Citation

Leplat J, Rasmussen J. Accid. Anal. Prev. 1984; 16(2): 77-88.

Copyright

(Copyright © 1984, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Methods for the analysis of work accidents are discussed, and a description is given of the use of a causal situation analysis in terms of a "variation tree" in order to explain the course of events of the individual cases and to identify possible improvements. The difficulties in identifying "causes" of accidents are discussed, and it is proposed to analyse accident reports with the specific aim of identifying the potential for future improvements rather than causes of past events. In contrast to traditional statistical analysis of work accident data, which typically give very general recommendations, the method proposed identifies very explicit countermeasures. Improvements require a change in human decisions during equipment design, work planning, or the execution itself. The use of a model of human behaviour drawing a distinction between automated skill-based behaviour, rule-based "know-how" and knowledge-based analysis is proposed for identification of the human decisions which are most sensitive to improvements.