Augmented Reality (AR) technologies are used as human–machine interface within various types of safety-critical systems. Several studies have shown that AR improves human performance. However, the introduction of AR might introduce risks due to new types of dependability threats. In order to avoid unreasonable risk, it is required to detect new types of dependability threats (faults, errors, failures). In our previous work, we have designed extensions for the SafeConcert metamodel (a metamodel for modeling socio-technical systems) to capture AR-related dependability threats (focusing on faults and failures). Despite the availability of various modeling techniques, there has been no detailed investigation of providing an integrated framework for risk assessment in AR-equipped socio-technical systems. Hence, in this paper, we provide an integrated framework based on our previously proposed extensions. In addition, in cooperation with our industrial partners, active in the automotive domain, we design and execute a case study. We aim at verifying the modeling and analysis capabilities of our framework and finding out if the proposed extensions are helpful in capturing system risks caused by new AR-related dependability threats. Our conducted qualitative analysis is based on the Concerto-FLA analysis technique, which is included in the CHESS toolset and targets socio-technical systems.

增强现实 (AR) 技术被用作各种类型的安全关键系统中的人机界面。多项研究表明，增强现实可以提高人类的表现。但是，由于新型可靠性威胁，AR 的引入可能会带来风险。为了避免不合理的风险，需要检测新型的可靠性威胁（故障、错误、失效）。在我们之前的工作中，我们设计了 SafeConcert 元模型（用于建模社会技术系统的元模型）的扩展，以捕获与 AR 相关的可靠性威胁（关注故障和故障）。尽管可以使用各种建模技术，但还没有详细研究为配备 AR 的社会技术系统中的风险评估提供集成框架。因此，在本文中，我们提供了一个基于我们之前提出的扩展的集成框架。此外，我们与活跃在汽车领域的工业合作伙伴合作，设计并执行案例研究。我们旨在验证我们框架的建模和分析能力，并找出提议的扩展是否有助于捕获由新的 AR 相关可靠性威胁引起的系统风险。我们进行的定性分析基于 Concerto-FLA 分析技术，该技术包含在 CHESS 工具集中并针对社会技术系统。我们旨在验证我们框架的建模和分析能力，并找出提议的扩展是否有助于捕获由新的 AR 相关可靠性威胁引起的系统风险。我们进行的定性分析基于 Concerto-FLA 分析技术，该技术包含在 CHESS 工具集中并针对社会技术系统。我们旨在验证我们框架的建模和分析能力，并找出提议的扩展是否有助于捕获由新的 AR 相关可靠性威胁引起的系统风险。我们进行的定性分析基于 Concerto-FLA 分析技术，该技术包含在 CHESS 工具集中并针对社会技术系统。