• Human Error Is the Start of the Question: James Reason and the System Behind the Mistake
    2026/08/17

    If an investigation ends with “worker error”, it probably has not gone far enough.

    In this episode of Safety Jon: The Real Risk Podcast, Jon examines the work of Professor James Reason and why his thinking remains central to understanding serious incidents, organisational accidents and the failures that sit behind the person closest to the consequence.

    The Swiss cheese model is well known across safety, but the useful part was never the cheese. Reason’s work challenges organisations to look beyond the final human action and examine the latent conditions, failed defences, management decisions and organisational weaknesses that allowed an error to become an injury, fatality or major event.

    The episode also looks at what happens when leadership takes the opposite approach. Pathological safety cultures blame workers, protect management decisions and treat adverse events as individual failures. That approach does more than produce poor investigations. Workers watch what happens to their colleagues, learn what the organisation rewards and punishes, then adjust what they report, challenge and disclose accordingly.

    Jon also draws on his former regulator experience, including responding to a serious incident at a food manufacturing workplace where a worker suffered a finger amputation after reaching into a heat-sealing machine. The employer’s immediate explanation was that the worker should not have reached into the equipment. The regulatory response focused instead on the failed system, including inadequate guarding.

    The worker was in hospital undergoing surgery. The machine still had to answer some questions.

    In this episode
    • Why “human error” describes an event but rarely explains it.
    • James Reason’s distinction between active failures and latent organisational conditions.
    • What the Swiss cheese model actually tells us about failed defences.
    • Why retraining a worker is often a weak corrective action when the underlying system remains unchanged.
    • How regulators examine plant, systems of work, supervision, guarding, training, design and organisational controls.
    • Why worker blame can become a symptom of pathological safety leadership.
    • How punitive responses affect reporting, consultation, trust and workforce behaviour.
    • Why workers must be treated as people operating within systems, not convenient endpoints for an investigation.
    • The difference between accountability and blame.
    • Why a fair investigation process matters even where a worker made an error.
    • How leaders influence safety culture through the way they respond when something goes wrong.
    • Why the question after an incident should not simply be “Who made the mistake?”, but “What allowed that mistake to produce this consequence?”

    The central point

    A worker can make an error and still not be the cause of the incident.

    People forget things. They misunderstand information, become distracted, make assumptions and sometimes make poor decisions. A functioning safety system anticipates that reality and provides effective defences between ordinary human fallibility and serious harm.

    When one mistake can place a hand into hazardous machinery, put a person in front of mobile plant or expose a worker to an uncontrolled fatal risk, the organisation has a system problem worth investigating.

    Human error is not the end of the investigation.

    It is where the investigation gets interesting.

    About Safety Jon: The Real Risk Podcast

    Safety Jon: The Real Risk Podcast looks at serious incidents, prosecutions, investigations, organisational failures and the safety concepts that actually matter in operational workplaces.

    No theatre, no corporate fog and no pretending another toolbox talk automatically fixes a failed system.

    Real safety. No theatre.

    Disclaimer

    This podcast provides general information and commentary about workplace health and safety, risk management and incident investigation. It does not constitute legal advice and should not be relied upon as a substitute for advice specific to your organisation, jurisdiction or circumstances.

    Legislation, regulatory guidance and legal obligations vary between Australian jurisdictions and may change over time. Where a workplace incident, regulatory matter or legal issue requires a decision, obtain appropriate professional advice and confirm the current requirements applying to the workplace.

    続きを読む 一部表示
    47 分
  • The Test That Wasn’t Done: The Bankstown-Lidcombe Hospital Gas Mix-Up
    2026/08/10

    In 2016, a catastrophic medical gas installation failure at Bankstown-Lidcombe Hospital in Sydney resulted in nitrous oxide being supplied through an outlet that should have delivered oxygen to a neonatal resuscitation unit.

    Two newborn babies were affected. One died and the other suffered serious injuries.

    This was not simply a case of two pipes being connected incorrectly.

    The NSW Chief Health Officer’s investigation identified failures in installation, testing and commissioning, along with broader problems involving project planning, risk management and governance. Procedures intended to detect exactly this type of error were either not followed or were ineffective.

    In Episode 4 of Safety Jon: The Real Risk Podcast, Jon Hillis examines how a physical installation error was able to survive multiple opportunities for detection before reaching the point of patient care.

    We look at:

    • how the medical gas cross-connection occurred

    • why testing and commissioning are safety-critical controls, not administrative formalities

    • the danger of relying on labels, documentation and assumptions without physical verification

    • contractor and client responsibilities when safety-critical systems are installed or modified

    • how failures across several layers of control can combine into a catastrophic outcome

    • why commissioning must establish that a system actually works as intended before it is placed into service

    • what this incident means well beyond healthcare, including plant, electrical systems, pressure systems, engineering projects and other safety-critical infrastructure

    The central lesson is simple.

    A system is not safe because the paperwork says it is safe.

    Where failure can kill or seriously injure someone, verification needs to demonstrate that the control exists, functions correctly and produces the intended result.

    That is the difference between completing a commissioning process and actually commissioning a safe system.

    Source material

    This episode draws on publicly available material concerning the Bankstown-Lidcombe Hospital medical gases incident, including the NSW Chief Health Officer’s Final Report and associated NSW Health material.

    Content note

    This episode discusses the death and serious injury of newborn babies and the circumstances surrounding a significant healthcare safety failure.

    Disclaimer

    Safety Jon: The Real Risk Podcast is provided for general information and education. It does not constitute legal, medical or professional advice. Legislative and regulatory requirements vary between Australian jurisdictions and should be checked against current applicable requirements.

    Safety Jon: The Real Risk Podcast

    Practical safety. Real incidents. Lessons worth learning.

    続きを読む 一部表示
    28 分
  • Dreamworld: When the Safety System Failed
    2026/07/30

    Content note: This episode discusses a fatal public safety incident and the deaths of four people.

    On 25 October 2016, four people died on Dreamworld’s Thunder River Rapids Ride. The subsequent coronial inquest identified significant deficiencies involving the ride’s design and risk assessment, engineering oversight, maintenance records, operating procedures, training, emergency arrangements and the treatment of earlier incidents.

    This episode examines how a foreseeable technical event was allowed to become catastrophic. It considers the difference between having safety documentation and having a functioning safety system, particularly where plant is modified over time, faults recur, responsibilities become fragmented and audits fail to test the controls that matter.

    The lessons extend well beyond amusement rides. They concern management of change, competent engineering review, critical control verification, maintenance governance, operator competence, emergency response, incident learning and officer oversight.

    Safety Jon examines serious incidents, prosecutions and operational failures to identify practical lessons for WHS practitioners, managers and duty holders.

    General information only. This episode does not provide legal advice.

    続きを読む 一部表示
    34 分
  • So… What is a Safe System of Work?
    2026/06/30

    Episode 2 of Safety Jon: The Real Risk Podcast looks at one of the most important and often misunderstood phrases in workplace health and safety: a safe system of work.

    A safe system of work is not just a procedure, SWMS, checklist, induction, toolbox talk or training record. Those documents may support the system, but they do not prove the work is actually safe. The real question is whether the work has been properly planned, controlled, supervised, consulted on, verified and reviewed under real operating conditions.

    This episode breaks down what a safe system of work means in practice, including the gap between work as imagined and work as actually done. It looks at why organisations get into trouble when they rely on paperwork, memory, worker experience or “old mate knows what he’s doing” instead of designing controls that hold up when people are tired, busy, under pressure, new to the task, dealing with change or working around equipment faults.

    The episode also considers safe systems across physical and psychological risk, including mobile plant and pedestrian separation, machine guarding and jam clearing, fatigue, psychosocial hazards, labour hire, consultation, supervision, leadership and field verification.

    In this episode:

    • What a safe system of work actually means in practical terms.
    • Why procedures, SWMS, JSEAs, permits and training records are not the same as control.
    • Why work as done matters more than work as imagined.
    • How documents should support a safe system, not replace it.
    • Why safe systems need to account for human error, fatigue, pressure, change and foreseeable workarounds.
    • How forklift and pedestrian interaction, machine guarding, fatigue, psychosocial risk and labour hire expose weak systems.
    • Why consultation with workers and HSRs matters.
    • Why supervisors are a control point, not just job allocators.
    • Why leadership must resource, support and verify the system.
    • How to test whether a task is genuinely controlled or just documented.

    Safety Jon: The Real Risk Podcast is general commentary only. It is not legal advice, consulting advice or a substitute for organisation-specific risk assessment, competent advice or professional judgement. Duties and regulatory requirements vary between Australian jurisdictions, and organisations should seek appropriate advice for their own operations, workforce, hazards and controls.

    続きを読む 一部表示
    35 分
  • The CEO Who Spooked Wall Street by Talking About Safety
    2026/06/23

    Episode 1 of Safety Jon: The Real Risk Podcast starts with Paul O’Neill, Alcoa, and one of the better-known safety leadership stories in corporate history.

    When O’Neill became Chairman and CEO of Alcoa in 1987, investors expected the usual executive language about profit, growth, capital and shareholder value. Instead, he opened with worker safety and his intention to pursue zero injuries.

    This episode looks at why that mattered. It was not safety theatre, and it was not paperwork wearing a hard hat. O’Neill treated safety as a diagnostic tool for the way the business actually operated, including how leaders received bad news, responded to risk, resourced controls, listened to workers and verified that the system worked in practice.

    The episode also connects the Alcoa case study to Australian WHS and OHS leadership duties, including officer due diligence, critical risk, control verification and the difference between genuine safety leadership and administrative confidence.

    In this episode:

    • Paul O’Neill, Alcoa and the business case for safety leadership.
    • Why safety is not separate from business performance.
    • How the Hudson Safety Culture Maturity Model explains pathological, reactive, calculative, proactive and generative cultures.
    • Why poor leadership flows through an organisation and shapes what workers, supervisors and managers tolerate.
    • What Ductus Exemplo, lead by example, means in a safety context.
    • Why dashboards, procedures and policies are not proof of control.
    • Why officers and senior leaders need to understand serious risks, critical controls, reporting pathways, resourcing and verification.

    Safety Jon: The Real Risk Podcast is general commentary only. It is not legal advice, financial advice, investment advice, consulting advice or a finding about any current Alcoa operation, officer, worker, site, system or incident. The Alcoa and Paul O’Neill material is used as a leadership case study, with confirmed historical facts separated from practical inference.

    続きを読む 一部表示
    29 分