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
Mersey  
#1 Posted : 22 February 2018 08:38:09(UTC)
Rank: Forum user
Mersey

Hi,

I have been asked to complete an investigation into an incident whereby an FLT's mast hit a roller shutter door causing extensive damage to the fabric of the building.

I loath to put in operator error and close the investigation there. From the info I am gathering thus far the person involved in the incident said he felt under pressure to keep a production line running and his shift was a man down consequently he was rushing here there any everywhere, the person has never had a bump on his FLT in 15 years  service etc....

Could anyone advise on the best type of investigative tool to find indirect causes? I struggle with the Five why concept I struggle to get to five whys and tie myself in knotts.

Are there any good websites out there explaining the methodology or other useful investiagtive tools?

Thanks

pl53  
#2 Posted : 22 February 2018 09:24:55(UTC)
Rank: Super forum user
pl53

5 WHYs - You don't always have to have 5  you may have 3 , 4 or 24 you just keep asking why until you no longer can get an answer.

FLT hits door WHY - Door was too low for mast WHY, Door halway down(?) WHY, Closure mechaism not working WHY POOR MAINTENANCE (Root Cause)

Driver not paying attention WHY Rushing WHY, under pressure WHY, Short staffed WHY, people off work WHY, not adequately staffed WHY, POOR PLANNING (Root Cause)

Driver not paying attention, WHY, concentrating on other things,  WHY, POOR BEHAVIOUR (Root Cause)

You can go on and on with this. Was the door stuck, was it left half open, could the driver see that the door was only partly open, was his mast extended, was the fLT suitable for the door, was the area well it, was it day time or night time

Another RCA tool you can use in this way is fault tree analysis but in general they all follow that same methodology.

     

thanks 2 users thanked pl53 for this useful post.
Mersey on 22/02/2018(UTC), jakatac1@gmail.com on 26/02/2018(UTC)
Granlund40055  
#3 Posted : 22 February 2018 09:45:46(UTC)
Rank: Forum user
Granlund40055

The HSE have a comprehensive accident investigation guide which provides a framework for using "WHY" questions and suggested formats for the report.

See http://www.hse.gov.uk/pubns/books/hsg245.htm

It may assist you identify the reasons why the driver was rushing without apportioning the blame as driver error.

thanks 1 user thanked Granlund40055 for this useful post.
Mersey on 22/02/2018(UTC)
aud  
#4 Posted : 22 February 2018 10:44:45(UTC)
Rank: Super forum user
aud

Looking for good examples? To learn more? Try thinkreliability.com

thanks 1 user thanked aud for this useful post.
Mersey on 22/02/2018(UTC)
Hsquared14  
#5 Posted : 22 February 2018 12:10:02(UTC)
Rank: Super forum user
Hsquared14

I agree with PL53 5 Whys is just a name - sometimes you need 5 million whys and sometimes only one or two.  I often use the 5 whys technique to gather information and then fishbone to organise it so that I can work with it.  Its all about understanding the story of the accident - what happened before the event to set the scene for the accident to happen and then the accident itself and finally the aftermath.  I sometimes use mindmaps, rich pictures and word cloud techniques too to help me to link and relate each bit of information together.  It isn't a matter of picking a technique and sticking to it exclusively its about finding a mixture of approaches which enables you to understand exactly what happened and why it happened and most importantly to understand why people do the things they do.  Operator error is a cop out people don't set out to make mistakes, circumstances combine to lead to the mistake and the consequence.  For what its worth I think the HSE approach is too rigid and mechanistic - the principals are sound but they don't help you to truly understand what happened, its just all about getting the facts and not understanding all the other factors.

thanks 3 users thanked Hsquared14 for this useful post.
Mersey on 22/02/2018(UTC), A Kurdziel on 22/02/2018(UTC), sideshow on 26/02/2018(UTC)
peter gotch  
#6 Posted : 22 February 2018 13:11:45(UTC)
Rank: Super forum user
peter gotch

In addition to comments above, one issue is why the incident resulted in the severity of consequence you describe and not simply damage to roller shutter door and/or fork truck.

Mersey  
#7 Posted : 22 February 2018 13:17:13(UTC)
Rank: Forum user
Mersey

Peter this is a very good point the design of the roller shutter door supports were badly designed.

Two brick pillars supporting a piece of  reinforced contrete that weighed in excess of 6 tonne to which the roller shutter doors were attached.

The doors were 20 years old too and there had been several near misses

Quote:

In addition to comments above, one issue is why the incident resulted in the severity of consequence you describe and not simply damage to roller shutter door and/or fork truck.

fairlieg  
#8 Posted : 22 February 2018 13:55:39(UTC)
Rank: Forum user
fairlieg

I have found rather than focusing on what went wrong and "why" to understand "how" the work is done.

Asking "how" changes the conversation a little.  I recommend two books Dekkers Field Guide to understanding Human Error and Conklins Better Questions.

Asking "how" in my experience looks at the operation both up and down stream.  It can really help understand the organisaiton and proprities and assumption that are made at each level.  "Why" focuses on the specific failures leading to an incident incident nothing wrong with that however, "how" can help you identify weak signals that may be precursers to other potential operations upsets.

The old mantra about why we investgate an accident (is to prevent a recurrence) is great but would it not be better to identify other potential problems in an operation?

5 why and fishbone etc are great for equipment failure they were borne out to 5S 6Sigma and Lean and they are fairly liniar approaches.  Accidents however, involve a range of dynamics and variability (management, interpritation of instruction Work as planned v's done, environment, operator etc).

I get the feeling you already know that there is more to this that human error.  Asking "how" might just help you broaden your investigation a little

thanks 3 users thanked fairlieg for this useful post.
Mersey on 22/02/2018(UTC), JavierGaspar on 24/02/2018(UTC), jakatac1@gmail.com on 26/02/2018(UTC)
A Kurdziel  
#9 Posted : 22 February 2018 14:59:04(UTC)
Rank: Super forum user
A Kurdziel

As you say the Five WHY’s is just a technique for accident investigation and to be honest I have in 15 years never used a formal system to investigate an accident. Instead I have tried to get down to what we like to call the Root Cause. There are 5 whys because that is generally seen as been the right depth of investigation that gets you to some form of management failure as the root cause eg POOR MAINTENANCE, POOR PLANNING , POOR BEHAVIOUR .  Anything less than 5  you end up at the operator error sort of conclusion which does not give much opportunity for improvement ( other than sacking the operator) or if you go too far 20 or 30 whys’ you can disappear into the realm of philosophy, metaphysics or thermodynamics.

The highly structured approach can make sense if you are investigating the failure of a part or a system but are less use when dealing with the human factor.

Edited by user 22 February 2018 15:00:13(UTC)  | Reason: If I could spell I'd be deadly

Kate  
#10 Posted : 23 February 2018 11:56:14(UTC)
Rank: Super forum user
Kate

You are right not to say 'human error' and leave it at that. I recommend the classic HSG48 'Reducing error and influencing behaviour' which shows how to analyse and classify human failures. 

http://www.hse.gov.uk/pubns/books/hsg48.htm

A Kurdziel  
#11 Posted : 23 February 2018 12:20:47(UTC)
Rank: Super forum user
A Kurdziel

Originally Posted by: Kate Go to Quoted Post

You are right not to say 'human error' and leave it at that. I recommend the classic HSG48 'Reducing error and influencing behaviour' which shows how to analyse and classify human failures. 

http://www.hse.gov.uk/pubns/books/hsg48.htm

HSG 48 is very good starting point for getting your head around accident investigation.

thanks 1 user thanked A Kurdziel for this useful post.
Mersey on 26/02/2018(UTC)
JavierGaspar  
#12 Posted : 24 February 2018 18:46:31(UTC)
Rank: Forum user
JavierGaspar

Mersey, Fairlieg has hit the nail on the head. I would also like to add, in answer to your original question, that should use the root cause analysis methodology.

Please note, this is an analytic methodology and not and investigative one; it focusses on 'how' things evolved while adopting a systems approach thus, considering and weighting all possible contributing factors and how these interact to one another.

Good luck.

thanks 1 user thanked JavierGaspar for this useful post.
Mersey on 26/02/2018(UTC)
Roundtuit  
#13 Posted : 24 February 2018 21:56:37(UTC)
Rank: Super forum user
Roundtuit

Each Why does not necessarily have a single answer so like the fishbone you may branch in different directions

Along each presented path ask Why until no answer appears (5 is just guidance)

You will likely end up with a web of contributory issues, and as you correctly identify not just "operator error"

The real skill is inversing this process to derive a comprehensible action plan for management to execute

thanks 2 users thanked Roundtuit for this useful post.
Mersey on 26/02/2018(UTC), Mersey on 26/02/2018(UTC)
Roundtuit  
#14 Posted : 24 February 2018 21:56:37(UTC)
Rank: Super forum user
Roundtuit

Each Why does not necessarily have a single answer so like the fishbone you may branch in different directions

Along each presented path ask Why until no answer appears (5 is just guidance)

You will likely end up with a web of contributory issues, and as you correctly identify not just "operator error"

The real skill is inversing this process to derive a comprehensible action plan for management to execute

thanks 2 users thanked Roundtuit for this useful post.
Mersey on 26/02/2018(UTC), Mersey on 26/02/2018(UTC)
andybz  
#15 Posted : 25 February 2018 14:39:54(UTC)
Rank: Super forum user
andybz

I would suggest that 5 Whys and Fishbone are not investigation methods.  They are tools to analyse incidents after you have done your initial investigation.  As with all tools, you need to use the right one for the right job, and you may need to use more than one to complete the job at hand.

You may find this useful for causal trees (same idea as 5-whys).

thanks 1 user thanked andybz for this useful post.
Mersey on 26/02/2018(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.