Monday, March 15, 2010

Toyota's Sudden Acceleration Problem

I have a little bit of a background in safety critical electronics and software design so I thought I would comment on this. Based on my current knowledge (Disclaimer: My knowledge is limited to what I know from news articles), given the condition that this problem is not transient (i.e. once the problem occurs, the problem stays with the vehicle and does not disappear after a power cycle), it is highly unlikely that the root cause is a software or electronic defect. If it was a software design defect, the issue would be permanent and reproducible on any given vehicle (since the software is the same). This is not the case. If it was a electrical EMI issue, it is highly unlikely that the fault would be a permanent fault. Rather, the user would feel a sudden increase in acceleration but then back to normal operation moments later. This is not the case either.

Here is my advice to Toyota engineer's on how to present this case to the public:

1. Be totally transparent.
2. Describe the process by which you use to derive root cause.
3. Provide the statistics and evidence gathered in a succient and understandable form that follows the process described.
4. Show how this evidence supports the root cause hypothesis.

I think if this was displayed in a clear and direct manner, rather than in the form of a mea culpa or obligatory advert (e.g. a Toyota engineer saying that she is confident of the vehicle, blah blah), Toyota would go a long way in reducing the amount of confusion and noise on this topic.

No comments:

Post a Comment