New product development (NPD) is a dynamic environment and the cycle times of the projects undertaken in this environment vary significantly. To develop and judge the performance of a theoretical model to adequately fit this environment requires the combination analytical methods. The data and model should include quantitative and qualitative characteristics. Certain statistical performance characteristics of a model are easily identifiable. Empirical evidence and expert opinion form a foundation for the model to ensure that the model performance represents the real world operating environment. This paper describes a modelling approach for predicting NPD project cycle time based on both statistical and fuzzy data. Statistical performance characteristics are used to determine the fit of a model. Fuzzy set theory is used to define the membership of the statistical performance in a well performing model and to aggregate the statistical and 'soft' performance characteristics to determine good overall model performance.