| Establishing and maintaining proper internal controls is an essential, ongoing, and complex process for today's organizations. Periodic checks by auditors evaluate a company's internal control system, probing it for weaknesses that need correction. Auditors use expert systems to assist in such work. However, it is managers, with their detailed day-to-day attention to company operations, who are in the best position to continually monitor for internal control problems. The literature contains no example of an expert system deployed to assist managers in recognizing potential problems in internal control systems. Yet, an intelligent system of this kind might prove to be highly beneficial.; This dissertation research reports on an expert system developed to assist managers, who do not have specific backgrounds in internal control concepts, evaluate potential weaknesses in internal control systems. To examine the feasibility of developing such systems, an expert system was constructed via a six-month series of interviews with an experienced auditor. The knowledge incorporated into the system represents primarily the expertise of one expert auditor who is a partner in a major international accounting firm. The development tool was an artificial intelligence environment that provides integrated rule management, relational database management, and structured programming capabilities.; Sixty-four practitioners participated in the study as novice internal control decision-makers to examine the utility of the system. Each was asked to play the role of a manager trying to detect potential weaknesses of the internal control system described in a series of case studies. Three treatments were used as values of the independent variable: internal control evaluation with no decision aid, internal control evaluation with support of the expert system, and internal control evaluation with no decision aid after being trained with the expert system. Accuracy and decision time were measured as response variables. Participants were given a post-experiment questionnaire to measure their perceptions with the assigned task and their satisfaction in using the expert system to support their decisions about internal control weaknesses. The experiment provides strong evidence that an expert system can be built to significantly improve the accuracy, efficiency, perceptions, and satisfaction of managers evaluating internal controls. |
