FUZZY FUNCTIONS OF EXPERT KNOWLEDGE ENCAPSULATED WITHIN STATISTICAL WORKFLOW DATA

Abstract

Taking over the skills of the human expert will make it possible to develop decision-making algorithms in conditions of uncertainty for industrial applications. Fuzzy sets are used in different fields and estimating membership functions is one of the most important issues in the design of fuzzy systems that depends directly on the identification of used method. This article presents an approach to this problem that can provide solutions in specific cases. In this context, a method of extracting the knowledge of the human expert is developed and it allows to retrieve specific expertise and the construction of algorithms for decision support systems. The conditions to apply the method and identify membership functions as well as the automation process of this stage are analyzed. There is proposed a method to determine trapezoidal or custom membership functions. The approach presented in this paper can be applied to the analysis and research of decision making in conditions of uncertainty. A case study is presented that reflects the applicability of the proposed method and algorithms.