An algorithm for the First Order Reliability Method (FORM)
1. Identify the distributions of your basic random variables X (typically marginal PDFs or CDFs, as well as correlations or joint distributions)
2. Decide which method is appropriate for transforming your basic random variables X into standard normal random variables U (using one of the four cases discussed).
3. Use the Hasofer-Lind Rackwitz-Feissler algorithm to find the design point u*
a. Choose an initial guess for the x associated with u*. This is x(0)
b. Find the u associated with the current value of x, using the transformation chosen in step 2.
c. Compute the gradient of g, evaluated at ( ) k x :
Sun | Mon | Tue | Wed | Thu | Fri | Sat |
---|---|---|---|---|---|---|
27 | 28 | 29 | 30 | 1 | 2 | 3 |
4 | 5 | 6 | 7 | 8 | 9 | 10 |
11 | 12 | 13 | 14 | 15 | 16 | 17 |
18 | 19 | 20 | 21 | 22 | 23 | 24 |
25 | 26 | 27 | 28 | 29 | 30 | 31 |