Picking scheduling for single picker to multi-workstations of the part-to-picker order fulfilment system

Abstract

To reduce human resource costs, the part-to-picker order fulfilment systems may have a single picker in charge of multiple workstations. And the picking speed of the picker becomes faster as the picking number increases due to the learning effect in the picking operation. In this paper, the scheduling problem to optimizing picking sequence of the picker is presented to minimize the maximum picking time, where one picker is responsible for multiple workstations. The learning effect and travel time between workstations are taken into account to improve scheduling accuracy. Two mixed integer programming (MIP) models are proposed to solve the problem, namely the rank-based model and disjunctive model. The performance of the two Mixed Integer Programming (MIP) models has been evaluated, and it has been found that they are only capable of solving small-scale problems. The rank-based model is limited to solving problems with up to 9 groups, whereas the disjunctive model can handle up to 20 groups. Therefore, the disjunctive model outperforms the rank-based model. Moreover, this paper proposes Interval Insertion NEH (IINEH) and iterative greedy (IG) algorithm to solve the large-scale problem. Numerical experiments demonstrate the effectiveness of the two methods to solve the problem, where IINEH operates faster while IG gives better results. Therefore, when faced with a large-scale problem, IINEH is recommended if a quick solution is needed. If better optimization results are needed, the decision maker can choose IG.