Assistant Professor Yan Zhenzhen, a mathematician specializing in optimization and data analytics, is developing decision models to help online retailers improve their logistics. Photo credit – M.Fadly
As more consumers move from brick-and-mortar stores to online shopping, e-commerce is booming. Behind the scenes, however, online stores face unique challenges caused by their business model. Unlike traditional retailing, items sold online are handled in small quantities delivered to individual consumers. The efficiency of the “order fulfillment” process therefore has a tremendous impact on profit margins. In 2018, for example, Amazon spent about USD$30B on shipping costs, against US$10B of net income.
Mathematicians have played a quiet but important role in helping to optimize the order fulfillment process. After an online order is placed by a consumer online, it goes through order-warehouse assignment, order-picking, packing, and delivery to the consumer. Sophisticated algorithms have been developed for dealing with many of these logistical tasks, including how to make order-warehouse assignments, perform order-picking and packing, and choosing whether to consolidate or split shipments.
Assistant Professor Yan Zhenzhen, a mathematician at Nanyang Technological University, Singapore (NTU Singapore), has recently been awarded the NOL Fellowship to develop a new type of order fulfilment model. The project, which began in August 2022, aims to create a set of optimized procedures that online retailers can use to strike a balance between cost reduction and responsiveness.
Assistant Professor Yan, who is a faculty member at NTU’s School of Physical and Mathematical Sciences, is an expert on optimization, data analytics, and the application of mathematical principles to practical domains such as supply chains and healthcare operations. Recently, her team has been investigating the problem of how long an online retailer should hold a consumer’s orders before sending them for delivery.
Assistant Professor Yan Zhenzhen’s research interest lies in the interplay between data optimization and analytics including data-driven pricing.
This “order-holding problem”, as Assistant Professor Yan calls it, involves a tricky balancing act. When a consumer orders consecutive items from an online retailer, the retailer can wait to package multiple items into a single shipment, thereby reducing costs. However, waiting too long risks frustrating the consumer. Developing a decision process for optimal order-holding turns out to be a surprisingly intricate mathematical problem, which Assistant Professor Yan and her research group will tackle with support from the NOL Fellowship.
Apart from building a mathematical model for the order-holding problem, Assistant Professor Yan also aims to develop a “sequential decision model” for consumers, which online retailers can use to craft personalized order-holding policies for individual customers.
“I aim to create predictive tools that are both accurate and easy to interpret, while requiring only a moderate amount of data,” says Assistant Professor Yan. “Using intelligent order fulfillment systems, online retailers will be able to better accommodate their customers’ needs, while reducing their own costs.”
A working paper, documenting some early results from Assistant Professor Yan and her group, was previously selected as a finalist in the 2020 MSOM Data-Driven Challenge, an industry-sponsored technical competition.
References
Acimovic, J., S. C. Graves. 2015. Making better fulfillment decisions on the fly in an online retail environment. Manufacturing & Service Operations Management, 17(1), 34-51.
Andrews, J. M., V. F. Farias, A. I. Khojandi, C. M. Yan. 2019. Primal-Dual Algorithms for Order Fulfillment at Urban Outfitters, Inc. INFORMS Journal on Applied Analytics, 49(5), 355-370.
De Koster, R., T. Le-Duc, K. J. Roodbergen. 2007. Design and control of warehouse order picking: A literature review. European Journal of Operational Research, 182(2), 481-501.
Gzara, F., E. Samir, U. Yildiz, and G. Baloch. 2020. Data-Driven Modeling and Optimization of the Order Consoli- dation Problem in E-Warehousing. Informs Journal on Optimization, 2(4), 273-296.
Jasin, S., A. Sinha. 2015. An LP-based correlated rounding scheme for multi-item ecommerce order fulfillment. Operations Research, 63(6), 1336-1351.
Wei, L., S. Jasin, R. Kapuscinski. 2021. Shipping Consolidation with Delivery Deadline and Expedited Shipment Options. Manufacturing & Service Operations Management, forthcoming.
Xu, P. J., R. Allgor, S. C. Graves. 2009. Benefits of reevaluating real-time order fulfillment decisions. Manufacturing & Service Operations Management, 11(2), 340-355.