EAGER: A Systems Approach to Predicting and Preventing Accidents During Operations

这个早期概念探索性研究（EAGER）项目将通过研究一种在事故发生前识别潜在事故的领先指标的新方法，采取第一个概念验证步骤，为国家的经济繁荣、安全和福祉做出贡献。尽管我们尽力开发安全的产品和服务，但事故和损失仍然时有发生。随着当今先进技术带来的潜在更大损失的增加，设计本质上安全的系统将变得更加困难，操作风险管理实践将变得更加重要。在重大事故之前总会有警告标志，但这些标志可能只有在事后才会被注意到和解释。与此同时，现在计算机可以收集大量的数据，但由于数据的大小和从数据中创建有用信息的困难，这些数据中的大多数没有被使用。这项研究将采用一种新的、独特的方法来分析运营数据，并确定风险何时增加。三家国际航空公司通过提供真实的运营数据和他们的专业知识参与了这项研究，以验证新的风险分析技术的可行性和有效性。虽然以航空公司运营为试验平台，但研究结果可以扩展到任何安全关键行业。本研究要评估的基本假设是，可以根据所采用的安全工程实践的基础假设和这些假设的脆弱性来确定有用的领先指标。这个过程将需要使用各种类型的操作数据，这些数据可以（而且经常）被收集，但需要一些过程来确定所收集数据的潜在后果。这项研究将建立在一个叫做STAMP（系统理论事故模型和过程）的事故原因新模型和一种新的危害分析技术（STPA或系统理论过程分析）的基础上，以对该模型进行研究。STAMP扩展了当前事故的因果关系，包括更复杂的原因，而不仅仅是部件故障和故障事件链或偏离操作预期（这是传统危害分析方法的基础）。STAMP结合了系统思维的基本原理，它基于系统论而不是传统的可靠性理论。一种名为Active STPA的新分析工具将被创建，该工具将使用在操作过程中通常收集的数据来识别在设计和开发产品或服务期间所做的假设在操作过程中是否被违反，从而可能是事故的前兆和风险可能增加的证据。测试平台将是航空公司的运营。虽然这项研究的主题仅限于确定与安全和事故相关的领先指标，但这些想法将适用于安全以外的系统属性的领先指标和风险管理，并可能为研究人员探索操作过程中的风险评估开辟一条新的途径。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

This EArly-concept Grant for Exploratory Research (EAGER) project will contribute to the economic prosperity, safety and wellbeing of the nation by taking the first proof-of-concept steps in investigating a new approach to identifying leading indicators signaling the potential for an accident before it occurs. Despite our best intentions to develop safe products and services, accidents and losses still occur. As the potential for greater losses involved in our advanced technology today increases, it will be harder to design inherently safe systems and operational risk management practices will become more important. There are always warning signs before a major accident, but these signs may only be noticeable and interpretable in hindsight. At the same time, enormous amounts of data can now be collected by computers but most of theses data are not used because of the size and difficulty in creating useful information from the data. This research will take a new and unique approach to analyzing operational data and identifying when risk is increasing. Three international airlines are participating in the research by providing real operational data and their expertise to validate the feasibility and efficacy of the new risk analysis technique. While airline operations are used as the testbed, the research results can be extended to any safety-critical industry. The basic hypothesis to be evaluated in this research is that useful leading indicators can be identified based on the assumptions underlying the safety engineering practices employed and on the vulnerability of those assumptions. This process will require the use of various types of operational data that can be (and often is) collected but needs some process for determining the potential consequences of the collected data. The research will build on a new model of accident causation called STAMP (System-Theoretic Accident Model and Processes) and a new hazard analysis technique (STPA or System-Theoretic Process Analysis) to work on that model. STAMP extends current accident causality to include more complex causes than simply component failures and chains of failure events or deviations from operational expectations (which is the basis for traditional hazard analysis methods). STAMP incorporates basic principles of systems thinking and is based on systems theory rather than traditional reliability theory. A new analysis tool called Active STPA will be created that will use data commonly collected during operations to identify when the assumptions made during design and development of a product or service are violated during operations and thus may be a precursor for an accident and evidence that risk may be increasing. The test bed will be airline operations. While the subject of this research is limited to identifying leading indicators related to safety and accidents, the ideas will apply to leading indicators and risk management for system properties other than safety and could open a new path to evaluating risk during operations for researchers to exploreThis award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.