Back to search

PETROMAKS2-Stort program petroleum

Analysis of human actions as barriers in major accidents in the petroleum industry, applicability of human reliability analysis methods

Awarded: NOK 8.8 mill.

The project is now finished. Martin Rasmussen defended his PhD successfully. The Petro-HRA method is now finished and complete, and a guideline has been issued, published at www.ife.no/petrohra. Petro-HRA is a "complete" human reliability analysis (HRA) method, including guidelines for how to do the qualitative phases of an HRA. This is unlike most other HRA methods that normally are pure quantification methods. 2016 was a year with lower effort than the previous years, where the main purpose was to update the Petro-HRA guideline after a thorough test in a First Use case. This First Use case was performed by DNV-GL for Statoil in the autumn, not funded by this research project. In this project, we interviewed the people who did the First Use analysis and used their feedback to update and improve the guideline and the method. The First Use work was finished late 2016 so the updated guideline is dated January 2017. Dissemination: Petro-HRA was presented in an ESRA seminar in September. Internationally, it has been presented at PSAM and ESREL. The consortium is in agreement to build on the project and provide a training course on Petro-HRA for the industry. As such we believe that the project will have great impact on the safety analyses in the industry by providing a usable and meaningful method for including human aspects of critical barriers into safety analyses.

The human contribution to the safety of petroleum installations has long been a concern for the industry and for the Petroleum Safety Authority of Norway. In the petroleum industry, quantitative risk analysis (QRA) has been used in order to predict the li kelihood of failure. However, this has mainly been focused on technical barriers. There are no standardized methods for how Human Reliability Analysis (HRA) is performed, nor a standard of how human error probability data should be integrated in the QRA. Statoil has begun exploring the usage of a simplified HRA method (SPAR-H) from the nuclear industry, revealing a method that is a good candidate for further development and use in the petroleum industry. The aim of this project is to test, evaluate, adjus t and standardize HRA to accident scenarios in the petroleum industry, thus improving the decision basis for managing risk in design and operations. This will be done by adapting performance shaping factors to the petroleum industry, developing guidelines for the qualitative HRA process, and testing HRA on human actions in major accident scenarios. Four work packages are defined: I. Evaluation and adjustment of contextual factors for human actions in accident scenarios in the petroleum industry. II. Task analysis and human error identification analysis (Human error analysis and reuse methodology). The purpose is to simplify the task analysis and human error identification analysis by looking at how much of the task analysis can be reused. III. Qualitativ e data collection: Interviews, observations and questionnaires. Evaluation and optimization of the qualitative data collection process. IV. Studies for the quantification by an expert group. Development of a systematic expert judgment process for evaluati on and validation of parameters in the HRA method, such as nominal values and the effect of performance shaping factors (multiplier values).

Publications from Cristin

No publications found

No publications found

No publications found

Funding scheme:

PETROMAKS2-Stort program petroleum