Citation

Le Coze J. Safety Sci. 2013; 51(1): 187-201.

Copyright

(Copyright © 2013, Elsevier Publishing)

DOI

10.1016/j.ssci.2012.06.008

PMID

unavailable

Abstract

This paper introduces what is defined as a sensitising model. This model has been developed to support empirical and methodological research and studies in different high-risk industries. Its purpose is to assist current and future practices of industrial safety assessments by taking into account input from the social sciences, reflecting in particular their insights on major accidents. For this purpose, the first objective of this paper is to take an overall view. There is not, in principle, only one way of introducing social sciences into safety assessment practices. In a historical approach four research traditions (safety management system, safety culture, high-reliability organisations, accident investigations and models) are introduced. A classification scheme is produced to make sense of the diversity of traditions according to three dimensions. It is argued that an integration of all the different research traditions would not be realistic. However, it is explained that some models offer complementary features worth exploring. The second objective of this paper is thus to present the design rationale of such an exploration in order to produce a sensitising model for safety assessments. This model is obtained by combining two generic models from the managerial and sociological (safety) literature. Safety is approached in this model as the dynamic interaction between several dimensions, including technological design and tasks, structural and functional features of organisations, but also cognitive, cultural and power issues, at several layers of analysis. Theoretical aspects with regard to the position and status of this model are discussed.