11 March 2010

"Toyotas Are Safe (Enough)"

One of the things I learned in my Risk Management course as an engineer was that there is no such thing as "safe."  There are only degree of safe. Hit me with some numbers, Robert Wright...
NY Times: Opinionator | Robert Wright | Toyotas Are Safe (Enough)

My back-of-the-envelope calculations (explained in a footnote below) suggest that if you drive one of the Toyotas recalled for acceleration problems and don’t bother to comply with the recall, your chances of being involved in a fatal accident over the next two years because of the unfixed problem are a bit worse than one in a million — 2.8 in a million, to be more exact. Meanwhile, your chances of being killed in a car accident during the next two years just by virtue of being an American are one in 5,244.

So driving one of these suspect Toyotas raises your chances of dying in a car crash over the next two years from .01907 percent (that’s 19 one-thousandths of 1 percent, when rounded off) to .01935 percent (also 19 one-thousandths of one percent).

I can live with those odds. Sure, I’d rather they were better, but it’s not worth losing sleep over. And I don’t think it’s worth all the bandwidth the Toyota story has consumed over the past couple of months.
I'm not sure where we want to be on the risk curve with respect to car safety. There is no single right place to be. But everyplace on that curve has associated levels of costs. And I don't just mean the monetary costs of installing safety features, I also mean that when you remove the electronic throttle system as a potential point of failure, you also lose the added fuel efficiency such a system contributes and a critical component of the electronic stability control system that helps to prevent other types of crashes.

Wright concludes with this:
But it worries me that this Toyota thing worries us so much. We live in a world where responding irrationally to risk (say, the risk of a terrorist attack) can lead us to make mistakes (say, invading Iraq). So the Toyota story is a kind of test of our terrorism-fighting capacity — our ability to keep our wits about us when things seem spooky.

Passing the test depends on lots of things. It depends on politicians resisting the temptation to score cheap points via the exploitation of irrational fear. It depends on journalists doing the same. And it depends on Americans in general keeping cool, notwithstanding the likely failure of many politicians and journalists to do their part.
It also worries me that this worries us so much, partly because it means people don't know math.  Look no further than the comments of Wright's article, which is full of people saying things that add up to "Lalalalalala My fingers are in my ears shut up about your stupid numbers it happened to this one person and I'm scared of it happening to me this is terrible somebody needs to DO SOMETHING!"

~ ~ ~ ~ ~

PS I would quibble with Wright's statement that "It’s pretty much impossible to anticipate all the bugs in a complex computer program," but only because he doesn't go far enough. Guaranteeing a computer program always works is impossible -- full stop. See Rice's Theorem, or for a more concise description, the Formal Methods section of the Wikipedia article on Static Code Analysis.

No comments:

Post a Comment