Data is a key component of smart manufacturing coming from various sources and in different forms. The amount of data produced in the Manufacturing Data Life Cycle (MDLC) could be vast and remain useless if not processed and mapped into information that is understood by users. Because of the data volume, velocity, variety,and veracity, it is not feasible to expect humans to analyze it using traditional manufacturing tools; business analytics solutions provide more efficient means. In this paper, first, we aim at identifying the latest research areas and trends of business analytics (BA) in the production management (PM) literature using text mining, bibliometric mapping, and visual analytics. We discuss the related research gaps, BA challenges and problems and give suggestions for possible BA applications and future research directions. Second, we selected the performance management and decision-making cluster for presenting a real-world case and we applied MDLC as a framework. The case study presents a big data analytics performance monitoring system from a multi-product refinery.