Understanding the Evolution of Safety Thinking in Organizations

When accidents occur in workplaces or industrial settings, the immediate response often focuses on identifying what went wrong at the operational level. However, over the decades, safety experts have discovered that understanding accidents requires a much deeper analysis that extends far beyond surface-level causes. The evolution of safety thinking has fundamentally transformed how organizations approach accident prevention, moving from simple linear models to complex systemic frameworks that consider technology, human behavior, organizational culture, and regulatory environments.

The foundation: analyzing cause and mechanism 🔗

The journey toward modern safety thinking begins with two fundamental questions: why did an incident occur (cause) and how did it happen (mechanism). These questions form the bedrock of accident investigation and prevention strategies. Fault Tree Analysis (FTA) represents one of the most powerful tools developed to answer these questions systematically. This top-down approach examines an undesired event and works backwards to identify all possible contributing factors through logical relationships.

Unlike simple checklists, FTA uses Boolean logic gates to map out how various events combine to produce accidents. An OR gate indicates that any one of several events could trigger the incident, while an AND gate shows that multiple conditions must exist simultaneously. This visual representation helps safety professionals understand not just what failed, but how different failures interact to create dangerous situations.

From reactive to proactive analysis 🔗

Early accident investigations focused primarily on identifying the broken component or the worker error that immediately preceded the incident. This reactive approach, while useful for addressing immediate hazards, failed to prevent similar accidents from recurring. Modern FTA techniques now incorporate evidence gates that allow investigators to trace backwards through chains of events, revealing not just what happened, but the underlying conditions that allowed it to happen.

Expanding the scope of cause analysis 🔗

Perhaps the most significant shift in safety thinking has been the progressive broadening of what we consider when analyzing accident causes. This evolution can be traced through distinct eras, each representing a fundamental change in perspective.

The technological age 🔗

During the early industrial period, safety management focused almost exclusively on equipment and machinery failures. The prevailing belief was straightforward: if the technology is safe, then workers will be safe. Organizations invested in engineering solutions like improved machinery guards, better ventilation systems, and structural improvements to buildings. While these technological improvements reduced certain types of accidents, they couldn’t prevent incidents rooted in human decision-making or organizational failures.

The human factors revolution 🔗

By the mid-20th century, safety experts recognized that human factors contribute to approximately 80% of occupational accidents. This realization sparked a new era focused on understanding human error, worker training, and behavioral safety programs. However, this human-centric approach sometimes led to an overemphasis on blaming frontline workers while overlooking deeper systemic issues.

The organizational dimension 🔗

Major disasters like Chernobyl in 1986 and the Exxon Valdez oil spill in 1989 revealed that accidents often result from latent organizational factors such as poor management decisions, inadequate safety culture, and flawed organizational processes. This understanding shifted focus toward examining how management decisions, resource allocation, and organizational climate create conditions that either prevent or enable accidents. Modern accident models like the Swiss Cheese Model illustrate how organizational defenses can have holes that, when aligned, allow accidents to occur.

Cultural influences on safety 🔗

The most recent expansion in cause analysis recognizes that organizational culture fundamentally shapes safety outcomes. Culture encompasses the shared values, beliefs, and norms that determine how people behave when nobody is watching. A strong safety culture means that workers at all levels prioritize safety not because they’re required to, but because it’s embedded in how the organization operates. Organizations with mature safety cultures treat safety not as a compliance burden but as an indicator of operational excellence and employee engagement.

Understanding how accidents happen 🔗

While identifying why accidents occur addresses root causes, understanding the mechanism reveals the sequence of events that transform potential hazards into actual incidents. Mechanism analysis examines the temporal progression of failures and the interactions between different system elements.

The importance of sequential analysis 🔗

Accidents rarely result from a single failure. Instead, they typically involve a cascade of events where one failure creates conditions for the next. By mapping these sequences, safety professionals can identify critical intervention points where the accident chain could have been broken. This understanding is essential for developing effective preventive measures and revising operational procedures.

Revising procedures based on mechanisms 🔗

When investigators understand exactly how an accident unfolded, they can design targeted interventions. For example, if an accident occurred because a backup safety system failed to activate when a primary system malfunctioned, the solution might involve redesigning the activation mechanism, adding redundancy, or implementing better testing protocols. Without understanding the mechanism, organizations might implement ineffective countermeasures that address symptoms rather than underlying vulnerabilities.

New explanatory models for complex systems 🔗

As our understanding of accident causation has deepened, safety experts have developed increasingly sophisticated models to explain how incidents occur in complex sociotechnical systems.

Beyond linear causation 🔗

Traditional accident models depicted causation as a linear sequence-like dominoes falling one after another. While simple to understand, these sequential models proved inadequate for analyzing accidents in modern complex systems where multiple factors interact dynamically. Contemporary approaches recognize that accidents emerge from the complex interplay of technical systems, human operators, and organizational factors.

Systemic accident models 🔗

Modern systemic models like STAMP (Systems-Theoretic Accident Model and Processes) view accidents as resulting from inadequate control of safety constraints rather than simple component failures. These models examine how safety is maintained through control structures involving feedback loops, organizational oversight, and regulatory processes. When these control mechanisms fail or become ineffective, accidents can occur even when individual components function correctly.

The role of performance variability 🔗

Another advancement recognizes that performance in complex systems naturally varies due to numerous internal and external factors. Functional Resonance Analysis Method (FRAM) examines how normal variability in everyday activities can occasionally align in ways that produce unexpected outcomes. This perspective shifts safety thinking from preventing failures to managing variability and enhancing system resilience.

Integrating multiple perspectives 🔗

The most advanced safety thinking today integrates technological, human, organizational, and cultural perspectives into comprehensive frameworks. Recent research examining major accidents across industries confirms that addressing multiple levels simultaneously-strengthening feedback loops, improving organizational oversight, and enhancing control processes-is essential for preventing severe accidents in complex systems.

This integrated approach recognizes that technology, people, and organizations don’t operate in isolation. They form interconnected systems where changes in one element ripple through the others. Effective accident prevention requires understanding these interdependencies and addressing safety holistically rather than through isolated interventions.

What do you think? How has your organization’s approach to accident investigation evolved over time? Are you analyzing accidents at the technological, human, organizational, or cultural level-or integrating all these perspectives for comprehensive prevention?

References

1. https://safetyculture.com/topics/fault-tree-analysis
2. https://sixsigmastudyguide.com/fault-tree-analysis/
3. https://www.hsestudyguide.com/fault-tree-analysis/
4. https://jsystemsafety.com/blog/can-fault-tree-analysis-be-used-effectively-in-accident-investigation/
5. https://www.avetta.com/blog/the-five-eras-of-safety-maturity
6. https://www.iloencyclopaedia.org/part-viii-12633/accident-prevention/item/895-human-factors-in-accident-modelling
7. https://www.mdpi.com/2071-1050/14/10/5869
8. https://www.sciencedirect.com/topics/medicine-and-dentistry/accident-causation-model
9. https://risktec.tuv.com/knowledge-bank/the-evolution-of-safety-culture/
10. https://www.axiomllc.com/blog/how-safety-evolved/
11. https://www.edrawmax.com/fault-tree-analysis/
12. https://www.sciencedirect.com/science/article/abs/pii/S0957582022005729
13. https://pmc.ncbi.nlm.nih.gov/articles/PMC12180727/