In high-hazard industries, it is not hazards that kill people, it is ineffective controls. Dr Sean Brady FIEAust CPEng unpacks the Brady Review into mining safety.

In July 2019, we commenced the Brady Review, an investigation into the causes of fatalities in the Queensland mining and quarrying industry since January 2000.

Within that 20-year period, 47 people had died, all in single fatality events. Of these, 15 died in vehicle incidents and 12 died from being hit by objects or being entangled in, or crushed by, machinery.

Ten who died in strata failures – being struck by falling rock. Four fell from heights. Four died while working on tyres. One died from a fire, and one died from entering an irrespirable atmosphere.

While the causes of these incidents may have differed, our review identified that one of the key contributing factors in the majority of fatalities was ineffective controls. 

Ineffective controls

What do we mean by ineffective controls? Like many organisations, mining companies deal with safety risk by identifying the hazards their workers are exposed to and putting controls in place to manage them. These controls can include engineering controls, such as conveyor guarding, or administrative controls, such as procedures. This is a standard approach to managing risk, and it appears quite sensible.

However, we found that fatalities typically occur because one or more of these controls – which the organisation believed were in place – were ineffective. If the controls had been effective, the fatality would likely have been avoided. It is not hazards that kill people, but ineffective controls.

What was the role of human error or bad luck in these incidents? While we did see plenty of both, we also saw that it was ineffective controls that allowed human error or bad luck to culminate in a fatality.

The word ineffective is key. These controls were typically ineffective, not missing, suggesting that organisations had assessed the hazard, determined that some form of control was required, and then put a control in place. But when needed, this control did not perform as expected. 

Many organisations will, of course, say they have programmes or processes to verify that their controls are in place, but simply checking that controls are in place is not the same as checking that they are effective.

So how do we move beyond simply checking they are in place?

 

High reliability organisations

One of our key recommendations from the review is that mining organisations should adopt the principles of High Reliability Organisations (HROs). 

The concept of HROs first appeared in the 1980s. HROs are organisations that have very few incidents, despite operating in hazardous environments.

Commercial airline companies are the most cited example: flying has become so safe that it is easy to forget how hazardous it is to climb into an aluminium tube and fly up into the sky.

So what is it that HROs do differently to other organisations? In very simple terms, they go much further than identifying hazards, implementing controls and checking the controls are in place.  

HROs understand and accept that it is not possible to design a perfect set of rules or controls to protect their people. They know that they cannot anticipate and plan for every scenario, they know that rules are often written for the wrong reasons (such as to comply with regulations), and they know that some rules are impractical.

They also know that even if they could design a perfect system, it will drift over time, and the controls that were once effective, will no longer be so.

So HROs put in place the best system they can, and then embrace continuous learning. They obsessively hunt for system weaknesses. They investigate incidents for the purposes of learning, not blame.

They view near-miss incidents as valuable warning signs that their system has failed to the point of almost causing serious harm. They actively engage with the workforce to understand how work is done in practice in order to identify gaps in their controls.

In a way, they take all the information companies traditionally think of as 'bad news' and use it to inoculate the system to make it stronger.

Our review showed that one of the next key steps that mining organisations can take to prevent fatalities is to recognise, then act on, the disconnect between how effective they believe their controls are on paper, versus how effective they are in real life. 

Author: Dr Sean Brady FIEAust CPEng. He is a forensic engineer and a Fellow of Engineers Australia. In 2020, he completed the Brady Review, which was tabled in parliament and made 11 recommendations to the regulator and mining companies on how to improve safety. In 2024, he completed the technical and organisational investigation into the May 25, 2021, incident at Callide C Power Station in Queensland.