The Procedure That Survived the Dead: Why Your Management System Didn't Change After the Last Fatality

A worker was fatally struck by a reversing haul truck on an open-cut mining site. The investigation found that the site's traffic management plan had not been updated following a layout change three months earlier. Pedestrian and vehicle segregation zones were outdated. The reversing alarm on the truck was functioning but inaudible over ambient noise. The operator had received no site-specific induction for the modified layout.

The investigation report was thorough. It identified the traffic management plan as the primary systemic failure. It recommended an immediate revision of the plan, installation of proximity detection systems, and a redesign of pedestrian exclusion zones. Fourteen months later, an external audit found: the traffic management plan had been revised on paper but the physical site layout remained unchanged. Proximity detection had been approved but not budgeted. The pedestrian exclusion zones existed in the revised plan but were not enforced on the ground. The procedure had survived the dead.

■ THE PAPER CHANGE TRAP Revising a procedure document is not the same as changing the system. When management treats the document update as the deliverable, rather than the operational change, the procedure becomes a liability shield, not a safety improvement.

Step 1: Create Urgency — And Sustain It

Incidents create instant urgency. The problem is that incident-driven urgency is emotional, not structural. It peaks in the first 72 hours and decays rapidly as production pressure reasserts itself. By the time the investigation report is finalised, the urgency has been replaced by fatigue. Kotter's research found that 50% of change efforts fail at this first step because leaders do not sustain urgency beyond the initial crisis.

In post-incident contexts, sustaining urgency requires a deliberate strategy: regular reporting to the board on implementation status, visible leadership engagement at the site, and — critically — tying the change to business continuity, not just moral obligation.

Step 2: Build a Guiding Coalition — With Authority

Most post-incident action plans are managed by the HSE department alone. This is structurally insufficient. HSE teams lack the authority to change production schedules, reallocate capital budgets, or redesign operational workflows. Without a coalition that includes operations, engineering, procurement, and senior leadership, recommendations die in the handover.

KEY TAKEAWAY If the post-incident change plan is owned by HSE alone, it will fail. Systemic change requires a coalition with budget authority, operational authority, and executive sponsorship. Anything less is a documentation exercise.

Step 3: Form a Strategic Vision for the Change

Post-incident recommendations typically list actions ("revise the procedure," "install barriers," "retrain staff"). They rarely articulate a vision for what the system should look like after the change. Without a vision, the actions become isolated tasks with no coherent design intent. The revised procedure is updated in isolation. The barrier is installed without integrating it into the traffic management system. The retraining occurs without updating the competency framework.

Condition 1 — Executive Ownership of Implementation: The COO or Operations VP must own the implementation plan, not the HSE Director. If the most senior operational leader is not accountable for implementation, the change will be filtered through operational convenience.

Condition 2 — Budget Allocation Within 30 Days: Any recommendation that requires capital expenditure must have budget approval within 30 days of the report. Recommendations without budgets are aspirations. The speed of budget allocation is a direct measure of organisational commitment.

Condition 3 — Operational Verification, Not Document Verification: Implementation must be verified by observing the change in operation, not by checking that the document has been updated. A revised traffic management plan that is not reflected in the physical site layout is not implemented — it is filed.

Condition 4 — 30-60-90 Day Effectiveness Reviews: At 30, 60, and 90 days post-implementation, verify that the change is producing the intended effect. Is the new procedure being followed? Has the hazard exposure been reduced? Are the barriers functioning? If not, the change has failed and requires redesign.

Condition 5 — Feed the Investigation Back into the Management System Review: Every investigation finding should trigger a review of the relevant management system element — not just the specific procedure. If the traffic management plan failed, the question is not just "what was wrong with this plan?" but "what is wrong with how we manage traffic management plans?"

■ THE KOTTER TEST Apply Kotter's 8 steps to your last fatality investigation's action plan. At which step did the change effort stall? That is where your management system's change capability breaks down — and that is what needs fixing before the next incident.