Mining SafetyVehicle InteractionOpen Cut

Vehicle Interaction Risks in Open Cut Mining: Lessons from Byerwen

Background

Vehicle interactions remain the leading cause of fatalities in Australian surface mining. Every year, regulators across Queensland and New South Wales publish safety alerts, investigation reports, and compliance notices related to collisions, near-misses, and rollovers involving mobile equipment. Despite decades of awareness campaigns, updated traffic management plans, and new technology, workers continue to die in incidents that follow patterns the industry has seen many times before.

The term “vehicle interaction” covers any event where two or more pieces of mobile equipment come into proximity in a way that creates risk. In open cut mining, this includes haul trucks passing light vehicles on ramps, dozers reversing near personnel, water carts operating alongside loaded trucks, and cranes working in active traffic zones. The scale of the equipment amplifies every mistake. A loaded haul truck can weigh over 500 tonnes. A light vehicle weighs roughly 3 tonnes. There is no contest in a collision.

Queensland’s Resources Safety and Health Queensland (RSHQ) has consistently identified vehicle interaction as one of the principal hazards requiring dedicated management under mining safety legislation. The obligation sits with Site Senior Executives (SSEs) and operators to maintain Safety and Health Management Systems (SHMS) that address these risks. And yet, the same failures keep appearing in investigation after investigation.

What Happened at Byerwen

Byerwen Mine is an open cut metallurgical coal operation in Queensland’s Bowen Basin, operated by QCoal Group. In August 2024, two workers died at the site in separate incidents less than three weeks apart.

On 3 August 2024, Chris Schloss was killed when a crane boom rolled over during lifting operations. The incident occurred during maintenance activities in an area where mobile equipment was operating. The circumstances pointed to failures in exclusion zone management and the interface between maintenance tasks and active mining operations.

On 22 August 2024, John Linwood was killed in a collision between a light vehicle and a haul truck. Light vehicle and haul truck interactions are among the most documented and warned-about scenarios in Australian mining. The size disparity alone makes these events catastrophic whenever they occur.

Two fatalities at a single site within three weeks is extraordinarily rare in modern Australian mining. The proximity of the two events raised immediate questions about whether the site’s safety management system was adequate, whether there were shared root causes, and whether the operator had responded effectively to the first incident before the second occurred.

RSHQ launched investigations into both incidents. The Queensland Mines Inspectorate issued directives, and the industry was forced to confront the reality that a site can have all the required documentation, conduct regular toolbox talks, and still experience catastrophic failures in basic hazard controls.

The Systemic Question

When a single fatality occurs, it is standard to investigate the immediate causes and contributing factors. When two occur at the same site in rapid succession, the focus shifts to systemic failure.

Systemic failure means the safety management system itself was not functioning as intended. This goes beyond a single worker making a single mistake. It points to breakdowns in planning, supervision, communication, risk assessment, and organisational culture. The question is not just “what went wrong in this incident” but “why did the system allow this to happen twice?”

Common systemic issues in vehicle interaction fatalities include traffic management plans that exist on paper but are not enforced on the ground, inadequate separation between light and heavy vehicle routes, poor visibility conditions that are not controlled, fatigue management failures that impair operator judgement, and a normalisation of risk where near-misses are not treated as precursors to serious events.

RSHQ’s own data reinforces the pattern. Between January 2022 and August 2023, the regulator recorded 19 separate dozer rollover incidents across Queensland mining operations. Each of these represented a potential fatality. The NSW Resources Regulator has published multiple warnings about overtaking incidents on haul roads where light vehicles attempt to pass trucks on grades, creating head-on collision risks.

Collision Avoidance and Proximity Detection

The technology to prevent or reduce vehicle interaction incidents exists and is improving rapidly. Collision avoidance systems (CAS) and proximity detection systems (PDS) use a combination of GPS, radar, LiDAR, and radio frequency to detect when equipment is approaching another piece of equipment or a person. These systems can provide warnings to operators, and in some configurations, can intervene directly by slowing or stopping equipment.

Industry adoption of CAS has increased. A 2024 survey found that 72% of Australian surface mining operations had some form of collision avoidance technology deployed, up from 62% in 2022. However, deployment does not equal effectiveness. Many sites run CAS in “alert only” mode, where operators receive warnings but the system does not intervene. Alert fatigue is a known problem: when operators receive constant proximity warnings during normal operations, they begin to ignore them.

The International Council on Mining and Metals (ICMM) committed in 2019 to the goal of eliminating vehicle-related fatalities at member company operations. This commitment was initially targeted at implementation milestones by 2025. Progress has been made, but the Byerwen incidents demonstrated that even in a jurisdiction with strong regulatory oversight, the gap between commitment and execution remains significant.

For CAS to be effective, it requires proper calibration to the specific site geometry, integration with traffic management plans, regular maintenance and testing, and a culture where operators trust and respond to the system rather than treating it as background noise.

Autonomous Haulage

The long-term trajectory for eliminating vehicle interaction fatalities in open cut mining points toward autonomous haulage systems (AHS). By removing the human operator from the haul truck cab entirely, AHS eliminates a major category of human error from the equation.

Rio Tinto operates the largest autonomous haul truck fleet in the world across its Pilbara iron ore operations. As of 2024, approximately 80% of their haul fleet runs autonomously. The safety record of autonomous trucks has been significantly better than manned operations, with zero fatalities attributed to autonomous truck movements.

Professor David Cliff from the University of Queensland’s Minerals Industry Safety and Health Centre has stated directly that “vehicle accidents are preventable.” His research group has studied mining fatalities across Australia and consistently found that the technology, systems, and knowledge to prevent these events already exist. The barrier is implementation, cost, and the willingness of operators to invest in change.

Autonomous haulage is not a silver bullet. It is expensive to implement, requires significant infrastructure investment (high-precision GPS, communications networks, integrated mine planning systems), and introduces new categories of risk around system failures and cybersecurity. Not every operation can justify the capital expenditure. Smaller mines and operations with shorter mine lives may never adopt full autonomy.

But the direction is clear. The sites with the best vehicle interaction safety records are those with the most advanced technology and the most disciplined management systems.

What Operators Should Be Doing Now

For operations that are not ready for autonomous haulage, the immediate priorities are straightforward but demand genuine commitment.

Separate light and heavy vehicles. Physical separation through dedicated roads, time-based separation through controlled access windows, or both. Every interaction that can be eliminated through design is one that cannot be failed through human error.

Implement CAS with intervention capability. Alert-only systems have limited value when operators are fatigued, distracted, or habituated to warnings. Systems that can slow or stop equipment provide a genuine last line of defence.

Treat near-misses as incidents. Every near-miss in a vehicle interaction scenario is a rehearsal for a fatality. If an operation is recording near-misses but not changing practices in response, the recording is performative.

Audit traffic management plans against actual practice. Plans that were written for the previous pit stage, or that do not reflect current haulage routes and shift patterns, provide a false sense of compliance.

Review shift handover processes. Many vehicle interaction incidents occur during shift changes, when traffic patterns shift, communication lapses, and equipment transitions between operators.

Invest in visibility. Lighting, reflective markings, camera systems, and clean windscreens are basic but frequently neglected. Dust suppression on haul roads is a visibility investment as much as an environmental one.

The Byerwen fatalities were not freak events. They were the predictable result of known hazards meeting inadequate controls. The industry has the knowledge and the technology to prevent them. What it needs is the discipline to apply both, consistently, across every shift, at every site.

Looking Forward

The investigation findings from the Byerwen incidents, once published by RSHQ, will provide detailed analysis of the specific failures involved. But the broader lessons do not require waiting for those reports. The patterns are well established. Vehicle interaction fatalities follow a consistent set of contributing factors that have been documented across hundreds of incidents over decades.

The question for every open cut mining operation in Australia is not whether they could experience a similar event. It is whether their current controls would prevent one. The answer to that question requires honest assessment, not optimistic assumption.