For decades, organisations have invested heavily in safety management systems, training, procedures, behavioural observation programmes and safety culture surveys. These approaches have undoubtedly improved workplace safety. Yet a persistent question remains: why can two experienced, competent and well-trained people, presented with essentially the same information, make very different decisions about risk?
The answer may lie in a layer of safety performance that organisations have traditionally struggled to measure: the thinking that occurs between perception and behaviour.
Safety climate and culture assessments can tell us how people perceive safety. Behavioural programmes can tell us what people do. Incident investigations tell us what happened. But these approaches do not always explain the cognitive process that led someone to decide that a particular risk was acceptable, that a procedure could be adapted, that additional information was required, or that production should continue.
This is where the emerging concept of Safety Decision Intelligence becomes important.
Risk is not simply a matter of knowledge
Competence is essential, but knowledge alone does not determine behaviour. Decisions in safety-critical environments are influenced by analytical reasoning, operational discipline, relationships, adaptability, perception of risk and leadership influence.
SDIA™ – Safety Decision Intelligence Assessment – approaches this through six interacting dimensions: Analytical Risk Intelligence, Operational Discipline Intelligence, Human Leadership Intelligence, Adaptive Thinking Intelligence, Risk Judgement Intelligence and Safety Leadership Intelligence.
The significance lies less in producing a score than in understanding the pattern.
Consider a highly analytical and adaptive leader. Such an individual may be excellent at identifying weaknesses in existing systems and developing better solutions. However, if operational discipline is comparatively lower, that same strength may create a vulnerability: the individual may become comfortable modifying established methods because they genuinely believe there is a better solution.
Conversely, someone with extremely strong operational discipline may reliably follow procedures but struggle when an unexpected situation requires adaptation.
Neither person is necessarily unsafe. They simply process risk differently.
That distinction matters.
History repeatedly demonstrates the importance of decision-making
The Space Shuttle Challenger disaster remains one of the clearest examples. Engineers had concerns about O-ring performance in unusually cold temperatures, yet the launch proceeded. Diane Vaughan's subsequent research described the normalisation of deviance: repeated exposure to anomalies without catastrophic consequences gradually changed how risk was interpreted. NASA itself subsequently highlighted normalisation of deviance, organisational silence and weaknesses in independent safety oversight as contributing factors.
The lesson is not simply that somebody broke a rule. The more interesting question is: how did intelligent, technically competent people collectively reach a point where the risk became acceptable?
A similar lesson emerged from BP Texas City in 2005, where 15 people were killed and 180 injured. Human-factors analysis identified latent organisational conditions and safety-system deficiencies that influenced operators' decisions and actions. Investigations also identified procedural deviations, equipment problems and wider cultural weaknesses.
Again, behaviour was visible. The deeper issue was the environment and decision process producing that behaviour.
The Deepwater Horizon disaster provides another example. Subsequent research found that employees did communicate safety concerns, but organisational response could be weakened by time pressure, limited resources and a powerful "can do" culture that prioritised getting things done.
These cases demonstrate why simply telling people to "speak up", "follow procedures" or "make safe decisions" cannot be the entire solution.
The same principle works positively
One of the most frequently cited corporate transformations is Alcoa under Paul O'Neill. Rather than treating safety as another corporate programme, O'Neill made worker safety a central organisational priority. Serious attention to injuries forced better communication between frontline employees and management and required organisations to understand why failures occurred and how systems needed to change. Productivity and financial performance subsequently improved alongside the cultural transformation.
Safety became what is often described as a keystone habit: changing one critical organisational routine triggered improvements elsewhere.
The Shell Eastern Petrochemicals Complex in Singapore offers another useful example. A major construction project involved approximately 15,000 workers from diverse cultural and construction backgrounds. Hearts and Minds tools were incorporated into contractor training and safety-adviser development, and the Energy Institute reports a lost-time-incident frequency of 0.6 per million working hours—a record-breaking result for a Shell project of that scale and for its managing contractor.
At Sakhalin Energy, the starting environment included behaviours such as working at height without fall protection, driving without seatbelts and entering confined spaces without gas testing. Hearts and Minds and Tripod Beta were among the approaches used to address a complex workforce involving multiple companies and contractors with different safety cultures.
Importantly, Hearts and Minds itself is built around the proposition that a safety management system is the starting point rather than the end point: culture improves when organisations understand the people, working environment, leadership, systems and conditions influencing performance.
Bridging perception, thinking and behaviour
This suggests a useful three-layer model for modern safety culture:
PERCEPTION → THINKING → BEHAVIOUR
Safety climate tells us: "How do our people experience safety here?"
Behavioural intelligence tells us: "What are people actually doing?"
Safety Decision Intelligence asks the missing question: "Why are different people making different decisions when confronted with risk?"
This is the gap that SDIA is intended to address.
Where SDIA has been applied organisationally, its potential value is not merely the individual report. Individual decision patterns can be aggregated to explore whether particular teams or leadership populations display common tendencies, for example, strong analytical capability but weaker operational discipline, high procedural discipline but limited adaptability, or strong leadership confidence alongside inconsistent risk judgement.
This should not be interpreted as predicting who will have an accident. SDIA is better understood as an evidence-informed leadership and organisational-development tool, providing hypotheses for coaching, development and organisational learning rather than a test of competence or employment suitability.
The financial argument is equally important
Most organisations already spend significant amounts on safety training.
The question is whether everybody requires the same training.
If 500 managers attend an identical two-day leadership programme, the organisation has assumed that 500 people have essentially the same developmental need.
They almost certainly do not.
One leader may need help making timely decisions with incomplete information. Another may need stronger procedural discipline. Another may require coaching around psychological safety and listening. Another may need to become more adaptive when conditions change.
Understanding those differences allows organisations to move from generic training to targeted development.
That changes the economics.
Instead of repeatedly treating symptoms through additional campaigns, courses and corrective actions, organisations can direct coaching and training toward the underlying decision patterns that require development.
The objective therefore becomes bigger than reducing incidents.
It is to make good safety decision-making a keystone organisational habit.
When people routinely challenge assumptions, recognise changing risk, listen to weak signals, know when procedure must be followed and when emerging conditions require escalation, organisations improve more than safety. They improve communication, operational discipline, leadership, learning and ultimately decision quality.
The next evolution of safety culture may therefore not be another procedure, campaign or behavioural checklist.
It may be understanding something much more fundamental: how people think before they act.