本研究係運用國民航局近年來建立之「主動提報」機制所取得之航空業者安全事件,運用人為因素分析及歸類系統(Human Factors Analysis and Classification System – Maintenance Audit, HFACS-MA)進行深入分析。主動提報意指業者在異常事件發生後,於民航局發現之前即主動向該局提報之,這是民航主管機關用以提升日常飛安監理工作之機制之一。透過檢視HFACS-MA架構中四種人為疏失層級,包括「不安全的行為」(Unsafe Acts)、「不安全行為的影響因素」(Preconditions For Unsafe Acts)、「不安全的管理行為」(Unsafe Supervision)以及「組織影響因素」(Organizational Influences),解讀分析業者主動提報事件,後續並使用包括卡方檢定及勝算比等統計檢定方法,探討四種疏失層級之間的關聯性與影響關係。分析結果顯示,在「不安全的行為」之中,以「技術型疏失」為最多,而「不安全行為的影響因素」則以「不良狀態」發生比例最高,「軟/硬體」問題次之。「不安全的管理行為」層級之中,以「規劃/組織」與「控制/改正」之管理問題較多,「組織影響因素」則以「安全監理」為最常發生之疏失類型。層級間之統計分析亦發現存在顯著關係,例如「組織資源管理」對於「規劃/組織」有所影響,如欲改善「規劃/組織」相關管理疏失,可考慮從組織資源管理著手。相較於其他類型之飛安資料,「主動提報」機制所獲事件資料因具有自發性,所分析出之疏失特性與關聯或可作為其他飛航安全管理資料之補充或參考,未來研究可根據本研究之結果,進一步探討主動提報資料之特性,並據以發展出更好的人為疏失改善方案。
The purpose of this study aimed to analyze the voluntary reports that were received by the Civil Aeronautics Administration from the domestic air transportation enterprises in Taiwan. Human Factors Analysis and Classification System – Maintenance Audit (HFACS-MA) was the exploring tool for analyzing the reports to find out significant faults or potential failures from the human factors perspectives. Four types of human errors would be examined, that are, Unsafe Acts, Preconditions for Unsafe Acts, Unsafe Supervision, and Organizational Influences. The result illustrated that “Skill-based Errors” took the highest proportion in Unsafe Acts. For the Preconditions for Unsafe Acts, the most common reasons were “Adverse States” and “Hardware/Software”. “Planning/Organizing” and “Controlling/Correcting” were the most significant issues of Unsafe Supervision. At the level of Organizational Influences, “Safety Oversight” was the most common matter. The result also revealed the causal relationships between different levels of human failures. For instance, “Resource Management” of Organizational Influences had significant influences to “Planning/Organizing” of Unsafe Supervision. Compared to other types of safety data in the aviation industry, the voluntary reports which were received from the airliners may be capable of disclosing some important findings that other data such as enforcement reports couldn’t detect. This study can be considered as part of the fundamental research of human factors in the flight safety field and could be took into account to enhance safety management in human error prevention.