Smart and efficient: Learning curves in manual and human-robot order picking systems

Order picking has been identified as the most labour-intensive, as well as costly activity within warehouse logistics and is experiencing significant changes due to new technologies in the forms of artificial intelligence (AI) and automation. One fundamental question concerns the employees learning progress in human-robot picking systems compared to existing manual technologies. Therefore, this paper presents an empirical analysis of learning curves in manual pick-by-voice (n=30 pickers) and semiautomated (n=20 pickers) order picking. Aspiring to measure the individual learning progress without a priori assumptions, this publication is the first to apply Data Envelopment Analysis and examine order pickers learning curves in real application scenarios. The findings indicate that automating human work accelerates the individual learning progress in human-robot picking systems.