Confirmation mechanisms shape order-picking performance in pick-by-light and pick-by-point systems

Abstract Order picking remains one of the most labor-intensive warehouse processes, and the design of human–system interaction may substantially affect both operational efficiency and process stability. This study compares six confirmation mechanisms embedded in Pick-by-Light and Pick-by-Point order-picking systems under controlled experimental conditions. Seventy participants completed standardized manual picking tasks using all tested variants across five repeated trials, resulting in 2,100 observations. The main performance measure was task completion time, while process stability was assessed using standard deviation and the coefficient of variation. The results showed significant differences between confirmation mechanisms, a significant short-term familiarization effect across repeated trials, and a significant confirmation mechanism-by-trial interaction. Pick-by-Light with infrared confirmation achieved the shortest mean completion time (20.74 s) and the lowest coefficient of variation (13.75%), whereas radar-based Pick-by-Point produced the slowest mean completion time (28.20 s) and the highest variability (CV = 27.91%). These findings indicate that performance differences are not determined solely by the general technology family, but also by the specific confirmation logic and interaction design. The study contributes to human-centered warehouse design by showing that low-friction confirmation mechanisms can improve order-picking speed, repeatability, and short-term user adaptation in manual picker-to-parts environments.