A data-driven work scheduling framework based on perceived physical fatigue in warehouse order picking.

This study addresses the problem of designing warehouse work schedules by incorporating workers' perceived physical fatigue. To better understand and mitigate workplace fatigue, a practical, data-driven methodological framework is proposed that integrates survey-based fatigue measurement, regression analysis and mathematical optimization. In a South Korean order fulfillment center, order pickers were surveyed to assess their perceived physical fatigue. The regression analysis identified six significant factors associated with fatigue, which were used as key inputs for schedule design. An optimization model, linked to the regression results, determines the timing and duration of work-break cycles. Four alternative schedules were generated to balance the objectives of minimizing the average fatigue level and the end-of-shift fatigue level. After managerial consultation, the schedule focusing on end-of-shift fatigue reduction was selected. Compared with the current schedule, it is expected to reduce the mean end-of-shift perceived fatigue level by 11.5% while maintaining the same total working hours.