Holistic context-sensitivity for run-time optimization of flexible manufacturing systems

Sebastian Scholze, Jose Barata, Dragan Stokic

Research output: Contribution to journalArticlepeer-review

16 Citations (Scopus)

Abstract

Highly flexible manufacturing systems require continuous run-time (self-) optimization of processes with respect to diverse parameters, e.g., efficiency, availability, energy consumption etc. A promising approach for achieving (self-) optimization in manufacturing systems is the usage of the context sensitivity approach based on data streaming from high amount of sensors and other data sources. Cyber-physical systems play an important role as sources of information to achieve context sensitivity. Cyber-physical systems can be seen as complex intelligent sensors providing data needed to identify the current context under which the manufacturing system is operating. In this paper, it is demonstrated how context sensitivity can be used to realize a holistic solution for (self-) optimization of discrete flexible manufacturing systems, by making use of cyber-physical systems integrated in manufacturing systems/processes. A generic approach for context sensitivity, based on self-learning algorithms, is proposed aiming at a various manufacturing systems. The new solution encompasses run-time context extractor and optimizer. Based on the self-learning module both context extraction and optimizer are continuously learning and improving their performance. The solution is following Service Oriented Architecture principles. The generic solution is developed and then applied to two very different manufacturing processes.

Original languageEnglish
Article number455
JournalSensors (Switzerland)
Volume17
Issue number3
DOIs
Publication statusPublished - 1 Mar 2017

Keywords

  • Context sensitivity
  • Cyber-physical systems
  • Flexible manufacturing system
  • Process optimization
  • Self-learning systems
  • Sensors for context extraction
  • SOA

Fingerprint

Dive into the research topics of 'Holistic context-sensitivity for run-time optimization of flexible manufacturing systems'. Together they form a unique fingerprint.

Cite this