Previous works have introduced digital assistance systems to provide non-machine-learning experts with a low-threshold way to approach the topic of machine learning (ML) in manufacturing environments. According to the literature, those systems are not only suitable for independent use, but also for so-called work-based learning. Literature further suggests that users need to be comprehensively involved in the development process to make the system human-centered. Therefore, this paper presents a procedure how digital assistance systems as medium for work-based learning in the context of ML applications can be extended with regard to human-centered design elements. The procedure follows the systematics of the ISO 9241-210 standard, where each step is taken such that the development is carried out from the perspective of potential future users. The procedure is then transferred to a previously published assistance system, whereby extending it accordingly. The final validation in learning factory workshops concludes that the system now has a measurably better usability. In summary, the research results provide learning factory operators with an innovative approach to empowering non-ML experts to handle ML applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
Author Keywords
Index Keywords
