This paper studies e-grocery order fulfillment policies by leveraging both customer and e-grocery-based data. Through the utilization of historical purchase data, product popularity trends, and delivery patterns, allocation strategies are informed to optimize performance metrics such as fill rate, carbon emissions, and cost per order. The study aims to conduct a sensitivity analysis to identify key drivers influencing these performance metrics. The results highlight that fulfillment policies optimized with the utilization of the mentioned data metrics demonstrate superior performance compared to policies not informed by data. These findings underscore the critical role of integrating data-driven models in e-grocery order fulfillment. Based on the outcomes, a grocery allocation policy, considering both proximity and product availability, emerges as promising for simultaneous improvements in several performance metrics. The study recommends that e-grocery companies leverage customer data to design and optimize delivery-oriented policies and strategies. To ensure adaptability to new trends or changes in delivery patterns, continual evaluation and improvement of e-grocery fulfillment policies are emphasized.

本文利用基于客户和电子杂货的数据来研究电子杂货订单履行策略。通过利用历史购买数据、产品流行趋势和交付模式，分配策略可以优化绩效指标，例如填充率、碳排放量和每笔订单成本。该研究旨在进行敏感性分析，以确定影响这些绩效指标的关键驱动因素。结果强调，与未通过数据通知的策略相比，利用上述数据指标优化的履行策略表现出卓越的性能。这些发现强调了集成数据驱动模型在电子杂货订单履行中的关键作用。根据结果​​，考虑到邻近性和产品可用性的杂货分配政策有望同时改善多个绩效指标。该研究建议电子杂货公司利用客户数据来设计和优化以交付为导向的政策和战略。为了确保适应交付模式的新趋势或变化，强调对电子杂货履行政策的持续评估和改进。