Pedestrian-Vehicle Interaction, an Improvement Notice and a Tonnage Bonus

Hierarchy of controls

• Engineered separation removes the interaction entirely - the highest-order control available short of eliminating the task. It is passive: it works when people are tired, distracted or in a hurry.
• Proximity detection is an engineering control, but it is a warning device that depends on the operator reacting, on the tag being worn, and on the unit being maintained and functional. In practice it sits closer to an administrative control.

Recommendation

1. Build the pedestrian drift first. $310 k, eight weeks, inside the 90-day notice, and it addresses the hazard at source.
2. Keep the detection system as a second-layer option for the remaining interfaces (blind corners, passing bays, the crusher tip) - but not as the primary control, and not as a substitute for separation.
3. Reject the framing that this is $1.8 M versus $310 k on cost alone. The comparison is $1.8 M spending most of the budget on a warning device, versus $310 k eliminating the exposure and leaving money for the remaining interfaces.

Also test the 90-day deadline against the capital project. The $1.8 M system has procurement, installation, commissioning and training; a plan that promises it inside 90 days is probably not credible, and a missed notice escalates the regulatory response.

What the numbers say

• Three lost-time injuries, all the same mechanism, on the same level, means the control was absent or not working for 14 months - a system failure, not three unlucky individuals.
• A near-miss reporting rate of 0.3 against a 2.0 target is not a good safety result; it is a reporting-culture red flag. Low reporting usually means under-reporting, not safety.
• The supervisor bonus tied to monthly tonnes directly rewards speed over traffic rules, and it penalises the supervisor whose crew stops the job and loses tonnes. That is the mechanism creating the exposure.

Actions

1. Redesign incentives: include leading indicators (near-miss reporting quality, inspection completion, take-5 compliance) in the supervisor scorecard, and explicitly protect the supervisor who halts work.
2. Run a reporting-quality audit - a short anonymous survey plus a review of the last 200 reports - to find out why near misses are not being raised, and close the loop visibly (show what changed as a result of reports).
3. Co-design the traffic-management plan with the workforce, including the returned worker and the consultation committee. People who walk that drift daily will identify the pinch points that a desk exercise misses.
4. Verify behaviour with supervised observation of the traffic rules, not just the paperwork.

90-day plan

• Days 0-14: risk assessment (bow-tie) of the 240 m level; interim controls immediately - speed limits with physical enforcement, exclusion zones during loader operation, radio call-up protocols at blind corners, mandatory high-visibility clothing, and a supervised trial measuring compliance.
• Days 14-56: build the pedestrian drift, refuge bays and cross-cuts; issue the traffic-management plan; train and assess every operator and every person who walks the level.
• Days 56-90: commission and verify - walk the route with the workforce, audit the tag board, run a drill, and document objective evidence of the control working for the regulator.
• Keep the proximity-detection decision open, with a scoped trial for blind corners only.

Justifying the spend

• Do not argue payback. The direct cost of three lost-time injuries (perhaps $150 k each in medical, compensation and lost production) is tiny against $1.8 M, so a payback case would reject both options.
• The real justification is the credible potential for a fatality or permanent disability at a pedestrian-loader interface, the legal duty to eliminate risk so far as is reasonably practicable, the improvement notice, and the licence to operate.
• Frame the business case around catastrophic-risk reduction and continuity of operations - a fatal incident or a prohibition notice would cost far more than either capital item.
• Make sure the safety culture change (incentives, reporting) is part of the formal response: installing hardware over a bonus system that rewards speed will reproduce the same injuries with better equipment.