An efficient data-driven method for storage location assignment under item correlation considerations

Abstract

This research investigates the storage location assignment problems of correlated-items under a realistic multiple-cross-aisle warehouse setting. To accommodate the shortest picking route decision and update customer orders in each replenishment cycle to capture the changing trend in customer preferences. An item-correlations considered fitness function is developed to evaluate the benefit of exchanging item locations and minimise the picking costs. A data-driven storage location assignment method called storage location assignment for correlated-item method is proposed to improve the order picking efficiency. The explicit considerations make this work distinct from existing studies: (1) correlation among items in customer orders, (2) penalty for crossing-aisles in warehouse traffic, and (3) real retail dataset adopted. Our method considers the effect of correlated items in customer orders, through a storage exchange benefit function to evaluate the fitness of storage location to minimise warehouse operation costs and enhance operation efficiency. With a real ecommerce dataset, the numerical study results show that our method can reduce the travelling distance by 5–10% compared with a conventional turnover-based storage policy. Our method not only outperforms in terms of travel distance. The picking time improvement is even more significant for large warehouse if a moderate penalty for crossing-aisles is considered.

KEYWORDS:

• Data-driven
• item correlation
• storage location assignment
• order picking
• cross-aisle penalty

Acknowledgement

The work of the corresponding author was supported in part by the National Science and Technology Council, Taiwan under Grant No NSTC 111-2410-H-035-055-MY2. The third author was supported in part by NSTC 110-2221-E-002-160-MY2 and 107-2628-E-002-006-MY3.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The dataset that supports the findings of this study is provided by the University of California, Irvine, and is openly available. It can be found at https://archive.ics.uci.edu/ml/datasets/Online+Retail.

Additional information

Notes on contributors

Ywh-Leh Chou

Dr. Ywh-Leh Chou is with the Department of Industrial Engineering and Systems Management, Feng Chia University, Taichung, Taiwan.

Vincent F. Yu

Dr. Vincent F. Yu is with the Industrial Management, National Taiwan University of Science and Technology.

Cheng-Hung Wu

Dr. Cheng-Hung Wu is with the Institute of Industrial Engineering, National Taiwan University, Taipei, Taiwan.