A Regional Transit Smart Card for Portland

Recently on the blog BlueOregon, it was suggested that Tri-Met (or somebody) should consider instituting a regional transit smart card in the Portland, Oregon metropolitan region, modeled after the Octopus card in Hong Kong. (Note: Hong Kong's Octopus card, developed I believe by ERG, is widely hailed as the best-implemented transit smart card in the world. It is the flagship implementation that ERG and other consultants use to sell the concept to new customers worldwide.)

While this might bring many benefits to the region over time, there may also be many pitfalls in the implementation process. For an example of this, I look to the nearest system to Portland to attempt to implement a Smart Card, the San Francisco Bay Area (as opposed to some metro area halfway around the world). The Bay Area has been attempting to implement the Translink Smart Card system to integrate fare collection for its 24+ transit operators... since the early 1990s. (The original pilot program for TransLink debuted in 1993 with mag-stripe cards, but it didn't work reliably due to problems with keeping card-readers operational on buses with high levels of constant vibration.) Currently, upwards of $150 million has been spent on the system (depending on how you calculate the costs), and it still is not operational for the core regional transit service: BART. (As of November 2006, it is up and running on Golden Gate and AC Transit services, and there is a planned eventual roll-out to the rest of the transit operators.)

But at what cost?

For the Bay Area -- which is admittedly a much larger region than Portland, with many more transit operators -- the cost is broken down as follows:

Phase 1 (Pilot Program for 6,000 users on BART, AC Transit, MUNI & Golden Gate Ferry, began in 2002 following the signing of a contract with Motorola/ERG in 1999. Users reportedly loved it!): $19 million
Phase 2 (Initial roll out using existing fare structures of the 24+ transit operators): $77 million
Agency Integration (between the two consultants-- ERG for MTC, and Cubic for BART -- & the various transit operators): $27 million
Design Changes: $17 million
Site Preparation: $5 million
Technical Support: $4 million

Grand total: $149 million (not including $77 million or so for a new set of faregates for BART that as delivered was not Translink-compatible, even though BART knew about the need for Translink as well as its specific design and technical requirements prior to ordering their faregates)

On the flip side, the operating cost for each transit operator will be about $0.04 per ride, which does not include the capital investment. This represents more or less a break-even fare collection cost to operators vis a vis their current fare collection systems.

TransLink is projected to serve 600,000 daily customers (1.2 million trips/day), or about 420 million transactions per year. In the future, it will offer limited use paper tickets for infrequent users, as well as explore the possibility of offering a Regional Transit Pass and integration with parking meters in participating jurisdictions throughout the region. Other future avenues of exploration, following final implementation of the transit fare card on all 24+ transit operators, may include payment of taxi fares and rewards programs with participating sponsors.

The TransLink rider experience gets rave reviews. TransLink automatically calculates rider discounts to give the rider the best deal possible -- monthly passes or other discounts first, then the E-cash second. It will auto-load when the balance dips below $10 in e-cash; as well as loading a new monthly pass the first time that the card is tagged during a month. Anonymous cards are available, but they cannot be replaced if lost (replacement cards are $10 for registered users). The maximum amount of value at one time on a cards is $300, due to on-board memory limitations.

As for operations: Translink is currently operational for a transit agency with a tiered (zoned) fare structure, but there's a hitch: Golden Gate Transit buses operate through six zones, but they do all loading and unloading through the front door only (the rear door is reserved for disabled passengers). All other current Translink-using bus operators operate within a single fare zone. This means that the particular implementation issues that Tri-Met would have if it decided to adopt a smart card with is existing fare structure (riders having to tag on and off each bus via either the front or rear doors) have not yet been fully tested.

Caltrain, which operates a commuter rail service mainly between San Jose (Silicon Valley) and San Francisco, represents the equivalent of MAX light rail operations, but it will not roll out TransLink service until late 2007. Riders will tag in at their origin platform, board the train, then tag out once they reach their destination platform. They will be charged the maximum Caltrain fare when they tag in, then the difference between the maximum and their actual fare will be given back to them when they tag out. This is probably the closest equivalent to the system that Tri-Met would implement, so Portlanders should watch its implementation closely.

I bring this up for two reasons:

1) Portland needs to very carefully examine the lessons from the Bay Area before embarking down the path of smart-card fare payment. I think that if you ask the Tri-Met managers about this, they will tell you that they are letting other regions take the lead on this issue, to see where the pitfalls are before they commit any of their own capital to the idea. And rightfully so. I'd much rather see one more streetcar, light rail or frequent bus service added to the system -- real additional functionality -- than a bunch of money wasted on what amounts to basically just a different way to collect the fare.

2) When all is said and done, the Translink system will be a *HUGE* benefit to the riding public in the Bay Area. It is currently *extremely* confusing, time-consuming and expensive to attempt to make multiple-operator trips from one part of the region to another. Heck, it's a P.I.T.A. just to travel on ONE operator, BART, due to the complexities of their own paper-ticket fare-collection system and the byzantine fare-collection machines, fare gates and unfriendly station agents associated with it. Translink will definitely help to simplify all this, and when I was a participant in the pilot program and able to use it for travel between selected stations on BART, when it worked, it worked satisfyingly well.

That is to say -- there is a benefit from these smart card transit fare collection systems, but there are also enormous pitfalls associated with implementing them. Aside from the initial capital expense, there is the difficulty of navigating entrenched bureaucracies who are wedded to the current way of doing things, as well as the ongoing operating expenses and the pitfalls of dealing with international consultants who may have their own agendas.

Tread this road with extreme caution, for the promises of benefit are huge and the potential for averse consequences even larger.