联系我们

论文查询结果	返回搜索

论文编号：	15456
作者编号：	2120233707
上传时间：	2025/6/11 21:22:05
中文题目：	不确定需求下零售商库存对收益影响的研究——基于消费者观察性学习的分析
英文题目：	Research on the Impact of Retailers’ Inventory on Profit under Uncertain Demand——An Analysis Based on Consumer Observational Learning
指导老师：	王永进
中文关键字：	观察性学习；随机需求；累积需求；库存管理；分数布朗运动
英文关键字：	Observational learning; Stochastic demand; Cumulative demand; Inventory management; Fractional Brownian motion
中文摘要：	需求不确定性是影响零售商收益与供应链稳定的关键因素。随着电子商务的普及，消费者越来越倾向于通过观察他人购买行为进行决策，这一过程被称为观察性学习。观察性学习不仅影响个体购买行为，还会引发需求的动态变化，进而影响零售商的库存管理。然而，现有研究在需求建模方面对观察性学习效应的考虑较为有限，且将其与库存管理相结合的研究较少，尤其在随机需求环境下，对需求的累积效应缺乏系统性的分析。因此，本文基于观察性学习理论构建需求模型，并在收益最大化框架下，分别探讨短销售周期与长销售周期产品的最优订货策略，同时通过数值分析验证模型的有效性并进一步讨论关键参数的影响。 本文采用几何分数布朗运动描述需求动态，该模型能够刻画需求的长记忆性和正相关性，反映消费者偏好高销量商品的行为，并通过Hurst指数（1/2 < H < 1）衡量观察性学习效应的强度。在短销售周期情境下，本文推导了零售商收益与库存水平的关系，并基于收益最大化原则求解最优订货量，进一步分析产品参数及需求动态参数对库存决策的影响。对于长销售周期产品，本文考虑需求的累积效应，采用加权几何平均法估计长期需求，并在此基础上优化订货策略。在此基础上，通过数值分析验证了模型的稳健性，并分析不同市场环境下零售商的收益及库存决策对产品和需求参数的敏感性。 研究结果表明，需求的增长趋势会推动最优订货量上升，而当需求波动性超过特定阈值时，风险加剧，零售商应调整订货量以降低库存成本。在波动率较低的情况下，较强的观察性学习效应（较高的H值）通常会提升最优订货量；但当波动率较高时，需求的不确定性增加，零售商可能采取更谨慎的订货策略。此外，在长销售周期环境下，若后期需求的权重增加，零售商倾向于提高订货量，但在高波动率和强观察性学习效应的情况下，这一趋势可能有所减弱。本研究不仅丰富了观察性学习与库存管理的交叉研究，也为零售商在电子商务环境下优化库存决策提供了新的理论支撑和实践指导。
英文摘要：	Demand uncertainty is a critical factor affecting retailers’ profitability and supply chain stability. With the widespread adoption of e-commerce, consumers increasingly rely on observing others’ purchasing behaviors to inform their own decisions, a process known as observational learning. This phenomenon not only influences individual purchasing choices but also drives dynamic demand fluctuations, thereby impacting retailers’ inventory management. However, existing research has paid limited attention to incorporating observational learning effects into demand modeling, and few studies have explored its integration with inventory management. In particular, the cumulative effect of demand in stochastic environments has not been systematically analyzed. To address this gap, this study develops a demand model based on observational learning theory and investigates optimal ordering strategies for both short-cycle and long-cycle products within a profit-maximization framework. Additionally, numerical analysis is conducted to validate the model’s effectiveness and examine the influence of key parameters. This study employs geometric fractional Brownian motion to model demand dynamics, capturing both the long-memory property and positive autocorrelation of demand. The model reflects consumers’ tendency to favor high-sales products, with the strength of observational learning measured by the Hurst index (1/2 < H < 1). Under the short sales cycle scenario, this study derives the relationship between retailers’ profitability and inventory levels and determines the optimal order quantity based on the principle of profit maximization. Furthermore, it examines the impact of product attributes and demand dynamics on inventory decisions. For long-cycle products, the study accounts for the cumulative effect of demand by employing a weighted geometric mean approach to estimate long-term demand and subsequently optimize ordering strategies. The robustness of the model is validated through numerical analysis, which also explores the sensitivity of retailers’ profitability and inventory decisions to key demand and product parameters under varying market conditions. The results indicate that an increasing demand trend drives a higher optimal order quantity. However, when demand volatility surpasses a critical threshold, heightened risk necessitates inventory adjustments to mitigate potential losses. Under conditions of low volatility, stronger observational learning effects (as indicated by a higher H value) generally lead to increased optimal order quantities. In contrast, when volatility is high, the greater uncertainty in demand prompts retailers to adopt more conservative ordering strategies. Moreover, in long-cycle scenarios, an increased weighting on later-period demand encourages retailers to raise order quantities, except in cases characterized by high volatility and strong observational learning effects. This study contributes to the intersection of observational learning and inventory management research and provides theoretical and practical insights for retailers seeking to optimize inventory decisions in e-commerce environments.
查看全文：	预览  下载(下载需要进行登录)