Welcome Guest! The IOSH forums are a free resource to both members and non-members. Login or register to use them

Postings made by forum users are personal opinions. IOSH is not responsible for the content or accuracy of any of the information contained in forum postings. Please carefully consider any advice you receive.

Notification

Icon
Error

Options
Go to last post Go to first unread
achrn  
#1 Posted : 25 February 2013 12:36:02(UTC)
Rank: Super forum user
achrn

It's another of those "I'm having a debate (argument) with colleagues..." ones: There is what is described as a common hierarchy of control which goes 'elimination - substitution - engineering - administrative - (behavior) - ppe'. See for example http://en.wikipedia.org/...rarchy_of_hazard_control . I've put behavior (sic) in parenthesis because it is sometimes omitted and sometimes included. I think this may be more common in the US than the UK - a hierarchy of control is more fuzzy in UK practice, and we tend to have different hierarchies in different scenarios, so there's probably a different one in work at height than in something else, for example. In construction design ERIC will probably be termed a hierarchy, or collective measures in preference to individual measures. However, the same hierarchy does appear in at least one HSE document - the leadership and worker involvement toolkit has it at http://www.hse.gov.uk/co...rarchy-risk-controls.pdf My question is, does anyone have a good definition of what is an engineering control rather than an administrative control. My view is that engineering control is a change to process or equipment that reduces the risk, and administrative tends to be doing the same thing but mitigating the risk by for example having fewer people around to get harmed. The contrasting view is that engineering controls are only physical changes - for example, physical modification to a machine. Administrative controls are anything where you tell someone what to do. To illustrate that, suppose there are two ways to skin a cat, each using a different machine. Machine A is quick but risky. Machine B is slower but safer. If you instruct your workforce to skin the cat in question using machine B, is that an engineering control (you've changed the process by changing equipment) or an administrative control (you've established a procedure that reduces risk). Assume machine A is still available to the workforce, because they need it for something else. My view is that instructing use of machine B is an engineering control. An administrative control would be instructing that only the machine operator is within 2m of the machine when it is operating, for example. I know it's a semantic question. I know what matters is that the cat is skinned in as safe a manner as is reasonably practicable. I know our job is to make people safe, not pick over fine print. I'd still like people's views.
imwaldra  
#2 Posted : 25 February 2013 12:53:23(UTC)
Rank: Super forum user
imwaldra

I suggest you don't waste too much time on this. Some controls can have more than one characteristic, there's no need to spend ages on definitions trying to make then exclusive. As you rightly say, the key thing is to apply the hierarchy in the right order. The theoretical difference is that engineering controls are always there once applied, whereas administrative rely on people continuing to follow them. But in reality ALL controls comprise a mix of hardware, paperwork/procedures and people, thugh in different proportions. So an engineering control has to be maintained by people, and meet a performance standard defined by some sort of standard or other document. Similarly an admin control will always have some kind of hardware associated with it - even if no more than an IT system on which it's stored and made accessible to users. None of these elements can be made 100% reliable, hence the need for defined performance standards and also more than a single control for most hazards.
thanks 1 user thanked imwaldra for this useful post.
A Kurdziel on 27/11/2019(UTC)
chris.packham  
#3 Posted : 25 February 2013 13:38:25(UTC)
Rank: Super forum user
chris.packham

I have the view that we should attempt to control the process rather than control the person, particularly with my specific interest in chemical exposure. In other words, if we can make the process intrinsically safe, irrespective of the behaviour of the person, then we have control. The moment we attempt to ensure that the person adopts certain working practices, use of PPE, etc., in order to eliminate or adequately control the process, then we do not have ultimate control. Controlling the process often makes it possible for us to design the system so that it is fail-to-safe. I would love to know how you make people fail-to-safe. If anyone knows......! Of course, in many situations what we will have is a combination requiring both types of approach, but the emphasis should always be on controlling the exposure. One approach that NIOSH in the USA has been pursuing is what they have called "Prevention through design", i.e. ensuring from the outset that the design of any buildings, plant and equipment, and processes is such that prevention is achieved from the outset. They have been running this project now for several years. The NIOSH website contains a recent report on progress. Chris
thanks 1 user thanked chris.packham for this useful post.
A Kurdziel on 27/11/2019(UTC)
pete48  
#4 Posted : 25 February 2013 13:53:55(UTC)
Rank: Super forum user
pete48

I concur with your definition of an engineering control but not with your assessment of the outlined process. The control applied is primarily an administrative one since it relies upon an instruction to use machine B whilst machine A still exists. Machine B may have better engineering controls but human failure (e.g. to follow the instruction not to use machine A) still exists. p48
David Bannister  
#5 Posted : 25 February 2013 14:02:14(UTC)
Rank: Super forum user
David Bannister

No formal definition from me I'm sorry. However, my interpretation is that an engineering control relates to a physical barrier, device or safeguard whilst and administrative control is one that relies on procedures, instruction, permits or training. Thus an interlocked gate is engineering and "follow your training and be very careful" is admin, whilst a LO/TO system is both. One could argue that PPE is an administrative control as it relies on information, instruction, training, selection, provision and enforcement. Current perceived wisdom is that engineering controls are to be considered before administrative. Skinning the cat in a machine may well require engineering controls although the old-fashioned manual method could rely on training in the safe use of knives and good housekeeping i.e admin controls, as I doubt that a safety knife (engineering) would do the job.
thanks 1 user thanked David Bannister for this useful post.
andrewcl on 20/12/2019(UTC)
damelcfc  
#6 Posted : 25 February 2013 14:17:18(UTC)
Rank: Super forum user
damelcfc

Engineering control = Physical entity. It's the meat and veg, the thing of substance, the actual physically touchable thing. An admin control is the information, instruction, training, supervision side and this sits lower in the hierarchy.
thanks 2 users thanked damelcfc for this useful post.
A Kurdziel on 27/11/2019(UTC), andrewcl on 20/12/2019(UTC)
SNS  
#7 Posted : 25 February 2013 21:52:14(UTC)
Rank: Super forum user
SNS

HSG 65 and HSG 48 give some fair information on risk and reduction. Make the safe way the easy way and people won't (generally) try to make a work-arounds. Make the safe way hard and work-arounds will be developed. Engineering controls make work arounds harder to achieve - if not impossible, but people being inventive can overcome many safe systems if they feel justified.
Kate  
#8 Posted : 26 February 2013 08:57:34(UTC)
Rank: Super forum user
Kate

Although it's true that the US have this hierachy, they have it the otehr way up, with PPE at the top.
Canopener  
#9 Posted : 26 February 2013 09:14:02(UTC)
Rank: Super forum user
Canopener

It's an interesting question and I don't think that you are likely to find a definitive definition (is that a tautology?). I am inclined to agree with something along the lines of #6 though. Reg 4 and Schedule 1 of MHASWR deals with the hierarchy of controls although I am not sure that there is much there in the guidance that is likely to help answer your question, and nothing within the interpretation of the regs.
boblewis  
#10 Posted : 26 February 2013 09:29:08(UTC)
Rank: Super forum user
boblewis

Look at it this way A process produces dust so applying the hierarchy you can 1) Find a different way altogether to achieve the end result - Eliminate 2) Find a different way to do the task to achieve the same ends without producing dust - Substitute 3) Remove the dust as it is produced - Engineering control 4) Ensure that people operate under a permit while the process is underway - Administartive controls etc etc I agree however that the hierarchy has to be viewed as a guide to prompt methods that avoid operatives needing to do something as far as possible. I do think that too often we slave away to avoid PPE when in fact it can be an adjunct to other methods. For instance with all the controls in the world I still wore PPE to sample liquid ammonia although in theory I should not have been exposed - belt and braces is not necessarily a bad thing. If it is why do the HSE expect hard hats on a construction site? Bob
achrn  
#11 Posted : 26 February 2013 10:55:02(UTC)
Rank: Super forum user
achrn

boblewis wrote:
A process produces dust so applying the hierarchy you can 1) Find a different way altogether to achieve the end result - Eliminate 2) Find a different way to do the task to achieve the same ends without producing dust - Substitute 3) Remove the dust as it is produced - Engineering control 4) Ensure that people operate under a permit while the process is underway - Administartive controls
I agree with all that. But if at step 3 you said "use this machine with dust extraction" is that an engineering control, or an administrative control? If you modified all the available machines so they had dust extraction, or only supplied machines with dust extraction, it looks fairly uncontroversially to be an engineering control. But what if you also have a machine on site that doesn't have dust extraction (because you need it for some other purpose). Then, is the instruction "don't use that machine" an engineering or an administrative control? I personally think it's still an engineering control. Nearly every control still needs administration, but that doesn't necessarily make it an administrative control.
Jane Blunt  
#12 Posted : 26 February 2013 11:13:37(UTC)
Rank: Super forum user
Jane Blunt

To me, where they can choose from two machines, one of which has the engineering controls and one does not, makes the instruction to use one, rather than the other, an administrative control. It readily fails to danger, which is something that engineering controls, properly designed and implemented, try to avoid.
thanks 2 users thanked Jane Blunt for this useful post.
A Kurdziel on 27/11/2019(UTC), andrewcl on 20/12/2019(UTC)
damelcfc  
#13 Posted : 26 February 2013 11:15:00(UTC)
Rank: Super forum user
damelcfc

Your now in the realms of the PUWER Hierarchy of FIAT Fixed. Interlocked. Adjustable. Trip devices. If the extraction is automatic it is the engineering control, if it needs ANY sort of human intervention for operational purposes it is an admin task. Only when the LEV is in the correct position with the correct capture velocity, switched on etc etc etc and chugging away happily on its own is it a fixed engineering control.
Kim Hedges  
#14 Posted : 27 February 2013 15:01:58(UTC)
Rank: Super forum user
Kim Hedges

Agreed Damel. I've liked PIGSRISE since I heard of it, similar to ERIC/PD. In reverse order of course: Eliminate, Substitute, Isolate, Reduce, Safe systems of work (ACOPs), Good housekeeping, Information Instruction Training Supervision, PPE.
Kishor  
#15 Posted : 27 November 2019 07:59:34(UTC)
Rank: New forum user
Kishor

I have a question related to this.

Is putting a barrication tape to prevent access to a work activity, lets say blasting and painting job, an engineering control? or is this an administrative control? 

Please do share your views

Thanks

achrn  
#16 Posted : 27 November 2019 08:28:54(UTC)
Rank: Super forum user
achrn

Originally Posted by: Kishor Go to Quoted Post

Is putting a barrication tape to prevent access to a work activity, lets say blasting and painting job, an engineering control? or is this an administrative control

As teh original poster (six years ago!) I think it's an administrative control, but teh point where such things become an engineering control is not clear to me.

For example, where on this list of options does something change from an administrative control to an engineering control:

1: Telling the workforce 'don't go near this operation that's going on'

2: Telling the workforce 'don't go into this defined area'

3: Painting a line on the floor and telling the workforce don't cross the line

4: Putting hazard tape up and telling the workforce don't go past it (i.e. what Kishor is asking)

5: Putting a lightweight free-standing barrier up and telling the workforce don't go past it

6: Putting a fixed low barrier in place and telling the workforce don't go past it

7: Putting a 2m high fixed fence with barbned wire on the top in place

I had a case were we were criticised (in a third party investigation into a dangerous occurence) because we'd done something analagous to 3 and the inspector said that was an adminstrative control - but if we'd done 5 that would have been an engineering control and that would have been OK.  The situation was a one-off operation within a single shift, with a specific briefing for everyone on site at the time, and it was actually somewhat more complex than painting-a-line, but within the constraints of what can be explained in a single posting here that will do.

We now, incidently, do something like 5 - but I'm still not convinced that that is automatically OK because it's an engineering control.

Roundtuit  
#17 Posted : 27 November 2019 08:38:09(UTC)
Rank: Super forum user
Roundtuit

For barriers / tapes / lines if I can pass with ease it is adminsistrative the barrier being indicative.

If I need tools (including keys to open doors, gates, padlocks) it is engineering the barrier preventing access.

thanks 2 users thanked Roundtuit for this useful post.
CptBeaky on 27/11/2019(UTC), CptBeaky on 27/11/2019(UTC)
Roundtuit  
#18 Posted : 27 November 2019 08:38:09(UTC)
Rank: Super forum user
Roundtuit

For barriers / tapes / lines if I can pass with ease it is adminsistrative the barrier being indicative.

If I need tools (including keys to open doors, gates, padlocks) it is engineering the barrier preventing access.

thanks 2 users thanked Roundtuit for this useful post.
CptBeaky on 27/11/2019(UTC), CptBeaky on 27/11/2019(UTC)
A Kurdziel  
#19 Posted : 27 November 2019 10:40:08(UTC)
Rank: Super forum user
A Kurdziel

An engineered control (six years ago!) is built into the process so that there is no real choice allowed to the operator. The equipment is so designed that the operator is protected from the hazard and cannot easily defeat the controls. A piece of tape is just a reminder to apply the correct administrative controls. As others have said separating out engineered controls from administrative controls is not always clear.  Many (most?) engineered controls can be defeated by determined staff and then you have to fall back onto administrative controls such as Information, training and supervision to make sure they don’t defeat the engineered controls.

hilary  
#20 Posted : 27 November 2019 11:10:27(UTC)
Rank: Super forum user
hilary

I've always understood an engineering control to be something that is physical - an extraction unit, a guard on a machine, etc, while an administrative control is a work instruction, sop, procedure, etc.

Well, that's what I've been doing for 25 years so if it's wrong I guess I misunderstood the basic premise a quarter of a century ago. haha

achrn  
#21 Posted : 27 November 2019 11:24:09(UTC)
Rank: Super forum user
achrn

Originally Posted by: A Kurdziel Go to Quoted Post

the operator is protected from the hazard and cannot easily defeat the controls.

The problem I have with this definition is how easily is 'easily'.

Stepping over a line is easy, but so (to lesser and lesser degree) is ducking under a tape, shifting a lightweight barrier, shifting a heavier barrier, and climbing over a 1.1m barrier.  In another context, sticking some duct tape over a microswitch, or tying down some other sort of interlock is often easy.

I think there's something about 'cannot inadvertently and without noticing, even if distracted...' because you don't inadvertently tape down an interlock.  Even then you couldn't walk through hazard tape without noticing (I think), though I'd still put hazard tape in the 'reminder of administrative controls' category.

thanks 1 user thanked achrn for this useful post.
A Kurdziel on 27/11/2019(UTC)
andybz  
#22 Posted : 30 November 2019 15:38:52(UTC)
Rank: Super forum user
andybz

As an engineer I am quite offended that barrication tape would be considered an engineered control. Surely, to be considered engineered it would have go through a design process, adhere to specifications, be subject to testing and inspection etc.

chris.packham  
#23 Posted : 30 November 2019 16:34:43(UTC)
Rank: Super forum user
chris.packham

My fundamental rule: Control the process rather than control the person.

Control the process by designing equipment etc. so that the operator cannot of his own volition be exposed to the hazard. I am in charge.

In may cases I can also design the process so that it is fail to safe.

Control the person. I am no longer in full charge. The person can elect to ignore or circumvent the controls that I have put in place. I am no longer in charge.

Controlling the person is almost always fail to danger.

Hence PPE (controlling the person and fail to danger) is the 'last resort'.

So the primary hierarchy is to control the process (prevent through design, add engineering controls) so far as is possible, only then control the person (administrative controls, training, monitoring, etc.)

Note 'monitoring' = management supervision = time = cost!

Chris

thanks 1 user thanked chris.packham for this useful post.
A Kurdziel on 02/12/2019(UTC)
Users browsing this topic
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.