Ensuring the resilience of contemporary society hinges upon achieving adequate reliability levels in structural systems during design and maintenance decisions. These processes are crucial for mitigating the risk of unforeseen failures, thus averting potential catastrophic damage or losses. Consequently, effective strategies for reliability assessment and optimization within the design phase of structures and other engineering systems are paramount. This paper delves into various reliability analysis methods, encompassing simulation-based, approximation, and Metamodel-based approaches, to enhance understanding of structural performance under uncertainties. The research reveals that while approximation methods may face accuracy challenges due to the nonlinearity of limit-state functions, they exhibit notable computational efficiency for designs with relatively small to medium-sized variables. Conversely, the Metamodel method may necessitate multiple design experiments for constructing accurate Kriging models, yet it demonstrates commendable computational efficiency and accuracy in structural reliability assessment. The paper further provides a comparative analysis of accuracy and efficiency through numerical examples of frame structure and grid-shell structure reliability assessment.