From the Red Line
Published in

From the Red Line

Should we scrap people movers?

And replace them with what, exactly?

Recently, several internet pundits have come out with several rather odd views of the people mover systems we’ve called LRTs. Sure, I might have agreed once, especially considering that a previous transport minister was the person who made it fashionable in the first place, but there is a significant amount of nuance that is being ignored in favour of feel-good NotJustBikes-like soundbites.

Some of it is justifiable. But other points are more reflective of how we build rail lines in general (or how we did so at the time), and they will also need to be addressed for the upcoming JRL. Others don’t make so much sense, but we will have to look at them anyway, since these come from the fact that we are not using the LRT system to its full potential.

Just because we’re realized our follies and are not building any more doesn’t mean we can’t try to fix whatever we already have. I expect to cover more on the nature of these fixes in a future blog post and how far they can go.

Picking the bones

In France around the late 20th century, there was also considerable interest in light automatic subway systems. I will let RMTransit explain more in the below embedded video. Does the tagline sound familiar?

It would thus not have been hard to think that when the LTA conducted their LRT scoping studies in the 1990s, they went to France and received a presentation from the French along the lines of the RMTransit video. As RMTransit says, capacity can be sharply increased by brute forcing trains in frequency — if they so wanted, the 10k pphd target of the RTS can easily be achieved by running 50 200-person 2-car LRT trains per hour — very easily doable as shown in Lille. Initial plans for the Marina Line also show a “4 car” system, likely people-mover based, capable of carrying 20k pphd.

This is where implementation fell short of the lofty visions espoused by Lille and other French cities. In Lille, since the Lille Metro is considered a full-fat subway system and not a cheap replacement for the inability to build proper MRT lines fast enough (though to be fair, JRL stage 3 is taking as long as the Circular Line in Taipei to build), there are underground sections and all. Even the Kelana Jaya line has a short cutting underground in the area of Sri Rampai station.

Had some of these been done in Bukit Panjang, perhaps the awkward ups and downs around Teck Whye, and perhaps Bangkit, might have been less painful to ride through. The LRT track could have passed under those bits with an underpass of sorts below the road, reducing the slopes to be climbed. But some people are complaining about cost, and LRT systems are supposed to be cheap and nasty. Having to dig up roads for short underground stretches would have added time and money to what was needed for the LRT construction.

And perhaps if a different technology had been chosen (maybe had we contracted Matra to build an actual VAL instead of using the CX-100), there might have been less issues, since most of the issues experienced have to do with the centrally mounted CX-100 power rail design. And had we remembered how to build train stations in the middle of the road as we did in the Initial System, perhaps there might not have been a need for misting windows as well.

But alas, what’s built is built, and some of the lessons from there have been incorporated into the construction of the SPLRT. Running in the middle of six-lane avenues means there is no need for misting windows as tracks are unlikely to be within sighting distance of apartments, and road-median stations also make them not much more painful than crossing an overhead bridge to get to a bus stop across the road.

It does, however, also suffer from urban design issues such as not being sufficiently walkable from existing estates or supplying too much parking. But that’s more a problem of the contemporary HDB, and not much can be done after the fact apart from repaving the “neighbourhood gardens” to provide walkable throughfares. Much as I like trams, a tram system with stops in the median of said six-lane avenues would also suffer from the same issues; the lack of staircases offset by the need to use a traffic light to access a median tram stop, and perhaps longer walking distances due to offset stops.

What LRT does not do

I disagree with the premise that ineffectual LRT systems directly contribute to car ownership. Sure, there may be more cars if people actively avoid the LRT and the LRT’s presence may mean lesser feeder buses are running. But it actually appears that to the contrary, Punggol feeders actually have quite good coverage; albeit at 8–10 minutes between buses for most routes, frequency isn’t that great. Though such wide coverage might be due to the underused potential of the LRT in dealing with the sheer density of Sengkang and Punggol.

That’s actually very good. (source:

Yes, it is also a fact that the Sengkang and Punggol LRT systems are built in the medians of six-lane avenues. But that does not mean they cannot coexist with bus priority measures, cycling lanes, and other fun stuff. In fact, their grade-separated nature taking the bulk of passengers up and away from the roads should mean faster travelling time for buses through the area, and more space for cyclists — especially as road lanes are reclaimed for that purpose. In fact, I could even argue that building cycling paths should actually be much easier in Bukit Panjang, by reusing land under the LRT viaducts where possible. And according to Minister Vivian Balakrishnan’s Facebook, that might already be happening.

source: Vivian B’s FB

Then of course there are social expectations — the five Cs and all. If you’re going to own a car, you might as well use it. Better still if you drive to the office in a shiny BMW or Tesla instead of some beat-up Toyota near the end of its COE. Gen Z may not understand, but this is how some people expect to live. And if this is the case, the sheer humanity in the northeast area that has built up over the years with the completion of BTO projects naturally means congestion. The CTE and KPE are both hotspots for accidents, after all.

And more importantly, you can’t argue that nobody uses public transport, as the NEL may also be our busiest MRT line based on passengers per kilometre, at 600k daily trips on average in 2019. That’s 219 million trips a year; one-third of what SMRT carried on the NSEWL in 2019, but with five times lesser trackage (NEL is 20km, NSEWL is 101km combined). It might speak volumes towards the benefits of automated, driverless systems enabling high frequencies, that I haven’t heard as much bellyaching about having to wait for the next train on NEL, while it was common on the NSEWL. I could be wrong.

But why build LRT lines in the first place? Yes, one possible reason is that buses cannot cope. Additionally, the focus on automation when done right means that operating costs become extraordinary low — no driver is still better than a tram driver, even if that tram driver replaces 3 bus drivers. The savings may well pay for themselves, especially if fewer LRT vehicles mean a reduced maintenance workforce is needed compared to a fleet of buses for feeder services.

In short, what other system could you have to handle the high capacities needed to serve dense neighborhoods? You could have a tram with double the capacity of a 2-car LRT train, but if it still comes at the same frequencies as a feeder bus, then you’ve made compromises that means people don’t get frequent public transport.

source: Department of Statistics, Population Trends 2021

It’s not entirely about urban design either

Please, let us fight the war on cars, but don’t turn and shoot your allies. Mindlessly slashing car lanes by government diktat and “let them take the bus” may be extreme policy that may end up backfiring, especially as the loss of road lanes to be used for other purposes affect bus travel as well. Just look at the upcoming changes to Senja Link. And even if the roads are untouched, the main Bukit Panjang Ring Road and parts of Choa Chu Kang Way are only 2 lanes each way. Hardly a stroad, and one has to note that Bricklands Road was constructed as a relief for CCK Way because it was simply getting too congested.

I said previously that the SPLRT runs in the median of six lane avenues. As it turns out, I lied.

source Google Maps Street View

The stroad argument rings even more hollow when you look at this perspective of Compassvale station — and there are several more similar stations along the SPLRT with similar cross sections. This road has two lanes in each direction, supposedly befitting its status as a major access road. The LRT track proper sits above the inner lanes of this road, with land space taken up by only supporting columns and lift machinery for the station above.

Looking down from above, this is essentially one lane of road and one lane of transit. You could do that with a tram. But buses would be poorly suited for sharing such median transit lanes because their doors are only on one side, so it becomes a tram only lane — and offset stops might be too long, as mentioned. And you’re not going to be doing street running trams at 2 minutes interval in peak hours. The outermost lanes may not work either considering the need to provide local access to developments — not only for cars, but for municipal service vehicles like what handles your garbage collection, as well as deliveries.

If you did that anyway, what then might happen is resident complaints asking the authorities to eat away the grass verge on the other side for another car lane to meet congestion. Is that really what we should want?

Or for saving road space, maybe this. (Fajar Road).

Fajar Road (source Google Maps)

Oops, I lied again. That’s a cycling path on the other side of the road, under part of the covered walkway.

A change in purpose?

Removing infrastructure has its costs. This is not like cutting a bus route and then bulldozing the remaining bus stops if no other service stops there. Infrastructure removal should only be tolerated if we are really sure that whatever it replaces can really be an improvement, or if no one is using the infrastructure as designed because it is irrelevant in the face of new network developments. Removing bus routes to be replaced with a train at least makes some sense if the train has several times the capacity.

Especially for the BPLRT, the opening of the DTL means that trips are shifted away from going the whole way to Choa Chu Kang, instead mostly going to and from the DTL station and the bus interchange. What does this tell me? Most of the LRT’s real time savings come from having a direct run from the interior of Bukit Panjang new town to the then-nearest MRT connection at Choa Chu Kang (around 10–15 minutes from inner Bukit Panjang to Choa Chu Kang around the loop, 976 takes 30 minutes). However, it also has to be pointed out that there are also Choa Chu Kang residents taking the LRT to the DTL, a job that LRT scrappers can say should be done by 67 or 983 or something.

It is entirely possible, due to the interchange configuration, that instead of going a handful of stops on the LRT, passengers choose to instead take a feeder bus to the interchange before taking the train, just like in any other town. This can be seen in how Senja and Petir stations are some of the quietest LRT stops, as passengers from there can easily walk or take a short bus ride to the MRT especially with Exit C near Block 604. Consequently, it might say something that apart from a brief articulated bus crossover in 2015 proving that it’s possible to operate bendies, 920 and 922 have never had high capacity bus deployments until 2018, with the introduction of Euro VI MAN A95 buses.

But the concerns about bus congestion from shifting 30k daily LRT trips (as of 2021, in 2018 it was 68k) to the buses do not just come from additional routes (such as BPS1/976), it comes also from the required frequency and capacity enhancements to other services. These include the aforementioned 67 and 983, maybe 160 and 180 for those heading to the west who may otherwise take the LRT to the JRL, so on and so forth. Is this really what we want, considering we’re replacing a driverless system with somewhere around two hundred drivers and half that amount of buses? Or its impact on the road network where more buses have to cram on the 4-lane roads?

An alternative explanation for the growth in 920 and 922 ridership, though, could be the off-peak suspension of Service A, which makes 920 and 922’s service areas more reliant on the buses. And if they’re taking the buses, they’re probably going to need some cajoling to go back to the LRT. Of course, the question remains on whether this suspension is only temporary to reduce daily mileage on the older vehicles and to hold them together, or it may be something more long-term.

Furthermore, there still remains ITE College West. As any NTU student will tell you, tertiary institutions are unable to be served solely by buses — that is why NTU gets its own JRL extension, and almost every other autonomous university and polytechnic has at least one MRT station to call its own. Even NP/SIM/SUSS might get a CRL station in their vicinity. In ITE’s case, there are 13k students, including 5k part time students, a large portion of which would use public transport to access the College West campus near Teck Whye LRT.

Transferring all these to buses would make an already bad situation even worse —apart from NTU, just look at College Central and everyone packing on 72/159 to get there; or Republic Poly and 902 prior to the opening of TEL Woodlands North station. And already I’m hearing talk amongst ITE College West students that the LRT is too small to meet the needs of students; going to buses would be an even greater disaster.

What now?

There’s a point made by the Youtuber-who-shall-not-be-named that urban design matters. In this aspect, the LRT suffers from the same urban design issues that MRT stations do, in that they simply do not have access from all directions compared to a bus stop. This can be somewhat easily fixed. It was fixed at Fajar, where an emergency exit was pressed into duty to serve as a side entrance to the station from residential blocks nominally at its rear.

If you really want to tear down the LRT, though, you should incorporate such features into its replacement from day one. Come 2045 after the currently-new systems also reach their end of life, they might be able to get away with extending the JRL Jurong branch from its terminus at Tengah, up to Brickland station and Teck Whye, then into Bukit Panjang perhaps following the Bukit Panjang Ring Road up to Segar or something.

This might be acceptable as in essence, it is gutting a small train and replacing it with an even bigger train, as well as one-seat (albeit indirect) service to Jurong Lake District. These efforts create a connectivity improvement sufficient to justify tearing down the LRT by removing the interchange at Choa Chu Kang from LRT to MRT. Larger JRL stations with more exits could also overcome the indirectness of how the LRT stations are accessed today, only from one end — so long as the exits of the JRL stations are easily accessible from surrounding estates and not only the one the exit directly adjourns.

Such a project would be similar to the Scarborough Subway Extension in Toronto, where a heavy rail extension is used to replace an end-of-life light metro system with similar ridership, also with buses to serve the 7-year gap between the light metro shutdown and the subway opening. Even if we were interested in running a BRT like they want over in Toronto, it would still take a while as the rail infrastructure was removed and the stations and tracks torn down, and the roads expanded to provide the required priority lanes, before any BRT service can be run, or to reconfigure the roads to accommodate a JRL extension. And like in Scarborough, a complex phasing plan involving buses and perhaps building whatever they can before the shutdown would also be necessary and involve large amounts of pain.

On the SPLRT, perhaps some limited rerouting of the Sengkang West Loop could be justified in order to connect the LRT to the 9th MRT line. The continued existence of parts of the Punggol West loop can also be re-examined in light of the construction of CRL Riviera station — though one might ask if going all the way down to underground CRL platforms is as “painful” as ascending to an elevated LRT platform, if not more. It can, after all, be argued that just like other towns with two or more MRT lines running through them, the CRL and buses can be enough to serve that bit of Punggol.

Or maybe we’ll finally have a common fleet of autonomous vehicles to be shared across the entire country as a magic bullet to solve all neighbourhood transportation issues, who knows. At least it’ll still be automated and not be Malaysian BRT. Talk about using the wrong tool for the job.

Look familiar? (photo by Sirap bandung, Wikimedia Commons)



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store


Sometimes I am who I am, but sometimes I am not who I am not.