The study of seismic safety of buildings is one of the basic requirements for its implementation. In recent decades, structural reliability theory to be allocated very extensive research. Monte Carlo simulation is a suitable method for structural reliability analysis. In this paper, a new method of Importance Sampling is proposed to reduce the number of limit state function computations. A feature of the proposed algorithm doesn’t need to know the position of the design point or shape of limit state function. The main idea of the proposed algorithm is that the average amount of sampling probability density function is changing in the simulation process. Accordingly, the concentration of sampling will be on the areas of problem space’s importance. One of the methods for calculating seismic structural reliability taking into account the structural uncertainties against deterministic earthquake records. Reliability analysis of crossing the structural threshold level for given earthquakes, will be determined as analysis of the level crossing is likely to exceed the structural response of a threshold value. Monte Carlo simulation method based on the importance sampling is one of the most powerful statistical methods for calculating reliability index, which improves and reduce the number of samples and increasing the speed of the operation. Due to the dependence of structural system capacity to the material properties and dimensions of the members, evaluation of uncertainties involved in any of these parameters is important.