当前位置: X-MOL 学术Transp. Res. Part E Logist. Transp. Rev. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Dynamic inventory and pricing control of a perishable product with multiple shelf life phases
Transportation Research Part E: Logistics and Transportation Review ( IF 8.3 ) Pub Date : 2025-01-16 , DOI: 10.1016/j.tre.2025.103960
Mohammad S. Moshtagh, Yun Zhou, Manish Verma

This paper investigates a dynamic inventory-pricing system with perishable products of multiple freshness levels. The firm may set different prices for items of different freshness levels, and customers either balk or choose to buy the freshness level that maximizes their utility. We model this inventory-pricing problem as a Markov decision process, where the assortment dynamically changes based on the freshness levels of the available items. Using the concept of anti-multimodularity, we characterize the structure of the optimal policy. Specifically, we show that the optimal production policy has a state-dependent threshold-based structure. The production decisions are more sensitive to the inventory of fresher items than less fresh ones. Moreover, the optimal price of a freshness level is nonincreasing in the inventory of items of any freshness level, and it is more sensitive to those of a closer freshness level. The structural properties enable us to devise three novel heuristic policies with good performance. We further extend the model by considering donations and a system with multiple freshness phases. Our research suggests that freshness-dependent pricing and dynamic pricing are two substitutable strategies, while freshness-dependent pricing and donation are strategic complements. The results further imply that the firm can benefit from high variability in freshness among items under dynamic pricing, but such variability may lead to a significant loss when single, static pricing is used. The results of our heuristic policies show that considering inventory and pricing decisions as a parametrized function of the inventory state leads to nearly optimal solutions.

中文翻译:


具有多个保质期阶段的易腐产品的动态库存和定价控制



本文研究了一个动态库存定价系统,该系统具有多个新鲜度级别的易腐产品。该公司可能会为不同新鲜度的商品设定不同的价格,客户要么犹豫不决,要么选择购买能最大限度地发挥其效用的新鲜度。我们将这个库存定价问题建模为马尔可夫决策过程,其中分类根据可用商品的新鲜度水平动态变化。使用反多模块的概念,我们描述了最优策略的结构。具体来说,我们表明最优生产策略具有基于状态的基于阈值的结构。生产决策对新鲜物品的库存比新鲜物品的库存更敏感。此外,新鲜度水平的最佳价格在任何新鲜度水平的商品库存中都不会增加,并且对新鲜度水平更接近的商品更敏感。结构属性使我们能够设计出三个性能良好的新颖的启发式策略。我们通过考虑捐赠和具有多个新鲜阶段的系统来进一步扩展模型。我们的研究表明,新鲜度依赖定价和动态定价是两种可替代策略,而新鲜度依赖定价和捐赠是战略互补。结果进一步表明,在动态定价下,公司可以从商品新鲜度的高度可变性中受益,但当使用单一的静态定价时,这种可变性可能会导致重大损失。我们的启发式策略的结果表明,将库存和定价决策视为库存状态的参数化函数会导致近乎最优的解决方案。
更新日期:2025-01-16
down
wechat
bug