Data Analytics Overview

Data analysis algorithms, knowledge management approaches and learning techniques will be applied at local and global level, focusing on embedding data analysis, learning algorithms in the GO0DMAN system. Agent technologies are used for local processing of information and knowledge extraction, complemented by data analytics and knowledge management techniques on global level to enable a continous, knowledge-driven production improvement.

Data Analytics Architecture

Data analysis relates to creating understanding and discovering of the usual/normal behavior of a system, based on the analytical processing on past data that can be multidimensional and very heterogeneous (challenges for the traditional quality control) and detection of anomalies as discovery in real-time, i.e. sensing early indicators that something might be strange/wrong with the process (in a particular stage, or in the process as a whole).

Knowledge Management introduces man-made observations on processes like expert experience, process calculation, process analysis, process simulation or ideal case envisioning. Because of the different level of knowledge expressions ranging from inform like natural text, semi-formal like tables or charts, to formal like strict process models, agent configurations of control thresholds, a hybrid modelling approach is proposed to merge formal data with informal know how.

Data Analytics Interaction

As real-world examples from use cases, GO0D MAN will allow:

  • A real-time analysis and knowledge extraction from the big data gathered from the shop floor to provide real time detection of anomalies or deviations, at a single stage or resulting from the effects of several stages. This mechanism allows to detect in real time the occurrence of possible defects or deviations, being able to select in real time proper strategies to prevent or mitigate their further occurrence.
  • Systemic support for root-cause analysis for different temporal and complexity dimensions as Knowledge Production Process.
  • Role-based visualisation techniques for measurements in real-time and historical results to determine in advance effects of changes and modifications performed.