Current research in Enterprise Wide Optimization (EWO) is oriented more towards studying the interface between chemical engineering and operations research. This investigation studies the role of industrial automation and data mining for leveraging EWO. In particular, the role of field device integration (FDI), data models, OPC Unified Architecture (OPC UA) and information models that promote vertical data integration, and data mining techniques that create knowledge from aggregated data in enhancing EWO is studied. Further, the investigation shows that, integrating data mining and optimization models in EWO results in more realistic optimization problem that encapsulate the disturbance and uncertainties faced by process industries. As a result, EWO integrated with data mining techniques lead to more realistic solutions that are capable of dealing with uncertainties. Two illustrative examples from a rolling industry on energy and asset optimization are studied in this investigation. Our study reveals that emerging models in industrial automation and data mining are the key enablers of EWO in process industries.
Enabling technologies for Enterprise Wide Optimization / Srinivasan, S.; Grobmann, D.; Del Vecchio, C.; Balas, V. E.; Glielmo, L.. - (2015). (Intervento presentato al convegno 10th IEEE International Conference on Industrial and Information Systems, ICIIS 2015 tenutosi a Peradeniya; Sri Lanka nel 17 December 2015 through 20 December 2015) [10.1109/ICIINFS.2015.7399051].
Enabling technologies for Enterprise Wide Optimization
Del Vecchio C.;Glielmo L.
2015
Abstract
Current research in Enterprise Wide Optimization (EWO) is oriented more towards studying the interface between chemical engineering and operations research. This investigation studies the role of industrial automation and data mining for leveraging EWO. In particular, the role of field device integration (FDI), data models, OPC Unified Architecture (OPC UA) and information models that promote vertical data integration, and data mining techniques that create knowledge from aggregated data in enhancing EWO is studied. Further, the investigation shows that, integrating data mining and optimization models in EWO results in more realistic optimization problem that encapsulate the disturbance and uncertainties faced by process industries. As a result, EWO integrated with data mining techniques lead to more realistic solutions that are capable of dealing with uncertainties. Two illustrative examples from a rolling industry on energy and asset optimization are studied in this investigation. Our study reveals that emerging models in industrial automation and data mining are the key enablers of EWO in process industries.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
