Wednesday, March 28, 2012

1.1 billion dollars more of buses will not help

I've waited a few weeks to post my opinion on the government's move to purchase buses to improve the public transportation system for a number of reasons, mostly because I wanted to wait to read the opinions of other bloggers.

As it turns out, the few people who blogged about this expressed opinions that differed little from the general sentiment of "It's a step in the right direction, but it comes a little late, and honestly, more should have been done earlier. But let's wait and see how well this turns out."

As always, I write only when I believe I have something fresh to offer, and this is my clear and unequivocal opinion:

I do not think adding more buses in the way that the government has envisioned will improve the public transportation system appreciably. And the train situation as it stands now will NOT improve because of this new addition of buses.

I take buses much more often than I do trains to get around, so I think I'm qualified to comment. And my reasons for being so unequivocally negative on this extra-budgetary measure are as follows:

1. If you read the puff piece that the Straits Times published a few weeks ago (follow the link in the first sentence of this post), the reporters wrote on how taking service 128 was in fact slightly faster and much more comfortable than going the feeder bus-MRT mode. The piece spoke specifically about how the bus was less crowded than MRT trains at peak hour.

A word to the wise: As anyone who builds or manages systems will tell you, adding more capacity to a mode of transportation that already has spare capacity (i.e. is less crowded) is not a strategy for improving operational outcomes.

It was this single fact alone that revealed the Straits Times piece for the cheerleading article that it was. Editors at a national propaganda machine should know better than this. Chances are, service number 128 was cherry-picked to "demonstrate" the merits of bus versus train travel.

2. Anyone who takes buses will know that journey times by bus are generally longer than journey times by MRT.

There are 4 basic components to journey time:
a) Walking times to and from bus stop / MRT station
b) Waiting times for arrival of public conveyance
c) Stopping times at bus stops / MRT stations
d) Actual travel time enroute to destination

MRT wins hands down over buses due to shorter (b), (c) and (d). It's no surprise that most people prefer trains for their speedier transit. Ergo, trains are crowded while buses less so.

There are groups of people for whom buses make sense over trains. These are people who can't stand crowds (yours truly), do not live sufficiently near to an MRT station to take advantage of its speed (and hence have to take feeder bus services or walk relatively long distances, yours truly again), or have to endure the hassle of making multiple transfers across rail lines even on MRT.

But by and large, for most people, the speed of the MRT wins out over buses.

Which brings me to this point: adding more buses on the roads generally only reduces (b). It does very little for the other three components that make up journey time by bus. MRT will still be faster for most journeys.

In addition, we have to consider the fact that even if the buses succeed in attracting passengers off the trains, they will only attract the marginal travellers who value space and comfort just that much more over shorter journey times. That's going to have limited impact on improving the train situation.

3. But wait, there's more.

Adding more buses on the roads will reduce waiting times, making journey times by bus shorter and hence a viable alternative to trains that more passengers will consider. That's what our public transportation planners would like to think.

As it turns out, there is already a way to reduce waiting times for buses to a minimum, as savvy bus travelers like myself know. Two things: schedules and bus transfers.

Going to work in the morning, buses travelling from the suburbs to the CBD area tend to arrive on time and on schedule because traffic is relatively light on the roads around the suburbs. After all, people live distributed all over the island. It's a different story in the evening when the roads out of the CBD are filled with vehicles all trying to get out at the same time, and arrival times for buses tend to become unpredictable.

When I head to work every day in the morning, I make it a point to reach the bus stop only 2-3 minutes before my bus arrives, a trunk service that brings me to within a 5 minute walk from my office building. No transfers, minimal waiting time. Granted, the bus takes a circuitous route and it's still slower than the feeder bus-MRT mode, but there's no jostling with the crowds and much less walking.

Going home in the evenings, I can take the same direct trunk service again to get home, but because of the aforementioned unpredictability in waiting times, I choose a different strategy instead.

I walk to a bus stop that has not one, but two bus services that take a very direct route toward home, but can only drop me off 4 bus stops short of my destination.

From there, I have two buses that can take me straight home, and another four that can take me 2 bus stops closer to my destination.

And finally, at this third bus stop, I have 4 buses that can take me the final 2 bus stops back to my home.

I call this extreme bus transferring. And it makes total sense after the transition to a distance-based fare system last year. At every stage of my journey, I maximize the number of buses I can take and minimize the waiting time. It's not uncommon for me to drop off a bus and take the next one that's either right behind it or in front of it at the bus stop. Total waiting time for my journey ranges from 2 minutes to perhaps 7 minutes, comparable to the MRT.

But even with all these stratagems, I have to say: the MRT is still faster. Getting from my office desk to across the doorstep of my apartment, or the converse, takes about 45 to 50 minutes by the walking-MRT mode. The same journey by bus takes perhaps 55 minutes to 1 hour 5 minutes. For reference, I work in the CBD and my home is in the Dover area.

So, despite the best intentions of our transport planners, the point I'm trying to make is that even with reductions in waiting time brought about by more buses, MRT still wins out.

Even with bus lanes and the newer concept of "bus hubs", it's unlikely stopping times and travel times will appreciably improve for buses and make them competitive with trains. Which is why I seriously doubt that more buses are going to alleviate the train situation.

The real kicker in this long post of mine discussing our woeful public transportation situation?

I run home from office twice a week, and I take just 1 hour 15 minutes to do so. Now that's fast.


Ponder Stibbons said...

I used to commute 20km to/from work by bicycle. It would take me ~55 minutes.

However, pre-Circle Line, the same trip would take me 60-70 mins by MRT.

orpheus said...

Can't agree more. Buses aren't the solution granted that the roads are already maxed out.

Even during off-peak hours, expressways are already jammed up. The only solution is to develop expansive (not expensive!) train networks which don't just take care of major roads, but are usually within walking distance.