Proceedings of the Eighth International Conference on Probabilistic Safety Assessment & Management (PSAM)
291 Measuring and Predicting Organizational Learning and Safety Culture (PSAM-0199)
Download citation file:
- Ris (Zotero)
- Reference Manager
In the world and society in which we live, errors and accidents occur, and as we see in our everyday lives, the opportunity for error and risk surround us. Events occur all the time, in all technologies that humans have invented and use.
In the field called “organizational learning”, the formal processes, individual models, and social behaviors all can contribute to the creation of knowledge and its use, and to the reduction of error. The overall idea is the obvious one: organizational learning is key, a major tool, and a focus on learning encourages greater safety, better error reporting and correction, and the development of more effective coping strategies. Similarly, practicing effective group communication enhances safety and understanding.
We are still faced with how to quantify what all this means in terms of risk prediction and reduction, other than we know that if we do not learn we have more risk.
We present a new exercise in the analysis and prediction of risk, accidents and the contribution of human error as applied to, and present in, organizational entities and homo-technological systems. We show how the Learning Hypothesis explains the trends in outcome rates (for events, accidents, near-misses, deaths, and injuries). The key results of the theory and data comparisons are: the quantified description of human error probability; the prediction of the variation of error and outcome rates with depth of experience; the key importance and measures of experience; and the quantification of learning at the individual and organizational levels.
Since the function and goals of (safety) management is the creation of order from disorder, we show that a suitable measure of the safety culture in an organization is the information entropy. This measure of disorder decreases with increasing learning rate and depth of experience, as expected. Hence, it is a specific and technically based quantitative measure of the degree of “safety culture” and of organizational learning. Such an integral part of probabilistic safety should be of utmost importance to both management and regulators.