In this study, stochastic analysis is aimed for space structures (satellite in low earth orbit, made of aluminum 2024-T3), with the focus on fatigue failure. Primarily, the deterministic fatigue simulation is conducted using Walker and Forman models with constant amplitude loading. Deterministic crack growth was numerically simulated by the authors developed algorithm and is compared with commercial software for accuracy verification as well as validation with the experimental data. For the stochastic fatigue analysis of this study, uncertainty is estimated by using the Monte Carlo simulation. It is observed that by increasing the crack length, the standard deviation (the measure of uncertainty) increases. Also, it is noted that the reduction in stress ratio has the similar effect. Then, stochastic crack growth model, proposed by Yang and Manning, is employed for the reliability analysis. This model converts the existing deterministic fatigue models to stochastic one by adding a random coefficient. Applicability of this stochastic model completely depends on accuracy of base deterministic function. In this study, existing deterministic functions (power and second polynomial) are reviewed, and three new functions, (i) fractional, (ii) global, and (iii) exponential, are proposed. It is shown that the proposed functions are potentially used in the Yang and Manning model for better results.