Detecting collusive fraud in enterprise resource planning systems

Abstract

As technology advances, fraud is becoming increasingly complicated and difficult to detect, especially when individuals collude. Surveys show that the median loss from collusive fraud is much greater than fraud perpetrated by individuals. Despite its prevalence and potentially devastating effects, internal auditors often fail to consider collusion in their fraud assessment and detection efforts. This paper describes a system designed to detect collusive fraud in enterprise resource planning (ERP) systems. The fraud detection system aggregates ERP, phone and email logs to detect collusive fraud enabled via phone and email communications. The performance of the system is evaluated by applying it to the detection of six fraudulent scenarios involving collusion.