Information Model to Advance Explainable AI-Based Decision Support Systems in Manufacturing System Design

Abstract

AbstractArtificial intelligence is currently being used in more and more areas of production. Be it in the field of industrial robotics, automated quality inspection or cognitive support for employees in production, artificial intelligence contributes to creating smart as well as sustainable manufacturing systems. In the area of manufacturing system design, decision support models are increasingly used to facilitate the work of system designers. In this paper, we address how information models can be used to design explainable artificial intelligence decision support systems. The paper will survey and describe the information that is necessary to communicate manufacturing system design requirements to meet customer needs and use cases. The objective is to propose an information model to express system design requirements with the goal to provide a transparent representation of decisions as well as alternatives of decisions to improve the description of artificial intelligence-based decision support systems during the manufacturing system (re)design phase. The purpose of the information model is to explore the requirements and technical solutions necessary to advance manufacturing systems without losing track of alternatives, and to be able to dynamically adapt them to changing conditions in the market or the production environment.KeywordsIndustry 4.0Smart manufacturingInformation modelExplainable artificial intelligenceDecision support systems