RECAST: An Open-Source Digital Twin Framework for Industrial Production Environments

Abstract

In the era of industry 4.0, it is essential for production facilities to be equipped with “smart” capabilities. The utilization of process data, which accumulates during production, is a critical component of this evolution. High potential is seen in the combination of artificial intelligence (AI) with this data to enhance plant productivity. However, these positive assessments are countered by the limited availability of rapid implementation options. Solutions ready for use in the realms of machine and process technology are predominantly offered by manufacturers alongside their respective systems. Consequently, complex AI projects often need to be executed as part of costly, individualized endeavors. For small and medium-sized enterprises (SMEs), these challenges pose significant financial barriers, particularly when the prospects of success remain uncertain. Within the RECAST project, a framework that is both user-friendly and adaptable for process support has been developed. This framework facilitates cost-effective upgrades of existing plants by digitally representing production processes. The objective is to enable companies to independently incorporate AI into current process chains. The cornerstone of this innovative framework is Supabase, an advanced open-source serverless solution. The utilization of its Functions as a Service (FaaS) feature allows for seamless interaction with the framework via a streamlined, lightweight, and intuitive web application. This setup empowers users to define complex processes effortlessly using a low-code interface, preparing them for immediate deployment in production environments.