Safety culture represents an ongoing concern and challenge for high-hazard industries worldwide. First identified as a causal factor in the Chernobyl nuclear disaster of 1986, safety culture weaknesses have consistently appeared in investigation findings following major incidents, including Piper Alpha in 1988, the Columbia space shuttle in 2003, the Texas City refinery in 2005, and Macondo in 2010. As a result, safety culture has served as a topic of increasing attention over the past two decades, with academics, consultants, governments, regulators, and corporations attempting to understand, operationalise, measure, and change the safety culture of organisations, as a means of improving safety performance.
Safety culture is a popular topic for research and publication; however a commonly accepted definition and model does not appear to exist. This has led to a highly diverse collection of safety improvement practices being labelled “safety culture improvement”. If the current variability in approaches to safety culture improvement continues, there is a risk that safety culture will be labelled a “fad”, and will be perceived as failing to add value to safety performance. Safety culture improvement initiatives which are poorly designed and implemented (i.e. they do not address safety culture) are unlikely to succeed in improving safety performance. Organisations and their members who have been involved in such unsuccessful attempts are then likely to conclude that the reason for the poor outcome is because safety culture as a concept is flawed, rather than recognising that the outcome was due to their approach to implementing the concept.
Safety culture has the potential to significantly impact safety performance positively or negatively, but only if approaches to operationalise the concept are implemented with rigour. Research conducted by NOPSEMA in 2012-13 suggests that, at present, such rigour is lacking amongst a significant proportion of the industry. To ensure that the potential power of safety culture is not lost in poor approaches to implementation, there is a need for a common definition and model of safety culture across the industry.
A definition of safety culture
NOPSEMA recommends the adoption of a common definition and model of safety culture across the industry, as a means of improving the rigour of safety culture improvement strategies and increasing their likelihood of success.
NOPSEMA defines safety culture as “the shared basic assumptions, held by most members of an organisation, which create and reinforce group norms of thoughts, language and behaviour in relation to major accident event prevention.”
Basic assumptions develop over time, as group members experience and solve internal and external challenges. When a problem is solved, and the solution worked well enough to be considered valid by the group members, this solution is then taught to new group members as the correct way to think, feel and act in relation to the problem. Over time, this correct way of thinking, feeling and acting becomes a basic assumption. It is no longer subject to challenge, and becomes an accepted part of ‘the way we do things around here’, informing future behaviour and organisational artefacts. This process is outlined in the diagram below.
A model of safety culture
This model highlights the process by which executive commitment to safety influences organisational safety outcomes. Safety outcomes exist as a direct result of organisational behaviour (at all job levels). Organisational behaviour is influenced by executive decisions and behaviour, as well as organisation-level factors of leadership practices, systems, and working environment. These three factors interact with each other, and are created and driven by executive decisions and behaviour. Executive decisions and behaviour are a reflection of executive commitment to safety; however this relationship is mediated by executives’ safety knowledge, and their understanding of organisational behaviour.
Executive decide and act in response to internal and external challenges. The assumptions that such decisions and actions are based upon are typically not explicitly stated, and so over time form the shared basic assumptions of the organisation. The implications of these decisions and actions, when successful, are embedded in systems, leadership practices, and the working environment. These organisation-level factors represent the espoused values and artefacts of the organisation. As systems, leadership practices, and the working environment deliver consistent and expected outcomes, the decisions and behaviour of executives are reinforced, and so basic assumptions are strengthened and reinforced.
This model is designed to be used as a common standard by which to design and evaluate safety culture improvement initiatives. Essentially, safety culture improvement initiatives should target each element of the safety culture model presented above. The way that this is done will necessarily vary between organisations. Each organisation is unique and so requires a fit-for purpose strategy which is appropriate to its culture, history, strategy and people. However it is critical that there is some degree of consistency in the way in which safety culture improvement initiatives are designed, if the concept of safety culture is to successfully deliver improvements in safety outcomes over time.
Note that operators are not obliged to use the model or definition presented above. These have been developed as a means of facilitating continuous improvement in the application of safety culture concepts to safety performance improvement, and do not represent a regulatory requirement. Operators preferring to utilise their own model or definition of safety culture are free to do so.