More "Bell the Hybrid" nonsense

No, Bob, unfortunately, you read that right. The study says, in effect, "advocates have told is there is a problem, other countries have also said there might be a problem, SAE considered it, seems reasonable on the face of it, so we're going to fix it."

They actually cite my study, then go on to say that, hey, sure we want to analyze the all-accident data, but here are all the reasons why analysis of the full analysis of all crashes is a) incomplete, b) hard to do. And the kiss of death, they are "waiting for more data" to be able to analyze the issue.

So in the meantime, rather than see if there is any objective basis, sure enough, they are going to plow right on ahead to determine how to fix the problem.

But it's not actionable. There's enough of a sop to objective analysis of actual accident rates that they can claim to have covered themselves.

So they are going to test whether it's harder to hear quiet cars than noisy cars.

And, of course, guess which one is true: 1) they are going to test several types of cars for noise level, to determine the true extent of the presumed problem, or 2) they simply know that only hybrids are quiet enough to present a threat, so what the hey, they are only going to test hybrids.

Hint: Which approach represents bad science, 1) testing only the answer you want to get, or 2) testing alternatives that might contradict your priors?

Answer, they're only going to test hybrids. So, even though the European tests posted here some months back show that the Prius is only in the top 10% of cars for quietness, they aren't going to address conventional cars that are as quiet as the Prius. Because -- take your pick -- either they don't know how that, they don't know how to design an experiment properly, or they don't want to end up suggesting that a bunch of higher-end cars will also require noisemakers under an objective standard for minimum noise.

Then, if I read that right, they are going to test whether quiet cars are harder to hear. Yep. But with great precision.

I see this kind of thing happen in health care sometimes. You'll get a civil servant or two who is convinced there is an issue, and you'll get one hard-headed solution, and absolutely nothing will stop it unless somebody higher up in the organization has the understanding and the willpower to get alternatives considered. All with the best of intentions. And it's all done by the book, they take "input" from all parties as required, but ... it's a done deal. And, basically, it's what the Congress pretty much asked for, which in general is your safest strategy to take anyway, considering that the Congress controls your budget.

I particularly liked the part in the crash data analysis, where they said, in effect, we're so unlikely to find enough incidents that we can't get the information we're after from the crash data. At that point, my response was pretty much, if these events are so rare that you can't find enough of them in the universe of accident reports to be able to analyze them, then ... why the hell are you doing this? But the answer is, clearly, because advocates for the blind have asked them to, as has the US Congress.

 

My informal survey on vacation last summer says that there are many late model cars that are very hard to hear over normal background noise. Add one idling diesel delivery truck, and you won't be able to determine if there are moving vehicles nearby. It has nothing to do with being a hybrid.

 

They should conduct a study. Put twenty blind people in a large parking lot and have someone in a blind fold drive a Prius around the parking lot and count how many people are run over.

I've said it before, and I'll say it again... in all fifty states, it is THE DRIVER's responsibility to not run over pedestrians. If a driver runs over a pedestrian, the DRIVER is the one at fault and the one that goes to jail. Regardless of whether the car makes noise or not, IT IS THE DRIVER'S RESPONSIBILITY TO WATCH WHERE THEY ARE GOING AND NOT RUN OVER PEOPLE!!!!

 

The thing that I fear the most is the "bad science" that Chogan2 mentioned. Lawmakers do not have the time or resources to thoroughly read the legislation that they vote on, and most of them do not have the technical background to recognize "bad science".
Add to that the fact that most politicians are easily corruptable by outside organizations, influencing them to make decisions that are not truly in the public's best interest.

Applying one of my favorite "laws", the law of diminishing returns, here's my big-picture take...

1. Will the outcome (presumably fewer deaf pedestrian injuries/deaths) justify the total effort (manpower and expense) to "bell the hybrid"?

2. Are there other pedestrian risks that deserve more attention to solving, compared to mitigating the risk of the quiet hybrid? (example - reverse rollover deaths to children, mitigated by rear-view cameras). In other words, would solving another risk save more lives, or result in a better benefit/cost ratio?

From my own engineer/scientist point of view, studies absolutely better include ALL quiet vehicles, not just hybrids. And the risk study absolutely must account for situations of loud general background noise from a nearby source (eg- an idling diesel delivery truck can overpower the noise from almost any car)

 

Pretty much covers it for me. At some point I did a back-of-the-envelope for (say) 2015, based on the research I did earlier for Bob.

If 2 million hybrids per year require a $25 noisemaker, that's $50 million per year. What does that $50M buy us?

First, it does not buy any reduction in deaths. Looking at the DOT FARS (fatality analysis reporting system) data, pedestrian deaths occur almost exclusively from higher-speed impacts. At those speeds, the hybrid nature of the Prius is irrelevant. (And, to date, there has been no reported case of a blind pedestrian death from being struck by a hybrid).

Instead, it is being done potentially to save injuries.

Using a different national database (the Hospital Cost and Utilization Project data from the Agency for Healthcare Policy and Research), I found about 40 blind pedestrians admitted as hospital inpatients per year. Of those, most surely were not hit (and will not be hit) by hybrids, and most were probably hit at high enough speeds that the hybrid nature of the vehicle would have been irrelevant anyway.

Assuming that the risk from hybrids is at most only modestly higher than the risk from conventional vehicles (if it were hugely higher, we have some data on that by now), I'm guessing that, optimistically, we might avoid a hospitalization or two per year, by belling the hybrid.

I could not track the corresponding ER figures (to get at blind pedestrians admitted to ERs), for lack of a national dataset that I could get my hands on. I had to estimate.

Nationally, there are about 4 ER visits for every for every inpatient admission. If that held true here, then, by analogy, I'm guessing we're avoiding another 4 to 8 ER visits per year.

So, bottom line, my best guess from the available data, I see us spending $50M possibly to avoid 2 hospitalizations and 8 ER visits per year, in the short term. Seems like a relatively poor return to me.

 

Driver go jail after killing a pedestrian?

[rant]

Not in and around Seattle. Certainly not the religious leader and habitual traffic offender, with a cell phone glued to his ear, who killed the chief of staff of a city councilman. Not even the elderly man, with no known political connections, who slaughtered a schoolgirl just around the corner from my house. When three of the four traffic lanes had stopped at the crosswalk, how was he to know that pedestrians were crossing, and that it was unsafe to change into the open lane and blow through?

I learned to drive in an area where pedestrian right of way was strongly enforced. But here, just look at the relative level of jaywalking tickets to failure-to-yield-to-pedestrian tickets. Real-life right of way belongs to the biggest.

[/rant]

The existence of a law does not indicate any intent to actually use it.

 

Bob,

I can't tell if you're gonna' like this or not. This on the recent London
intro of the new all-electric Smart Car, the BRABUS:

"Positioning itself as a 'sports car' is more tongue-in-cheek than fact –
although it does look the business with sports suspension and
'Monoblock VII' alloy wheels in a five double-spoke design – with
Smart adding an artificially produced running sound to the car to
replace the steady hum of the electric engine.

"Interestingly, the makers have gone for a 'race car engine' sound but
this can be replaced by virtually any sound in due course.

"Of course it's a gimmick, but the electric car needs all the help it can
get to grab headlines in the UK."

Sounds like Mercedes is trying to get a leg up on hybrid/EV noise making
by pretending its all a game.

Or, are they poking fun at the whole idea of hybrid/EV noise making. They
could have a lot at stake as their gas/petrol cars are some of the quietest
I can just barely hear.

The whole article is here.

Note: BRABUS is apparently an intentional pseudo-reference to the "Barabus,"
a MB uber-super car:

Barabus (Not Brabus) Sports Car - 1005 hp, 270 mph

 

I seem to keep reading these articles, but I haven't seen any stories or examples of blind people being hit. I think their hearing is probably more sensitive sense they rely on it. If they install bells or some stupid "engine simulator" noise then it will be the first thing I rip off the car. I think getting rid of the noise is moving forward.

 

>I think getting rid of the noise is moving forward

I couldn't agree more... this is a side of the coin almost nobody has mentioned.

Assuming that hybrids are quieter (and pure electric cars are almost always naturally quieter) and assuming these vehicles are our future, then they offer a real chance for cutting down the noise pollution in our inner cities.

 

Bob,

Little did I now when I wrote that study that NHTSA staff were just going to sit on their hands for a few years. This is the classic contrast between a policy analyst (which is what I was) and a data analyst (which is what they are). The data analyst says, the information does not yet meet our typical quality standards. The policy analyst says, the Congress is going to make a decision anyway, tell me what you know and how much confidence you have in it.

Somebody really needs to prod them to release whatever they can infer from their all-incidents databases (not just fatal accidents data). Even if all they can say is that they can only find an extremely low count of relevant accidents, that, by itself, officially said in writing, would be useful information.

You'd think the Congress would be asking these questions. I sure handled enough calls for information on health-related topics when I worked for the government. Was glad to, considering that the alternative was to have the Congress just take a guess.

But this one? There's almost zero federal budget implication, so there's not even the check from the agencies who total up the cost of the legislation. I think this one's going to go based on a vague feeling that they are doing good.

Chris