Generation of structure function based on ambiguous and incompletely specified data using fuzzy random forest

Analysis of any system requires its mathematical model. One of the possible mathematical models of reliability engineering is the structure function. The structure function shows a correlation between system performance levels and its components. The structure function determines system performance level according to the states of the system components. In this paper, the new method of structure function construction is proposed. The fuzzy random forest is used to transform data of real system behavior to structure function. This method allows handling incompletely specified and uncertain data.