A framework for digital assembly instructions as a step towards manufacturing inclusiveness

Industrial manufacturing processes require precise understanding of instructions, which can be challenging for neurodiverse operators with reading difficulties. To bridge this gap, a digital instruction framework using object detection and natural language processing is proposed in this research. The framework uses an intelligent vision system to monitor task execution, coupled with the automatic generation of personalised voice instructions via large language models. This approach aims to improve accessibility and inclusivity in assembly lines. A case study on the assembly of a horizontal bare-shaft centrifugal pump demonstrates the effectiveness of the framework in reducing assembly errors and improving operational efficiency, making it particularly beneficial for neurodiverse individuals and promoting an inclusive work environment.