Fuzzy logic for risk assessment in auditing

Abstract

The assessment of the risks that an entity's internal control system may fail represents a significant challenge to independent auditors. The methodologies used to audit financial statements are usually supported by classical logic, also called binary logic, departing from the relatively simplistic premise that risk factors are either present or not in a certain kind of control process. This study aimed to conceive a risk assessment model for an entity's internal control system, using the fuzzy logic approach, to take into account the diffuse elements that compose the factors of this type or risk, which are analyzed in financial statements auditing. The conceived model was conceptually validated though interviews and debates with financial statement auditing experts and relevant bibliography. We concluded that the use of fuzzy logic to support risk assessment models not only eliminates the binary restriction imposed by classic logic, but also allows for the quantitative treatment of ambiguous concepts through a psychometric scale, to refl ect adjectives like "very good", "good", "reasonable", "of great importance", "of little importance" etc. This approach makes it possible to produce broader results that are closer to reality.