<p>Target failure possibility-constrained safety life analysis can effectively guarantee the structural safety service under fuzzy uncertainty, but it is a challenge to improve the computational efficiency of estimating safety life. Based on equivalent constraint method, a new sequential decoupling optimization method is firstly proposed in this paper to efficiently estimate the safety life under the required target failure possibility, and then the Kriging surrogate model of performance function is adaptively embedded for a further improvement of the efficiency. In the sequential decoupling optimization method, the safety life analysis under the required target failure possibility is transformed into a sequential cycling process consisted of the minimum performance target point tracking and the safety life searching corresponding to the minimum performance target point. The proposed method not only inherits the excellent feature of the existing equivalent constraint method-based safety life analysis methods without requiring the time-dependent failure possibility estimation, but also speeds up the convergence resulted from adopting the sequential decoupling strategy. As regard to the adaptive Kriging model, the key contribution of this paper is to adopt the current expected improvement learning function to reduce the number of evaluating the performance function while tracking the minimum performance target point and searching the safety life under the given minimum performance target point. As a result, the computational effort and time of the sequential decoupling optimization method are significantly less than those of other analogous methods, which is sufficiently verified by the numerical examples as well as aeronautical structure example.</p>