I read this report some days ago and just scratched my head in wonderment! I'm sure most people have heard the phrase "to err is human".

Errors of judgement etc made by people always will occur, and will / do contribute to accidents.

Typically, in cases like this, "human error" is offered as an excuse, being unforeseeable contributor to the event which should be taken into account when assessing culpability.

Whether or not human error is a valid "cause / contributor" to the event the key issue is "how can causal factors, including actions / inactions of people (errors) be controlled / minimised / prevented? 20 or 30 possible causal factors could probably be identified in the case in point. Identifying as many contributing factors as possible, then assessing which factors offer some possibility for control (change) is the purpose of an investigation.

A Risk Assessment should evaluate the likelihood that contributing factors may re-occur. Introducing rules, procedures, education, supervision, audits etc as possible controls may all help in some way. The Risk Assessment process should serve to evaluate the likely effectiveness of the proposed controls, i.e how much reduction in risk do they provide, and is that level of reduction reasonable in the circumstances?

So I think it doesn't really matter what labels are given to causal factors. The focus must be on predictability of occurrence, and practicability of control.

The term Human Error is considered distasteful because it smacks of victim blaming, and suggests uncontrollability. In the same vein the term "accident prone" is used. I suspect most practitioners would avoid using those terms. Yet rules and procedures are still religiously introduced in our workplaces in an effort to constrain and control human behaviour, much like the road speed limits are intended to constrain driver behaviour. The success rates are probably about the same.

Engineering out the opportunities for human fallibility (error) to become a causal factor in undesirable events is the legal imperative. Conducting Risk Assessments is the means by which the practicability of the possible controls can be assessed and subsequently justified.

The whole point and essence of a well thought out Health and Safety Plan including whatever instruments the creators want to use: is that it takes any forseeable instance of 'human error' into account and creates controls.

So, no! "Human error" is a comletely unacceptable conclusion in an investigation. It would by 'safety system failure'.

Were they using the term "human error" as a polite way of saying the person was incompetent? I note the report goes on to give at least a half dozen additional factors as well as human error. Can even great safety systems, and controls prevent an incompetent person from doing a stupid thing? They don't really get to this point, and it is perhaps the most relevant reason for the injury. Was the person adequately trained in using a lathe (it says lack of instruction)? It mentions fatigue which is a very real factor for a factory worker; the energy drinks were probably to keep the worker awake. I agree with the writer, human error is not an adequate reason, and I think if someone does something stupid to cause an accident, then reports need to be clear about it, to avoid doubt.

I think Jan has also been reading Sydney Dekker's book where he advocates putting "human error" in quote marks because it is an attribution we use after the event!
"Human error" is a cop out for a PCBU that failed to think about the human factors, the performance-shaping factors, that might contribute to some variation from the intended outcome.
The trouble is that almost nobody is trained in human factors or dealing with the performance-shaping factors. And too many HSPs are in that category.
By the way, how many readers own a copy of IEC62508:2015 "Human aspects of dependability"?

Hi All, first post to the forum so hopefully I don't start off on the wrong foot. I have been working in human factors for almost 15 years now, investigated/co-investigated 100s of incidents, utilised a range of methodologies for investigative and risk management purposes, and even the proud owner of humanerror.co.nz.

However, I do not subscribe to the term human error and especially the way it is used by many organisations as a fait accompli. I recall a recent "simple case of human error" that was quoted in the news (I get daily google feeds of news items with the term human error in them), whereby a nurse administered methadone to a baby - I found this case particularly upsetting as a father to two children - and utterly unfair on the nurse who must have been beside themselves with the awfulness of their actions. My experience of working with nurses suggest there could have been many reasons for this mistake happening and not necessarily associated with ineptitude driven by the discourse of the article I read.

I do have a lot of time for the Safety II / resilience engineering understanding of challenges to individual resiliency as it favours the idea that humans are not an erroneous liability but actually a commodity, managing the day to day challenges of the workplace. No two days are the same, each day has its challenges to our resilience to manage our day, and it is only when our ability to be resilient goes beyond its limits, do errors happen. I am sure many of you reading this may feel I am preaching to the converted - I hope so. I believe Safety II and its associated tools embraces the human factor much better than any older concepts associated with human error / swiss cheese / HFACS systems that seem to provide more aesthetic value than a true understanding of the human element in the workplace.

Happy to discuss further over coffee. Have been working closely with Erik Hollnagel (father of Safety II) for a few years now and have learnt a lot with using his FRAM methodology to increase workplace resilience and eradicate old concepts of human error. I have learnt a lot from Erik and would love the opportunity to share that learning.

Have a great day.

Karl

Time for us to get real. Humans make mistakes. They make errors. They contribute to accidents. Individuals are often the authors of their own misfortune (or that of others).

Are they the sole cause of an incident - often not. But more often than not its a contributing factor.

The degree or extent of contribution will help determine any consequence that human should shoulder post event

The problem is that concluding "human error" caused an incident begs far too many additional questions.
Humans make errors. That is why it is called "human error". Errors are inevitable and are made every day.
A better approach is to say: I know the people working in Department X will make errors, because they are human. There are some dangerous machines or processes or chemicals used in Department X, so I know the consequences of those errors could be seriously harmful. Therefore, how can the work be designed, in collaboration with those people, so that their inevitable errors cannot result in significant harm?
That puts the onus back where it belongs - on management - to assist the people who face the risks in designing out the possibility that harm could arise from error.
So any organisation which concludes "human error" was the cause of an injury is just flagging its own failure to manage risk.

I see lots of incident reports where human error is identified as the immediate cause of whatever incident has occurred. Some but by no means all of those reports go on to look at the underlying root systemic causes.

Interestingly, in my experience WorkSafe is starting to push back on duty holders who get to human error and stop their analysis there.

Just doing a serious near hit investigation at the moment. Its not looking good. Pretty much all causes are pointing to a human making an error.

A human making an error is not a 'cause'. It is something that happens frequently in everyday work. The question to ask is: how is the work designed so that when the inevitable human errors happen no one gets hurt? That is a much harder question but one that every manager needs to address.

Interesting comment in the media today. To paraphrase ""so far as reasonably practicable" cannot have intended employers and PCBU's to identify employer or PCBU features which can only become a hazard in circumstances of operator incompetence, carelessness or non compliance...…". Seems to be opening the door to allow for human error causing the hazard to me.

I would urge people to take up Karl's offer of a discussion over coffee (sorry Karl if you get inundated). There are few people in NZ who genuinely understand human factors as anything much beyond 'don't forget people aren't perfect'. It is also worth reading someone like Steven Shorrock who provides some insightful knowledge in an accessible way https://humanisticsystems.com/author/stevenshorrock/ . There are lots of specialists who don't like the term 'human error' at all. At best, it is a useful marker that your investigation has further to go. At worst, just another tool in the blame the worker tool kit so beloved of many businesses.

Theres always a worker to blame. Just depends how far up the organisation you are prepared to aim

My personal opinion about the topic is that any human error has a cause. Unless the person sets out to injure themselves, the reasons can be almost infinite. From lack of knowledge, to work pressure, lack of tools or equipment, stress, distraction and simply not enough physical safeguards. Humans don't normally behave in a random way unless the brain has a fault that triggers illogical or involuntary behaviour.

And most accidents have a sequence, or combination of failures.

Therefore, in almost all cases, there is a cause involved but we generally don't have enough investigative skills to find the intangible ones like fatigue, stress, work pressure. On the other hand, my observation over years of viewing the average investigation, is we don't have the skills to even find obvious physical causes either. Accident investigation is something of a hidden art, available to very few people.

We need to remember that the article was written by a journalist (no criticism intended Peter). The terms "human error" and in particular, "complacency" are appallingly archaic . Sounds the same as "Told him/her to be more careful in future". However, the paragraph that lists root causes sounds like a good investigation. There were also additional causes of lack of understanding and instruction, fatigue, dietary options (energy drinks) and the injured worker's arrogant attitude."

Leaving aside the vacuous statement about arrogant attitude, (they knew he had an attitude and let him continue to use a lathe, then when the poo hit the fan, they pulled it out of the drawer. I'd have kept quiet. That's having your cake and eating it too). None of the causes were, in fact, human ERROR. They appeared to be evidence of lack of knowledge, or human FACTORS, like stress and fatigue.

Testing notifications out of the Forum, please disregard. Simon

Test response.

Whenever I consider this issue I often think about motor vehicle accidents. Is driver error ever a factor in motor vehicle accidents? Insurance companies contemplate this every day when they make a decision on which insurance policy is to be credited (or perhaps debited) with the cost of the repairs.

Just a couple of observations.

What came to my mind was the Challenger incident. Was this not human error? The NASA management over ruled the engineers who had doubts about the cold weather and the 'O' rings that eventually failed, alternatively the engineers didn't speak up load enough!

And again in the Gulf of Mexico incident, was it not people, at all levels, who ignored the the tell tale signs staring them in the face?

I often use this quote when speaking to groups.

“Everyone, and that includes you and me, is at times careless, complacent, overconfident and stubborn. At times, each of us becomes distracted, inattentive, bored and fatigued. We occasionally take chances, we misunderstand, we misinterpret and we misread. These are completely human characteristics. Because we are human and because of all of these traits are fundamental and built into each of us, the equipment, machines and systems that we construct for our use have to be made to accommodate us the way we are, and not vice versa.” (Chapanis, 1985)

It doesn't excuse workers, nor does it excuse PCBU's (who are actually humans as well).

The reality is that we ARE human and sometimes we stuff up!

Keep up the good work.

Exactly right John. Which is why modern motor vehicles have all the hi-tech systems designed to reduce the likelihood people will lock their keys in the car. It's still possible, but a pretty good balance between achieving vehicle security (locking the vehicle) and human fallibility ( forgetting the keys are in the car (ignition, handbag, centre console, boot etc) has been achieved. At some cost, of course, but avoiding the cost of call-outs or damaging the vehicle. An engineering solution to an acknowledgement of human fallibility (aka human error).

The trouble is, most people have failed to grasp the implications of "reasonably practicable" and they therefore stop any investigation with the convenient post-event attribution "human error".
Perhaps we should start with the premise that an unintended action of one component (a human being) of a system is known to be error-prone and such errors can cause severe consequences. We could then ask if any risk treatments are practicable (ie, "capable of being done") that would minimise risk ("effect of uncertainty on objectives" in ISO31000). Those treatments might start with redesign of the overall system or components of the system. The cost of any treatments that are found to be practicable can then be calculated (along with any additional benefits they may offer) to ascertain if they would be reasonable.
From a quality management point of view, repeated failures to achieve the requirements for product safety, quality, etc would result in quality improvement. Why not for OHS?
There is research evidence that organisations with an ISO9001 quality management system find it easier to introduce an OHS management system. I have been arguing in the current NZISM ISO45001 Masterclasses that the structure of management system standards provides OHS practitioners/professionals with an opportunity to introduce integrated HS/Q or HSEQ management systems with economic benefits for their PCBUs.
And to reduce "human error" and injuries.

The focus should be examining the cause of the human error. In too many companies this is used as a part of a blame culture, instead of a learning culture and determining what behaviors led to the incident occuring and what behaviour's need to change to chnage to prevent re-occurance

People make mistakes, or decide on a course of action, inside of a context. This context is based on experiences, organisational pressures, what they can interpret from their environment, and a hundred other variables. The decision usually seems like the right one at the time, and with the power of hindsight, we can see that it was incorrect or inappropriate. When we use language like "human error", we automatically place our focus on the person involved, and that they were at the source of the failure. In reality, the person is just another cog in the machine, where the machine itself has not been designed perfectly. We could focus on the error, or the reasons that the error was made. I know which one is more likely to result in improvements

Well said Paul Robertson, except I'd go one further and flip that first sentence around.
People decide on a course of action within a context, based on their knowledge, experience, competence and ability to determine (judgement) what they deem to be the best course of action at the time. Once the outcome is known, most of the time it is wanted - deemed success. Sometimes it is unwanted - deemed failure and then thought of as a mistake or human error.
To me, the 'mistake' or 'human error' label only exists afterwards, but does not apply at the decision making time, so really shouldn't be used, other than a descriptor leading to investigation of the context of the situation.

Well... I started at the top thinking human error, dependent on context was a possible cause and operator attitude can certainly be an influence.

Having now arrived at the bottom I now have changed my opinion?

My reflection now is human error is nothing more than a catogory of causes which need to be worked through to unveil all of our contributing factors. In this instance when we arrive at human error, we simply continue the questioning - Why?

Now attitude on the other hand is a big player in whether we achieve the desired result of not. Poor attitudes contribute to many a failure and when experienced at a senior level within a business, it can undertermine everything a H&S rep or committee are working towards.

In order to be successful we need many things, a couple of important ones are believing and being invested. In many respects attitude is key.

A very interesting discussion.
We had a near fatal incident last year. 4 different PCBU's had failures that contributed to the event, each of which may have prevented the event if the failure had not occurred. The final failed defence, however, was a driver who did not follow the correct procedure in securing his load. To this day he is unable to explain why he failed to do so. There is no apparent reason for him to have not done so.

Our drivers work unsupervised and we have to be able to rely on them doing their job properly. We know that they are capable of violations. But apart from identifying the potential for that and setting Safe Operating Procedures to minimise the risk there is little we can do about it. Our pre-employment assessments weed out the majority of those we feel unable to trust. Random monitoring of driver performance and site safety observations identifies others for further coaching. But the bottom line is anything more than that is not 'Reasonably Practicable'. WorkSafe asked us to consider an engineering solution that would physically prevent each part of the load from coming off the truck until the crane was hitched up to it. Given the variety of shapes and sizes of the loads, this is not just impracticable - it is impossible!

If the workplace is fully automated and people are removed from the workplace it will be incident free. Human error is one of the identifiers to the cause of any event minor to catastrophic and the eternal question WHY must be asked at every step of the way - in the case in question - WHY had he not used a lathe for several weeks, WHY was the job inappropriate for the lathe, Why was an energy drink being consumed ?
Lets run with the last one - was the employee tired - Why - is he working overtime Why - does he have a second job why - is he always thirsty why - is his attitude linked to something else why.
It may be that from the answers to these questions an unstable diabetic is identified who is not sleeping well as he is up every hour or so to the toilet, sugar levels are up and down which is why he needs the energy - SO human error may be the biggest contributing factor but once again ask why and look at what the employeer and employee are doing to maintain the health of the worker

5 — Jan Hall

Not merely Sydney Dekker ( "Been reading"? Some time ago now!)
I think most comments have nailed it! "Human error" really only means "a human did it".. The real question is: now let's look at the circumstances that enabled it.

Anyone else notice, when encountering 'human error' written on an accident investigation, that the writer,frequently (invariably?) has little appreciation of safety systems and precautions? And is also possibly the product of a fairly unsophisticated health and safety background?

But if there'd been a STRONGER system in place so that the engineers' doubts were recognised and immediately checked????
Same with Gulf of Mexico: categoricl instruction to RESPOND to any doubtful signal. That is: a good h&s system recognises human fallibility and makes provision for it.

For those people engaged in this thread this is a good reference on Human Factors from the UK HSE.

Hi Richard: I'm pleased to see that HS(G) 48 is still free of charge at http://www.hse.gov.uk/pubns/books/hsg48.htm . Sadly IEC62508 "Human Reliability" (which is also very useful) is over $400. I've recommended both to people on the ISO45001 Masterclasses when talking about clause 6.1.2 (risk assessments) and the definition of hazard in the HSWA2015. The Act strongly implies we must include human factors when considering if risk has been minimised SFAIRP.