Document Type



Information systems (IS) outsourcing is an important issue of IS practices in organizations. To achieve better IS outsourcing performance, it is critical for organizations to develop appropriate outsourcing strategies. A prediction model is useful to forecast the possibility of a failure or to point out unforeseen problems. Based on a prediction model of IS outsourcing success, this study applied an constraint-based evolutionary approach for backward tracking the optimal values of organizational IS attributes that best approximate the target success level of IS outsourcing. One hundred forty-six real IS outsourcing cases, each with 22 features and 8 outcome features, are collected. The proposed system demonstrated that valuable suggestions can be made regarding the increase or decrease levels of adjustable IS attributes for organizations outsourcing their IS functions.