【24h】

Towards Defense-in-Depth of Human-Machine Interaction

机译:走向人机交互的纵深防御

获取原文
获取原文并翻译 | 示例

摘要

The paper presents the potentiality of Combinatorial Context Model (CCM) for context quantification of the cognitive human erroneous action. In order to demonstrate the crucial impact of violations on the post-accident context and human reliability a Violation of Objective Kerbs (VOK) method is proposed. The VOK provides a way of accounting for the probability of aberrant circumstances (prior to the initiating event) in cognitive function. The VOK adjustment of CCM to the new Performance Evaluation of Teamwork (PET) method for obtaining non-response probability curves is presented. The CCM employs the Human Cognitive Reliability (HCR) correlation idea for taking into account the context influence through the correlation between the changes of 'time window' and 'context window' rates. The new point here is that the 'context window' rate is used as a mirror image of cognitive function. The illustrative example, of the numerical analysis by VOK and PET methods, is presented in order to understand the cognitive error sensitivity to violation. The context quantification together with reliability models of the cognitive process (Rasmussen's SLM) and group communication process included in the PET method makes it possible to find, interpret and assess new characteristics of human actions, and to accomplish Defense-in-Depth of HMI.
机译:本文介绍了组合上下文模型(CCM)在认知人类错误行为的上下文量化方面的潜力。为了证明违规对事故后环境和人类可靠性的关键影响,提出了一种违反客观遏制(VOK)的方法。 VOK提供了一种方法来解释认知功能中异常情况(在发起事件之前)的可能性。提出了将CCM的VOK调整为新的团队绩效评估(PET)方法以获得无响应概率曲线的方法。 CCM采用人类认知可靠性(HCR)相关性思想,通过“时间窗口”和“上下文窗口”速率变化之间的相关性来考虑上下文影响。这里的新点是“上下文窗口”比率用作认知功能的镜像。提出了通过VOK和PET方法进行数值分析的说明性示例,以了解认知错误对违规的敏感性。 PET方法中包括的上下文量化以及认知过程(拉斯穆森的SLM)和群体交流过程的可靠性模型,使得可以发现,解释和评估人类行为的新特征,并实现人机界面的深度防御。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号