Abstract

With increasing availability of consumer data and rapid advancement and application of technologies, online retailers are gaining better knowledge of customers’ shopping behavior and preferences. Thus more and more retailers are providing personalized product assortment to better match the needs of customers and generate more sales. In this article, we study a two-stage revenue management model. In the first stage, the retailer decides non-personalized price discounts for each product. In the second stage (upon the arrival of customers), the retailer offers a personalized assortment to each type of customer. Based on this assortment, the customer then makes a purchase decision according to the Multinomial logit choice model. We employ a robust approach for the joint discounts and personalized assortment optimization problem in order to handle data uncertainty from estimating customer preferences and distribution of different customer segments. We analyze the structural properties of the problems and propose efficient computational methods to solve the problems with/without a cardinality constraint on the assortment. In certain cases, our algorithm converges at a superlinear rate. When there is a cardinality constraint on the assortment, we find that the retailer should offer greater discounts as the constraint becomes more restrictive. We also discuss the value of our robust solution and the extension of when the customer discount sensitivity function is also uncertain. Finally, our extensive numerical study shows that the solutions under the robust approach perform very well when compared to the one assuming accurate information, and are robust when there is uncertainty.

摘要

随着消费者数据可用性的提高以及技术的快速发展和应用，在线零售商正越来越了解客户的购物行为和偏好。因此，越来越多的零售商提供个性化的产品分类，以更好地满足客户的需求并产生更多的销售额。在本文中，我们研究了一个两阶段的收入管理模型。在第一阶段，零售商决定每种产品的非个性化价格折扣。在第二阶段（客户到达时），零售商为每种类型的客户提供个性化的商品。根据此分类，客户然后根据多项式logit选择模型做出购买决定。我们针对联合折扣和个性化分类优化问题采用了一种可靠的方法，以便通过估计客户偏好和不同客户群的分布来处理数据不确定性。我们分析问题的结构特性，并提出有效的计算方法来解决有/无基数约束的问题。在某些情况下，我们的算法以超线性速率收敛。当分类存在基数约束时，我们发现随着约束变得越来越严格，零售商应提供更大的折扣。我们还将讨论可靠解决方案的价值，以及不确定客户折扣敏感性函数时的扩展。最后，