AI-Supported Marketplace For Industrial Capabilities

Abstract

The current efforts to digitize production processes in the industrial environment are associated with several challenges. While there is already a wide range of standards for mapping components and products, there is still a lack of solutions for communicating concrete product requirements and given capabilities of a plant or production hall. Previous approaches, which rely on standardized feature catalogs or rigidly defined semiotics, do not offer a satisfactory solution. More desirable is an intervention-free data exchange and a negotiation process that mediates between product requirements and setup conditions. To this end, this paper proposes an architecture of a marketplace whose components compare the capabilities of an industrial system with the product requirements and evaluate them in terms of their suitability to meet the desired requirements. An approach for the rapid recognition of equality in chemical compounds using GCN is examined in detail and transferred to an industrial context.