evaluation of the methods for calibration of optimal load and resistance factors based on structural reliability theory

In code calibration based on structural reliability theory, the purpose is to provide partial safety factors for obtaining safe and economical structural design. this goal can be achieved by taking statistical characteristics of loads and resistance parameters into account. in this paper, an algorithm is proposed for code calibration based on structural reliability theory, which leads to a uniform level of reliability for a variety of limit states or design situations. this method is based on computing derivatives of limit state function with respect to random variables for reliability assessment which results in increasing the efficiency of the algorithm. in this context, the optimal partial safety factors for a limit sate with different design situations and probability distributions are computed to achieve a target reliability for the design. in this context, for a linear limit state function, the effects of an increase in kl, the ratio of live to dead load, on the load factors including dead, live, and wind load and resistance factor are investigated. the results show a slight increase in resistance and dead load factor, a significant increase in live load factor, and a significant decrease in wind load factor. finally, the optimal partial safety factors for the limit state with different design situations are computed by two different methods, namely minimum penalty function method (method 1) and weighted factors method (method 2). the comparison of the results shows that the minimum penalty function method (method 1) leads to the partial safety factors obtained by only one design situation with dead-to-live load ratio of 1. on the other hand, the weighted factors method (method 2) leads to the partial safety factors contributed by all design situations. consequently, the weighted factors method is more suitable for problems with high number of design situations and reliability analyses.