The condition-based maintenance (CBM) focuses on the prediction of aging, degradation, and failure process of data center at the levels of components and systems. The benefits of CBM are increasing system availability, mission effectiveness, and reducing maintenance costs. In this paper, we propose an innovative concept of decision support methodology for system failure diagnosis and prognosis in complex systems of data center power distribution systems. This paper proposes an action research of a new decision support methodology for system failure diagnosis and prognosis in data center power distribution systems. Shifting from time-based maintenance (TBM) to CBM using automated prognostics and diagnostics to identify and resolve issues before they become problems of data center downtime costs.