Quote of the Day: Accidents

 

There’s an old saying: “If you think safety is expensive, try an accident.”

The first choice after an accident is to say “How can we improve the design so this can’t happen? How can we remove the opportunity for errors?”

In a well-designed plant, a simple error like closing the wrong valve shouldn’t result in an accident.

For a long time, most people were saying that most accidents were due to human error. This is true in a sense, but it is not very helpful. It’s like saying falls are due to gravity.

— Trevor Kletz (1922-2013)

Trevor Kletz was the founder of process safety management/engineering. Process Safety deals with major accidents and defects in the process, as opposed to accidents on a personal scale. The ideal process is designed to be resilient and capable of dealing with human error.

When you change a process, you need to carefully consider how the change affects the other processes at the plant. If you don’t, a seemingly minor change could have terrible results. Replace that damaged reactor with stretch of pipe with expansion bellows, it will let you keep the system running; the result was Flixborough. Deactivate the flare stack and scrubber while the plant is shut down – it should not be a problem. Don’t tell workers about the potential for catastrophic events if the chemicals react with water. The result was Bhopal.

I highly recommend his book What Went Wrong? Case Studies of Process Plant Disasters to anyone who works in manufacturing, processing, etc. Rent it from your local library, as the Amazon price is insane. It’s written for the operator and foreman, with clear descriptions of how things can go off the rails. At the end of the book, he discusses is how analysis of change is applicable outside of the chemical industry. ICI corporation engaged in massive acquisitions without considering the effects on their viability. The result was bankruptcy and the company getting carved up.

You could say that the field of process safety is inherently conservative, in the broad sense. The famous Chestertonian Fence is basically a management of change review. Conservatism is the reminder that new does necessarily mean improved. That new program might just make things worse or waste money to no effect. Why, yes, despite our health care system having problems, we could get worse. Much worse. It is no accident that liberal plans tend to fail – they do not stop to consider the unintended consequences. That also could describe a good portion of the disasters in What Went Wrong?

The Constitution is designed deliberately to be resilient against minor errors, and remain enduring. Conservative policy should be designed similarly — carefully crafted to be resistant to mistakes, corruption, and leftists.

There are 14 comments.

Become a member to join the conversation. Or sign in if you're already a member.
  1. Vectorman Inactive
    Vectorman
    @Vectorman

    OmegaPaladin: The ideal process is designed to be resilient and capable of dealing with human error. When you change a process, you need to carefully consider how the change effects the other processes at the plant.

    Why can’t Communists, Socialists, Leftists, and wacko “Liberals” understand the same effect with government? If people were “perfect,” then a simple but resilient government would be all that is needed.


     

    • #1
  2. OmegaPaladin Moderator
    OmegaPaladin
    @OmegaPaladin

    If you are curious about process safety, I recommend the US Chemical Safety Board videos.  They are excellent documentaries with decent CGI animations of what happens. https://www.youtube.com/user/USCSB

    Here is an example that features Illinois corruption:


    Join other Ricochet members by submitting a Quote of the Day post, the easiest way to start a fun conversation. There are many days available on the February Signup Sheet   We even include tips for finding great quotes, so choose your favorite quote and sign up today!

    • #2
  3. Steve C. Member
    Steve C.
    @user_531302

    I’ve always thought there should be a federal agency, the Bureau of Unintended Consequences. It would publish an annual report titled, “We Told You So”.

     

    • #3
  4. Arahant Member
    Arahant
    @Arahant

    Process and change management. You’re singing my songs.

    • #4
  5. Stad Coolidge
    Stad
    @Stad

    Arahant (View Comment):

    Process and change management. You’re singing my songs.

    Change control and configuration management are crucial elements in project management.  CM involves what we call the three Ps: plant, procedures, and people.  Change control is how we make sure any change to a P is well thought out, evaluated, officially approved, and properly implemented.

    • #5
  6. Tex929rr Coolidge
    Tex929rr
    @Tex929rr

    There has a been a push in public safety for the last few years to no longer refer to motor vehicle crashes as accidents.  The acceptance term now is crash, because generally the reason isn’t accidental, but rather some bad choice.  I find it kind of interesting, because I don’t know how it started but it’s pretty widespread now.

    We all still refer a crash as an “MVA” on the radio.

    • #6
  7. Gumby Mark (R-Meth Lab of Demo… Thatcher
    Gumby Mark (R-Meth Lab of Demo…
    @GumbyMark

    Terrific post.  We spent a lot of time on process safety, change management and improving our accident investigation process to get at underlying system problems and beyond human error.  And I like the comparison to public policy making.

    One of the things triggering my defection from liberalism was no matter what problems happened with a policy or program the solution was always the same – spend more money, more laws, more regulations, better regulators.  There never seemed to be any deeper reflection about why problems were occurring.  Liberalism turned out to be remarkably bad at learning from experience.  And then it got worse.

    • #7
  8. GLDIII Temporarily Essential Reagan
    GLDIII Temporarily Essential
    @GLDIII

    Tex929rr (View Comment):

    There has a been a push in public safety for the last few years to no longer refer to motor vehicle crashes as accidents. The acceptance term now is crash, because generally the reason isn’t accidental, but rather some bad choice. I find it kind of interesting, because I don’t know how it started but it’s pretty widespread now.

    We all still refer a crash as an “MVA” on the radio.

    Pilots have always joke that the FAA & NTSB has the accident reports pre filled out with “Pilot Error”. The rest of the form is just for the particulars.

    • #8
  9. Steve C. Member
    Steve C.
    @user_531302

    GLDIII Temporarily Essential (View Comment):

    Tex929rr (View Comment):

    There has a been a push in public safety for the last few years to no longer refer to motor vehicle crashes as accidents. The acceptance term now is crash, because generally the reason isn’t accidental, but rather some bad choice. I find it kind of interesting, because I don’t know how it started but it’s pretty widespread now.

    We all still refer a crash as an “MVA” on the radio.

    Pilots have always joke that the FAA & NTSB has the accident reports pre filled out with “Pilot Error”. The rest of the form is just for the particulars.

    It usually is. Isn’t it?

    They didn’t coin “Controlled flight into terrain”, for no reason.

    • #9
  10. Mark Camp Member
    Mark Camp
    @MarkCamp

    Gumby Mark (R-Meth Lab of Demo… (View Comment):
    Liberalism turned out to be remarkably bad at learning from experience.

    I think that it’s been very good at learning from experience, according to its fundamental economic theory, which is called “economic positivism” or “economic empiricism”.

    Not so according to the competing theory, which is called “Austrian”, “logical”, or “causal realist” economics, where

    • it’s not possible to learn true economic laws from experience
    • it’s unnecessary to try, because they can be known with certainty by common sense, so the question of whether the positivists have been good or bad at it is meaningless: the question has a false implicit assumption.
    • if assertions are tentatively supported by the results of an experiment, but contradict what is known a priori to be true, then the tentative conclusion is false.

    Economic positivists like Krugman, the Fed board, and other establishment economists,  believe that economic laws can only be discovered by the same process as laws of the physical sciences: creating hypotheses and testing them in an infinite series of experiments on society as a whole, where it is known beforehand that the effects of the test variable will be more or less swamped by variables which one hasn’t controlled for.

    One is good at learning, under this doctrine, if one keeps tweaking his one’s collectivist/totalitarian/elitist/utopian policies (one’s hypotheses) after each experiment and performing another poorly controlled and thus not-very-conclusive experiment, ad infinitum.

    That is exactly what the opponents of economic liberalism (whom you refer to as “liberals”!) do.

    • #10
  11. Stad Coolidge
    Stad
    @Stad

    Tex929rr (View Comment):
    The acceptance term now is crash, because generally the reason isn’t accidental, but rather some bad choice.

    It’s been done in criminal justice for years:

    “My client is not a bank robber.  He just made some bad choices in life.”

    • #11
  12. CarolJoy, Above Top Secret Coolidge
    CarolJoy, Above Top Secret
    @CarolJoy

    Just remarking “Thank You!” so I can find my way back to this topic later on.

    • #12
  13. Reese Member
    Reese
    @Reese

    CarolJoy, Above Top Secret (View Comment):

    Just remarking “Thank You!” so I can find my way back to this topic later on.

    Me too.  We have a “safety minute” fad in routine meetings where I work.  The Kletz quote will be fine to get me through when it’s my turn. 

    • #13
  14. Arahant Member
    Arahant
    @Arahant

    So, we have to keep commenting to keep it alive so y’all can find it? Wouldn’t it be easier to bookmark it?

    • #14
Become a member to join the conversation. Or sign in if you're already a member.

Comments are closed because this post is more than six months old. Please write a new post if you would like to continue this conversation.