Hi, I want to map a fault tree into bayesian network on GeNIe, but I got wrong posterior probability after updating.
The fault tree is like that:
I gave the probability of basic events (a, b, c) and conditional probability table of intermediate events(E1, E2) and top event.
The conditional probability table for 「or」logic gate is like that:
And here is the result after mapping into bayesian network and updating:
I supposed the probability of 'TOP' event is:
False: 40.8% (60% * 68% TOP is false only when both E1 and E2 are false )
True: 59.2%
The result calculated by GeNIe is "True:46%, False:54%" of TOP event, which was different with what I calculated.
However, the results of E1 and E2 don't have this problem, the results of posterior probability are the same as which I calculated by myself.
This situation only occurred when different child nodes have common parent node on bayesian network(ex. b to E1 and E2) , but I really don't understand why... Hope someone can help me, thanks!
Promblem with mapping fault tree into bayesian network on GeNIe
-
- Posts: 1
- Joined: Fri Jul 29, 2022 9:22 am
-
- Site Admin
- Posts: 430
- Joined: Tue Dec 11, 2007 4:24 pm
Re: Promblem with mapping fault tree into bayesian network on GeNIe
Your error is assuming that E1 and E2 are independent. They are not, as they have a common ancestor, b. Because they are dependent, you cannot just multiply probabilities as you did in your calculation.
I hope this helps,
Marek
I hope this helps,
Marek