Exploring the Model of Accident Phenomenon: A Path to Safer Work Environments

Every year, millions of workers are injured on the job – and many of these accidents are not random. They follow a pattern. They build up through a sequence of decisions, conditions, and missed warning signs. In 1969, researcher Jean Surry proposed a model that maps this exact process: the Model of Accident Phenomenon. Rather than viewing accidents as isolated events, Surry’s framework treats them as the end result of a chain of human and environmental interactions – a chain that can be interrupted at multiple points. Understanding this model is one of the most practical tools available for making workplaces genuinely safer.

The origins of Surry’s model 🔗

Jean Surry published her foundational work, Industrial Accident Research – A Human Engineering Appraisal, in 1969. Before developing her own model, she reviewed five broad categories of existing frameworks: chain-of-multiple-events models, epidemiological models, energy-exchange models, behaviour models, and systems models. Her conclusion was that none of these were mutually exclusive – each simply highlighted a different aspect of the same problem. This led her to synthesise them into one comprehensive model, while openly acknowledging it was provisional and not meant to be the final word.

Surry’s model is classified as a decision model – one that traces the full progression from hazard formation to injury or damage through the lens of human perception, cognitive processing, and physiological response. At its core, it operates within a “person plus environment” system and examines how the decisions people make – or fail to make – release danger.

The three principal stages of the model 🔗

Surry’s model works through a hierarchy of sequential questions. The answer to each question – essentially “yes” or “no” – determines whether the situation escalates toward an accident or is safely resolved. The model has three distinct stages, connected by two similar cycles.

Stage 1: The person-environment system 🔗

The starting point is the total environment in which a person operates – physical surroundings, equipment, workplace conditions, and the individual’s own characteristics. This stage accounts for all relevant situational and personal variables that together create the backdrop against which danger can emerge. The potential injury agent (a machine, a chemical, a height) is also identified here. Through the actions or non-actions of the individual, dangers grow out of this environment – they are not simply present from the start, but develop as the situation unfolds.

Stage 2: The danger build-up cycle 🔗

This is the first of the two analytical cycles in the model. A series of questions probes whether the person perceives the growing danger, understands its significance, and has both the ability and motivation to respond. If any of these questions receives a negative answer, the danger escalates toward becoming imminent. Surry’s model frames accident occurrence in terms of two classes of variables: situational and individual. Situational factors include noise, atmospheric conditions, sharp tools, heavy equipment, and the physical layout of the workplace. Individual factors include vision, age, perceptual and motor skills, emotional state, attitudes, and personal issues. The occurrence of any accident can be linked to one or both of these variable classes.

This stage is particularly important for prevention because it is where the widest window for intervention exists. If a worker perceives a hazard early and understands its consequences, the chain can be broken before any real danger materialises.

Stage 3: The danger release cycle 🔗

The second sequence of questions – the danger release cycle – links the level of danger to possible alternative outcomes once the danger is triggered. Once a hazard has built up and becomes active, the model again asks: Can the person observe it? Do they perceive it as a threat? Are they able to physically avoid it? Depending on the answers, the outcome may be an accident with injury, an accident without injury (a near-miss), or a safe resolution. Critically, the model also allows a distinction between deliberate risk-taking and unintentional negative outcomes – which has significant implications for how organisations design both their safety systems and their disciplinary approaches.

Why situational and individual factors both matter 🔗

One of the most valuable aspects of Surry’s model is its dual focus. Workplaces often default to blaming either the worker or the environment, but the model resists this binary. Behavioural models focus on inherent aspects of personality and attitude as well as situational aspects like tension and anxiety that may be at the origin of accidents, and research has shown that situational aspects are preponderant in explaining human errors.

This matters enormously in practice. Safety ultimately depends on factors that control human conduct – whether the individual has the knowledge, the skills, the opportunity, and the will to act safely in the workplace. Knowledge means workers must first be aware of the types of risk they face; the opportunity to act means the environment must actually allow safe behaviour; and the will to act depends on motivation, culture, and individual psychology. All of these connect back to Surry’s model – the danger build-up cycle will not be interrupted unless all three elements are simultaneously present.

Research has consistently identified risk perception as the most influential individual-level factor in unsafe behaviour, followed by work experience, skill level, knowledge, and risk-taking propensity. Surry’s model anticipated this finding: if a person cannot perceive that danger is building, no amount of skill or knowledge will prevent the accident.

The model as a framework for intervention 🔗

Because Surry’s model breaks down the accident process into discrete, answerable questions, it reveals exactly where intervention can be most effective. Rather than waiting for an accident to occur and then investigating retrospectively, organisations can use the model to identify which stage of the cycle is most vulnerable in their specific context.

Intervening at the danger build-up stage 🔗

Interventions at this stage focus on improving hazard perception and situational awareness before danger becomes critical. Safety information – delivered through campaigns, courses, or workplace materials – can change behaviour by providing workers with the right knowledge about hazards and their consequences, which in turn shapes attitudes and promotes safer behaviour. Training is the most direct tool here. Surry noted that past experience significantly reduces accidents, and that training can provide new employees with some of that experience before they encounter real hazards on the job.

Personnel selection also plays a role. Tests such as vision assessments, psychomotor evaluations, and job-sample tasks can help identify individuals whose profiles indicate higher accident risk on specific jobs, allowing targeted support before incidents occur.

Intervening at the danger release stage 🔗

Once a hazard becomes active, the remaining window for prevention is narrow – but it still exists. Engineering controls (machine guards, protective gear, environmental redesign) can reduce the severity of harm even after the danger is released. Organisations can improve outcomes through proper task design that accounts for human physical and mental limitations, effective communication channels, and a workplace culture that genuinely prioritises safety over speed.

Behaviour-based safety programmes operate directly within this stage. Among investigated variables, safety training was consistently identified as the most important factor influencing employees’ behaviour at the workplace, and high-quality training programmes can substantially reduce the rate of high-risk unsafe behaviours.

Limitations and later developments 🔗

Surry herself described the model as provisional, and subsequent researchers have refined it. The Swedish Work Environment Fund’s WEF model, developed in 1973, built on Surry’s framework by adding the concept of “objective danger” – hazards that exist independently of whether any individual perceives them. An evaluation of the WEF model by Swedish researchers in Malmö found that models focused primarily on individual behaviour risk creating an illusion of free choice: some dangers are not avoidable by the worker alone but only by management or through system redesign. This important critique reinforces that Surry’s model should be used as a complement to structural and organisational safety measures, not a replacement for them.

Surry’s model was among the early human information processing models of accident causation; later researchers including Hale, Wigglesworth, Lawrence, and Anderson developed similar frameworks, all emphasising human factors. Today, systemic models such as STAMP, CREAM, and FRAM extend this thinking further – accounting for organisational culture, regulatory environments, and complex sociotechnical interactions. Yet Surry’s core insight remains relevant: an accident is not simply an “event” but a process that can be analysed as a series of steps, and those steps are where prevention lives.

Applying the model in real workplaces 🔗

For safety managers and organisational psychologists, Surry’s model provides a structured way to audit workplace risk. Rather than asking only “what went wrong after the accident?”, it prompts the more useful question: “at which point in the build-up cycle did the system fail?” Was it that the worker lacked the knowledge to recognise a hazard? That the environment made safe behaviour structurally difficult? That the danger signal was present but not perceptible? Each of these failure points has a different intervention – and the model makes them visible.

The ILO’s guidance on accident prevention aligns with this thinking, noting that estimating workplace risk prospectively means evaluating the presence of risk factors before accidents occur – not counting injuries after the fact. Surry’s question-based, sequential structure is precisely designed for this kind of prospective analysis.

What do you think? If you were to apply Surry’s model to a workplace you’re familiar with, at which stage – the danger build-up or the danger release cycle – do you think most accidents go uninterrupted, and why? And do you think current workplace safety training does enough to address individual factors like perception and cognition, or does it focus too heavily on situational controls?

References

1. https://safetyengineering.softecks.in/492/
2. https://www.sciencedirect.com/topics/medicine-and-dentistry/accident-proneness
3. https://egyankosh.ac.in/bitstream/123456789/23924/1/Unit-4.pdf
4. https://www.sciencedirect.com/topics/social-sciences/accident-causation-model
5. https://www.iloencyclopaedia.org/part-viii-12633/accident-prevention
6. https://www.mdpi.com/2313-576X/12/1/14
7. https://pmc.ncbi.nlm.nih.gov/articles/PMC9159701/
8. https://www.hseblog.com/human-factors-and-their-influence-on-safety-performance/
9. https://pmc.ncbi.nlm.nih.gov/articles/PMC7191013/