Exploring the relationship between major hazard, fatal and non-fatal accidents through outcomes and causes

• Database of 23,030 serious occupational accidents in a socio-technical model used.
• Fifteen accident triangles for selection of different hazards show lethality of hazards.
• Fatal versus non-fatal accident within hazards are compared.
• Case studies of 2 big accidents (Amercentrale and Buncefield) safety barrier failures.
• Major, fatal and non-fatal accidents within hazards found to have same underlying causes.

Smaller severity more frequent accidents can provide information about the direct and underlying causes of bigger severity more catastrophic accidents but only if looking within the same hazard category. Use is made of a database of around 23,000 Dutch serious reportable accidents 1998–2009 that have been analysed in Storybuilder™ in 36 hazard specific bow-ties using a management-task-safety barrier model of failure causation. The data are first developed as hazard specific accident triangles to show differences in lethality. Then comparisons of fatal and non-fatal accident causes are carried out, showing commonality in causes. The same is done for two case studies of catastrophic accidents – the Amercentrale power station scaffold collapse in the Netherlands and the major chemical accident at the Buncefield oil storage depot in the UK. Results indicate that, provided accidents from different hazard bow-ties are not mixed together, small severity more frequent accidents can be used to consider the causation and hence prevention of the bigger severity rarer accidents. This leads to the conclusion that the analysis of occupational accidents can help in addressing major ones providing it is restricted to the same hazard type, contradicting the view that personal and process safety are totally unrelated.