Probability of Fault Tree "Top Event" ASP/CSP
Vložit
- čas přidán 3. 06. 2021
- For ASP / CSP exams candidates are expected to know a little about fault tree analysis, along with a basic understanding of probability. In this example data are provided for basic and intermediate events. From these data the probability of a the "top event" can be calculated. The presenter is Dr. Doug Martin. Dr. Martin earned his Ph.D. from Oklahoma State University and is a CSP. In addition Dr. Martin worked as a safety manager in the construction industry for 20 years. For more information about the EHSM program at Northeastern you may contact Dr. Martin at marti223@nsuok.edu. Northeastern State University's Environmental, Health, and Safety Management program has been recognized as a Qualified Academic Program by the Board of Certified Safety Professionals (BCSP).
Perfect video man! ty so much
Nicely explained ☺️
thankyou...valuable info
Where does the base events probability come from? How do I determine them for any given project? 🤔
Excellent question. The probabilities used in this example are made up to illustrate the process. This is a deficiency with quantitative fault tree analysis. Numerical data are not usually available. Most fault trees are qualitative, based on the judgement of experts.
The failure probabilities for some components can be obtained from manufacturer data and is often expressed as the mean time between failure (MTBF). Third party "testing" laboratories, such as Underwriter's Laboratory (UL), may also be a source. If you have years of data available within your company you may also be able to come up with a good estimate of failure probabilities for critical components. You could also do your own testing, but in most cases this is not going to be practical.
Great question.
the calculated value for the replacement value was suppose to be 0.7009. Thank you for sharing this video
But from where 1 get values that are mentioned under each event in RED. I can understand probability calculation, but what about the value
They are made up to show you how the calculations work. Calculating what the probability is of a particular event happening is a completely different thing. They are not relevant right now
But shouldn't the OR operation rather be calculated with P_parent = 1 - (1 - P_childA) * (1 - P_childB)?
It's not disjunct, you could get over 1 otherwise when just summing up.
You are absolutely right. This is a simple problem designed to give CSP (Certified Safety Professional) test takers a heads up for the exam. For detailed coverage of addition and multiplication rules I have several other "probability" tutorials for students in my statistics classes.