ABSTRACT
Employing advanced statistical techniques to intelligent data analysis scenarios within the Retail IT project management industry against an underlying phenomenological Cynefin framework, this study proposes the use of the complementary techniques of factor, cluster, and regression analysis on a stratified random sampling of a large and historical dataset to measure and predict the optimal combination of variables leading to successful project outcomes. Statistical analysis techniques will enable inference from the data that is not otherwise directly measurable and where subjectivity is inherent in human analysis. Delivering statistical validation of the ability to identify Cynefin complexity category at Initiation and predict the optimal assignment of Project Manager and delivery methodology is expected to correlate to project success outcomes. Automating this data analysis in subsequent studies using a cognitive machine learning approach and Robotic Process Automation (RPA) tools will assist in providing recommendations for optimal alignment of complexity ranking, delivery methodology, and project manager skills prior to project commencement
