The risk of risk assessments: Investigating dangerous workshop biases through a socio-technical systems model

Organisations are required to manage industrial safety by implementing risk treatments to prevent hazardous events. In the absence of objective failure data, risk estimates are generated in facilitated workshops that rely on the judgement of diverse technical opinion from subject matter experts. These estimates generate the performance targets for the engineering controls referred to in the Hierarchy of Controls embedded in work health and safety regulations. This novel study investigated biases within the context of facilitating risk assessment workshops which can lead to understatement of risk estimates. Such biases can be dangerous and potentially difficult to prevent or mitigate. Interviews with facilitators and technical experts (N = 23) were conducted, with experience covering eight substantive industry sectors including oil/gas, mining, and chemical processing. Conventional content analysis identified five biases at a group level and one at an individual level and ranged from cultural or societal forces, industry norms, company pressures, as well as pressures associated with occupational roles, and highly dominant individual views. These were mapped across Rasmussen's (1997) socio-technical systems model to identify potential mitigative/preventive measures that might improve risk assessments from levels higher in the system. Practical implications and future research directions are drawn.