The awful incident at North Melbourne last week highlighted the need to continually improve rail safety. It may have been a freak accident, but one with huge consequences for the individual involved and all who witnessed it, as well as flow on delays to thousands of other passengers.
The exact circumstances will need to be investigated, and that will help determine causes and measures that could help prevent similar future occurrences.
But meanwhile, some have been asking about whether Platform Screen Doors could help. They prevent interactions between people on the platform and trains and the track, except when the train is ready for loading.
We’ve got them at the new Metro tunnel stations, so many Melburnians have now seen how they work. They’re also common around Europe and Asia, and Sydney has them on their Metro.
Platform Screen Doors are not cheap or easy to install. And until recently in Melbourne there was a blocker: the PSDs need to match the train doors, so they need train fleets with consistent door positioning.
On an old legacy rail network this is difficult, because of a mixed train fleet. But increasingly the fleet is being standardised on some lines.
- Sunbury, Cranbourne, Pakenham (Metro Tunnel line) – all use the HCMT fleet – though a few stations share platforms with V/Line trains
- Mernda, Hurstbridge, Lilydale, Belgrave, Alamein, Glen Waverley – all use the X’Trapolis (original) trains
At present the other lines (including those that run via North Melbourne) run a mix, but the introduction of the new X’Trapolis 2.0 trains will see that change that over time.
Some cities have retrofitted PSDs, with Singapore being a prime example, but also Seoul, and Paris into some of their stations.
Obviously I snapped this pic at a quiet time!
Could Melbourne start by retrofitting PSDs to the stations served by only one type of train? It’s not a simple job – the PSDs also depend on signalling and fleet changes to allow trains to stop accurately.
There’s a version used in some cities involving ropes or nets strung along the platform that move out of the way when the train has stopped. This also requires less precision but provides less protection. Other technologies are also in development.
Meanwhile, some networks have installed simple fixed fencing, which provides partial protection. It’s very low tech, but would help, as it reduces the areas of platform without any kind of barrier to the track, and likely discourages people from walking along the edge of the platform – at the cost of some platform capacity.
Something like this is worth considering in the short term.
As for full blown PSDs, they could take decades, but given the huge safety benefits they bring, it’s definitely worth having a plan to progressively deploy them, as stations, train fleets, signalling systems all get upgraded over time.
Also this week:
- Today it’s 20 years since the Meeting Our Transport Challenges plan was released. Melbourne On Transit marks it in this post, and you can also read mine from ten years ago. It made a big difference to “safety net” basic bus services (particularly evenings and weekends) but less so to trams and trains.
- The Age reports strong patronage growth during the free period (paywall). How much of this is due to petrol prices vs free fares is still unknown (a bit from each?), and whether the growth will be sustained, and whether we’ll see extra services/capacity to help.
- PTUA members can attend a member meeting with Public Transport Minister Gabrielle Williams on 28th May. Details and registration here. You need to be a member to attend; you can join here.
21 replies on “Platform Screen Doors”
The incident was just another case of sheer stupidity, and we have the parents blaming the system again. The only way that could have occurred, was if the child was on the edge of the platform, facing backwards to the train. There was no other way a backpack could have got caught on the train mirror.
I can’t see retrofitting ever working in Melbourne, due to vandalism, just look at the recent spate of glass bus shelters destroyed. The City Loop would be only viable place, if they added complete walls, but too many train variations there.
I think the rope-based barriers would be ideal for high density stations (the loop, North Melbourne, Richmond, Caulfield, Footscray etc). I’ve seen them working in Tokyo and they seem like a no-brainer.
If somebody’s determined to get around them, they will. But these would prevent a multitude of potential accidents such as the North Melbourne one the other week, strollers rolling off platforms etc.
They could probably start with Footscray / West Footscray, as those platforms are covered by CBTC and only have the HCMT trains stopping there.
I did ask reddit about Malvern -> Westall but apparently even though V/Line trains don’t stop at those stations they still could in an emergency but I’m not completely sure if that’s the case (the other panels push out from track side in an emergency)
Does Melbourne have the yellow line and frequent announcements to stand behind the yellow line like Sydney does? It helps, though still not a physical barrier.
Retrofitting PSDs is a desirable long-term goal, but will be eye-wateringly expensive.
In the meantime, they could lower the risk of some accidents by tackling the task of aligning platforms with train doors, both horizontally and vertically. That would be expensive too, but not as much as PSDs.
At some stations (e.g. Essendon), you have a huge step down from the train, making the station difficult or impossible for disabled people. I can still manage it – just – but if I lose any more mobility, I’ll have to cross Essendon off the list of stations I can use, because I don’t want to risk falling and breaking a bone.
At other stations, there is a large horizontal gap between platform and train. I was on a train earlier this year on the Mernda line. I forget exactly which station – West Richmond, maybe? I was sitting within sight of the door, and I watched a mother pushing a baby in a pram and coaxing her pre-schooler over the large gap. I wasn’t close enough and I’m not able-bodied enough to lift the child over the gap myself, and I had my heart in my mouth watching him board the train, worried he’d fall into the gap and disappear underneath the train.
@Travis: we don’t know enough yet about the circumstances of the incident to know whether it was stupidity or not. Certainly, it is unwise to stand near the platform’s edge at any time, and especially with your back to the tracks.
But it was school let-out time, and Platform 6 is a single platform with a wall behind it. I’ve been at North Melbourne at that time of day. Platform 6 can get VERY crowded. It could be that the platform was so crowded, the boy was progressively forced closer and closer to the edge. Until the full details of this incident are known and made public, I think we should refrain from judging the boy.
Perhaps we could use an education campaign to encourage people with backpacks to take them off and hold them in front when in risky situations like this. It would also lower the risk of seated passengers being smacked in the face with someone’s backpack, which has happened to me many times on trains and trams.
Google image search Melbourne Train Platform versus Sydney, most of the Sydney photos have “Mind the Gap” and “Stand Behind the Yellow Line” whereas Melbourne usually doesn’t.
Something to address? Paint is much cheaper than doors and we won’t be waiting 30 years.
@Karen: The last report from ABC specifically mentioned the bag was caught on the mirror, and I thought they still had the warnings to stay behind the yellow line. I think in general school children just tend to get too close, as they crowd and push in, as soon as the train doors open, at least locally.
We desperately need an accessibility upgrade program for older train stations, even if it is just for first carriage (not just first door, see below). One of my local train stations used to have the very large gap, and I would see that extreme danger with prams too. The other local train station has level platform for the first door / carriage only, but the train drivers sometimes fail to align correctly, so you get an unexpected gap, which can be even worse.
Being against platform screen doors is like being against kittens.
But the risk is demonstrably very, very, low. And the cost to retrofit is very high. It’s notable that many installations exist because the system is driverless (higher risk), new (lower cost), the platforms can get very crowded (again higher risk), or they’re driven by non-safety issues (e.g. airconditioning costs or security).
This is why a “not so safe” system of platform screens may be a much more effective solution. The risk is already low, so you don’t need a super effective solution to reduce the risk to essentially zero. Note, however, this might not address other drivers (security, airconditioning) that might actually be of greater interest to the operator for the introduction of screen doors.
Perhaps I am the only one who has unexpectedly clouted someone on a crowded train because I didn’t realise how far my backpack extended back. (I take it off now :-) And I don’t wear a supersize schoolkids backpack.
Melbourne stations do have a yellow line on the platform. They also have intermitent messages to stand behind the line and to mind the gap, but they are not automatic as the are in Sydney. Frequent automatic announcements, of course, are a proven non-starter in getting people to pay attention to a safety issue.
The gap between the train and the platform has a number of issues, none of which are normally easily fixed. The difference in height is largely due to platforms constructed to old standards. Increasing the height is difficult because of the station buildings on the platforms and obstructions under the track. Excessive gaps between the train and the platform are normally due to curved platforms, again very difficult to fix. This is why they spend a lot of money when constructing new platforms to make them straight. (I’d note that Sydney platforms are generally more curved than Melbourne platforms and have a bigger problem here.)
@Travis, I’d have to say I don’t find that line of argument very helpful. It’s not about blame, it’s about cause and prevention.
People sometimes make mistakes, sometimes behave unpredictably, and circumstances/context is often a factor. Safe systems are designed to prevent or mitigate this. The same reason level crossings have bells plus warning signs plus boom barriers, and also a prime reason level crossings get removed.
And – again – variations of train designs is not an excuse, even in the City Loop. Two of four tunnels always use X’Trapolis trains.
@Dan, yes, coming up with something for the busiest stations would make a lot of sense.
@Michael, great point – Footscray in particular is a very busy station, and even if only platforms 1&2 were fitted, it’d bring safety benefits.
@Tony/@Andrew (1), here’s an example of yellow line, “Mind the gap” signage. Many stations also have tactiles for the yellow line. (This is Spotswood; they’ve probably refrained from installing tactiles because the whole station is set to be rebuilt soon.)
https://www.flickr.com/photos/danielbowen/55276622601/
@Karen, absolutely there are other things they should be doing in the meantime. I overheard a train driver recently saying they were very surprised Essendon hadn’t been fixed. Sounds like it’s one of the worst on the network.
@Andrew (2), it’s also apparent that a safer (not 100%) cheap easy system (eg basic static screens) that can be achieved quickly is a good option against one that is very difficult to implement and would take decades. Even if it’s only a stopgap.
Unfortunately the X’Trapolis 100 and the X’Trapolis 2.0 doors don’t line up. The XT2s have much more even door spacing, similar to the Comeng, whereas the end doors of the XT100s are much closer to the carriage ends (and therefore the end doors between carriages are closer together than the distance between the middle and end doors) so the door spacing is totally uneven. And of course, the Siemens Nexas has a different number of doors altogether…
The advantage of the XT2 doors is that it could be possible to run 7 car sets with the end doors of the train just on the platforms at Parliament, which currently have the shortest platforms on the network that can’t easily be extended.
@dbowen: No, we live in a society that lacks any personal responsibility more and more, the family just literally blamed the system, and demanded upgrades to the system. One of the reasons for the level crossing program was impatient people dashing through, but I still constantly see people do the same on the nearby road crossings, no matter how frequently they change (i.e. Mentone) on the newer builds. Add mobile phones, and we have people taking no notice at all of their surroundings (footpath signs galore around Mentone, due to schools). Add all the safety measures possible, but there is no stopping stupidity.
My problem is I take complete notice of my surrounds, and see these problems (impatience, and lack of spatial awareness) constantly getting worse and worse.
That seems like a colossal expenditure of taxpayer funds just to save one person every few years from the fatal consequences of their own stupidity.
While weโre at it, why donโt we force trains to crawl along at 5km/h at every level crossing and platform? Why donโt we fence off every metre of railway track in the state? Why not fit every locomotive nose with big bouncy balloons?
This is OUR money – vast quantities of it – youโre talking about. Letโs spend it on stuff thatโll benefit the sensible 99.9% rather than the foolish 0.1%.
I wonder about taking a holistic approach to transport safety, rather than just a modal one.
Money can be spent on either retrofitting and maintaining PSDs or on infrastructure and service upgrades that would get people out of cars and onto rails. I’m guessing the latter would see a much bigger safety return on investment.
@Travis/@Chris, even if you think this is down to individual behaviour, and have no sympathy for that person and their family, there is still value in finding ways to prevent these incidents – so other people don’t have to witness them, or be caught up in hours of disruption because of them.
Pursuing better safety doesn’t have to be unaffordably expensive. The goals of automatic closing/locking doors on trains and other public transport vehicles, for instance, were not achieved overnight, they took decades, but they improved safety.
Ditto on level crossings – even leaving aside their removal, the crossings themselves didn’t always have gates or boom barriers. That too took decades (and isn’t completed yet in some regional areas.)
Noting Dan’s good point about the safety benefits of mode shift, regular major disruptions are something that discourages people from switching from driving to PT. So how do we reduce or prevent them?
Great write up Daniel on PSDs and international examples that we can follow. I saw this parliamentary petition recently, but it did not get enough signatures to be presented: https://www.parliament.vic.gov.au/get-involved/petitions/safety-barriers-and-security-fencing-for-rail-infrastructure.
Like others have said, it will probably take more time before the public and state Gov are motivated enough to make further plans on PSD retrofitting.
@dbowen: We don’t even have an accessible transport system though, we need to make the whole system more usable, before considering spending any money on fail safe features like PSDs. A person with a pram at older train station doesn’t have any options, while a person has a choice whether to ignore danger warnings. Are prams even allowed to request the ramp? I can’t recall ever seeing anyone do that before. I still find it rather shocking, we are only just getting trams and tram stops near the Royal Show Grounds, that are actually accessible.
I still remember someone coming to school in the old days, warning us about the dangers of jumping out of the trains.
The biggest benefit of the level crossing removals, is the new modern stations, that have complete DDA access. The large train gap I mentioned earlier, was at Mordialloc, and was one of the worst (quote from train driver). I’m really lucky to use the Frankston line, with so many train stations replaced, but other areas aren’t so lucky.
It’s not just eye-wateringly expensive, PSDs are intrinsically linked to signalling performance. Even if you can agree to a door-spacing distance, you then need train stopping position that can reliably stop in the same location, +/- about 10cm. This is in contrast to current MTM platform standards (expecting +/-5m) and V/Line too (+/-10m). That’s Metres, not cm. This is why platforms are about 10-20m longer than the train and stop marks are about 5m from platform end.
This is why it took until the tunnel got CBTC before it also got PSDs. In fact, it wouldn’t surprise me if PSDs were a driving factor to include CBTC, and why the tunnel is completely driverless. The driving performance of our current fleet is sufficiently sloppy that there would be no hope of stopping accurately in place without making a line section functionally driverless.
And even if you went for door-agnostic ‘rope fences’, you then have to contend with our overly cautious touch potential standards, which are the driving force behind why we have so few platform shelters, and why they don’t actually provide any shelter to our platforms. Our network is Neurotic about touch potential to our overhead traction power system and won’t let any new structure get within 2m of the overheads; and also insulate any platform assets that a passenger could hypothetically touch even if its many metres from a hypothetical train. They’re terrified of a crowd forming a human chain from a poorly earthed train body to some stauncheon 3m from the coping. More chance for a snowball to get approved on the network than a structure at the platform’s edge that rises towards the overhead structure unfortunately.
the platform screen doors in melbourne look so ugly. For some reason they made the door have a large gap between the platform door and the train. No other place has a door this ugly. Every other system is modern and flat.
@Mike – the gap between platform door and train is because we’re running a legacy railway system where there needs to be a large clearance between the train and anything it could hit, hence the platform screen doors are set back some way from the edge of the platform, with the rubber panels attached to the back of each door to fill the gap between it and the train.
Give it another 25 years when the Siemens and X’Trapolis (100) trains are gone. Besides, the only stations that are even remotely ready for PSDs are the three underground City Loop stations, which still has the problem of having four different train types going through it as of 2026. PSDs in the loop will only be feasible once all the Comengs are gone and the Siemens’ are exclusively on the Cross City group. The rest of Melbourne is plagued with century-old low-height curved platforms with uneven terrain.
Spend the current PSD feasibility study money on actually improving today’s services and abolishing Sunday timetables throughout the entire state. Every 10 Minutes to Everywhere needs to make a comeback.
Be that annoying fly that politicians can’t swat, whenever they talk to the media on how awesome our trains and trams are, get in frame and show the media the 40-minute trains on the Sandringham line at 9AM on Sundays, the complete lack of almost any train whatsoever at 7AM on Good Friday, buses in Mooroolbark on both Saturdays and Sundays, trains at Bayswater at midday on a Tuesday, crush loaded half-hourly Burnley group trains after all of 7PM on the weekend (no expresses and no extra Alamein/Blackburn/Ringwood services from the city), crush loaded half-hourly Sunday evening trams even on the 86, 96 and 109 (the latter also running A class trams because “E is for Everyone” as long as you live in the inner north or acquired a $12 million mansion in Toorak).