Analysis of metamodels for model-based production automation system engineering

Abstract

In the current Industry 4.0 era, automated production systems (aPS) comprising of multi-disciplinary artefacts all closely interwoven are required to adapt to various and varying requirements from customers and environment which introduce additional complexity. Model-based engineering on the premise of metamodelling is regarded as a promising paradigm to handle this complexity to engineer aPS. Although various metamodels appear to solve problems in different viewpoints on systems, the absence of a core metamodel causes inconsistencies between the metamodels and hinders common understanding of the system and model reuse. In this study, the authors analyse existing metamodels from different research groups and present inconsistencies among them explicitly which support the necessity of the core metamodel. Considering properties of aPS together with relevant standards, the authors present a demonstration of analyses on exemplary metamodels and a set of criteria to understand the various aspects of the aPS metamodels as the first step towards the core metamodel. Feasibility of creating a universal metamodel of aPS domain is discussed, and the authors claim the necessity of having a common understanding of core concepts of aPS to support the cross-disciplinary reuse of existing metamodels and, accordingly, the compatibility of the instalment and operation of components.