Reason's Hazards–Defences–Losses scheme frames an accident as an uncontrolled trajectory from a hazard, through a defended sociotechnical system, to a loss. Losses occur when the defences erected between a hazard and its victims or assets are either absent or penetrated by a combination of active failures at the sharp end and latent conditions that lie dormant inside the organisation.
In Managing the Risks of Organisational Accidents Reason integrates earlier work on error and defences into a single causal picture. Organisations deploy a portfolio of defences — hard (guards, interlocks, alarms, automatic shutdowns) and soft (procedures, licensing, training, supervision, safety culture). Each defence can fail for two distinct reasons: active failures, the unsafe acts of front-line operators whose effects are immediate, and latent conditions, the resident pathogens created upstream by designers, procedure writers, regulators and senior managers.
Latent conditions create local workplace factors (time pressure, undermanning, poor tools, fatigue) and error-producing conditions that translate into violations and slips at the sharp end. When a hazard is released, the trajectory only causes a loss if the defences-in-depth are simultaneously weakened. The model is the organisational elaboration of the Swiss-cheese metaphor and is the conceptual ancestor of HFACS and most modern SMS investigation standards.
Moves analysis beyond the individual operator to include the designers, managers and regulators whose decisions create the conditions in which sharp-end errors occur.
Active failure, latent condition, defence, and pathogen have become lingua franca across aviation, healthcare, rail and nuclear — enabling cross-industry learning.
Focuses remediation on defences and workplace conditions that can be engineered or managed, rather than on blaming individuals for unsafe acts.
Maps naturally onto safety assurance processes: hazard identification finds gaps; risk management strengthens defences; promotion addresses latent cultural conditions.
The defences-in-depth picture implies a single trajectory; critics including Dekker, Hollnagel and Leveson argue that complex, nonlinear couplings are poorly captured by a sequential metaphor.
Latent conditions are only legible after a loss. Applied prospectively the model risks a proliferation of plausible-but-unverifiable "pathogens" limited only by the analyst's imagination.
It is difficult to decide where the organisation ends and the regulator, contractor or customer begins — ambiguity that complicates both investigation and accountability.
Real defences are actively maintained by people; treating them as discrete slices underweights the continuous work of monitoring, adapting and repairing them during normal operations.
Reason's HDL scheme reframes accidents as organisational phenomena: losses occur when a hazard trajectory punches through defences whose integrity has already been compromised by latent conditions seeded upstream and active failures at the sharp end.
Reason, J. (1990). Human error. Cambridge University Press.
Reason, J. (1997). Managing the risks of organizational accidents. Ashgate.
Reason, J. (2008). The human contribution: Unsafe acts, accidents and heroic recoveries. Ashgate.
Reason, J., Hollnagel, E., & Paries, J. (2006). Revisiting the Swiss cheese model of accidents (EEC Note No. 13/06). EUROCONTROL.
Shappell, S. A., & Wiegmann, D. A. (2000). The human factors analysis and classification system — HFACS (DOT/FAA/AM-00/7). FAA.
International Civil Aviation Organization. (2018). Safety management manual (Doc 9859, 4th ed.). ICAO.
Dekker, S. (2006). The field guide to understanding human error. Ashgate.
Hollnagel, E. (2004). Barriers and accident prevention. Ashgate.
Leveson, N. G. (2011). Engineering a safer world: Systems thinking applied to safety. MIT Press.
Perneger, T. V. (2005). The Swiss cheese model of safety incidents: Are there holes in the metaphor? BMC Health Services Research, 5, 71.
Underwood, P., & Waterson, P. (2014). Systems thinking, the Swiss cheese model and accident analysis. Accident Analysis & Prevention, 68, 75–94.