Development of IT Architecture for Generative AI-Driven Assembly Assistive Technologies

This thesis discusses the creation of a system intended to assist neurodivergent workers in manual assembly tasks within the framework of Industry 5.0, specifically in the manufacturing sector. The main objective is to develop an intelligent assistance system that can identify errors in real time and offer tailored advice to operators so they can quickly fix their errors and drastically cut down on the time between error detection and resolution. The system focuses on five distinct forms of neurodivergence: attention deficits, logical reasoning difficulties, dyslexia, memory problems, and dyscalculia. Cameras and a collaborative robot (cobot) that can identify mistakes made during assembly are used to keep an eye on operators. In response, specific modules offer customized instructions according to the operator's condition: light projections with detailed instructions for attention-related difficulties, automatic component delivery for memory problems, visual flowcharts for logical difficulties, LED indicators for dyscalculia, and audio feedback for dyslexia. Now, the system consists of five separate modules, each of which addresses a different neurodivergence. To effectively manage them, there isn't a single system architecture, though. Creating a system architecture that combines these modules into a logical, adaptable, and scalable environment is the aim of this thesis. An Event-Based Microservices Architecture is suggested to accomplish this, enabling future system extensibility, real-time responsiveness, and modular interoperability. The suggested architecture intends to serve as a basis for inclusive and human-centered design in smart manufacturing environments in addition to increasing operational efficiency and lowering cognitive load for neurodivergent operators. By putting diversity and human well-being at the center of technological innovation, this work hopes to support the larger vision of Industry 5.0. Delivering a strong and inclusive technological framework that improves human well-being and productivity in smart manufacturing contexts is the ultimate objective. This work supports the larger goal of Industry 5.0, where innovation is propelled by human-centric values like personalization, accessibility, and social sustainability, by creating a workplace that actively adjusts to individual cognitive differences. Future advancements in assistive technologies and adaptive human-machine collaboration systems across a range of industrial domains may also be based on the findings of this study.