SLIM在数据获取方面有其独特的优势,但其过分依赖于专家的判断,使得其应用性和可靠性存在明显的不足。本文针对SLIM的关键环节,使用CREAM支持行为形成因子的确定、贝叶斯方法修正专家对行为形成因子权重的判断和模糊集确定任务等级,对SLIM进行重大改进。在保留SLIM本身特点的同时,使之在人员可靠性相关数据取得方面发挥更大的作用。
SLIM in data acquisition has its unique advantages.This paper analyzes SLIM's advantages and disadvantages on the basis of its description,and then for its obvious deficiencies to be improved.The work includes three aspects:the determination of the PSFs,the correction of experts' judgement of the weight of the PSF using Bayesian method and the determination of task hierarchy using fuzzy sets.Through the above work,SLIM's advantages will be retained.At the same time,it will play a greater role in the data collection of human reliability.