Human Factors Analysis and Classification System (HFACS, Wiegmann & Shappell, 2003) based upon Reason's (1990) model described human error as active failures and latent failures. HFACS framework consists of 4 levels and 18 categories, the first level classifies events of 'unsafe acts of operators' including 4 categories; the second level concerns 'preconditions for unsafe acts' including 7 categories; the third level is 'unsafe supervision' including 4 categories and the fourth level is 'organizational influences' including 3 categories. This research analyzed 545 accident reports between 1978 and 2008 by applying the HFACS framework. The statistically-described paths by Chi-square and Goodmann & Kruskal tau (τ) shows that human errors at the first level of HFACS been significantly affected by inadequacies at the second level, and more remote levels at the supervisory and organizational levels. Fallible decisions in upper management levels were found to affect supervisory practices, creating preconditions for unsafe acts and hence impaired performance of pilots leading to accidents. The HFACS framework is proven as an effective tool for identifying the path of failure leading to an accident, and it is the foundation for developing accident prevention strategies in the future.