A rule based fuzzy reasoning system for assessing the risk of management fraud

Statement on Auditing Standards (SAS) No. 53 requires that the audit be designed to provide a reasonable assurance of detecting management fraud. Touche Ross (1974), Coopers and Lybrand (1977), and Price Waterhouse (1985) released a list of red flags based on business and economic factors that can be used as a set of warning indicators. The purpose of this paper is to demonstrate the use of a rule based fuzzy reasoning system to assess the risk of management fraud. Based on the Bell et al. (1993) study we identified statistically significant red flags for each assessing the risk of management fraud. Using the statistical significance of each red flag and the theoretical model we developed membership functions and fuzzy rules. The rule based system was implemented in XpertRule. Then the fuzzy reasoning system tested using the same fraud data as in Bell et al. and the results obtained were similar to statistical models. The fuzzy reasoning system can be used as an intuitive decision aid in assessing the risk of management fraud and in structured decision making