Modelling of Safety Barriers Including Human and Organisational Factors to Improve Process Safety
Markert, F.
Duijm, N.J.
Thommesen, J.
Download PDF

How to Cite

Markert F., Duijm N., Thommesen J., 2013, Modelling of Safety Barriers Including Human and Organisational Factors to Improve Process Safety, Chemical Engineering Transactions, 31, 283-288.
Download PDF

Abstract

It is believed that traditional safety management needs to be improved on the aspect of preparedness for coping with expected and unexpected deviations, avoiding an overly optimistic reliance on safety systems. Remembering recent major accidents, such as the Deep Water Horizon, the Texas City explosion, and the Mont Blanc Tunnel Fire, such an approach may have helped to maintain the integrity of the designed provisions against major deviations resulting in these disasters. In order to make this paradigm operational, safety management and in particular risk assessment tools need to be refined.
A valuable approach is the inclusion of human and organisational factors into the simulation of the reliability of the technical system using event trees and fault trees and the concept of safety barriers. This has been demonstrated e.g. in the former European research project ARAMIS (Accidental Risk Assessment Methodology for IndustrieS, see Salvi et al 2006). ARAMIS employs the bow-tie approach to modelling hazardous scenarios, and it suggests the outcome of auditing safety management to be connected to a semi-quantitative assessment of the quality of safety barriers. ARAMIS discriminates a number of different management issues such as competence management, dealing with conflicts, management of maintenance and inspection, and management of procedures. Shortcomings in these management processes effectuate increased probabilities of failure-on-demand (PFD) of the safety barriers, depending on the type of safety barrier (passive, automated, or involving human action). Such models are valuable for many purposes, but are difficult to apply to more complex situations, as the influences are to be set individually for each barrier.
The approach described in this paper is trying to improve the state-of-the–art, and it is based on the understanding that certain human and organisational factors may be seen as a kind of common cause failures that influence the performance of several barriers. Therefore, the model links the performance of a barrier with the necessary set of specific activities to maintain and/or to control that barrier. These specific activities are executed within one of the aforementioned management processes, and the efficiency of the activity will depend on the quality of this management process.
Download PDF