Engineering accident investigations are systematic inquiries into the facts and causes of engineering accidents. The aims of an engineering accident investigation include identifying significant truths about an accident, learning lessons to prevent similar future accidents, and authoritatively communicating the investigative results to the stakeholders. An important normative dimension along which an engineering accident investigation can be evaluated is its degree of success in fulfilling these aims. In this paper, I propose criteria for evaluating the degree of success of an engineering accident investigation using a question-centered framework, and then argue for the relevance of this proposal to the actual engineering practice. The basic idea of my proposal is that an engineering accident investigation is successful to the extent that (1) questions that should arise in the investigation do arise, and (2) questions that arise—especially the more significant ones—are resolved satisfactorily by the end of the investigation. The first part of this paper unpacks my proposal by analyzing the following three concepts and illustrating them using examples from the TWA Flight 800 accident investigation: The (satisfactory) resolution of questions, the significance of questions, and the arising of questions. The second part of this paper argues for the relevance of my proposal to the practitioners and stakeholders of engineering accident investigations. First, I argue that my proposal is sensitive to the aims of the investigators and stakeholders regarding engineering accident investigations, and that it helps them navigate competing and conflicting aims. Second, I go beyond the TWA 800 case study and argue that my proposal explains the strengths and limitations of different types of accident causation models used in investigations.

工程事故调查是对工程事故的事实和原因进行系统调查。工程事故调查的目的包括查明事故的重要真相，吸取教训以防止未来发生类似事故，并向利益相关者权威地传达调查结果。评估工程事故调查的一个重要规范维度是其实现这些目标的成功程度。在本文中，我提出了使用以问题为中心的框架来评估工程事故调查成功程度的标准，然后论证了该建议与实际工程实践的相关性。我的建议的基本思想是，工程事故调查成功的程度是：（1）调查中应该出现的问题确实出现了，（2）出现的问题——特别是更重要的问题——得到了令人满意的解决。调查结束。本文的第一部分通过分析以下三个概念并使用环球航空 800 航班事故调查的示例来阐述我的建议：问题的（令人满意的）解决、问题的重要性以及问题的产生。本文的第二部分论证了我的建议与工程事故调查的从业者和利益相关者的相关性。首先，我认为我的建议对调查人员和利益相关者在工程事故调查方面的目标很敏感，并且它可以帮助他们应对相互竞争和冲突的目标。其次，我超越了 TWA 800 案例研究，并认为我的建议解释了调查中使用的不同类型事故因果模型的优点和局限性。