Machine failures cause adverse impact on operational efficiency of any manufacturing concern. Identification of such critical failures and examining their associations with other process parameters pose a challenge in a traditional manufacturing environment. This research study focuses on the analysis of critical failures and their associated interaction effects which are affecting the production activities. To improve the fault detection process more accurately and efficiently, a conceptual model towards a smart factory data analytics using cyber physical systems (CPS) and Industrial Internet of Things (IIoTs) is proposed. The research methodology is based on a fact-driven statistical approach. Unlike other published work, this study has investigated the statistical relationships among different critical failures (factors) and their associated causes (cause of failures) which occurred due to material deficiency, production organization, and planning. A real business case is presented and the results which cause significant failure are illustrated. In addition, the proposed smart factory model will enable any manufacturing concern to predict critical failures in a production process and provide a real-time process monitoring. The proposed model will enable creating an intelligent predictive failure control system which can be integrated with production devices to create an ambient intelligence environment and thus will provide a solution for a smart manufacturing process of the future.