Omnichannel Strategies and Data-Driven Robust Inventory Policies with Demand Uncertainties
Yue Sun,Ruozhen Qiu,Songshi Shao,Minghe Sun,Zhi-Ping Fan
DOI: https://doi.org/10.1016/j.cor.2024.106830
2025-01-01
Abstract:This work investigates retailer omnichannel strategies and inventory policies when selling multiple products through online and physical stores with demand uncertainties. Three profit maximization models are developed under the buy-online-pickup-in-store (BOPS), the ship-to-store (STS) and the BOPS+STS omnichannel modes. A data-driven robust optimization approach based on support vector clustering (SVC) is applied to cope with demand uncertainties. By mathematical derivation, the developed robust optimization models are reformulated as tractable linear programming models and the optimal solutions are obtained. Numerical experiments are conducted to evaluate the effectiveness and practicality of the SVC-based robust optimization approach in coping with demand uncertainties, and to explore the conditions for the retailer to benefit from the BOPS, the STS and the BOPS+STS strategies. The results indicate that the proposed approach outperforms the traditional robust optimization approaches in hedging against demand uncertainties, and the developed models can be solved efficiently. The retailer’s optimal strategy is always BOPS+STS when the store pickup hassle cost is high, is STS when this hassle cost is low, but is never BOPS. Consumer’s channel selections are significantly influenced by the hassle costs associated with different channels and, thus, the retailer should conduct a comprehensive analysis of consumers’ behaviors to make optimal ordering decisions and order fulfillment strategies. The results further suggest the retailer to maintain a record of daily sales data and to implement measures, such as advanced technology and procurement from low cost suppliers, to reduce operating costs.