LAUSR

Abstract The solution of today’s complex decision-making for smart manufacturing are dependent on the ability to: a) realistically model the manufacturing system, b) easily and timely integrate valid and consistent plant data, c) solve the problem efficiently with reasonable computational efforts, and d) incorporate feedback to continuously improve the decision-making process over time. In such a context, advanced analytics such as the predictive, prescriptive and detective analytics are the foundation of smart manufacturing in the information age. Predictive analytics examines raw data to be augmented with the purpose of concluding the behaviour of the systems, by estimating and anticipating what is likely to happen within the forthcoming future. Prescriptive analytics automates the decision-making of any physical system concerning its design, planning, scheduling, control and operation using any combination of optimisation, heuristics, machine-learning and cyber-physical systems. Detective analytics makes diagnostics on data to improve both the predictive and prescriptive analytics. In the former, by identifying and eliminating gross-errors for better predictions. In the latter, by uncovering and rectifying infeasibilities and inconsistencies for optimal prescriptions. We construct a plot of the connections of the advanced analytics at their time-spaces considering the well-established, the current and the next generation of analytics techniques. An example of an automated application of advanced analytics considering a multi-unit real-time estimation and optimisation engine relying on data integration and integrity for better decision-making is highlighted.