The evaluation of internal control is one component of the audit process. It establishes to what extent controls in a company's accounting system prevent, correct or detect system exposure to loss or misrepresentation. This information is used for determining what tests to subject accounting procedures to. Internal control evaluation is a complex process in terms of the judgments necessary and the lack of normative criteria for specifying these judgments. The main advantages to be gained from automating part of the evaluation process include reducing the variability in the quality of evaluation, and enabling the auditor to concentrate of more complex judgment areas. Our goal is to explore the feasibility of developing an intelligent audit assistant for internal control evaluation. This paper describes a computational model for internal control evaluation which uses a knowledge-based approach.