Automatic translation of plant data into management performance metrics: a case for real-time and predictive production control

• Published in: International journal of production research Vol. 55; no. 17; pp. 4862 - 4877
• Main Authors: Mousavi, A., Siervo, H.R.A.
• Format: Journal Article
• Language: English
• Published: London Taylor & Francis 02.09.2017; Taylor & Francis LLC
• Subjects: Accuracy; brewery; Business metrics; Computer simulation; control; Control systems; Decision making; Mathematical models; overall equipment/line effectiveness; Parameter estimation; Performance measurement; predictive; Predictive control; Production controls; Real time; Simulation; Supervisory control and data acquisition; Transfer functions
• ISSN: 0020-7543; 1366-588X
• DOI: 10.1080/00207543.2016.1265682

A scalable and repeatable solution for linking shop-floor control system to a discrete event simulation (DES) model is presented. The key objective is to automatically translate the real-time data from the control system (e.g. supervisory control and data acquisition, SCADA) into KPI transfer functions of the production process. Such a seamless translation allows for the integration of engineering data emitted at plant level to higher level information system for decision-making. The solution provides a platform for researchers and practitioners to utilise the capabilities of real-time DAQ and control with that of discrete event simulation to accurately measure the key manufacturing systems performance metrics. In addition to the real-time capabilities, the predictive capabilities of the solution provide the managers to look ahead and to conduct What-if scenarios. Such capability enables line management to optimise performance and predict destabilising factors in the system ahead of time. A fully operational version of the designed solution has been deployed in a brewery's live production system for the first time. The brewhouse production line model measures the utilisation of resources, Overall Equipment Effectiveness, and Overall Line Effectiveness in real-time and fast-forward mode simulation. The results of the predictive models (What-if-Scenarios) have been validated and verified by statistical means and direct observations. The accuracy of the estimated parameters is highly satisfactory.