Ontology Validation of Manufacturing Execution Systems Through the Analysis of Semantic Descriptions

Abstract

Current manufacturing systems are comprised of heterogeneous software and hardware components that exchange information on various levels. These levels have distinct functionalities and target different timeframes but they have to communicate for the effective and efficient operation of an enterprise. On one hand, the present trend in industry 4.0 promotes smart manufacturing systems. On the other hand, new product variants, assets, machinery, and diverse manufacturing technologies are constantly added to the manufacturing systems. Hence, the capability of a manufacturing system to follow the dynamic changes of the industry and customers becomes essential. In order to realize this, integration is required to link those individual levels, such as Enterprise Resource Planning (ERP) and Manufacturing Execution Systems (MES), and subsequently perform physical operations in the shop floor. In that sense, using standards becomes significant in order to avoid inconsistent and redundant systems and integration architectures. The ISA-95 standard, from the International Society of Automation (ISA), describes the interface needed for integration of enterprise and control levels by specifying a uniform terminology and a coherent collection of concepts and models. The objective of this thesis work is to demonstrate an approach for designing a generic manufacturing systems model using a Knowledge Representation and Reasoning (KR&R) formalism, i.e., an ontology, conformant to ISA-95 that allows easy extendibility. The main contribution of the approach lies in the addition of standard and use case specific semantic rules that connect the core concepts and increase the expressivity and reasoning capabilities of the model. Ontologies are flexible and easy to update and enable the reuse of knowledge, which should be considered with the abundance of data available in modern systems. The proposed model describes the system based on products, processes, and resources involved in manufacturing. The applicability, extendibility, and reusability of the proposed model has been validated by its application in an industrial use case as a proof of concept.