31 December 2009

Reflections on the myki debacle

The Victorian Government has finally performed a pseudo-launch of its dreadful “smart” public transport ticketing system. Pseudo-launch because, almost three years after it was scheduled to go into service on 1 March 2007, covering all transport modes, Public Transport Minister Lynne Kosky has launched it for use on trains only.

The project is $350 million (70%) over its initial budget estimate of $500 million. Even at $500 million the project sounds very expensive to me as the cost of reinventing something that has been done before and better many times in major cities all over the world.

As a former Secretary to the Department of Defence I cannot help comparing this performance to the Collins Class submarine project, which has received such a bad press over the years that it is often cited as an example of cost overruns in major Defence projects.  In fact, the total cost “overrun” (additional money that had to be spent to meet contractual requirements) was $143 million in a $5 billion project – not bad by any standard, certainly a lot better than a $350 million overrun on a project a tenth of the cost. And it would have to be said that developing a new generation submarine (rather like developing an underwater aeroplane) is more of a challenge than developing yet another public transport ticketing system.

On the subject of defence projects more generally, Defence Minister Senator Faulkner told the Australia and New Zealand School of Government on 13 August that 83% of the 200 defence projects closed out in the last decade were on or below budget, so Defence is not in fact the principal generator of failed public sector projects.

Another thing I cannot help noticing about myki, 45 years after first laying my hands to a computer, is that the developers (or writers of the specifications?) seem to have fallen for the classic error of computerising the old way of doing things rather than asking the fundamental question of how one would design the system from the bottom up given the capacity to access and process mountains of data in real time.

I refer in particular to the fact that the zone system (and the two-hour ticket system) seems to have been faithfully translated into this expensive white elephant. Given that the division of the city into zones is a proxy for pricing by distance in the absence of a satisfactory way of measuring distance, I would have hoped that after spending $850 million we could have a genuine user pays system in which people are charged by the kilometre on the basis of where we swipe our cards on and off. Rather, it seems that we will persist with a system in which I can travel seven stops to the city and ride around the city loop for two hours for a zone one ticket price, but if I want to make a single journey six stops in the opposite direction to pick up my car after a service, I have to pay a great deal more because I cross a zonal boundary. Why did we want to enshrine that idea in the new “smart” system? What is smart about that?

Minister Kosky says that we will all learn to love myki. I doubt it.

1 comment:

James said...

Change is always going to be difficult, and while I agree with your major points re: cost, timeliness, and the roll-out - don't be surprised if the system does allow for "genuine user pays" down the track. Introducing those sort of changes along with the ticketing hand-over would have your run-of-the mill PT user frothing at the mouth. A person can only endure so much change.