Shorter product life cycles and an ever-growing variety of product variants require manufacturers to perform production ramp-ups more frequently. These ramp-ups, as well as the subsequent series production, are becoming increasingly complex. The associated productivity losses, especially during ramp-up, threaten companies’ competitiveness and must therefore be minimized. In largely manual processes, worker competence is the key lever for reducing these losses: the speed and quality with which workers perform a task depend primarily on their individual skills. This paper presents a methodology that integrates established learning- and forgetting-curve models to predict individual competence development as a function of the production program. Unlike existing approaches, it primarily focuses on the individual task-based execution time of one worker. Based on this, execution times can be examined in detail at the task, product, and workstation levels. The proposed methods and visualizations provide valuable insights for actively managing competence development and thereby effectively reducing productivity losses, enabling conclusions about task sequencing, employee assignment, and learning intervals that can accelerate ramp-up performance in practical applications.
Mastroianni et al. (Tue,) studied this question.