Developing a Method for Creating and Validating the Safety of Interface Design

技术变得无处不在，软件越来越多地在更复杂的环境中实施，在这些环境中，人类软件交互更有可能导致死亡、受伤、诉讼、软件重新设计成本和重新实施成本。该研究计划的总体目标是开发一种方法来创建和验证基于证据的安全启发式方法，以减少界面设计中出现的错误数量。迄今为止，人在环仿真尚未用于验证安全启发法或指南。这种方法对于防止技术引起的错误至关重要。 该研究计划的目标是：（1）根据已发表的涉及人类操作员与系统交互的实证文献制定基于证据的安全指南，（2）在人机交互领域专家的帮助下验证这些指南，（3）开发几种界面设计原型（有和没有）指南，（4）使用从公开可用的事件数据中提取的人在环模拟来测试用户界面设计的安全性，（5）验证指南防止技术引起的错误的能力。 该研究利用了软件安全和界面设计方面的现有知识，并将此类工作扩展到： 创建基于证据的安全启发法以防止技术引起的错误；开发一种涉及人机交互模拟的方法，以测试和验证新开发的界面设计指南，以确保其能够在实际实施之前减少人机交互所产生的错误。这项研究将为工作环境的技术设计、定制和采购提供信息，在这些工作环境中，实证研究（系统实施后）记录了技术无意中导致工人犯错误（即技术引起的错误）的情况。

Technology is becoming ubiquitous and software is increasingly being implemented in more complex environments, where human software interactions are more likely to lead to death, injury, litigation, software redesign costs and re-implementation costs. The overall goal of this research program is to develop a methodology for creating and validating evidence-based, safety heuristics to reduce the number of errors arising from interface designs. To date, human-in-the-loop simulation has not been used to validate safety heuristics or guidelines. This approach is essential to preventing technology-induced errors. The objectives of this research program are to: (1) develop evidence-based, safety guidelines based on the published empirical literature involving human operators interacting with systems, (2) validate those guidelines with the assistance of experts in the field of human computer interaction, (3) develop several prototypes of interface designs (both with and without) the guidelines, (4) test safety of the user interface designs using human-in-the-loop simulations drawn from publicly available incident data, (5) validate the guidelines for their ability to prevent technology-induced errors. The research draws upon existing knowledge in software safety and interface design, and extends such work to: the creation of evidence-based safety heuristics to prevent technology-induced error; and the development of a methodology involving human-in-the loop simulations to test and validate newly developed guidelines for interface designs for their ability to reduce errors arising from human computer interaction before real-world implementation occurs. This research will inform the design, customization and procurement of technologies for work environments where empirical studies (after systems have been implemented) have documented instances where technology has inadvertently caused workers' to make errors (i.e. technology-induced errors).