Zero, Zero, Nothing: A Vision with Zero Need of Zero-Goals

A translation of the article that I wrote for the February 2018 edition of NVVK Info, which has ‘Zero’ as the main subject.

‘Zero accidents’ has a major attraction. It intuitively appeals to us; after all, we do not want anyone being hurt or killed at work, or because of our activities. Recently, these was this online-discussion that started with the sincere question “How safe is safe enough?”. The very first reaction - applauded from various sides - was “Zero” accompanied by a tale about the enormous reduction in lost time incidents that a particular company had achieved thanks to a safety campaign. And so on. You know the stories. However, ‘zero’ comes packed with problems.

Does Zero mean safe?

Firstly, we have to ask ourselves whether ‘zero accidents’ really is a measure of safety. More fundamentally, what does the number of accidents actually tell us about safety? Say I decide to blow-dry my hair while taking a bath. When I manage to get out of the tub without being electrocuted, that means that I have realised ‘zero’, but was this operation safe? The classic example is obviously Deepwater Horizon with seven years without a lost-time accident before the installation went up in flames and smoke.

Absence of accidents is not the same as the presence of safety. Outcome-indicators like the number of fatalities, injuries or accidents are at best weak indicators and goals for safety. We can read little out of this kind of metrics. Maybe ‘zero’ means that the work happened safely, maybe they were just lucky (like yours truly in the tub), maybe they did not measure properly, or incomplete with regard to the risks (like on the DWH) or maybe there is just massive underreporting.

The foundation

The foundations of ‘zero’ are weak. It builds upon the unrealistic premises that all accidents are preventable. That is only possible in a utopic, ideal world where everything can be controlled, things are perfectly predictable and there are no surprises. Man, machine and systems work like a Swiss clock (even better than that) and external influences are completely absent, just like variables like human mood swings, distraction, conflicting objectives, and shortage of resources, time and the like. [2]

Obviously, that is an unrealistic scenario. People and organisations do have limited knowledge and resources; all the while, they have to handle a number of challenges simultaneously. Given these limitations one has to make choices, as well as possible under the circumstances. We just cannot prevent everything. Even stronger, we do not want to prevent absolutely everything. Some things we can live with perfectly all right or we must learn to live with them. That is a clear reason for having a ‘reasonably’ criterion. In management language: a ‘zero goal’ is not SMART!

Zero what?

Besides, in what area are you aiming for ‘zero’? Fatalities? Injuries? A noble aspiration surely, but one problem is that consequences in many scenarios are rather random. Hard to predict and hard to steer. Except if you aim for risk reduction to ‘zero’, which is simply impossible.

Some organisations go as far as claiming to strive for ‘zero incidents’ or ‘zero errors’. That is definitely an unlucky approach. Learning of ‘failure’, ‘errors’ or ‘events’ is an important way to work towards improvement. If you are steering towards a low metric for these things, this might lead to ‘zero reports’ which might dry up your most important source of information.

Another problem with zero-accident-goals is that they often only deal with ‘hard’ safety. The biggest problem for employees, however, may not be in the realms of safety, but rather be related to occupational health. Exposure to harmful chemicals, biologic substances, and physical or psychological stress are a much greater problem when we look at long-term effects, lost time and fatalities. [6]

Finishing line?

Some organisations manage to reach ‘zero’ - at least during a certain period. The question then arises: Now what? One problem is that when you make it to ‘zero’, this could give people a wrong impression and lull them into sleep. It suggests that the situation is safe as long no accidents happen. That might not be the case at all. Maybe there have been no accidents by sheer luck, or because other circumstances compensated all that time.

Even if we regard ‘zero’ as a commitment to continuous improvement. What direction to go as soon as you have reached ‘zero’? Basically, there is only one way to go, and that is back. Is that what you want? Of course not. To hold a certain level at all cost can be extremely frustrating, however, and not exactly motivating. Besides, people will be afraid to spoil the seemingly perfect performance.

You can also wonder how balanced a zero vision really is. Organisations must always take other objectives and interests into account, because believe it or not, not one organisation exists with the prime goal of being safe.

Side effects

As the history of humanity has taught us, ambitious goals contribute to reaching exceptional performances. Numerical objectives can be a useful tool to get people and organisations in motion and keep the forward movement. At the same time, it can be highly frustrating if you raise the bar to unrealistic levels, especially when we are talking about absolute measures like ‘zero’.

Only a minor disturbance is needed to fail. The ‘score’ is the ruined for the remainder of that period. This does not really motivate and can lead to ‘creative’ use of definitions and categories, and that ‘bad’ news is censored out. Just think of the Volkswagen emission scandal.

René Amalberti has researched the added value of striving for even more safety within ultra-safe systems and reached a couple of interesting conclusions. For example, that unrealistic expectations with regard to safety very well might end to speeding up the end of an ultra-safe system. [1]

Ethics!

An often heard argument is that ‘zero’ can be the only ethically and morally acceptable goal, because you just cannot have an objective that allows harm or even death. That is complete nonsense for a number of reasons. For example, because this totally disregards the principle of ‘reasonably’, the need to balance diverse and conflicting objectives and life itself: if you get out of your bed in the morning to participate in what we call living (and besides, also if you decide to stay in bed) you accept implicitly that you may encounter some harm. This may be highly unlikely, but still.

Those who feel that ‘zero accidents’ is the only morally acceptable ‘goal’, should realise that we do not need goals for accidents at all. Let us rather find other, better, preferably positive, performance indicators. And let us monitor incidents and accidents purely for the sake of mandatory statistics, use them as opportunities for learning and maybe follow them as an indicator, but not as a goal.

Speaking about ethics, how morally acceptable is it really to require perfection (= ‘zero’) from fallible humans, machines and systems? Perfectionism has positive sides, because it can motivate individuals to strive for ambitious goals. On the other hand, it can also be a destructive factor, and it even is a psychological disorder. Not exactly something to strive for thus. [4]

In connection to the Vision Zero movement, I have seen statements that accidents do not belong to the identity of the organisation. [7] That kind of nonsense annoys me massively. Although there definitely is a number of mistakes in my life that I rather not would have committed, also these experiences have contributed to making getting me where I am today. The very same applies also to organisations and accidents. Aviation, railways or automobiles would not have today’s level of safety if it were not for the accidents that gave us an opportunity to learn. No matter how sad the consequences have been, these accidents have absolutely been an essential part of the identity, and I think it is hugely disrespectful of the victims to deny this!

So, it ‘zero’ completely useless?

Let us be clear: I am not against the striving for no accidents, or even against a zero vision. This, however, must never turn into a target. There is a world of difference between a target (that must be concrete and SMART) and a vision (‘I have a dream’).

A vision is a dream about where we want to be. It is about striving for improvement on the (very) long term. Like reaching for the stars, it is something of which we must realise that we most likely never will reach it, but striving for it could be worthwhile nevertheless. The right approach is then to “combine an ambitious, inspiring long term vision with quantified interim targets” [3], instead of running unthinkingly behind some nice sounding slogan.

A vision has most of all a symbolic value: it can be used to communicate commitment for continuous improvement. This must be balanced with realism, however. A major problem is that many do not get the symbolism and take the ‘zero’ all too literal. That turns ‘zero’ immediately in some kind of a target, with all kinds of unwanted side effects, including not reporting incidents, creative accounting (e.g. by stretching definitions) or the emergence of a culture in which people blame each other for ruining a perfect score.

Often zero vision and ambitious targets are said to play an important role in realising safety improvements. [8] We should ask ourselves, however, if we are rather talking about correlation instead of a causal connection. An organisation that gives her utmost will most likely also have a vision. Embracing just a zero vision because many other organisations do the same will not lead to improvement in itself, for that you will first need a solid safety program.

And so?

Again, you do not need ‘zero’ as a target. As we saw, the number of accidents or injuries is a weak measure of safety. Striving for improvement and caring for our people should be sufficient. We cannot prevent everything, but that does not diminish the responsibility to do our utter best - within reasonable limits.

James Reason puts it very well:

‘Safety is a guerrilla war that you will probably lose (since entropy gets us all in the end), but you can still do the best you can.’ [5]

Let us take these wise words at hard, and get working. We cannot prevent all accidents and reach ‘zero’ in a sustainable way. At least we can prevent a great many accidents by working systematically and structurally. Hopefully, we will manage to prevent the ones that matter the most. Good luck!

Literature

[1] Amalberti, R. (2001) The paradoxes of almost totally safe transportation systems. Safety Science, 37 (2): 109-126.

[2] Busch, C. (2016) Safety Myth 101. Mysen: Mind The Risk.

[3] Groot-Mesken, J. de (2014) De Waarde van Nul. Nulvisies en Verkeersveiligheidsbeleid. (R-2014-8). Den Haag: Stichting Wetenschappelijk Onderzoek Verkeersveiligheid.

[4] Long, R. (2012) For The Love of Zero: Human Fallibility and Risk. Kambah, ACT: Scotoma Press.

[5] Reason, J. (2008) The Human Contribution. Farnham: Ashgate.

[6] Townsend, A.S. (2014) Safety Can’t Be Measured: An Evidence-based Approach to Improving Risk Reduction. Farnham: Gower Publishing.

[7] Zwetsloot, G.I.J.M. (2016) De Nul Ongevallen Visie: een nieuw perspectief voor het verbeteren van veiligheid. Tijdschrift voor Toegepaste Arbowetenschap, 29 (2): 61-64.

[8] Zwetsloot, G.I.J.M. et.al. (2017) The importance of commitment, communication, culture and learning for the implementation of the Zero Accident Vision in 27 companies in Europe. Safety Science, 96: 22-32.