Stock allocation in a two-echelon distribution system controlled by (s, S) policies

Abstract

Stock allocation is an important decision in distribution systems, which determines how the on-hand inventory of an upstream stock is allocated to its immediate downstream stocks if their replenishment orders cannot be fulfilled completely. In the literature, almost all studies assume each stock in a distribution system is controlled by a base stock policy, although a real system may be controlled by (s, S) policies because of the existence of fixed ordering costs. In this paper, we study stock allocation in a real distribution system of Alibaba and propose two easy-to-implement Allocation-to-S methods that are more effective than three well-known rationing rules and a rationing rule currently used in Alibaba. For the 300 instances tested, one of the methods can reduce the expected cost per period by more than 1.5% and improve the service level by more than 1.3% on average with respect to the best of the four rules. In addition, a general procedure is provided to solve the imbalance problem of linear rationing rules which include the rules.

KEYWORDS:

• Inventory management
• distribution system
• stock allocation
• rationing
• (s
• S) policy

Acknowledgment

This study is supported by the Alibaba Innovative Research Project entitled ‘Optimization of Safety Stock Placement in Supply Chains with Demand and Lead Time Uncertainty’.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Haoxun Chen

Haoxun Chen received his B. S. degree in applied mathematics from Fudan University, China, in 1984, his Master and Ph.D. degree in systems engineering in 1987 and 1990, respectively, from Xi’an Jiaotong University, China. He was with Xi’an Jiaotong University as a Lecturer from 1990 to 1992 and as an Associate Professor from 1993 to 1996. He visited INRIA-Lorraine, France, as a Visiting Professor, in 1994, University of Magdeburg, Germany, as a Research Fellow of Alexander von Humboldt Foundation, in 1997 and 1998, and University of Connecticut as an Assistant Professor in 1999 and 2000. He joined University of Technology of Troyes, France, in 2001 and has been a Full Professor since 2004. His research interests include supply chain network design and performance evaluation, collaborative logistics, inventory management, production planning and scheduling, combinatorial auctions, game theory, and data envelopment analysis. He has published more than 170 papers in international journals and conference proceedings. He received King-Sun Fu Memorial Best Transactions Paper Award from IEEE Robotics and Control Society in 1998 and National Natural Science Award (China) of Second-Class in 2005. He was responsible for more than ten academic and industrial research projects such as a French ANR (Agence Nationale de la Recherche) project on collaborative logistics and an AIR (Alibaba Innovative Research) project on inventory management.

Bo Dai

Bo Dai obtained his PhD degree from University of Technology of Troyes in France. He has published more than 20 papers in well-known international journals and conference proceedings on logistics and supply chain management, including Omega, JORS, CAIE, EJIE, ORIJ. His current research interests include inventory management, supply chain optimisation, smart manufacturing and cooperative game theory.

Yuan Li

Yuan Li joined Alibaba Group as an algorithm engineer in 2017, where he has been working on inventory control in E-commerce supply chains for over two years. He received his Ph.D. degree in Logistics and Optimization of Industrial Systems (LOSI) at University of Technology of Troyes in 2017.

Yidong Zhang

Yidong Zhang joined Alibaba Group as an algorithm engineer in 2016. In 2018, he led the inventory optimisation team to deliver inventory optimisation decision-making tools for both internal and external clients. Since then, the team has worked and implemented optimisation solutions to reduce turnover days and improve the service level via improving inventory replenishment and transshipment policies. Before joining Alibaba Group, he worked at BASF and Sabre Airline Solutions, practicing operations research in solving varieties of optimisation problems in chemical and airline industry. He completed his PhD in Operations Research at University at Buffalo in 2012 and obtained BE in Mechanical Engineering at University of Science and Technology of China in 2003.

Xiaoqing Wang

Xiaoqing Wang is a senior algorithm expert in Alibaba group, and responsible for leading R&D algorithm team in inventory and fulfillment optimisation. Before joining Alibaba Group, he was a manager of supply chain FX department in Amazon and a research scientist in IBM Research, and responsible for analysis and optimisation work in Supply Chain field. He has rich research experience in the areas of logistics and retail. He has published 30+ academic papers and 20+ patents in America and China.

Yuming Deng

Yuming Deng got his OR/IE Ph.D. degree in Univ. of Texas at Austin in 2009. After graduation he joined Amazon.com with an emphasis on algorithm design for Supply Chain & transportation optimisation. He joined Tmall.com under Alibaba Group since 2014. Currently He focused on algorithm design in Dept. of Supply Chain Platform, Alibaba Group.