基于CREAM的船舶引航员人因可靠性预测研究

摘要

In order to precisely predict human reliability of ship pilots, this paper proposes a model to analyze human reliability based on cognitive reliability and error analysis method (CREAM).An improved decision-making and trial evaluation laboratory (DEMATEL) method is utilized to obtain the weights of common performance conditions (CPCs) by using triangular fuzzy numbers.Then, a fuzzy comprehensive evaluation method is introduced to evaluate the performance CPCs, which can reduce the influences of subjective assessment when considering judgments from experts.Afterwards, the relationship between situational environment index and human error probability is established for a precise prediction.Finally, the results of modified CREAM and basic CREAM is compared under both actual and predefined scenarios.The result demonstrates that the modified CREAM is more reliable and sensitive than the basic CREAM.%为有效预测船舶引航员人因可靠性,以认知可靠性和失误分析方法(CREAM)基本法为基础,建立了人因可靠性预测模型.使用三角模糊数改进决策实验与评估实验方法(DEMATEL),获得了CPC因子权重;将模糊综合评价理论引入CPC因子绩效效应评价,降低了专家判断的主观因素对评价结果的影响;通过构建情景环境指数和人因失误概率的函数关系,实现了人因可靠性的精确预测.分别采用CREAM基本法和模型分析了实际情景环境下和设定情景环境下引航员的人因可靠性,结果表明:模型较传统的CREAM基本法有更好的可靠性和敏感性.