Not rapid
The trials and tribulations of the BPLRT teach the rest of the rail network a lesson.
Rail reliability and maintenance is a complex topic. But that complexity should thus warrant close political observation, in order to ensure that the needs of public transport users are balanced against KPI-chasing and “over-maintenance”.
But this also extends to rail projects, how they’re run, and what kind of impacts they have on the passenger. Rail reliability is not just about avoiding incidents, it is about minimising inconveniences, planned or unplanned.
In this coming New Year, with plenty of projects closing in on the finishing line, let’s hope this lesson is well learnt.
Political pressure
With the LRT, Bukit Panjang is in fact already a 15 minute town. Walking to and from the station — five minutes. Wait for the LRT — up to five minutes. The actual LRT trip — up to five minutes within Bukit Panjang itself. However, all this has flown out the window, as we have now entered the fifth year of single track operations within the Bukit Panjang loop. I might even go as far as to say that we have now spent more time single-tracking, than benefiting from the 13 C801A vehicles in expanding system capacity.
And well, local MPs have to look like they’re doing something. May and June 2023 were some of the darkest days for the LRT. Apart from the reliability issues of the old vehicles, someone decided that close to the aftermath of the Cove LRT tragedy, all LRT trains had to slow to a near halt and crawl into LRT stations. While correlation does not imply causation, it seems strange to me that resumption of normal service tied in with the introduction of SMRT’s track intrusion video analytics system.
It also didn’t help that trains spent loads of time waiting at stations instead of moving. This resulted in a full round trip taking up to 40 minutes, a big difference from the usually scheduled 30-ish minutes, and potentially the cause of the additional waiting times. And it’s even slower than the originally promised 27 minutes back in the 1990s.
Looking like they’re doing something amidst all this may also mean asking for something else. Perhaps they have written off the LRT as a lost cause in dogmatically advocating for direct bus services? While it may be “counterintuitive” to go back to Bukit Timah/Woodlands Road for MRT service instead of driving out to the BKE via Bukit Panjang Road, the LRT makes this a faster option. Fixing the LRT also benefits people living in Fajar and Bangkit as well, and to visit amenities within other parts of Bukit Panjang and Choa Chu Kang, or for those going to Jurong and Tuas who can use other rail services from Choa Chu Kang. So why the difference?
Service 972 takes a tour of Bukit Panjang before it enters the expressway, providing residents of the entire town with direct bus access to Orchard Road. But those in Zhenghua ward have to sit through the Bukit Panjang SMC stretches before entering the highway — the time loss from this may well be the same as going to the MRT. So what happens is that the people queueing at Bukit Panjang ITH for 972 may only be using it as an alternative to 920/922, and aren’t going to town. All the more since outside of peak hours, LRT single tracking means it doesn’t operate in their favour.
There is value in fixing the LRT, since many more people in western Bukit Panjang and in Choa Chu Kang benefit from LRT improvements, compared to a minority nearer to the BKE who benefit from direct bus services. At least the LTA has announced that 2 more LRT vehicles will arrive in 2024; but to which I ask, what about the two that are already here? What about the remaining 15?
That it has taken so long to fix the LRT, prolonging the pain without providing interim alternatives, understandably drives people to look elsewhere. Perhaps this is why the MPs and residents think the way they do. At the very least, it can be possible to decouple some 2-car trains and operate a few 1-car trains at tighter headways to provide trains to restore 2-way loop service, but even that’s not done.
The new lines lesson
There may also be a need to think about how we deliver new lines and extension projects, especially since no more in Singapore history will we be opening an entire new line at one go. New lines will also have the same kind of investments as current lines, with things like CBTC Simulation Facilities and backup control centres becoming the standard and allowing for a level of off-line testing of new systems. With these investments in mind, is there a better alternative to modified service hours on weekends in order to perform system migration testing?
In Singapore, system migration to incorporate new extensions is chiefly done at night. It might explain that while early closures for the CCL and TEL have to be taken, the NEL’s Punggol Coast extension to come in 2024 doesn’t seem to require such closures — not yet, at least, because the line is short enough that everything can be done in engineering hours. And in fact, the early part of the Circle Line did not require such changes. This however changed with DTL3.
In addition to the usual three engineering hours per night to run tests, two more hours are required for these test runs as one full round-trip loop between Bukit Panjang and Expo stations will take about 2.5 hours. Beyond running test trains, additional time is also needed for preparatory works prior to commencement of each test, and readying the systems for passenger service.
There is nothing wrong with the phased approach. In fact, it may be better as the phased approach gives the systems time to stabilize before more stations and trains are added in. Both the JRL and CRL will officially have at least three phases; the CRL also the Punggol and Changi T5 extensions to come. And both lines also have their phases with the most stations and tracks open first, allowing comprehensive testing to be done across the entire line to improve stability and reliability, without inconveniencing people who’ve come to rely on it.
With the ability to do more offline, perhaps some of the risk assessments can be adjusted. The first instinct should not be “close the line early”. Perhaps it should be “avoid closing the line early”. In fact, this can refer to what was done during the re-signalling project of 2017–18, where system proving was also done during the last hours of the day. Instead of closing the line at 11.30pm, it may be possible to switch configurations to permit trains to enter the extensions at 11pm or so, maintaining service while allowing for more testing hours, before progressing to full-day operational trials.
This is also what the Putrajaya Line in KL did. Some trains were dedicated to Phase 2 testing (a practice also done in Singapore) and did not operate in passenger service before the opening of Phase 2. But otherwise they were tested in service hours, acting as if they were in-service trains, if not for barriers preventing entry. A full ride on the Putrajaya Line takes around 82 minutes — by the LTA’s SOP, they’d need to close the line to do testing of this scale. RapidKL avoided doing so because RapidKL wants people to take the train, even if with less people using the Putrajaya Line, it’s easier for them to get away with it.
It’s also interesting to note that the approaches between the DTL and TEL also differ. The DTL late opening back in mid-2017 was scheduled in such a way that everyone can be aboard a train by 7.35am, just five minutes after opening. The TEL late opening, not so much; first train services are similar to a regular service day, which means the first southbound train only arrives at Gardens by the Bay around 8.30am. Great for jogging, not for a morning run.
And it will be disaster for all of eastern Singapore if they try to do the same thing in 2025. The LTA must do its best to avoid what is likely to be major bustitution works, especially for the combination of both the DTL3 Extension and TEL5.
You can perhaps see the same mindset with CCL6, where single tracking will be implemented for four months. Would it make better sense to just close Telok Blangah station for a week or two, to complete all construction activities of the new turnout track in one go? Marina Bay got away with it because single platform operations to remove the access to the Marina Bay Maintenance Facility took only half a month and didn’t affect the service, but not this time.
But with the current plans, we can only just hope whatever needs to be done is finished ahead of schedule. Preferably before NUS’ exam period. For what it’s worth, the full closure approach will have to be taken at Tanah Merah sometime this year, in order to connect the new depot and new tracks to the EWL. One hopes this can be minimized.
Please see a therapist
While some may laud this, it is also not healthy.
The reason why Malaysia doesn’t do this is because of optics. If there are signs at Surian MRT saying that you can take bus 780 to Pasar Seni, why even take the MRT when you can just take bus 780? Well, bus 780 sucks with the maximum frequency being around half an hour. No matter how bad the Kajang Line is, you’re still likely to see at least two Kajang Line trains pass you before bus 780 shows up.
You also won’t see the MTR do this either because the MTR has a vested interest in you not taking the bus. MTR instead places a premium on incident management and quick restoration of services. For what it’s worth, hundreds of buses fighting for limited road capacity would be needed to accommodate passengers displaced from eight-car MTR trains, after all, so getting trains running again is of paramount importance.
On the other hand, while this is really only useful in a disruption, such eye-catching design also reinforces, psychologically, on a normal day, existing biases and perceptions that the MRT is overcrowded and unreliable. I call it rail reliability PTSD, as such prominent display of information — more prominent than actual MRT emergency features! — is clearly a symptom of post traumatic stress disorder.
Ideally, all these information should pivot to only be shown in event of a disruption or serious congestion in stations — only on the pulldown boards that get displayed only when the relevant SOP is activated. Similarly, digital information screens used to show train arrivals can have ad space overwritten with QR codes and other digital information allowing passengers to customise their own journeys. In the CBD, walking routes to alternative lines can also be shown.
We already have these information on the pulldown boards, so there’s not much net cost involved. The regular space can then be better used to show other MRT service information, such as train frequencies, travel times, and connecting bus service to local areas — the latter which SMRT used to do, but not so much the LTA.
Tell the world
I may have many things to criticise the Khaw ministry for, but I might now have to re-examine the need for performative politics to make it look like we’re doing something.
I could easily make a drinking game out of seeing claims from the public, and even from transport enthusiasts who ought to know better, that the TEL still has 9–10 minute off peak service. Yes, the 5 minute peak service did not change from TEL2. But off peak service has improved to 6 minutes, and along with the 11 new stations, that’s in line with the near-tripling of daily users on the line — even if despite that, the TEL is still much less intensely used than other lines. It still could be better, and that can be charged to the marketing budget too.
To be fair, our approach is close to the way things are done in France. And the French also have much more radical closures — the worst being Line 14, completely shut for two entire weeks in February 2024 to accommodate a new extension to Orly Airport. Line 4 is also seeing closures, but to be fair, they’re completely automating Line 4, so it is much more challenging. But are the French really the example worth following?
The case for MOT and LTA senior management to monitor the progresss and strategies of rail projects, and to push for better ways to minimize overall inconvenience to commuters, directly relate to public transport sustainability. Improved accessibility to rail does not just mean building more stations and more exits.
It also requires the LTA to equip public transport operators with the ability to demonstrate that a frequent and reliable rail service is being run. This can be best done by implementing the same level of open data accessibility for trains as is currently done by bus services. The costs of implementing this may well be a minor marketing expense compared to the costs of maintaining parallel modes that compete with each other for years.
Instead of asking people to change from bus 980 to bus 132, both with similarly horrible frequencies, emphasise the rail line. Show them that they can get to places quickly using new rail services, instead of building stations to tick boxes and call it a day. This might also reassure passengers concerned about long bus waiting times — take the train, it comes much more often. For what it’s worth, direct services to the CBD will also become increasingly harder to demand as pedestrianization and active mobility efforts reduce road capacity, and more road closures for events like Christmas light-ups reduce availability. This is why we built the MRT.
Most importantly, telling a better story on rail service — not just numbers— does not only persuade the man on the street. It also means that local political forces have less of a leg to stand on when it is pointed out that rail alternatives are available. It doesn’t matter that the Downtown Line has the highest MKBF of all Singapore’s MRT lines, if public perception says that the MRT and LRT are overcrowded and unreliable, giving local politics ammunition to insist on duplication which harms the overall network.
Those seeking to eke out extra efficiency on the public transport network should start by examining the impact of projects that affect the daily travels of public transport users, and to do something about them. The bottom line is that projects cannot be allowed to run amok and cause much larger impacts on services than they should; and they should be held more accountable than they do now.
Like what you read? Join the Telegram Channel for updates!
