Citation

Baker AR. J. Occup. Accid. 1982; 4(1): 33-45.

Copyright

(Copyright © 1982, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

The normal diagnostic methods of investigating accidents do not always provide a convincing explanation of what happened. A convincing explanation may be necessary to allow precautions to be taken to prevent repetition of the accident or because it is required in court as a basis for the allocation of responsibility for the accident. In such cases consideration may be given to carrying out simulation experiments. These may be designed to confirm or demonstrate that an event or sequence of events could have occurred, to determine a mechanism or to determine the conditions under which a mechanism can operate.

Some of the uses and limitations of simulation experiments and their design are discussed and illustrated by eight examples, mainly involving fires and explosions.Full-scale simulation experiments are often time-consuming and expensive. Small-scale experiments are adequate in many cases, but in some fires and explosions quantitative simulation on the small scale can become difficult or even impossible because the scaling laws applicable to the significant factors may be different.

Ways in which the number or extent of simulation experiments can be limited are discussed. Usually there will be no need to simulate more than one or two of the events in a sequence and successive events can often be simulated separately. There may be no need to extend the experiment to the point at which the dangerous event could occur and there may be no need to match the effects of the accident very closely.