Accident Causation Model
Definition - What does Accident Causation Model mean?
An accident
causation model is a systematic method of ascertaining the causes of an
accident. An accident is a complex coincidence of activities or
phenomena in a single time and space. Therefore, determining the causes
leading to an accident can be quite difficult, as there are so many
variables to consider.
Accident causation models vary from simplistic linear models to complex non-linear models.
Accident causation models vary from simplistic linear models to complex non-linear models.
Safeopedia explains Accident Causation Model
Systematic examination of causes of
accidents began in the early 20th century. The development started with
the simple linear "domino model," explaining the individual’s behavior
and circumstances surrounding an accident. It gradually advanced to
complex linear models, and further to complex non-linear models
considering the time sequences. Varying models contunue to be developed
still.
There are three basic types of accident causation models:
There are three basic types of accident causation models:
- Simple linear models (Heinrich, 1931) - Presumes that an accident is the end result of a series of sequential events playing out like dominos. The sequence begins with the social environmental factor, individual factor, unsafe acts, mechanical and physical hazards, accident, injury, etc.. It is expected that, the elimination of one of the dominos may prevent the accident
- Complex linear models - Presumes that an accident is a combination of a number of unsafe conditions and factors where an individual interacting close to the system is at the risk of an accident. It is expected that an accident could be prevented by setting appropriate controls. Some varieties of this type of model are time sequence models, generic epidemiological models, systemic models, Reason’s "Swiss Cheese" model and models of systems safety
- Complex non-linear models (Hollnagel, 2010) - Expresses that the accidents are caused by mutually interacting variables in real time environments. Accidents could be prevented through understanding these multiple interacting factors. Examples of such models are theoretic accident model and process (STAMP) and functional resonance accident model (FRAM)
No comments:
Post a Comment