The Deliverable 2.2, entitled “Ontology Definition”, has been submitted. This document presents the developments carried out in order to design and implement the GO0DMAN data model, with the purpose of harmonizing the data coming from the different solution’s domains at the agent level, enabling a seamless integration and data exchange between each of GO0DMAN’s solution actors and the Multi-Agent based Cyber-Physical System.
The main classes of the GO0DMAN Data Model are:
- GMResource: it represents a single resource inside the multi-stage manufacturing system and may have specific skills, values, events, connectors and rules associated with it.
- GMSubsystem: it represents a group of resources, or possibly other subsystems, which work together to achieve a common goal.
- GMSystem: it abstracts the entire system and extends the GMSubsystem class
- GMProcess: it provides a description of all the operations necessary to produce a given product.
- GMQualityTest: it encompasses all the information related to a specific quality control inspection (e.g. gap & flush), such as who performed it, on which product it was performed, when the inspection was carried out and also which measurements were taken.
- GMManagementRule: rules are meant to be used by the MAS-CPS to evaluate whether certain parameters extracted from the shop-floor are within their normal operational conditions, or if corrective action needs to be carried out (e.g. maintenance or machine parameter adaptation).
The GO0DMAN Data Model has been implemented using AutomationML, providing a clear way to generate instance model blueprints of different elements of the GO0DMAN solution, assisting the MAS-CPS not only in the deployment phase, but also in adapting in runtime (i.e. loading management rule descriptions, required parameter adaptation specifications, component configurations).