Abstract
Performance analysis of safety barriers and how they affect accident risk have paramount importance in risk analysis, as suggested by regulations and standards. Despite the fact that conventional barrier analysis mainly focuses on technical factors, recent investigations on major accidents demonstrate also the influence of operational and organizational factors. Therefore, research efforts have been focused on methods to incorporate such factors in risk management, as factors affecting the safety barrier performance. An example is the Risk OMT method, which aims to assess the effect of technical, operational and organizational Risk Influencing Factors (RIFs) on major accident probability. It employs Bayesian Networks (BNs), analyzing dependencies among variables, redundant failures and multi-state variables. Moreover, recent studies focus on the use of performance indicators to assess risk variations for better usability. A representative result of such studies is the Risk Barometer technique. The use of indicators may be challenging as the results strongly depend on the input data and their mathematical aggregation is not consistently defined. This paper presents a comparison of the techniques aforementioned highlighting their relative advantages and limitations. Potential alternatives and related improvements will be suggested and discussed. A case from the offshore oil and gas industry is used in the comparison, namely LOC prevention accounting for operational maintenance procedures. © 2017 Taylor & Francis Group, London.