The JRL and NTU

With three stations on campus, NTU should reconsider its transport strategy. But how?

Universities and polytechnics are widely known as key sources of public transportation usage. In Singapore it’s no different — the most famous story would be apparently how SP students and staff lobbied the MRTC/LTA for the construction of Dover station.

Recent MRT expansions have also given other such higher education institutes stations they pretty much can call their own — NUS has Kent Ridge station, SUTD has Upper Changi, Temasek Poly has Tampines West, Nanyang Poly has Yio Chu Kang, Republic Poly Woodlands North; even the Ngee Ann/SIM/SUSS area may also get a CRL station of its own. SIT’s future campus in Punggol will also have Punggol Coast *and* Teck Lee stations.

Attentive readers may note that I have made two major omissions. The first is SMU, which need not really be counted given how its campus is integrated into the CBD landscape — but anyway it has Bras Basah and Bencoolen serving its campus, with SMU itself even serving as the pedestrian link for anyone who wants to use that as an interchange.

The Island in the island

The second is NTU. NTU has the most unique transportation needs — with the amount of boarding places it offers, and how most faculties are centralized around the North and South Spines, it may be considered as a uniquely self-contained ecosystem — with a “CBD” (the Spines) and “suburbs” (the halls). It even might have the population to boot —around 24,500 undergraduates, 7500 graduate students, and then there’s faculty as well as researchers and other support staff.

Almost in recognition of this, the JRL is routed similarly. Of its three stations, JW3 stations serves the “suburbs” where the halls are, and the other two serve the North and South Spines respectively — JW4 station also serves NIE, which may or may not be included in NTU’s statistics depending on who you ask.

However, this isn’t really a silver bullet. Several halls, mostly in the single digits, still will not have access to the JRL. And the phasing of the JRL itself also does not easily allow for a comprehensive rethink to be done at one go — after all, in phase one, the JRL stops short at JW2 station. It would not be that meaningful for NTU to reorganize its bus routes, but for the two intervening years operate another bridging bus along the JRL route.

That also won’t solve the problem of external access. While NTU offers two guaranteed years of hall to those who ask for it (and who can then participate fully in its closed transport ecosystem), and internships/exchange may reduce the real amount of students on campus, there is still a significant demographic who commute in. Only after the whole JRL is complete will these students be able to use the line, but otherwise a minimum of two transfers including one to a shuttle bus may be off-putting.

Seasons come and go

Another more long-term problem with giving NTU so much MRT to call its own is that ridership is highly seasonal. Unlike SIT or SUTD’s trimester systems, NTU adopts a traditional academic calendar structure with two semesters and a nice long summer break where classes are optional. During summer, ridership can drop even further because of students being away for summer school, internships, or the fact that people just don’t have to go to class. The research centres also can’t really generate so much ridership by themselves; it would be a fraction of the traffic seen during term time.

In fact, the difference in demand is easily demonstrated by bus 179A, an express variant of 179 that forms basically every other departure from Boon Lay interchange in peak hours. It simply doesn’t have to run in NTU holidays because there’s few enough people going that way that the normal frequency can support. In fact, West Virginia University operates a PRT system that might be comparable to the JRL NTU branch, and the PRT is completely closed during university holidays.

This issue may well persist with the JRL as well. JW2 station serves Cleantech Park, as a business district that might be able to generate sufficient demand. The problem is what happens past that; it may mean that much like the TWE, the JRL operator may have to short-turn trains that aren’t going so far during the off-peak hours in vacation time, since the branch can’t be fully decoupled seeing as it lacks its own tracks between Boon Lay and JS7 stations. Additional facilities may also have to be provided for this somewhere along the line.

Of course, these are based on today’s numbers with the current size of the university; if it has significant ambitions to grow, it may be well needed. But even growth may not be enough — Kent Ridge station only saw 25k in average daily entries during NUS’ 2019/20 Semester 1, and that’s even after including patients and visitors to NUH, of which there should be a steady stream. With 38k students and 12k faculty, after removing students who stay on campus and students who might prefer taking a bus from Clementi Road, that is not a lot in the final accounting.

What is to be built

So what can NTU do? How do we make such a significant investment count? Firstly, it should not assume that government largesse will continue to provide 179 and 199 services after the opening of JRL stage 3; which means that its own shuttle bus providers will have to step up their game. The saving grace is that the Campus Rider and services further out can be abandoned as well, again with the expectation that like the rest of Singapore, long distance trips are to be done by MRT. That way, NTU can refocus its resources on providing transport service to parts not reasonably covered by MRT.

An integrated transport strategy may work, but nobody likes to be asked to take a bus then transfer to a train for just one or two stops. Thus, one might say that for it to work inside NTU, subsidized fares may have to be provided, with either the government or the university picking up the tab for anyone travelling with valid university ID. The current NTU ID is apparently “Nets Flashpay ready” so it’d be fair to say that NTU is using the CEPAS standard already. They could simply add some kind of free/discounted entry pass for the JRL stations, the hard part may be the politics of such an arrangement, given that there’s a sizeable group of students already willing to pay for 179/199 for on-campus travel.

Like elsewhere, NTU would also have to look at how the area around JW3 station can be tweaked to promote through-block walkability from some of the halls further out to the station, such as 8–11 and the Nanyang Crescent halls. They would probably find better success that way.

But yes, they can’t get rid of the shuttle bus. Halls 14 and 15, and the Lien Ying Chow Drive area still won’t have convenient access to the JRL, not being within 400m of any JRL exit. Though of course with a fair bit of traffic being shunted to the JRL, it should be possible to improve the hilariously poor reliability of the NTU shuttles. Their routes can even be amended to avoid overlapping too much of the JRL service area.

The chief problem with the NTU shuttles appears to be the lack of a defined terminal area — while NUS shuttles have Kent Ridge bus terminal and PGP bus terminal, NTU appears to only have the Hall 4 bus stop. If reports on NTU Confessions are any indication, even that isn’t respected and the shuttle buses just go off service and skulk off from whenever. As part of JRL-related infrastructure works or even today, a dedicated terminal, even if just a few lots, should be constructed somewhere as a defined terminating point for the Red and Blue loops and some discipline instilled. We can then play with the loops come 2029.

Loopy ideas are (somewhat) fine

Then again, that’s assuming the shuttles in their current form continue existing past 2029. Yes, 2029 is a long way away, and it may seem woefully premature for me to write this now. But the way I see it, two things can and probably may happen, by the time 2029 rolls around.

Firstly, there’s a possibility, however remote, that NTU is able to work with the LTA and the contractor for an earlier completion of JW3 and JW4 stations, be it either in stage two or a separate mini-stage just in time for the beginning of the academic year. This would allow NTU to reap the full benefits of the JRL faster, since I’d think running a shuttle from JW2 station would result in an unpopular shuttle, and perhaps student grievances if 179/199 were cut and the students forced to transfer twice or more. Remember, Bahar may factor into the equation.

Secondly, if Hall 14/15 and Lien Ying Chow Drive turn out to be far too little traffic for even shuttle buses, the university would be in a good position to try out driverless mini-shuttles or even larger ones. It’s important that driverless vehicles are used, lest the mistakes of LTA’s on demand bus trials repeat itself, but I’m sure a university is a much more controlled environment, and with the technology being kickstarted by the same trial it’ll probably mature soon.

I’d think this is more likely to happen with Hall 14/15 given its position in a no man’s land between areas walkable from the JRL stations. Then again, it appears they did, although nothing really seems to be happening to that, so it might either not really be workable or they just forgot with bigger problems going on.

With a lack of a highly pronounced peak hour relative to the actual CBD, on-demand transportation options with no fixed route might even work better for these use cases — just like how the WVU PRT has similar on-demand modes, just on a dedicated guideway instead. The coverage of such shuttles could even extend to Cleantech Park, providing access to JW2 for areas around it. Or the opposite could happen, and JTC extends its JID podcar system to serve NTU, both of which would be a more targeted solution for the Lien Ying Chow area.

No quick fix for geography

Of course, even with a MRT connection via the JRL, the isolation of the NTU campus still means a significant trek out, be it daily commuters or dorm residents booking out for the weekend — 15 minutes from NTU to Boon Lay is the quoted ballpark figure, though it may change as designs develop and the various system parameters are adjusted to determine the true travel time.

I don’t see any easy way to make this less painful — since even to get anywhere meaningful, one would have to trek out to Boon Lay. Even if you talk about faster travel from using the CRL, it may be a larger detour to get to any potential JRL/CRL interchange south of Boon Lay, than to take the EWL at Boon Lay and change to the CRL at Clementi.

But the one thing I can think of would be to extend the JRL to Gul Circle, using the originally planned Tuas South Extension platforms as a terminal. After all, if the Tuas South extension is run as part of the CRL, and the CRL is likely to be deep underground in that area, those platforms may no longer be of use.

With this, the CRL can be 1–2 stops away from NTU — albeit with almost nothing in the area since it’s just military land and the PIE, so such an extension might be hard to justify. Co-locating a JRL facility with the CRL depot, likely to be in the area, also makes little sense given how much space is already being cleared for the Peng Kang stabling facility, unless that is cancelled at the last minute seeing as there’s already more than enough space for 62 trains at Tengah Depot.