What does "safe" mean to you? 6

[Haf]

As engineers, it is expected that we hold paramount the safety of the general public. Everyone has heard the expression "safety first."

One definition for "safe" (Webster's) is "free from harm or risk." I would argue that no product or process whatsoever is completely free from harm or risk.

So how do we really decide if a product or process is safe? Obviously, at some point, you have to accept some level of risk. How do we decide what level of risk is acceptable?

With some products or processes, there are codes or standards that must be met for safety, which can make our jobs easier. But did you ever stop to think where those codes and standards came from? Who developed them and what was their reasoning? More importantly, do you agree with their reasoning?

Anyway, before I ramble too much (maybe it's too late), I'd be interested to hear (read) peoples' thoughts on this.

Haf

 

[IRstuff]

What's sometimes done, as in the case of the Pinto, is a calculated cost-benefit margin.

In other cases, such as the Space Shuttle, there was a calculated and acceptable level of risk.

In yet other cases, such as when we speed or do something stupid on the freeway, no consideration is taken for level of risk, even though we may consciously be aware of the risk.

In yet other cases, such as when we wake up, we simply ignore the risk and the possibility of risk altogether.

TTFN

 

[jetmaker]

Nice post IRstuff.

 

[PSE]

To IRstuff's post I would add the caveat of doing the cost-benefit/risk analysis based upon an established performance envelope. Using the Pinto as an example, I would contend that it is reasonable to assume that auto accidents may occur. Therefore, the design should be capable of accommodating accidents to a certain level of severity. While I do not know the requirements for the Pinto, it appears that it will be remembered historically as insufficient. If the explosions and fires had been caused by accidents with semi's or trains, perhaps a different result.

The shuttles appeared to be one case of use outside of the established performance envelope (launch in conditions too cold for the O-rings seals), and a case of reality proving that "theory" is exactly that. Theory: Foam impact should not cause damage. Modified Theory: Foam impact might cause damage. Reality: Self explanatory.

Note that these are my opinions only. For any design, there is a need to accomodate an expected performance envelope. With in this realm, we are bound ethically. Use or misuse outside of this envelope is beyond the control of designers and engineers.

Regards,

 

[GregLocock]

You are conflating two different accidents.

The O ring problem ,as Feynman showed, was that the system was routinely being stressed beyond its design limits.

I don't think anyone has really demonstrated what happened with the foam, yet, to the same sort of standard.






Cheers

Greg Locock

 

[JOM]

Haf,

I think your post implies the engineer knows the risks associated with the product or process. There is a clear responsibility for the people who use the product or are employed to operate the process are advised of the risks.

There are plenty of court cases that determined this did not happen. That's not answering your question, but ensuring people know the risks is part of the equation?



J.

 

[CanEngJohn]

Whether or not we agree with the reasoning is insignificant. The code or standard is now law and we must abide by it.

I believe a good example of this is the OHSA in Ontario. The approach is taken that an operator will at some point go out of their way to do something that will injure themself. The safety of a machine being operated is approached from this standpoint and every effort is made to ensure that it is impossible for the operator to injure themself on the equipment even if they are trying to injure themself.
ie. guarding must go low enough and light curtains mounted low enough that an operator cannot reach under them and touch moving equipment. The opposite is true as well.

In our manufacturing environment if I say "well it looks safe enough..." I know more work needs to be done.

In terms of the general public I believe the term is called due diligence. As an Engineer you are responsible for assuming that an item will be pushed beyond its design limits by the general public in its normal application. You are expected to design for this (translate as factor of safety).

For unusual applications not forseen by the Engineer you sometimes must demonstrate that it should have been obvious that the item would fail in this application (translate as the guy was an idiot for trying that and should have consulted an Engineer).

The second item rarely occurs but there have been two cases in which it was required to demonstrate that the individual should have known to consult with an Engineer.

All cases are relevant to Ontario Law and may not be the same where you live.

So safe to me means Darwin is spinning in his grave.

 

[IRstuff]

The problem is that the "law" is insufficiently precise to allow an engineer to make that type of design decision. Will you allow 3 interlocks to fail? 4? 5? 6?

Do you design a $100,000 Pinto that's absolutely safe and too expensive for your "general public" to buy?

It's nice to think that engineering is clear cut, but it's not. You have a certain price range that you KNOW your customer is willing to pay. You then fit all the required features into the design and find that it's double the price and you still haven't gotten it perfectly safe.

Perfect safety is a myth and to buy into that notion and ultimately deny your customers a useful product because you want to design for the 10th sigma is a disservices to the public.

The bottom line is that the public makes and allows some level of risk and couples that with a firm belief that it won't happen to them. Otherwise, you'd never get out of bed.

Don't want to start a flame war, but consider that MILLIONS of Americans still smoke, even though the occurence and mortality rate from cancer is absurdly high. Clearly, the smoking public makes a trade between cost and benefit as well.

TTFN

 

[GregLocock]

Just a thought, you do realise that there is a safety critical system in a car that has NO redundancy, no backup, and if it fails the result will be unnerving at best, and catastrophic in many cases?

This is quite an interesting case to me, when I am thinking about system design. How do you design a practical system that quite simply cannot be allowed to fail?



Cheers

Greg Locock

 

[IFRs]

I see "safe" as:
1) complying with minimum design loads and criteria, as dtermined by long term observation and experience.
2) including some redundancy in the design can increase the degree of "safeness" but is often not a quantifiable change.
3) something is "safe" if it conforms to generally accepted practice which assumes a certain (again not quantifiable) amount of risk that has been determined as acceptable.
4) if I design a device and include a manual of operation, it is "safe" if operated within those boundaries. I have communicated clearly what those boudaries are.
5) My designs are not made in a vacuum, but with input from our procurement, fabrication, installation, sales and legal departments. Everyone contributes and has veto power. Engineering has the last veto for anything not cosmetic.
6) If design guidelines are not available then full-scale testing is often the only way to know.
7) I'm sorry - life is risky!!

PS - I think Greg is referring to the Human Driver.

 

[IRstuff]

Depends on the expected failures and number thereof.

If it were a purely mechanical system, you cuold certainly overdesign by what you consider to be an adequate margin. Electronics are often dealt with through redundancy.

But, in the end, your system would neither be practical nor simple.

From a statistical perspective, the concept of "not" or "impossible" is anathema. Consider the humble bolt. How would you keep in from failing and under what set of conditions? You'd probably wind up with a bolt that's 3 times the size and weight that you'd otherwise use.

TTFN

 

[GregLocock]

The steering wheel, column, intermediate shaft and rack are all sub systems that have no redundancy and are safety critical.

Cheers

Greg Locock

 

[IRstuff]

So's the driver ;-) and I'd guess that more drivers than steering columns have failed

TTFN

 

[StressGuy]

In that case, your back up system is the seatbelt and airbags.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

All opinions expressed here are my own and not my company's.

 

[etch]

Safe is to me

FIT FOR PURPOSE

 

[IRstuff]

That seems to be more indirection, since now we need to know what "fit" means and by how much.

The original Shuttle SRBs would be "fit" for their purpose of launching the Shuttle, but they were clearly unsafe in certain conditions.

TTFN

 

[Fizzhead]

How safe is safe? How red is red? Obviously something one cannot specifically define in any legal document or engineering guideline. What I accept as safe (say driving 90 on the interstate) may be perceived as horrifyingly risky to another. And that person's perception can and does change with the situation. You might be very accepting of my driving 90 if I am taking you to the hospital. This is a subject that keeps lawyers occupied every day. An engineer designs something, it's used in an unanticipated way, or in a way that, at least to others, seems to go against common sense, and something bad happens. Along comes a lawyer and bingo! Lawsuit and another long winded debate on the "standard of care" that should be employed by engineers. Unfortunately, nothing of lasting value (except maybe for the attorney and plaintiff) is accomplished with all of this. Problem is that you can't define it. You have to rely on professional judgement and that is always subject to question by others. It's a never ending circle.

Don't get me wrong, I beleive that we need product liability laws etc. to protect the general public because our profession does have it's bad apples. But, when we see hair dryers with the instruction "Do not use while asleep", or "Do not use in the bathtub or shower" don't you just wonder a bit if we are not somehow short circuiting evolution? The woman who spilt hot coffee on her lap while going through the drive through and then sued McDonalds, her lawyer and the judge who let such a judgement out the door, absolved us from having to be responsible for own unwise actions. They and others opened a Pandora's box that we will never get closed. Common sense has escaped us and will not be easily regained. The points made here about over designing and redundancy are good ones. However, do any of us think that such measures can protect us from lawsuits in today's world? I remember in college hearing a professor say "the problem with trying to make something fool-proof is that the world is always busy inventing a better fool". Good words to live by on this issue I think. To my mind, the real problem is that Americans, as appears to be our nature, have taken a good concept and pursued it to the bitter end. Will we ever learn?

 

[IRstuff]

But, I think that's why there should be a Hell for designers who design soap dispensers in bathrooms that are unusable.

Common sense, likewise, should be applied to designers and suppliers. Up until recently, most cars did not have cupholders as a built-in feature. When I was in college and driving home on Coke and No-Doze, I didn't have a cup holder and holding the can between my legs was a normal occurrence. Even ignoring that aspect, a car is a moving vehicle, subject to sudden or unintended movements, and allowances should be made for a higher probability of spillage. Even ignoring that aspect, it was shown that the McDonald's coffee was kept at so high a temperature than even a normal spillage would have potentially caused severe burns.

McDonalds was clearly negligent in providing coffee that did not need to be that hot, particularly in the drive-through.

The jury award is a separate issue.

TTFN

 

[profengmen]

Anyone that drives 90 mph on the interstate is equally as stupid as a person the uses a blow dryer in the bathtub.

 

[ewh]

There must be a whole lot of stupid people in Atlanta, then. A few years ago, a traffic study was done on the loop around town, and the AVERAGE rush hour speed was 85 mph.
In good, dry conditions, with light traffic, in a sound vehicle, 90 mph on an interstate is less dangerous than 35 mph.