A multi-criteria prediction model for project risk classifications

Abstract

Project distress predictions are essential in project management. Developing appropriate methods to classify projects and building prediction models for multi-criteria decisions requires empirical methods to minimise misclassification errors. This paper carries out multi-criteria analysis to classify projects risks using a preference disaggregation method, utilites additives discriminantes - UTADIS. The UTADIS requires predefined classification which is implemented using critical path analysis. The methods are applied on three projects and result in no misclassification error and an effective prediction model.