Introduction
Vehicle interaction is one of the defining fatal risks of Australian surface mining. A rear-dump haul truck of the class used in Pilbara iron ore operations carries a payload above 200 tonnes and has blind areas extending several metres forward of the machine, so a light vehicle within them is invisible to a competent operator. Mobile plant remains a leading source of traumatic injury nationally (Safe Work Australia 2023) and on Western Australian mining operations (DEMIRS 2023).
This case study investigates a serious injury arising from such an interaction at Kurra Downs, a hypothetical open-pit iron ore operation south-east of Newman run by the hypothetical Pilbara Ridge Iron Pty Ltd; the operation, personnel and events are invented for this unit. It moves beyond the operator error attributed in the initial shift report by applying the Incident Cause Analysis Method (ICAM), re-scoring the risk, proposing controls mapped to the hierarchy, and identifying the duties engaged under Western Australian law.
The Incident
Kurra Downs runs an 18 unit haul fleet around the clock, with approximately 490 personnel including contractors on 12.5 hour shifts, fly-in fly-out from Perth and Port Hedland. The Ramp 4 intersection, where the Stage 3 access ramp meets the main haul road, is the operation’s busiest point of contact between light vehicles and loaded trucks.
On the night of 14 May a contractor maintenance technician was tasked to attend a dewatering pump that had tripped offline in the Stage 3 sump. Travelling alone, the technician stopped at Ramp 4, requested access on the general radio channel rather than the pit channel, received no acknowledgement, and entered the intersection, where a loaded haul truck contacted the light vehicle inside its blind area. Table 1 reconstructs the sequence.
Table 1: Reconstructed incident timeline, night shift 14 May (local time)
| Time | Event | Evidence source |
|---|---|---|
| 17:45 | Night shift pre-start. Ramp 4 lighting plant out of service; no alternative control assigned. | Pre-start record |
| 18:12 | HT-142 begins ore haulage from the Stage 3 pit to the run-of-mine pad. | Fleet system |
| 21:30 | Pump P-06 trips offline; the pit supervisor requests maintenance attendance. | Control room log |
| 22:47 | LV-08 departs the workshop, driven alone by a contractor technician. | Telematics |
| 23:02:41 | LV-08 stops at Ramp 4 and requests access on channel 1 (general) rather than channel 3 (pit). No acknowledgement. | Radio log; camera |
| 23:02:52 | LV-08 enters the intersection 11 seconds later, without a repeat call. | Telematics; camera |
| 23:02:57 | HT-142, loaded at 38 km/h in a 40 km/h zone, contacts LV-08 inside the forward blind area. | Fleet system |
| 23:03 | The truck operator stops and transmits an emergency call. | Radio log |
| 23:21 | Emergency response on scene; the driver is conscious and is stabilised. | Response report |
| 03:10 | Royal Flying Doctor Service retrieval departs for Hedland Health Campus. | Retrieval record |
| 06:20 | Regulator notified; incident site isolated and preserved. | Notification record |
The technician sustained multiple lower limb fractures, remained in hospital for eleven days and was absent for four months. There was no fatality. The rollover protective structure retained its shape and the driver was restrained, so severity was determined partly by defences that happened to be intact, which is not the same as the risk being controlled (Dekker 2019).
Analytical Framework
ICAM is the structured investigation method used across most of the Australian resources sector. It operationalises Reason’s organisational accident model, in which an incident occurs when weaknesses in successive layers of defence momentarily align, and classifies findings at four levels: organisational factors, task and environmental conditions, individual and team actions, and absent or failed defences (Reason 2016). The first two are latent conditions, the third active failures, the fourth the barriers that should have arrested the sequence. Figure 1 applies the framework to this incident.
The framework resists the first available explanation. The initial shift report recorded the cause as failure to follow the positive communication procedure, which names the last act in the sequence and stops there. Analysis of 508 mining incidents in Queensland using a comparable taxonomy found unsafe acts were routinely preceded by identifiable organisational deficiencies (Patterson and Shappell 2010), so the driver’s actions are treated below as a finding to be explained.
Absent and Failed Defences
Five defences that should have separated the two vehicles were absent or ineffective. Most consequentially, HT-142 carried no proximity detection: an approved retrofit to an intervention-capable standard, the highest tier of the vehicle interaction control framework developed by the Earth Moving Equipment Safety Round Table in Brisbane (EMESRT 2019; ICMM 2020), had reached only 12 of 18 trucks. The communication protocol depended entirely on human compliance and offered no means of detecting non-compliance. There was no bypass lane, the windrow and stop bar had been degraded by grading after rain (DMIRS 2019), the intersection was unlit, and its stop sign sat in a backlog of 214 open defects. The driver’s competency verification had expired six weeks earlier, and because contractor records sat outside the site database it did not block the callout.
Three defences did operate: the rollover protective structure and restraint contained the occupant space and are the proximate reason the outcome was a serious injury rather than a fatality; the beacon and warning whip were serviceable, though of little value once the vehicle entered the blind area; and emergency response was on scene within eighteen minutes. All three mitigate consequence rather than prevent the event, so every barrier positioned to prevent the interaction failed.
Contributing Factors
Table 2 records the contributing factors established by the investigation, coded to the four ICAM levels. The pattern is characteristic of organisational accidents: managerial decisions taken months earlier, none of which appeared unsafe at the time, created the conditions in which a single lapse became consequential.
Table 2: Contributing factors classified by ICAM level
| ICAM level | Code | Contributing factor |
|---|---|---|
| Organisational factors (latent) | OF1 | Collision avoidance retrofit deferred two quarters after a capital reprioritisation; six of 18 trucks unfitted, no compensating control. |
| OF2 | A pit redesign four months earlier raised Ramp 4 light vehicle transits by 41 per cent and renumbered the pit channel; the traffic plan was not reviewed. | |
| OF3 | Contractor competency records held outside the site access system, so a lapsed verification did not block haul road authorisation. | |
| OF4 | Only 61 of 96 scheduled critical control verifications completed last quarter; road defects absorbed into a 214 item backlog. | |
| Task and environmental conditions (latent) | TEC1 | Night operation, intersection lighting out of service, no portable lighting substituted. |
| TEC2 | Windrow and stop bar removed by grading after rain; the approach grade shortened sight distance. | |
| TEC3 | Driver in hour 11.6 of a 12.5 hour night shift, day 12 of a 14 day swing; hours-of-work exceptions not reviewed. | |
| Individual and team actions (active) | ITA1 | Access request sent on the general channel; intersection entered without acknowledgement or a repeat call. |
| ITA2 | Supervisor tasked a lone night callout without confirming competency or nominating an escort. | |
| ITA3 | Truck operator was within the speed limit and could not detect a vehicle inside the blind area; a design limitation, not an error. | |
| Absent or failed defences | AFD1 | No proximity detection or collision avoidance fitted to HT-142. |
| AFD2 | Communication protocol not completed, with no engineered means of detecting non-compliance. | |
| AFD3 | No segregation, bypass lane, stop bar, signage or lighting at the intersection. |
Risk Re-assessment and Performance Measurement
The site register scored this interaction on a 5×5 matrix in which risk score equals likelihood multiplied by consequence, with likelihood L1 (rare) to L5 (almost certain), consequence C1 (insignificant) to C5 (catastrophic, meaning a fatality or permanent disabling injury), and bands of Low (1 to 4), Medium (5 to 9), High (10 to 16) and Extreme (17 to 25). It recorded L2, assuming protocol compliance. Fleet records for the 90 days before the incident do not support that assumption:
- Average daily transits = 4,140 / 90 = 46 per day
- Protocol compliance, from 120 sampled transits of which 78 showed a completed exchange = 78 / 120 = 0.65, or 65 per cent
- Transits without completed positive communication = 4,140 x (1 – 0.65) = 1,449 over 90 days, roughly 16 per day
Sixteen unprotected exposures per day at an intersection carrying loaded haul trucks supports L4, not L2. Consequence remains C5, because the energy involved has fatality potential irrespective of the outcome realised.
- Register score before the incident: Risk = L2 x C5 = 10 (High)
- Reassessed on the evidence: Risk = L4 x C5 = 20 (Extreme)
- Residual with administrative controls only: Risk = L3 x C5 = 15 (High)
- Residual with isolation and engineering controls added: Risk = L1 x C5 = 5 (Medium)
The arithmetic makes the hierarchy of controls argument quantitatively: rewriting the procedure moves the score five points, removing reliance on it fifteen. Consequence cannot fall below C5 while both vehicle classes share a road, so the residual settles at Medium.
Injury frequency rates were also recalculated. The total recordable injury frequency rate (TRIFR) expresses recordable injuries, comprising lost time, restricted work and medical treatment cases, per million hours worked. The operation recorded nine recordable injuries across 1,200,000 hours in the 12 months to the incident:
- TRIFR = (recordable injuries x 1,000,000) / hours worked = (9 x 1,000,000) / 1,200,000 = 7.50 per million hours
- TRIFR including this incident = (10 x 1,000,000) / 1,200,000 = 8.33 per million hours
- Change = 8.33 – 7.50 = 0.83, an increase of 11.1 per cent
- Lost time injury frequency rate = (4 x 1,000,000) / 1,200,000 = 3.33, up from 2.50
Recordable rates aggregate high frequency, low consequence events and are a poor proxy for control over low frequency, high consequence hazards; organisations managing to the rate rather than the hazard have been surprised while their lag indicators improved (Hopkins 2019). Critical control verification completion of 61 / 96 = 0.635, or 63.5 per cent, described the state of the defences; a potentially fatal event that moves TRIFR by 0.83 does not.
Corrective Actions
Corrective actions follow the hierarchy of controls in the model code of practice on risk management, which obliges duty holders to adopt the highest order control that is reasonably practicable (Safe Work Australia 2018). Table 3 maps each action to a hierarchy level, an owner and a verification measure, so closure is evidenced by effectiveness rather than sign-off.
Table 3: Corrective action plan mapped to the hierarchy of controls
| Ref | Corrective action | Hierarchy level | Owner and due | Verification measure |
|---|---|---|---|---|
| CA1 | Dedicated light vehicle service road to the Stage 3 dewatering infrastructure, removing the haul road crossing. | Elimination | Mining Manager, 90 days | Zero Ramp 4 light vehicle transits |
| CA2 | Telemetry and remote restart on pit dewatering pumps so most trips clear without attendance. | Substitution | Maintenance Superintendent, 120 days | Callouts down at least 70 per cent |
| CA3 | Reinstate separation at all haul road intersections: bypass lane, windrow to 1.5 times the largest wheel height, hard stop bar, one-way routing. | Isolation | Mining Manager, 45 days | Weekly road audit with photographic record |
| CA4 | Complete the collision avoidance retrofit across all 18 trucks; restore lighting and fit vehicle-activated beacons. | Engineering | Site Senior Executive, 180 days | Fitment 18 of 18; availability above 95 per cent |
| CA5 | Reissue the traffic management plan and vehicle interaction hazard management plan with a review trigger on any pit design change; one access channel with mandatory acknowledgement. | Administrative | Health and Safety Manager, 60 days | Each design change carries a plan review record |
| CA6 | Integrate contractor competency records with site access so a lapsed verification blocks haul road authorisation. | Administrative | Contracts Manager, 60 days | Access block tested against a lapsed record |
| CA7 | No new personal protective equipment control; restraint monitoring, rollover protection and beacons retained, since such equipment cannot control a vehicle interaction. | Personal protective equipment | Health and Safety Manager, ongoing | Restraint compliance above 98 per cent |
Regulatory Duties
Western Australian mining operations moved from the Mines Safety and Inspection Act 1994 (WA) to the harmonised Work Health and Safety Act 2020 (WA) and Work Health and Safety (Mines) Regulations 2022 (WA) in 2022, administered by the Department of Energy, Mines, Industry Regulation and Safety through its mines safety inspectorate and aligned with the Safe Work Australia model laws.
Section 19 imposes the primary duty to ensure health and safety so far as is reasonably practicable, and section 18 defines that standard by weighing likelihood and degree of harm, what the duty holder knew or ought to have known, and the availability, suitability and cost of controls. An intervention-capable collision avoidance system was available, suitable and already funded, which makes its deferral difficult to defend and engages the officers’ due diligence duty in section 27. Sections 46 to 49 required the operator and the contractor to consult and coordinate, so the disconnected competency systems at OF3 are a coordination failure rather than a clerical oversight.
Because the injury required treatment as an in-patient, the event was notifiable under section 35, triggering notification under section 38 and site preservation under section 39, both discharged as Table 1 records. Vehicle operating areas are a principal mining hazard under the mines regulations, which require a documented management plan and its review after a related incident, so the omission at OF2 is a non-conformance independent of the incident. No fatality occurred, so the industrial manslaughter provisions are not engaged and the conduct falls to be assessed as a Category 2 offence.
Conclusion
The finding that a worker failed to follow the communication procedure is defensible as a description and useless as an explanation. ICAM instead identifies a deferred engineering control, a pit redesign never carried into the traffic management plan, a competency system disconnected from site access, and a defect backlog holding the signage and lighting at the very intersection where exposure had risen by 41 per cent.
Two conclusions follow. The risk was mis-scored rather than knowingly accepted: an evidence-based likelihood of L4 against a catastrophic consequence yields 20, and the register erred because it credited a control that had never been verified. Second, the actions in Table 3 will reduce the residual score to Medium, but only the elimination, isolation and engineering measures do substantive work; the administrative measures make the remaining exposure visible rather than smaller.
References
Dekker, S 2019, Foundations of safety science: a century of understanding accidents and disasters, CRC Press, Boca Raton.
Department of Energy, Mines, Industry Regulation and Safety (DEMIRS) 2023, Safety performance in the Western Australian mineral industry, DEMIRS, Perth.
Department of Mines, Industry Regulation and Safety (DMIRS) 2019, Guideline: safe operation of mobile equipment on Western Australian mining operations, DMIRS, Perth.
Earth Moving Equipment Safety Round Table (EMESRT) 2019, Vehicle interaction control framework, EMESRT, Brisbane.
Hopkins, A 2019, Organising for safety: how structure creates culture, Wolters Kluwer CCH, Sydney.
International Council on Mining and Metals (ICMM) 2020, Critical control management: good practice guide, ICMM, London.
Mines Safety and Inspection Act 1994 (WA).
Patterson, JM and Shappell, SA 2010, ‘Operator error and system deficiencies: analysis of 508 mining incidents and accidents from Queensland, Australia using HFACS’, Accident Analysis and Prevention, vol. 42, no. 4, pp. 1379-1385.
Reason, J 2016, Organizational accidents revisited, CRC Press, Boca Raton.
Safe Work Australia 2018, How to manage work health and safety risks: model code of practice, Safe Work Australia, Canberra.
Safe Work Australia 2023, Key work health and safety statistics, Australia 2023, Safe Work Australia, Canberra.
Work Health and Safety Act 2020 (WA).
Work Health and Safety (Mines) Regulations 2022 (WA).