Anticipatory shipping versus emergency shipment: data-driven optimal inventory models for online retailers

ABSTRACT

The inventory levels of pickup points play an important role for the same-day or next-day pickup and delivery services. The previous inventory optimisation research usually makes an assumption about demand distribution, does not use the real dataset or consider shipping strategies for this problem. In this study, we introduce a new strategy, mixture of anticipatory and emergency shipping, and propose forecasting-optimisation integrated approach to optimise multi-items' inventories in each pickup point based on big data analysis. We explore a real dataset including 23,808,261 records with 54 pickup points and 4018 items. We first cluster the dataset based on the distances between pickup points and the warehouse, then, implement the forecasting-optimisation integrated algorithms to select the more profitable strategy for each group. The result indicates that compared with the original algorithms, our proposed approach can effectively increase the profits, particularly, the novel algorithm, Long Short-Term Memory networks – Quantile Regression, performs better. Additionally, we find that the 100% anticipatory shipping is not necessarily superior to emergency shipment, when the pickup point is farther from the warehouse, the advantage of emergency shipment is more significant. However, the mixture of anticipatory and emergency shipping can contribute to higher profits for online retailers.

KEYWORDS:

• Anticipatory shipping
• emergency shipment
• forecasting
• inventory management
• data-driven decision
• deep learning

Acknowledgements

The authors would like to thank the 10th IFAC MIM 2022 conference for providing a platform to present the brief version of this study (Ren et al. 2022), and thank the experts for their valuable comments and suggestions, which help to improve the quality of the paper greatly.

Disclosure statement

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

Data Availability Statement

The data that supports the findings of this study is openly available on Kaggle Competition platform at http://www.kaggle.com/competitions/favorita-grocery-sales-forecasting/data.

Additional information

Funding

This study was supported by the National Natural Science Foundation of China (Grant Nos. 71971095, 71821001, 71620107002).

Notes on contributors

Xinxin Ren

Xinxin Ren is a Ph.D. candidate of management science and engineering at Huazhong University of Science and Technology. She is a visiting Ph.D. in AIM Institute, Emlyon Business School. Her research interests include decision science, machine learning, big data analysis and decision, electronic commerce, and logistics management.

Yeming Gong

Yeming Gong is a professor of management science at Emlyon Business School. He is the institute head of AIM (Artificial Intelligence in Management) Institute and the director of BIC (Business Intelligence Center). He published 100+ papers in journals such as International Journal of Production Research, Production and Operations Management, Transportation Science, European Journal of Information Systems, International Journal of Research in Marketing, European Journal of Operational Research, International Journal of Production Economics, Journal of Business Research, Transportation Research Part E, International Journal of Information Management, OMEGA, Annals of Operations Research, and Journal of the Operational Research Society, among others.

Yacine Rekik

Yacine Rekik is a professor of decision sciences at ESCP Business School. His work has appeared in International Journal of Production Research, Decision Sciences, European Journal of Operational Research, International Journal of Production Economics, Production Planning and Control, International Journal of Systems Science, and Transportation Research Part E: Logistics and Transportation Review, among others.

Xianhao Xu

Xianhao Xu is a professor of management science and engineering at Huazhong University of Science and Technology. His work has appeared in Transportation Science, European Journal of Operational Research, International Journal of Production Economics, International Journal of Information Management, Journal of the Operational Research Society, Computers & Industrial Engineering, Transportation Research Part E: Logistics and Transportation Review, and International Journal of Production Research, among others.