Dataset of workers operation durations in assembly tasks

As manufacturing systems increasingly integrate human-centric strategies with smart technologies, understanding and modelling operator performance becomes essential. However, one of the major bottlenecks in this research area is the lack of publicly available, highquality datasets capturing real operator performance under authentic
production constraints. This data descriptor introduces a dataset collected from a semi-automated assembly line featuring two workstations, one manual and one collaborative, where four workers executed standardized tasks in a real factory belonging to Silverline Endustri Ve Ticaret A.S. (Silverline). The dataset comprises detailed,
time-stamped operational durations, with over 180 task-level records per worker. By making this dataset openly accessible, the aim is to provide a critical resource for validating human performance models and fostering reproducible research in the emerging field of humancentric industrial analytics.