Re: [PSES] Risk assessment versus HBSE

[Mark Ortlieb]

All,

I appreciate the responses to this topic, and yes I'm going to spend 
some time with that study. Doug, yes, the AGV scenario you described is 
exactly the type of situation I'm doing research for, except for AMPs 
(autonomous mobile platforms) and in a commercial environment where 
children could be present. There is some data out there for impact with 
adults in industrial environments, but precious little... well, nothing 
that I've found so far... involving contact with children.


Thanks,

Mark

On 2/23/22 8:01 PM, Douglas E Powell wrote:

Looks very interesting, I'll have to fit it into my reading list

Doug



On Wed, Feb 23, 2022 at 5:42 PM Richard Nute  wrote:

Hi Doug:

See:


https://www.tandfonline.com/doi/pdf/10.1080/01446193.2016.1274418?needAccess=true

Best regards,

Rich

*From:* Douglas Nix 
*Sent:* Wednesday, February 23, 2022 8:57 AM
*To:* Richard Nute 
*Cc:* EMC-PSTC@LISTSERV.IEEE.ORG
*Subject:* Re: [PSES] Risk assessment versus HBSE

Hi Rich,

Your points are well taken.

There are some good ISO standards that relate to exposure to hot
and cold temperatures. These standards take the type of material,
and therefore the transfer rate, of heat into account and are
quite useful IMO.

You are right about the HBSE model, and I think that we need both
approaches, that is HBSE and “conventional” risk assessment. The
issue is always this: what do you do when you don’t have any data?
Practitioners must start with what they know, and that usually
means starting with qualitative risk assessments. This is
especially true in the broader OHS sector where these types of
assessments are used for workplace inspections. The problem is
that when we attach arbitrary numeric values to qualitative scales
people start to believe that the math is somehow “right”
regardless of how arbitrary in the input data. Even if the math is
correct, GIGO. This is what plagues the application of
conventional risk assessment techniques and why Cox, Quintino and
others have been raising the alarm for so many years.

We’re now in a place where the EU has a semi-quantitative (yes I
hate that term too, but it’s descriptive) risk assessment tool
built into the RAPEX directive for use on consumer products and
white goods. This same tool has been adopted by the US CPSC,
although they are not making this widely known. Health Canada has
their own methodology for the same purpose, and I’m quite sure
that if we were to examine the methods used by the national health
and safety agencies in any country that has such an entity, we
would find that they too have some method like the RAPEX/CPSC or
Health Canada methods. So, for now we are stuck with what we have.
At least we are getting manufacturers to think about risk, rather
than “just” hazards.

Progress comes slowly…

--

Doug Nix

d...@mac.com

(519) 729-5704

"Opportunity is missed by most people because it is dressed in
overalls and looks like work." - Thomas A. Edison



On 22-Feb-22, at 19:03, Richard Nute  wrote:

Hi Doug:

Thanks for your comments.

What bothers me about risk assessment is that the committees
that have written the standards requiring risk assessment have
not critically evaluated the risk assessment process.  If they
had done so, we would not have the process as we know it today.

Actually, I do not fully agree with the Gibson finding that
energy causes injury.  I can show that the injury parameter is
energy per unit time, e.g., joules/second.  The body can
absorb energy slowly without injury, but not quickly. Consider
that a car with people in it can brake or stop without causing
injury to the passengers, but cannot “crash” to a stop where
injury is likely.  In both cases, the kinetic energy to stop
is the same, but the kinetic energy per time to stop is low in
braking, but high in crashing.

The attached picture is that of catching three objects and
assumes the deceleration time is the same for each object. 
Note that when we catch an object, we can catch it “slowly”
and distribute the energy over a longer time than catching it
“directly.”  I submit this as proof that energy per unit time
is the parameter that causes injury.

HBSE does indeed have (or can have) energy criteria for each
form of energy.  However, I agree that some energy data is not
readily available and must be researched.  And, using the
energy model can be quite complex.  For example, injury from
thermal energy is often simply taken as accessible
temperature, sometimes including a time of contact.  Using a
single parameter, temperature

Re: [PSES] Risk assessment versus HBSE

[Douglas E Powell]

Looks very interesting, I'll have to fit it into my reading list

Doug



On Wed, Feb 23, 2022 at 5:42 PM Richard Nute  wrote:

>
>
>
>
> Hi Doug:
>
>
>
> See:
>
>
>
>
> https://www.tandfonline.com/doi/pdf/10.1080/01446193.2016.1274418?needAccess=true
>
>
>
> Best regards,
>
> Rich
>
>
>
>
>
> *From:* Douglas Nix 
> *Sent:* Wednesday, February 23, 2022 8:57 AM
> *To:* Richard Nute 
> *Cc:* EMC-PSTC@LISTSERV.IEEE.ORG
> *Subject:* Re: [PSES] Risk assessment versus HBSE
>
>
>
> Hi Rich,
>
>
>
> Your points are well taken.
>
>
>
> There are some good ISO standards that relate to exposure to hot and cold
> temperatures. These standards take the type of material, and therefore the
> transfer rate, of heat into account and are quite useful IMO.
>
>
>
> You are right about the HBSE model, and I think that we need both
> approaches, that is HBSE and “conventional” risk assessment. The issue is
> always this: what do you do when you don’t have any data? Practitioners
> must start with what they know, and that usually means starting with
> qualitative risk assessments. This is especially true in the broader OHS
> sector where these types of assessments are used for workplace inspections.
> The problem is that when we attach arbitrary numeric values to qualitative
> scales people start to believe that the math is somehow “right” regardless
> of how arbitrary in the input data. Even if the math is correct, GIGO. This
> is what plagues the application of conventional risk assessment techniques
> and why Cox, Quintino and others have been raising the alarm for so many
> years.
>
>
>
> We’re now in a place where the EU has a semi-quantitative (yes I hate that
> term too, but it’s descriptive) risk assessment tool built into the RAPEX
> directive for use on consumer products and white goods. This same tool has
> been adopted by the US CPSC, although they are not making this widely
> known. Health Canada has their own methodology for the same purpose, and
> I’m quite sure that if we were to examine the methods used by the national
> health and safety agencies in any country that has such an entity, we would
> find that they too have some method like the RAPEX/CPSC or Health Canada
> methods. So, for now we are stuck with what we have. At least we are
> getting manufacturers to think about risk, rather than “just” hazards.
>
>
>
> Progress comes slowly…
>
>
>
> --
>
> Doug Nix
>
> d...@mac.com
>
> (519) 729-5704
>
>
>
> "Opportunity is missed by most people because it is dressed in overalls
> and looks like work." - Thomas A. Edison
>
>
>
> On 22-Feb-22, at 19:03, Richard Nute  wrote:
>
>
>
>
>
>
>
> Hi Doug:
>
>
>
> Thanks for your comments.
>
>
>
> What bothers me about risk assessment is that the committees that have
> written the standards requiring risk assessment have not critically
> evaluated the risk assessment process.  If they had done so, we would not
> have the process as we know it today.
>
>
>
> Actually, I do not fully agree with the Gibson finding that energy causes
> injury.  I can show that the injury parameter is energy per unit time,
> e.g., joules/second.  The body can absorb energy slowly without injury, but
> not quickly.  Consider that a car with people in it can brake or stop
> without causing injury to the passengers, but cannot “crash” to a stop
> where injury is likely.  In both cases, the kinetic energy to stop is the
> same, but the kinetic energy per time to stop is low in braking, but high
> in crashing.
>
>
>
> The attached picture is that of catching three objects and assumes the
> deceleration time is the same for each object.  Note that when we catch an
> object, we can catch it “slowly” and distribute the energy over a longer
> time than catching it “directly.”  I submit this as proof that energy per
> unit time is the parameter that causes injury.
>
>
>
> HBSE does indeed have (or can have) energy criteria for each form of
> energy.  However, I agree that some energy data is not readily available
> and must be researched.  And, using the energy model can be quite complex.
> For example, injury from thermal energy is often simply taken as accessible
> temperature, sometimes including a time of contact.  Using a single
> parameter, temperature, or including time of contact parameter, does not
> address the difference between an aluminum block and aluminum foil (which
> is the issue some members of IEC TC108/HBSDT are addressing).   Or the
> difference between an aluminum block and a plastic block.
>
>
>
> Best regards,
>
> Rich

Re: [PSES] Risk assessment versus HBSE

[Richard Nute]

 

 

Hi Doug:  

 

See:

 

https://www.tandfonline.com/doi/pdf/10.1080/01446193.2016.1274418?needAccess=true

 

Best regards,

Rich 

 

 

From: Douglas Nix  
Sent: Wednesday, February 23, 2022 8:57 AM
To: Richard Nute 
Cc: EMC-PSTC@LISTSERV.IEEE.ORG
Subject: Re: [PSES] Risk assessment versus HBSE

 

Hi Rich,

 

Your points are well taken.

 

There are some good ISO standards that relate to exposure to hot and cold 
temperatures. These standards take the type of material, and therefore the 
transfer rate, of heat into account and are quite useful IMO.

 

You are right about the HBSE model, and I think that we need both approaches, 
that is HBSE and “conventional” risk assessment. The issue is always this: what 
do you do when you don’t have any data? Practitioners must start with what they 
know, and that usually means starting with qualitative risk assessments. This 
is especially true in the broader OHS sector where these types of assessments 
are used for workplace inspections. The problem is that when we attach 
arbitrary numeric values to qualitative scales people start to believe that the 
math is somehow “right” regardless of how arbitrary in the input data. Even if 
the math is correct, GIGO. This is what plagues the application of conventional 
risk assessment techniques and why Cox, Quintino and others have been raising 
the alarm for so many years.

 

We’re now in a place where the EU has a semi-quantitative (yes I hate that term 
too, but it’s descriptive) risk assessment tool built into the RAPEX directive 
for use on consumer products and white goods. This same tool has been adopted 
by the US CPSC, although they are not making this widely known. Health Canada 
has their own methodology for the same purpose, and I’m quite sure that if we 
were to examine the methods used by the national health and safety agencies in 
any country that has such an entity, we would find that they too have some 
method like the RAPEX/CPSC or Health Canada methods. So, for now we are stuck 
with what we have. At least we are getting manufacturers to think about risk, 
rather than “just” hazards.

 

Progress comes slowly…

 

--

Doug Nix

d...@mac.com <mailto:d...@mac.com> 

(519) 729-5704

 

"Opportunity is missed by most people because it is dressed in overalls and 
looks like work." - Thomas A. Edison





On 22-Feb-22, at 19:03, Richard Nute mailto:ri...@ieee.org> > 
wrote:

 

 

 

Hi Doug:

 

Thanks for your comments.

 

What bothers me about risk assessment is that the committees that have written 
the standards requiring risk assessment have not critically evaluated the risk 
assessment process.  If they had done so, we would not have the process as we 
know it today.  

 

Actually, I do not fully agree with the Gibson finding that energy causes 
injury.  I can show that the injury parameter is energy per unit time, e.g., 
joules/second.  The body can absorb energy slowly without injury, but not 
quickly.  Consider that a car with people in it can brake or stop without 
causing injury to the passengers, but cannot “crash” to a stop where injury is 
likely.  In both cases, the kinetic energy to stop is the same, but the kinetic 
energy per time to stop is low in braking, but high in crashing.  

 

The attached picture is that of catching three objects and assumes the 
deceleration time is the same for each object.  Note that when we catch an 
object, we can catch it “slowly” and distribute the energy over a longer time 
than catching it “directly.”  I submit this as proof that energy per unit time 
is the parameter that causes injury.

 

HBSE does indeed have (or can have) energy criteria for each form of energy.  
However, I agree that some energy data is not readily available and must be 
researched.  And, using the energy model can be quite complex.  For example, 
injury from thermal energy is often simply taken as accessible temperature, 
sometimes including a time of contact.  Using a single parameter, temperature, 
or including time of contact parameter, does not address the difference between 
an aluminum block and aluminum foil (which is the issue some members of IEC 
TC108/HBSDT are addressing).   Or the difference between an aluminum block and 
a plastic block. 

 

Best regards,

Rich

 

 

 

From: Douglas Nix mailto:d...@mac.com> > 
Sent: Monday, February 14, 2022 12:46 PM
To: Richard Nute mailto:ri...@ieee.org> >
Cc: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> 
Subject: Re: [PSES] Risk assessment versus HBSE

 

Hi Rich,

 

I have to admit that I’ve been thinking about your reply all weekend.

 

As you know, I teach machinery risk assessment and consult in this area 
regularly. I want to stipulate that there are some significant issues with risk 
assessment the way it is most commonly applied in industry, see my list of 
references on this topic at the end of my message. 

 

The inherent subjectivity of ris

Re: [PSES] Risk assessment versus HBSE

[Douglas E Powell]

Mark,

I did something along those lines back in 2012 when a question came up
about the need for a finger guard on a low power fan. In order to better
understand the possibility of injury, I estimated the surface contact area
of the blade edge against a finger, rotational speed and mass of the fan
impeller in terms of mechanical energy. I think I converted to J/mm^2. In
any case, the result was quite low and in my opinion "safe". To prove it
out, I was willing to first try jamming  a #2 pencil, and then my finger
into the fan while running at full speed. So far, 10 years later, I still
have my fully intact finger, but sadly the pencil was eventually used or
otherwise lost.

-Doug

Douglas E Powell
Laporte, Colorado USA
doug...@gmail.com
LinkedIn <https://www.linkedin.com/in/coloradocomplianceguy/>

(UTC -07:00) Mountain Time (US-MST)


On Wed, Feb 23, 2022 at 4:04 PM Mark Ortlieb 
wrote:

> Hello all,
>
> I am keenly interested in this discussion on energy vs energy/time with
> respect to causing injury, especially from a physical contact perspective.
> A couple of other considerations that are worthy to be included in such a
> discussion, which would add another layer of complexity would be:
>
>- The geometry of the object that comes into contact with the body.
>Take a needle as an example, or a knife edge. Let's say the needle is very
>light (a gram or two) and moving very slowly (1 mm/hour), at some point
>severe injury could take place through piercing of skin. Real world
>scenario aside, this illustrates a different kind of potential harm that
>can be inflicted. I don't know if there is a practical way to incorporate
>such a factor into an equation (such as the Energy Transfer equation), but
>certainly it needs to be included as part of a thorough analysis.
>- The kind of "victim" person being considered, for example a child vs
>an adult. A child may not be able to withstand the same level of energy
>transfer as an adult, or may be subject to a different kind of injury as
>the result of an energy transfer, that is, a secondary injury, such as
>being more easily pushed down and hitting their head on the floor. In fact,
>I am interested to know if there are others out there who are aware of any
>data or studies regarding impact with children.
>
> Thanks,
>
> Mark
>
>
>
>
>
>
> Hi Doug:
>
>
>
> Thanks for your comments.
>
>
>
> What bothers me about risk assessment is that the committees that have
> written the standards requiring risk assessment have not critically
> evaluated the risk assessment process.  If they had done so, we would not
> have the process as we know it today.
>
>
>
> Actually, I do not fully agree with the Gibson finding that energy causes
> injury.  I can show that the injury parameter is energy per unit time,
> e.g., joules/second.  The body can absorb energy slowly without injury, but
> not quickly.  Consider that a car with people in it can brake or stop
> without causing injury to the passengers, but cannot “crash” to a stop
> where injury is likely.  In both cases, the kinetic energy to stop is the
> same, but the kinetic energy per time to stop is low in braking, but high
> in crashing.
>
>
>
> The attached picture is that of catching three objects and assumes the
> deceleration time is the same for each object.  Note that when we catch an
> object, we can catch it “slowly” and distribute the energy over a longer
> time than catching it “directly.”  I submit this as proof that energy per
> unit time is the parameter that causes injury.
>
>
>
> HBSE does indeed have (or can have) energy criteria for each form of
> energy.  However, I agree that some energy data is not readily available
> and must be researched.  And, using the energy model can be quite complex.
> For example, injury from thermal energy is often simply taken as accessible
> temperature, sometimes including a time of contact.  Using a single
> parameter, temperature, or including time of contact parameter, does not
> address the difference between an aluminum block and aluminum foil (which
> is the issue some members of IEC TC108/HBSDT are addressing).   Or the
> difference between an aluminum block and a plastic block.
>
>
>
> Best regards,
>
> Rich
>
>
>
>
>
>
>
> *From:* Douglas Nix  
> *Sent:* Monday, February 14, 2022 12:46 PM
> *To:* Richard Nute  
> *Cc:* EMC-PSTC@LISTSERV.IEEE.ORG
> *Subject:* Re: [PSES] Risk assessment versus HBSE
>
>
>
> Hi Rich,
>
>
>
> I have to admit that I’ve been thinking about your reply all weekend.
>
>
>
> As you know, I teach machinery risk assessment and consult in thi

Re: [PSES] Risk assessment versus HBSE

[Mark Ortlieb]

Hello all,

I am keenly interested in this discussion on energy vs energy/time with 
respect to causing injury, especially from a physical contact 
perspective. A couple of other considerations that are worthy to be 
included in such a discussion, which would add another layer of 
complexity would be:


 * The geometry of the object that comes into contact with the body.
   Take a needle as an example, or a knife edge. Let's say the needle
   is very light (a gram or two) and moving very slowly (1 mm/hour), at
   some point severe injury could take place through piercing of skin.
   Real world scenario aside, this illustrates a different kind of
   potential harm that can be inflicted. I don't know if there is a
   practical way to incorporate such a factor into an equation (such as
   the Energy Transfer equation), but certainly it needs to be included
   as part of a thorough analysis.
 * The kind of "victim" person being considered, for example a child vs
   an adult. A child may not be able to withstand the same level of
   energy transfer as an adult, or may be subject to a different kind
   of injury as the result of an energy transfer, that is, a secondary
   injury, such as being more easily pushed down and hitting their head
   on the floor. In fact, I am interested to know if there are others
   out there who are aware of any data or studies regarding impact with
   children.

Thanks,

Mark



Hi Doug:

Thanks for your comments.

What bothers me about risk assessment is that the committees that have 
written the standards requiring risk assessment have not critically 
evaluated the risk assessment process.  If they had done so, we would 
not have the process as we know it today.


Actually, I do not fully agree with the Gibson finding that energy 
causes injury.  I can show that the injury parameter is energy per 
unit time, e.g., joules/second.  The body can absorb energy slowly 
without injury, but not quickly.  Consider that a car with people in 
it can brake or stop without causing injury to the passengers, but 
cannot “crash” to a stop where injury is likely.  In both cases, the 
kinetic energy to stop is the same, but the kinetic energy per time to 
stop is low in braking, but high in crashing.


The attached picture is that of catching three objects and assumes the 
deceleration time is the same for each object.  Note that when we 
catch an object, we can catch it “slowly” and distribute the energy 
over a longer time than catching it “directly.”  I submit this as 
proof that energy per unit time is the parameter that causes injury.


HBSE does indeed have (or can have) energy criteria for each form of 
energy.  However, I agree that some energy data is not readily 
available and must be researched.  And, using the energy model can be 
quite complex.  For example, injury from thermal energy is often 
simply taken as accessible temperature, sometimes including a time of 
contact.  Using a single parameter, temperature, or including time of 
contact parameter, does not address the difference between an aluminum 
block and aluminum foil (which is the issue some members of IEC 
TC108/HBSDT are addressing).   Or the difference between an aluminum 
block and a plastic block.


Best regards,

Rich

*From:* Douglas Nix 
*Sent:* Monday, February 14, 2022 12:46 PM
*To:* Richard Nute 
*Cc:* EMC-PSTC@LISTSERV.IEEE.ORG
*Subject:* Re: [PSES] Risk assessment versus HBSE

Hi Rich,

I have to admit that I’ve been thinking about your reply all weekend.

As you know, I teach machinery risk assessment and consult in this 
area regularly. I want to stipulate that there are some significant 
issues with risk assessment the way it is most commonly applied in 
industry, see my list of references on this topic at the end of my 
message.


The inherent subjectivity of risk assessments that are performed 
without empirical data is unquestioned. The difficulty is that for 
many areas of human endeavour we have no empirical data, and try as we 
might we cannot calculate without numeric data. Nevertheless, we must 
be able to make risk-based decisions when designing products and 
equipment, and so we muddle along with the best tools that we have, 
hopefully while recognizing their flaws.


The HBSE model is a good one, and it fits machinery applications as 
readily as does risk assessment, however, the risk assessment methods 
that are used today have a history that stretches back to the 1960s, 
while the HBSE model is much younger. This doesn’t take away from HBSE 
in any way for me, but it does have an impact on the broader 
acceptance of the method since it is not yet as widely known as 
“conventional” risk assessment. None of the the standards in the 
machinery safety sector recognize the method as yet, so getting 
regulators and users to consider the method is a challenge.


HBSE also suffers from issues with lack of data when it comes to 
characterizing some hazards, leaving the 

Re: [PSES] Risk assessment versus HBSE

[Douglas Nix]

Hi Rich,

Your points are well taken.

There are some good ISO standards that relate to exposure to hot and cold 
temperatures. These standards take the type of material, and therefore the 
transfer rate, of heat into account and are quite useful IMO.

You are right about the HBSE model, and I think that we need both approaches, 
that is HBSE and “conventional” risk assessment. The issue is always this: what 
do you do when you don’t have any data? Practitioners must start with what they 
know, and that usually means starting with qualitative risk assessments. This 
is especially true in the broader OHS sector where these types of assessments 
are used for workplace inspections. The problem is that when we attach 
arbitrary numeric values to qualitative scales people start to believe that the 
math is somehow “right” regardless of how arbitrary in the input data. Even if 
the math is correct, GIGO. This is what plagues the application of conventional 
risk assessment techniques and why Cox, Quintino and others have been raising 
the alarm for so many years.

We’re now in a place where the EU has a semi-quantitative (yes I hate that term 
too, but it’s descriptive) risk assessment tool built into the RAPEX directive 
for use on consumer products and white goods. This same tool has been adopted 
by the US CPSC, although they are not making this widely known. Health Canada 
has their own methodology for the same purpose, and I’m quite sure that if we 
were to examine the methods used by the national health and safety agencies in 
any country that has such an entity, we would find that they too have some 
method like the RAPEX/CPSC or Health Canada methods. So, for now we are stuck 
with what we have. At least we are getting manufacturers to think about risk, 
rather than “just” hazards.

Progress comes slowly…

--
Doug Nix
d...@mac.com
(519) 729-5704

"Opportunity is missed by most people because it is dressed in overalls and 
looks like work." - Thomas A. Edison

> On 22-Feb-22, at 19:03, Richard Nute  wrote:
> 
> 
> 
> Hi Doug:
> 
> Thanks for your comments.
> 
> What bothers me about risk assessment is that the committees that have 
> written the standards requiring risk assessment have not critically evaluated 
> the risk assessment process.  If they had done so, we would not have the 
> process as we know it today.
> 
> Actually, I do not fully agree with the Gibson finding that energy causes 
> injury.  I can show that the injury parameter is energy per unit time, e.g., 
> joules/second.  The body can absorb energy slowly without injury, but not 
> quickly.  Consider that a car with people in it can brake or stop without 
> causing injury to the passengers, but cannot “crash” to a stop where injury 
> is likely.  In both cases, the kinetic energy to stop is the same, but the 
> kinetic energy per time to stop is low in braking, but high in crashing.
> 
> The attached picture is that of catching three objects and assumes the 
> deceleration time is the same for each object.  Note that when we catch an 
> object, we can catch it “slowly” and distribute the energy over a longer time 
> than catching it “directly.”  I submit this as proof that energy per unit 
> time is the parameter that causes injury.
> 
> HBSE does indeed have (or can have) energy criteria for each form of energy.  
> However, I agree that some energy data is not readily available and must be 
> researched.  And, using the energy model can be quite complex.  For example, 
> injury from thermal energy is often simply taken as accessible temperature, 
> sometimes including a time of contact.  Using a single parameter, 
> temperature, or including time of contact parameter, does not address the 
> difference between an aluminum block and aluminum foil (which is the issue 
> some members of IEC TC108/HBSDT are addressing).   Or the difference between 
> an aluminum block and a plastic block.
> 
> Best regards,
> Rich
> 
> 
> 
> From: Douglas Nix 
> Sent: Monday, February 14, 2022 12:46 PM
> To: Richard Nute 
> Cc: EMC-PSTC@LISTSERV.IEEE.ORG
> Subject: Re: [PSES] Risk assessment versus HBSE
> 
> Hi Rich,
> 
> I have to admit that I’ve been thinking about your reply all weekend.
> 
> As you know, I teach machinery risk assessment and consult in this area 
> regularly. I want to stipulate that there are some significant issues with 
> risk assessment the way it is most commonly applied in industry, see my list 
> of references on this topic at the end of my message.
> 
> The inherent subjectivity of risk assessments that are performed without 
> empirical data is unquestioned. The difficulty is that for many areas of 
> human endeavour we have no empirical data, and try as we might we cannot 
> calculate without numeric data. Nevertheless, we must be 

Re: [PSES] Risk assessment versus HBSE

[Douglas Nix]

Hi Douglas,

Glad I could help. :-)

If you are interested in some hard-core debunking of the risk matrix/decision 
tree approach, start with [17], [18], and [19]. The authors are not fans of 
these approaches as they do not hold up mathematically, even, as you say, many 
want them to seem mathematically rigorous.

WRT information for safety engineers on hazards, apart from the IEC standards 
related electric shock, and the ISO standards related to contact with hot and 
cold surfaces, I use the ACGIH TLVs and BEIs as my guidance. It doesn’t cover 
everything, but it covers all of the common hazards. 
https://portal.acgih.org/s/store#/store/browse/cat/a0s4W0g02f3QAA/tiles

I’m always happy to discuss any of this stuff, so feel free to reach out.

--
Doug Nix
d...@mac.com

“If there are no dogs in Heaven, then when I die I want to go where they went.” 
-Will Rogers



> On 14-Feb-22, at 18:21, Douglas E Powell  wrote:
> 
> Hi Doug,
> 
> This looks like a very good summary and mentions a few of the things I was, 
> in my poor attempt, trying to point out. One of my concerns about RA, and 
> FMEA in particular, is that this method does have a lot of numeric 
> computation for what is essentially a qualitative process. As such, it gives 
> the "appearance" of quantifiable due diligence, which many decision makers 
> want to see. Sad to say, when I've been involved in meetings with various 
> SMEs, I've occasionally heard discussions on how to adjust the parameters in 
> order to keep a particular risk listed on the spreadsheet but not trigger any 
> corrective actions. As with any system, there are those who would like to 
> manipulate it to their own advantage. It is for this reason that, when I am 
> leading the team effort and using a severity scale of 1 to 10, I always press 
> for mandatory action when severity is a 9 or 10 (disabling injury or worse), 
> regardless of the other parameters.
> 
> Oh, and that's a nice bibliography as well. I think I just received my summer 
> reading list.
> 
> Best,  Doug
> 
> Douglas E Powell
> Laporte, Colorado USA
> LinkedIn 
> 
> (UTC -07:00) Mountain Time (US-MST)
> 
> 
> On Mon, Feb 14, 2022 at 1:45 PM Douglas Nix 
> <0bb8ff993b10-dmarc-requ...@listserv.ieee.org 
> > wrote:
> Hi Rich,
> 
> I have to admit that I’ve been thinking about your reply all weekend.
> 
> As you know, I teach machinery risk assessment and consult in this area 
> regularly. I want to stipulate that there are some significant issues with 
> risk assessment the way it is most commonly applied in industry, see my list 
> of references on this topic at the end of my message.
> 
> The inherent subjectivity of risk assessments that are performed without 
> empirical data is unquestioned. The difficulty is that for many areas of 
> human endeavour we have no empirical data, and try as we might we cannot 
> calculate without numeric data. Nevertheless, we must be able to make 
> risk-based decisions when designing products and equipment, and so we muddle 
> along with the best tools that we have, hopefully while recognizing their 
> flaws.
> 
> The HBSE model is a good one, and it fits machinery applications as readily 
> as does risk assessment, however, the risk assessment methods that are used 
> today have a history that stretches back to the 1960s, while the HBSE model 
> is much younger. This doesn’t take away from HBSE in any way for me, but it 
> does have an impact on the broader acceptance of the method since it is not 
> yet as widely known as “conventional” risk assessment. None of the the 
> standards in the machinery safety sector recognize the method as yet, so 
> getting regulators and users to consider the method is a challenge.
> 
> HBSE also suffers from issues with lack of data when it comes to 
> characterizing some hazards, leaving the user to estimate the 
> characteristics. This brings in the biases of the person(s) doing the 
> estimating just as surely as conventional risk assessment methods.
> 
> The absence of a probability parameter in the HBSE model is an interesting 
> one, since the probability aspect is the one most subject to error in 
> conventional risk assessment. Humans are notoriously bad at estimating 
> probability. It appears to me that the absence of that parameter implies that 
> the presence of a hazard will inevitably lead to harm, which I don’t disagree 
> with. CSA Z1002, OHS risk assessment, actually states that this is the case, 
> and recommends that hazards are eliminated on this basis whenever possible.
> 
> So we’re left with this situation, I think:
> 
> 1) Risk assessment, when done quantitatively using sound statistical 
> techniques and valid data is a useful and relatively objective method to 
> provide data to decision makers,
> 2) Conventional risk assessment using subjective opinions and risk matrices 
> or decision trees are

Re: [PSES] Risk assessment versus HBSE

[Douglas E Powell]

Hi Doug,

This looks like a very good summary and mentions a few of the things I was,
in my poor attempt, trying to point out. One of my concerns about RA, and
FMEA in particular, is that this method does have a lot of numeric
computation for what is essentially a qualitative process. As such, it
gives the "appearance" of quantifiable due diligence, which many decision
makers want to see. Sad to say, when I've been involved in meetings with
various SMEs, I've occasionally heard discussions on how to adjust the
parameters in order to keep a particular risk listed on the spreadsheet but
not trigger any corrective actions. As with any system, there are those who
would like to manipulate it to their own advantage. It is for this reason
that, when I am leading the team effort and using a severity scale of 1 to
10, I always press for mandatory action when severity is a 9 or 10
(disabling injury or worse), regardless of the other parameters.

Oh, and that's a nice bibliography as well. I think I just received my
summer reading list.

Best,  Doug

Douglas E Powell
Laporte, Colorado USA
LinkedIn 

(UTC -07:00) Mountain Time (US-MST)


On Mon, Feb 14, 2022 at 1:45 PM Douglas Nix <
0bb8ff993b10-dmarc-requ...@listserv.ieee.org> wrote:

> Hi Rich,
>
> I have to admit that I’ve been thinking about your reply all weekend.
>
> As you know, I teach machinery risk assessment and consult in this area
> regularly. I want to stipulate that there are some significant issues with
> risk assessment the way it is most commonly applied in industry, see my
> list of references on this topic at the end of my message.
>
> The inherent subjectivity of risk assessments that are performed without
> empirical data is unquestioned. The difficulty is that for many areas of
> human endeavour we have no empirical data, and try as we might we cannot
> calculate without numeric data. Nevertheless, we must be able to make
> risk-based decisions when designing products and equipment, and so we
> muddle along with the best tools that we have, hopefully while recognizing
> their flaws.
>
> The HBSE model is a good one, and it fits machinery applications as
> readily as does risk assessment, however, the risk assessment methods that
> are used today have a history that stretches back to the 1960s, while the
> HBSE model is much younger. This doesn’t take away from HBSE in any way for
> me, but it does have an impact on the broader acceptance of the method
> since it is not yet as widely known as “conventional” risk assessment. None
> of the the standards in the machinery safety sector recognize the method as
> yet, so getting regulators and users to consider the method is a challenge.
>
> HBSE also suffers from issues with lack of data when it comes to
> characterizing some hazards, leaving the user to estimate the
> characteristics. This brings in the biases of the person(s) doing the
> estimating just as surely as conventional risk assessment methods.
>
> The absence of a probability parameter in the HBSE model is an interesting
> one, since the probability aspect is the one most subject to error in
> conventional risk assessment. Humans are notoriously bad at estimating
> probability. It appears to me that the absence of that parameter implies
> that the presence of a hazard will inevitably lead to harm, which I don’t
> disagree with. CSA Z1002, OHS risk assessment, actually states that this is
> the case, and recommends that hazards are eliminated on this basis whenever
> possible.
>
> So we’re left with this situation, I think:
>
> 1) Risk assessment, when done quantitatively using sound statistical
> techniques and valid data is a useful and relatively objective method to
> provide data to decision makers,
> 2) Conventional risk assessment using subjective opinions and risk
> matrices or decision trees are unrepeatable and therefore unscientific,
> however. despite their flaws, they provide a means to help guide decision
> makers,
> 3) HBSE improves on some aspects of conventional risk assessment by
> eliminating the probability parameters, but is still subject to some
> subjectivity, and is still not widely accepted enough for some decision
> makers.
>
> I wish there was a more utopian perspective to take on the topic, but I
> have yet to find my way to it.
>
> *References*
>
> [1] E. S. Levine, “Improving risk matrices: The advantages of
> logarithmically scaled axes,” *J. Risk Res.*, vol. 15, no. 2, pp.
> 209–222, 2012.
>
> [2] R. Long, “Calculators , Matrices and Mumbo Jumbo Risk Assessment,” 
> *Safetyrisk.net
> *, 2016. [Online]. Available:
> http://www.safetyrisk.net/calculators-matrices-and-mumbo-jumbo-risk-assessment/.
> [Accessed: 03-Feb-2016].
>
> [3] D. J. Ball and J. Watt, “Further Thoughts on the Utility of Risk
> Matrices,” *Risk Anal.*, vol. 33, no. 11, pp. 2068–2078, 2013.
>
> [4] C. Bao, D. Wu, J. Wan, J. Li, and J. Chen, “Comparison of Different
>

Re: [PSES] Risk assessment versus HBSE

[Douglas Nix]

Hi Rich,

I have to admit that I’ve been thinking about your reply all weekend.

As you know, I teach machinery risk assessment and consult in this area 
regularly. I want to stipulate that there are some significant issues with risk 
assessment the way it is most commonly applied in industry, see my list of 
references on this topic at the end of my message.

The inherent subjectivity of risk assessments that are performed without 
empirical data is unquestioned. The difficulty is that for many areas of human 
endeavour we have no empirical data, and try as we might we cannot calculate 
without numeric data. Nevertheless, we must be able to make risk-based 
decisions when designing products and equipment, and so we muddle along with 
the best tools that we have, hopefully while recognizing their flaws.

The HBSE model is a good one, and it fits machinery applications as readily as 
does risk assessment, however, the risk assessment methods that are used today 
have a history that stretches back to the 1960s, while the HBSE model is much 
younger. This doesn’t take away from HBSE in any way for me, but it does have 
an impact on the broader acceptance of the method since it is not yet as widely 
known as “conventional” risk assessment. None of the the standards in the 
machinery safety sector recognize the method as yet, so getting regulators and 
users to consider the method is a challenge.

HBSE also suffers from issues with lack of data when it comes to characterizing 
some hazards, leaving the user to estimate the characteristics. This brings in 
the biases of the person(s) doing the estimating just as surely as conventional 
risk assessment methods.

The absence of a probability parameter in the HBSE model is an interesting one, 
since the probability aspect is the one most subject to error in conventional 
risk assessment. Humans are notoriously bad at estimating probability. It 
appears to me that the absence of that parameter implies that the presence of a 
hazard will inevitably lead to harm, which I don’t disagree with. CSA Z1002, 
OHS risk assessment, actually states that this is the case, and recommends that 
hazards are eliminated on this basis whenever possible.

So we’re left with this situation, I think:

1) Risk assessment, when done quantitatively using sound statistical techniques 
and valid data is a useful and relatively objective method to provide data to 
decision makers,
2) Conventional risk assessment using subjective opinions and risk matrices or 
decision trees are unrepeatable and therefore unscientific, however. despite 
their flaws, they provide a means to help guide decision makers,
3) HBSE improves on some aspects of conventional risk assessment by eliminating 
the probability parameters, but is still subject to some subjectivity, and is 
still not widely accepted enough for some decision makers.

I wish there was a more utopian perspective to take on the topic, but I have 
yet to find my way to it.

References
[1] E. S. Levine, “Improving risk matrices: The advantages of logarithmically 
scaled axes,” J. Risk Res., vol. 15, no. 2, pp. 209–222, 2012.

[2] R. Long, “Calculators , Matrices and Mumbo Jumbo Risk Assessment,” 
Safetyrisk.net, 2016. [Online]. Available: 
http://www.safetyrisk.net/calculators-matrices-and-mumbo-jumbo-risk-assessment/.
 [Accessed: 03-Feb-2016].

[3] D. J. Ball and J. Watt, “Further Thoughts on the Utility of Risk Matrices,” 
Risk Anal., vol. 33, no. 11, pp. 2068–2078, 2013.

[4] C. Bao, D. Wu, J. Wan, J. Li, and J. Chen, “Comparison of Different Methods 
to Design Risk Matrices from the Perspective of Applicability,” Procedia 
Comput. Sci., vol. 122, pp. 455–462, 2017.

[5] C. Peace, “The risk matrix : uncertain results?,” Policy Pract. Heal. Saf., 
vol. 0, no. 0, pp. 1–14, 2017.

[6] B. Ale and D. Slater, “Risk Matrix Basics,” 2012.

[7] P. Gardoni and C. Murphy, “A Scale of Risk,” Risk Anal., vol. 34, no. 7, 
pp. 1208–1227, 2014.

[8] P. Baybutt, “Guidelines for Designing Risk Matrices,” Process Saf. Prog., 
vol. 00, no. 0, p. 7, 2017.

[9] H. J. Pasman, W. J. Rogers, and M. S. Mannan, “Risk assessment: What is it 
worth? Shall we just do away with it, or can it do a better job?,” Saf. Sci., 
vol. 99, pp. 140–155, 2017.

[10] X. Ruan, Z. Yin, and D. M. Frangopol, “Risk Matrix Integrating Risk 
Attitudes Based on Utility Theory,” Risk Anal., vol. 35, no. 8, pp. 1437–1447, 
2015.

[11] S. Albery, D. Borys, and S. Tepe, “Advantages for risk assessment: 
Evaluating learnings from question sets inspired by the FRAM and the risk 
matrix in a manufacturing environment,” Saf. Sci., vol. 89, pp. 180–189, 2016.

[12] P. Thomas, R. B. Bratvold, and J. E. Bickel, “The Risk of Using Risk 
Matrices,” SPE Annu. Tech. Conf. Exhib., no. April 2015, 2013.

[13] F. Gauthier, Y. Chinniah, D. Burlet-Vienney, B. Aucourt, and S. Larouche, 
“Risk assessment in safety of machinery: Impact of construction flaws in risk 
estimation parameters,” Saf. Sci., vol. 109, no. June

Re: [PSES] Risk assessment versus HBSE

[Charles Grasso]

Hello Rich

I am somewhat alarmed by a paragraph in your email!! In it you indicated
that:

 "When I evaluate a product, I look for the physical energy sources, *and
then determine if the energy sources are hazardous or no*t.  Unlike Risk
Assessment, this is easy and repeatable and not subjective.  For example,
all primary circuits are hazardous energy circuits that can cause injury
(electric shock, thermal, fire, and maybe more) and safeguards must be
provided. "

Does this effort not expose the safety engineer to litigation?

Personally I liked using  the UL (NRTL) specs as one had an objective
source.

On Sat, Feb 12, 2022 at 3:00 PM Richard Nute  wrote:

> * This message originated outside of DISH and was sent by: ri...@ieee.org
>  *
> --
>
>
>
>
>
> I don’t like the Risk Assessment process because it is highly subjective
> and not very repeatable.
>
>
>
> When I was with Hewlett Packard, three of us developed “Hazard Based
> Safety Engineering,” HBSE.  The basis for HBSE was James J. Gibson’s
> (Cornell University) research into child injury from auto accidents.
> Gibson said:
>
>
>
> “Injuries to a living organism can be produced only by some energy
> interchange. Consequently, a most effective way of classifying sources of
> injury is according to the forms of physical energy involved. The analysis
> can thus be exhaustive and conceptually clear. Physical energy is either
> mechanical, thermal, radiant, chemical, or electrical.”
>
>
>
> In a moving automobile, the automobile and its passengers have kinetic
> (mechanical) energy.  In an accident, the kinetic energy of the automobile
> is dissipated in crumpling parts.  The kinetic energy of the passengers is
> dissipated in injuries to the body.  Seat belts transfer the passenger
> kinetic energy to the automobile.  Air bags slow the rate of kinetic energy
> transfer to the automobile.
>
>
>
> HBSE identified the magnitudes each kind of physical energy necessary to
> cause injury.  We called this “hazardous” energy.  Then, HBSE went on to
> specify safeguards that would attenuate or prohibit hazardous energy
> interchange.
>
>
>
> When I evaluate a product, I look for the physical energy sources, and
> then determine if the energy sources are hazardous or not.  Unlike Risk
> Assessment, this is easy and repeatable and not subjective.  For example,
> all primary circuits are hazardous energy circuits that can cause injury
> (electric shock, thermal, fire, and maybe more) and safeguards must be
> provided.
>
>
>
> Best regards,
>
> Rich
>
>
>
>
>
> *From:* Douglas E Powell 
> *Sent:* Friday, February 11, 2022 11:37 AM
> *To:* EMC-PSTC@LISTSERV.IEEE.ORG
> *Subject:* Re: [PSES] EN 62368-1 : 2020 Ed 3
>
>
>
> In my view, the Risk Assessment should never be treated as a 'get out of
> jail' card or panacea. Instead, it is only a starting point for a safe
> design and should be done near the beginning of a project, not the end. I
> agree with what Rich says, I've seen a lot of subjective assessments by
> cross-functional teams, with variability based on personal risk tolerance
> or risk aversion.  There are any number of articles pointing to why humans
> are not very good at assessing risk (Google search
> 
> ).
>
>
>
> When using FMEA for risk assessment, I always stress that the RPN factors
> of probability of occurrence, severity, and detection be quantified
> separately without regard to the other factors, not an easy task. There is
> also the problem of RPN vs Criticality (severity x occurrence).  If using
> the RPN, there is the possibility that Detection can dilute the RPN number
> to a point below the threshold for action. So in my view, Criticality alone
> should be used to trigger action.
>
>
>
> Kenneth Ross wrote a very good article last month on Navigating the Safety
> Hierarchy; for me, it was an excellent refresher on how I should use
> risk assessment more effectively (
> https://incompliancemag.com/article/navigating-the-safety-hierarchy/
> 
> ).
>
>
>
> -Doug
>
>
>
> Douglas E Powell
>
> Laporte, Colorado USA
> -
> 
>
> This message is from the IEEE Product Safety Engineering Society emc-pstc
> discussion list. To post a message to the list, send your e-mail to <
> emc-p...@ieee.org>
>
> All emc-pstc postings are archived and searchable on the web at:
> http://www.ieee-pses.org/emc-pstc.html
> 
>
> Attachments are not permitted but the IEEE PSES Online Communities site at
> http://product-compliance.oc.ieee.org/
> 
> can be used for graphics (in well-used formats), large files, etc.
>
> Website: http://www.ieee-pses.org/
> 
> Instructions: http://www.ieee-pses.org/list.html (including how to
> unsubscr