The Art of Conducting a Safety Risk Analysis on In-Service Problems 2008-01-2228
This technical paper presents the author’s recommended approach to one aspect of managing flight safety - conducting Safety Risk Analyses (SRA) on in-service problems that may threaten flight safety. The author did not develop this statistically based approach for assessing the risk of future events, but has helped to improve it and highly endorses it. In conducting a safety risk analysis, the analyst might decide to perform a “quick” SRA and will need a minimal amount of information that will show the relative level of flight safety risk. When the analyst decides a complete safety risk analysis is needed, the possible approaches and level of details included in the SRA can vary greatly from company to company.
Knowing what specific information is needed for the inputs and assumptions, how best to use that information, what missing data needs to be estimated, selecting the most appropriate estimation technique, is what makes conducting a Safety Risk Analysis an Art, as well as a science. Process knowledge is not enough. It takes creativity, foresight and experience. The analyst must have an understanding of characteristics of the problem and the available data. The approach described herein has been proven over time to produce accurate results, i.e. a great track record in the industry. In the interest of flight safety as the top priority for all of us in commercial aviation, the author wishes to share this SRA process, so that other safety risk analysts can consider having it as an option for their use.
This paper briefly discusses the need for a company to establish flight safety risk criteria levels as part of a safety policy, and the creation of a safety board to manage the application of that policy. It also describes the safety risk analysis process and why a company should adopt such a process as part of its safety policy. As part of the safety risk analysis process, there is a brief explanation of some safety risk analysis tools, such as statistical trend analysis and Weibull analyses, the use of checklists for mistake proofing, and some basics of Monte Carlo simulation modeling used in assessing the risk of future events. The author hopes that each aerospace company will give careful consideration to this safety risk analysis framework, and decide to put into practice any appropriate segments that will help enhance their efforts to achieve and maintain flight safety.