Fuzzy cognitive map for the design of EDI controls

Abstract

EDI control design is ill-structured and demands consideration of the complex causal relationships among various components of the controls, which may be broadly classified into formal, informal, and automated types. Each of these can, in turn, be categorized as internal or external. However, it is difficult even for EDI experts to predict the causal effects of one control on another. In order to aid the design of EDI controls, the application of a fuzzy cognitive map, EDIFCM (EDI-Control Design using a Fuzzy Cognitive Map), was developed. Structural equation modeling was used to identify relevant relationships among the components and indicate their direction and strength. A standardized causal coefficient from structural equation modeling was then used to create a fuzzy cognitive map, through which the state or movement of one control component was shown to have an influence on the state or movement of others. Thus, EDI auditors were able to enhance their understanding of the causal relationship of controls and effectively design them. (C) 2000 Elsevier Science B.V. All rights reserved.