The question behind the paper
From 1 June 2026, Resources Safety and Health Queensland tests critical control management by whether each control holds under production conditions, not by whether it appears in a bowtie or on a verification schedule. Most operators arriving at that date have the documents. Far fewer can say how reliable a given control is when the shift is short-handed, the ground changes or the schedule is behind.
Binary verification cannot answer that question. A control can score fully compliant on the verification protocol while it is routinely worked around at the workface, poorly matched to how the work is actually done, or so demanding of attention that it fails under load.
The paper describes Control Effectiveness Assessment (CEA), a method developed by The Art of Work to see those failure modes. CEA is the intellectual property of The Art of Work.
Best Paper, #IMHS2026
AusIMM International Mine Health and Safety Conference
Pan Pacific Perth, 5 to 7 October 2026
The method in brief
• Robustness. The engineering strength of the control, where it sits in the hierarchy of controls, and the maintenance, spares and staffing it needs to work as designed. Scored by subject matter experts from design documents, inspection records and digital indicators.
• Human Dependency. How much focus, memory and skill a person needs for the control to work, in normal and abnormal conditions. Controls that lean hard on people tend to hold in steady state and fail under congestion or at shift change.
• Alignment. Whether the control helps or hinders the person doing the work, and whether it is felt as a support or a punishment. A sound control that crews see as surveillance gets worked around.
• Adaptability. How the control behaves outside its design envelope: contractors, night shift, weather, disruption. Many controls are designed for the steady state and break down in the conditions where the energy event is most likely.
Each dimension is scored from 1 to 5 and the four sub-scores stay visible rather than being collapsed into one number. Frontline crews, including night shift and contractors, score Human Dependency, Alignment and Adaptability in a separate workshop run on learning-team principles. The SME and frontline views are then reconciled, not averaged, because where they disagree is usually where the most useful findings sit.
What the case study showed
The method was field-tested on the driving-safety portfolio of a large Australian asset-intensive operator. The case is not a mine, but the controls it assessed (vehicle specification, driver monitoring, journey management, competency, fitness for work) are the same families that dominate mobile-plant and traffic-management risk on site.
• Coverage gaps. The two engineering controls at the top of the hierarchy scored as reliable and would have passed any binary check. Neither covered the grey fleet or delivery-partner vehicles, where a large share of the actual exposure sat.
• Alignment failure. The GPS driver-monitoring system scored 3.5 on Robustness and 2.7 on Alignment. The technology worked, but drivers saw it as a performance-management tool. Collision-avoidance and proximity-detection systems are exposed to the same failure.
• Self-managed fatigue. Fitness-for-work and fatigue management scored 2.2 on Robustness. It rested on self-assessment under deadline pressure, with no anticipatory monitoring. Mining rosters and shift-end fatigue tools share the pattern.
• State checks mistaken for capability. A licence check confirmed that the driver held a licence, not that they could handle off-bitumen or long regional driving. The same gap appears wherever a signed permit stands in for competently done work.
The paper translates these findings to geotechnical and strata control, isolation and lockout, gas monitoring and ventilation, and collision avoidance.
Questions for mining leaders
• For each critical control on your top material unwanted events, can you say under what conditions it fails?
• Does your verification confirm that a control exists, or that it worked on the shift?
• When was a night-shift crew or a contractor last asked to score a control?
• How many of your listed critical controls are procedures, permits or training packages standing in for engineering that was never funded?
• If a control's reliability was slipping this year, how would you know before the incident?
Hard controls and the operating platform
Procedures, permits, training packages and supervisory routines are routinely labelled critical controls alongside the engineering that actually interrupts the energy pathway. CEA separates the two using the Energy Damage Model: a hard control is an object or system that stops, redirects or absorbs the energy, and everything else is control support. The label matters because it steers investment. The Brady review found that after serious accidents in Queensland mines, 62 per cent of corrective actions were administrative controls only.
CEA tells an operator which controls are fragile and why. Whether those controls stay sound depends on the conditions the site runs under. Where an operation lives in chronic catch-up, with supervisors replanning and expediting, the strongest reliability profile will not survive long contact with production, and the easy path leads back down the hierarchy to another procedure.
That is the ground covered by the Fatality Plateau paper and the Capacity to be Safe briefing on this site. The two lines of work fit together: one measures the controls, the other stabilises the platform they have to work on.
Frequently asked questions
Is CEA a Stratflow method?
No. Control Effectiveness Assessment is the intellectual property of The Art of Work and was developed by Kelvin Genn. Hendrik Lourens is a senior consultant with The Art of Work and co-authored the paper. Stratflow's own work concerns the operating conditions the controls have to function under.
Has it been used on a mine?
The published case study is a driving-safety portfolio at a large Australian asset-intensive operator, reported under confidentiality. The paper sets out how the method applies to the main mining critical-risk families.
Does it replace bowties, performance standards or verification?
No. It sits on top of them and tests whether they hold. Existing bowtie and PHMP or SHMS work stays in place.
What does the regulator get out of it?
A traceable record: how the critical controls were identified, how each was scored and by whom, where the SME and frontline views diverged and how that was resolved, and what was done about each fragility.
Where can I read the paper?
The full paper is available as a PDF from this page. It was presented at the AusIMM International Mine Health and Safety Conference, Pan Pacific Perth, 5 to 7 October 2026.
About the authors
Kelvin Genn is the founder of Art of Work Solutions in Melbourne and the author of the Control Effectiveness Assessment methodology. His work applies human and organisational performance, learning teams and Safety-II to field-led control assessment across utilities, telecommunications, transport and other asset-intensive sectors.
Hendrik Lourens, FAusIMM, COHSProf, is CEO of Stratflow Australia, a senior consultant with The Art of Work and a member of the AusIMM Health and Safety Society Committee. His work applies Theory of Constraints and flow-based production to the operational stability that critical controls depend on.