The great game

From the Red Line
Published in
10 min readMar 25


Travel Smart Journeys is back. What are they trying to achieve?

The LTA announced an expansion of the Travel Smart Journeys scheme to cover the City Direct services in the northeast. This is a rather interesting change, in that I find the goals of the CDS are considerably different from those of Service 43e, the original testbed.

The original mission remains, which is supposedly to decongest the NEL, by far the busiest MRT route by passengers per kilometre. Every 43e and CDS passenger is someone not on the NEL, helping make NEL trains just that bit less crowded.

Will it help? I’m of two minds. The concept is sound, but as currently implemented, one might just be trading the unpleasantness of crowded trains for the chaos of the expressways. I’m not sure which is worse myself.

Laws of attraction

Eligible boarding stops for CDS and 43e (image:

The idea behind Travel Smart Journeys is to help incentivize the use of express buses as an alternative for certain demographics of passengers to avoid congestion on the NEL. Service 43e also plays a bit of a double duty in that it also allows passengers to skip a change at Serangoon to the Circle Line to get to Paya Lebar.

This works in an incentive point system. Every qualifying passenger gets 150 points per qualifying ride. After four rides and 600 points, participants can cash out 500 points as a credit to their travel cards. You’d thus need to ride every working day for two weeks in order to fully redeem the 1500 points earned, as can be seen in the below table.

Of course, you need to front the 60 cent express fare surcharge, and for earlier departures, you also stand to lose the 50 cent discount for exiting the MRT system before 7.45am. At the end of the day, the overall incentive can be as low as 40 cents a day. Is 40 cents a day worth rushing for a bus at a given interval, looking at the fixed departure times of many CDS services? At least CDS 654 can boast departures every 5 minutes from 7.30am onwards. KPE willing, one might be able to get downtown by 8.30am.

Considering, though, that the critical point of peak loading is the 8–9am rush hour, many of the target demographic for TSJ won’t be getting that 50 cent early-exit discount anyway. That sweetens the economics to a 90 cent discount. Still, this doesn’t change the fact that there will be a sunk cost in participating in TSJ as currently designed. But perhaps that’s the point, to hopefully switch over some passengers to use the City Direct network on a longer term basis, with sunk costs ensuring longer participation.

To me, this raises the question; why didn’t passengers already use such services? This question needs answers; for our financial sustainability debates hinge heavily on what those answers are. The LTA may find that imposing express fares on expressway services, even in a bid to improve the economics of such services, might result in ridership going away and towards the MRT network.

This could also be a likely reason why for the fourth year running, we have seen service cutbacks on 12e, 851e, and 960e, all of which follow MRT lines. Or perhaps one can ask why 74e and 151e were discontinued as well, even despite them being a much better value proposition in avoiding a NSL/NEL -> CCL -> DTL -> bus triple transfer. I guess you could just take the local bus.

This phenomenon is a double edged sword. Sometimes we want to empty buses and fill MRT trains, and sometimes we don’t want to. And sometimes, we want to fill other MRT trains.

For want of a train

There will continue to be growth in Sengkang and Punggol, and adding bus services will not be a solution for anything more than the short term, considering the high manpower costs of bus operations; or even just the logistics of getting enough bus drivers to operate such express routes. No one likes sitting in highway jams, whether as a car driver, bus driver, or bus passenger.

The medium term solution has to be improving NEL capacity, a much easier ask than a 9th MRT line — though I still expect that to happen eventually. We’ll need to take stock of existing projects first in other to understand whether this will work. Curiously, it’s going to be almost two years since the C851E trains were delivered, yet the LTA appears to be sticking to its guns on launching the trains with the Punggol Coast extension.

Still, recent updates from JTC around the end of 2022 show what is largely still a concrete shell, and doesn’t lend cause for optimism that the station will be able to open in the earlier portion of 2024. This is in contrast to TEL4, where architectural works are well underway at several stations. PSD installation is even done at Katong Park as of September 2022! I’m actually going to be putting my money on TEL4 opening before Punggol Coast.

But more seriously and more relevant to the discussion, this makes it a hard ask for the LTA to wait for the opening of Punggol Coast station before launching the C851E trains. The second factor in this is the ongoing C751A Refurbishment project. Currently, as things stand, it’s taking them approximately 4–5 months to refurbish one train. And sometimes the project can take out two trains at the same time.

It means that even after accounting for standby trains and maintenance requirements, SBST may have a difficult time pulling out additional trains to increase NEL service levels, without the C851E trainsets. Improvements in train availability from higher spending on rail reliability, probably only helps to make it easier to operate the existing timetable; and there will likely not be any space to add more trains.

Of course, one also has to ask why newer NEL trains don’t have the standing space of the C751A. Older train cars on older lines with removed seating are spiritually replaced on the R151 fleet by the removal of the 2-seaters at car ends, as well as some 7-seater benches being shortened to 5-seater benches and multi purpose spaces; refurbished NEL trains also keep their large standing areas. But so far it doesn’t appear that this has happened on the newer NEL trains (and perhaps on the CCL too), be it removal of whole 7-seater benches or the 2-seaters.

Likewise, there is a need for longer trains to increase service capabilities of the SPLRT, but those will also take time to come.

Cheap wins

Policy moves like Travel Smart Journeys may be able to help in the short term while capacity improvement projects are delivered. By how much, though? It may not be sustainable in the long run, but as the Khaw ministry will tell you, an emergency is an emergency.

A count of departures and vehicles used during the TSJ duration (7am-9am) (fleet info from Land Transport Guru) (as of 27th March)

Let’s assume the average capacity of a single decker bus is 85 persons, and a double decker bus holds 130. There are 14 single decker departures and 9 double decker departures during the entire 2-hour period. Assuming all buses are filled, this gives us a rough maximum of (1190 + 1215) = 2405 passengers being transported out of the entire northeast area by CDS services during the peak.

What does this actually look like? The planning capacity of a 6-car MRT train is 1500 passengers. 2405 passengers between 2 hours barely means 1 train per hour of passengers; though perhaps averaged out across the peak-of-peak between 8am and 9am it’s likely to be a bit more pronounced. Perhaps we might be able to say that one less trainload of passengers need be carried. It might help, perhaps averaging out across the hour to around 10 less passengers per train car.

It might also help that the running cost of this program seems rather low. 2405 passengers a day, at a $1.50 rebate each, means only about $3600 a day, at most, keeps this programme running. Not only might buses not be full to the capacities I’ve stated, some of them might also be existing 43e/CDS passengers who won’t qualify for the rebate anyway, which is likely to further lower the cost of the rebates the LTA has to pay out.

And of course, limitations on bus capacity and road capacity may make this difficult to scale to a meaningful level without significant and drastic bus priority measures taken along the KPE or CTE. Or even to reroute these CDS services to use Upper Serangoon Road directly. And what’s not to say that the second-order effects of such bus priority might further inflame traffic congestion?

To break the wheel

The question now is whether this might be able to kickstart a cycle of increased demand. I’m personally not so sure myself. Apparently the LTA claims over 53k commuters signed up to be a part of TSJ, but how many of them actually found themselves on a 43e bus on the regular, or did they sign up just in case they need that bus and can get some free money out of it while they’re there? And because of this program, did Service 43e see additional demand, resulting in additional service being required?

Still, I can’t help but wonder whether not only can TSJ use existing bus resources to support where we’re not investing in rail expansion, can the concept also be used to drive ridership on the still-relatively quiet Thomson-East Coast Line, helping create the case for more frequent trains on that line as well? If we’re going to have to increase service levels come TEL4 for the east side, we’ll have emptier trains to fill anyway.

Rewards could be paid out for passengers who enter and/or exit from TEL stations where they previously used the NSL. Examples include rides from Mayflower to Shenton Way, or Woodlands South to Maxwell, instead of Ang Mo Kio to Raffles Place, or Admiralty to Tanjong Pagar. If Woodlands and Ang Mo Kio residents are encouraged to use the TEL instead, it might leave more space for Sembawang and Yishun residents on the NSL.

Perhaps, also, some northeast-to-TEL bus routes stopping at Springleaf or Lentor such as Services 86 and 163 could also be strengthened, with incentive systems to get people to use them instead of using the NEL and changing to the TEL at Outram Park; or to access the CCL from Caldecott instead of Serangoon, which could be of use to westbound passengers. If you need to take a bus/LRT to Sengkang or Punggol anyway, might as well, right? Bus lanes along Yio Chu Kang Road could shorten travel times on these routes and make them comparable to the existing SLE/TPE bus services.

There might also be a case for doing the same on the DTL to help meet SBST’s target of 660k passengers a day on the line, and also justify any more service than the current 2.5–3 minutes (which, it bears repeating, is still lesser service than the 6-car lines). Bukit Batok residents could be incentivized to take a bus to Beauty World; Bedok residents to Bedok North/South or Bedok Reservoir stations.

As long as it works

A much larger-scale expansion to cover underused rail lines might cost the LTA more on paper, but with the lower operating costs of rail systems to begin with compared to parallel bus routes all the way to the CBD, it might be worth it. Where we have the parallel corridors already, we should make good use of them and improve them where lacking. Just as Montreal did.

Tools like bus lanes can be more easily deployed to support “orbital” service or some other form to get people to other corridors, compared to direct expressway bus routes, and especially as intertown travel demand recovers much faster than CBD-bound demand. These might help more people than we think, as we see in the case of 43e.

But for such modal transfers and corridor transfers to properly succeed, we will need to consider if planners are asking themselves the right questions. Are they being exclusionary, where only a single transport mode is tweaked in response to other external factors? Are changes in other forms of public transport service, such as a new MRT line, also regarded as such external factors for a given mode that only justify changes in the frequency domain and not in route planning?

Or are these changes even considered irrelevant, with complementary modes of travel deemed ineffective as the data shows people sticking to existing routes and paradigms? Incentive programs might help change this mindset for passengers, but what about policymakers? Worst case, perhaps our greater political environment may be dealing in absolutes, where it doesn’t matter what service is provided as long as the route is there?

At least Malaysia knows what needs to be done. Up north, new MRT openings, including the recent opening of the MRT Putrajaya Line Phase 2, have been also accompanied by a rollout of feeder bus services to bring passengers to the MRT. They’re not entirely good, but it’s the thought that counts, and bad service levels and route design can be improved.

Using parallel corridors, no matter the mode, to decongest existing rail corridors is something that needs multimodal thinking. Service 43e can be such a positive example as a parallel corridor, but I don’t think discounting CDS services is the smart thing to do. But an emergency is an emergency, and if it works— no matter how little — then it works.



From the Red Line

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