30 year gap and back to an incident).
-I saw a signal go from no recorded history to multi-spad in 1 month. Never seen that before in over 20 years.
On specific incidents.
-The Spital overspeeds. Two carbon copy incidents at the same signal in 13 months
Now I get that lightning can strike twice but that signal has likely been there for decades.
In the 2000s We had an incident with a driver who had a spad at a high risk signal on his second solo turn and his first at that location.
That incident is so ingrained into the depot memory that even today new starters know of how “Joe Bloggs” did that and they must be careful there. Even though he retired a decade or more ago and they never knew him.
So it concerns me. After the first Spital incident every driver on that route would be hyper aware of that incident and signal. Yet it happened again. I’m not judging the driver. Just saying that in my career I haven’t seen carbon copy incidents so close together.
-We recently had a local incident. Driver approaches junction on yellow. Normally would be less restrictive. Controls train as he usually would. Realises he’s approaching red too fast, throws brake in, no grids tripped, no SPAD but he missed AWS magnet. That was enough to flag it up. Conflicting move had just cleared junction.
I was talking to a guard who tells me a story that happened on his train recently. He basically describes the same incident, but this time the driver cancelled the AWS so it didn’t get officially flagged.
Was that just coincidence? Was Spital? How many things are happening that don’t make the stats but we get ‘lucky’.
I have been a passenger on two trains in the last year that clearly hit curves above line speed and the driver made a heavy brake application. One spilt coffee everywhere. Before that I went over 20 years without ever feeling a train wasn’t under full control on routes I sign. Was I previously lucky. Was my recent experience a blip?
But overspeeds are an accepted industry issue atm. There have been some massive ones. And I genuinely don’t remember being briefed on any of such magnitude in the rest of my career. It’s regular now.
We always had derailments but many were land slips or points/infrastructire.
We’ve now had 3 high profile collisions. Two involved low adhesion. The other is under investigation.
This may be coincidence or bad luck.
But on the first two. While the RAIB does an excellent job of reports I wonder how we got lucky for decades before. After all autumn happens every year.
I’m not trying to be flippant, offend or otherwise. But I think things are worse. But haven’t translated into being noticeably worse.
And when it comes to the reasons given. One is often “the railway is busier”. This may be statistically true. We have loads of passenger trains running on tight headways on fairly predictable patterns. But we don’t have all the slow steel or coal trains or various shunting moves. If anything given many SPADs traditionally were ground signals there is an argument the multi Spad list should be dropping.
My feeling is there is something in wider society. I read aviation forums and anecdotally there have been a number of high profile incidents there. But the stats could still say things are great.
I once saw a post on a forum (purely unscientific I know). It was an air traffic controller asking if anyone had noticed more mistakes in readbacks lately as he felt it was now much higher. Unfortunately it didn’t become a discussion as no one replied.
But do we focus on the cause of individual incidents while there is unnoticed rot in the system? You can ascribe a technical cause to any incident without asking why that technical cause never affected anyone in decades but now seems to affect multiple trains/aircraft within a short timeframe.
But again, I’m not qualified and only have feelings to go on.
Finally I ask this. If there was something fundamentally worse at a societal level or certainly a deeper level. Would it be flagged quickly particularly if it wouldn’t be politically expedient to do so and while the statistics were still good enough that we can hope it’s a blip and will right itself.
One factor influencing junction overspeeds is likely to be the much better acceleration of modern units such as the 80x when on electric power and the 700, compared with the types of trains approaching them a decade or two ago.
It's generally accepted (other than by "deniers") that climate change is leading to more extreme weather, with heavy rainfall directly causing floods and landslips but long dry spells also weakening Victorian earthworks. Strong winds may also bring down leaves and cause unexpected adhesion problems outside the traditional autumn season.
One thing that I completely unscientific believe is a large factor is that we build infrastructure to the absolute minimum that allows running a given set of services. We refuse to pay more to get spare capacity and everyone hating on HS2 and it being cut right back means we don't get the capacity boosts there.
All this does is lead to pushing the risk envelope eeking everything out of the system, drivers must be used to pushing speeds etc to keep to overly tight timetables. Then when something goes wrong they've got no headway because they've already been expected to push to the limit.
I've seen people say that delays were caused by a driver being 'overly cautious' in extreme weather, that culture of almost blame must surely be a big factor in all this.
Like said above the fact trains reach such high speeds so fast must surely contribute to some of these issues too. If risk assessments were done assuming a 158s acceleration and then suddenly an electric 800 comes along how often does that get re-assessed?
But I wonder if we start to reach a point whether the law of diminishing returns kicks in, and the tough question of how much it costs and is it worth it kicks in? Weren't Boeing the first company to put a price on human life, based on pay outs to bereaved families?
UK railways certainly have, and if you do have money to spend it could be spent far more effectively on safety in other areas of life. IDK about Boeing, but I worked in an industry with potentially serious danger to the public, and we had a price on life for the purpose of evaluating the cost/benefit of safety measures. I attended meetings with the Health and Safety Executive where these things were discussed. I am not going to state the value (which was not based on payouts) because any value will upset some people and it will have changed by now anyway.
But that does not mean we should not investigate accidents thoroughly and look for (and evaluate) safety improvements.
But just to show the potential absurdity (a reductio ad absurdum argument), you could increase railway safety (even though very slightly) by having two signal posts where there are currently one, in case one fell over. Or why not three in case two fell over? Or a hundred in case 99 fell over? You have to draw the line at throwing money at some point.
But just to show the potential absurdity (a reductio ad absurdum argument), you could increase railway safety (even though very slightly) by having two signal posts where there are currently one, in case one fell over. Or why not three in case two fell over? Or a hundred in case 99 fell over? You have to draw the line at throwing money at some point.
Absolutely, and deciding where to draw the line can be rather tricky if you can't put a value on lives saved even if the idea seems morally repugnant to some.
Well the railway definitely has a number they use, called the Value of preventing a fatality (VPF) the number in 2025 was £2,856,000 (happy to be corrected).
That is the kind of thing that concerns me as well.
Some of the things I’m seeing that concern me…..
-Local Multi Spad list is the highest it’s been in my career.
-Most Spad bulletins seem to be “first recorded spad at this signal.
-A lot of other signals that do have previous spad history returning to being spaded after many years or even decades without.
(It appears records go back to c1985. I’m seeing things “this is the 4th recorded spad at this signal. The others were 1989/1991 and 1993” Then a 30 year gap and back to an incident).
-I saw a signal go from no recorded history to multi-spad in 1 month. Never seen that before in over 20 years.
On specific incidents.
-The Spital overspeeds. Two carbon copy incidents at the same signal in 13 months
Now I get that lightning can strike twice but that signal has likely been there for decades.
In the 2000s We had an incident with a driver who had a spad at a high risk signal on his second solo turn and his first at that location.
That incident is so ingrained into the depot memory that even today new starters know of how “Joe Bloggs” did that and they must be careful there. Even though he retired a decade or more ago and they never knew him.
So it concerns me. After the first Spital incident every driver on that route would be hyper aware of that incident and signal. Yet it happened again. I’m not judging the driver. Just saying that in my career I haven’t seen carbon copy incidents so close together.
-We recently had a local incident. Driver approaches junction on yellow. Normally would be less restrictive. Controls train as he usually would. Realises he’s approaching red too fast, throws brake in, no grids tripped, no SPAD but he missed AWS magnet. That was enough to flag it up. Conflicting move had just cleared junction.
I was talking to a guard who tells me a story that happened on his train recently. He basically describes the same incident, but this time the driver cancelled the AWS so it didn’t get officially flagged.
Was that just coincidence? Was Spital? How many things are happening that don’t make the stats but we get ‘lucky’.
I have been a passenger on two trains in the last year that clearly hit curves above line speed and the driver made a heavy brake application. One spilt coffee everywhere. Before that I went over 20 years without ever feeling a train wasn’t under full control on routes I sign. Was I previously lucky. Was my recent experience a blip?
But overspeeds are an accepted industry issue atm. There have been some massive ones. And I genuinely don’t remember being briefed on any of such magnitude in the rest of my career. It’s regular now.
We always had derailments but many were land slips or points/infrastructire.
We’ve now had 3 high profile collisions. Two involved low adhesion. The other is under investigation.
This may be coincidence or bad luck.
But on the first two. While the RAIB does an excellent job of reports I wonder how we got lucky for decades before. After all autumn happens every year.
I’m not trying to be flippant, offend or otherwise. But I think things are worse. But haven’t translated into being noticeably worse.
And when it comes to the reasons given. One is often “the railway is busier”. This may be statistically true. We have loads of passenger trains running on tight headways on fairly predictable patterns. But we don’t have all the slow steel or coal trains or various shunting moves. If anything given many SPADs traditionally were ground signals there is an argument the multi Spad list should be dropping.
My feeling is there is something in wider society. I read aviation forums and anecdotally there have been a number of high profile incidents there. But the stats could still say things are great.
I once saw a post on a forum (purely unscientific I know). It was an air traffic controller asking if anyone had noticed more mistakes in readbacks lately as he felt it was now much higher. Unfortunately it didn’t become a discussion as no one replied.
But do we focus on the cause of individual incidents while there is unnoticed rot in the system? You can ascribe a technical cause to any incident without asking why that technical cause never affected anyone in decades but now seems to affect multiple trains/aircraft within a short timeframe.
But again, I’m not qualified and only have feelings to go on.
Finally I ask this. If there was something fundamentally worse at a societal level or certainly a deeper level. Would it be flagged quickly particularly if it wouldn’t be politically expedient to do so and while the statistics were still good enough that we can hope it’s a blip and will right itself.
Especially the point that it's what we don't know about that is the real risk factor. Things that happen in the real world outside our picture defined by recorded statistics.
Well the railway definitely has a number they use, called the Value of preventing a fatality (VPF) the number in 2025 was £2,856,000 (happy to be corrected).
I remember reading a while ago (which does of course not mean that it's true even if I have remembered correctly) that the value was set higher for rail than road on the grounds that society views rail deaths as less acceptable than those on roads.
Of course on the roads there are a very large number of people making safety decisions every day and they are not generally doing so in a formal fashion.
We are aware of an issue with emails from the Forum to Microsoft-based email accounts (hotmail/outlook/live.com email addresses). This is being looked into currently, thanks for your patience meanwhile.
