COMPARISON OF QUASI- AND PSEUDO-MONTE CARLO SAMPLING FOR RELIABILITY AND UNCERTAINTY ANALYSIS