摘要：随着科学技术更新换代的不断加速，顾客越来越追求得到个人心理上的满足，产品市场需求愈发难以掌控。而在多变的市场环境下，除了产品市场需求本身是一个随机变量外，其分布的参数也是随机的，由此零售商需要对顾客需求做出快速响应。两次采购策略的研究就是随着快速响应理论的发展而产生的，这种采购策略更能适应快速变动的市场。

本文选用易逝品作为研究对象，在多阶段需求的背景下，研究双重不确定信息下的双源采购策略。不同于一次性需求，多阶段需求考虑的是零售商一边销售、一边采购的情况，这就要求零售商在进行本销售周期的采购决策时，既要考虑上一销售周期的订货批量以及期末剩余产品的数量，又要考虑在销售过程中收集到的实际需求信号。本文采用逆推法建立了基于贝叶斯需求预测更新的双源采购策略模型，并进行了最优解的证明和求解分析。为了比较贝叶斯信息更新的优劣性，本文对无需求预测更新时零售商的采购策略也进行了类似的建模分析，通过对比分析，发现需求预测更新能够有效帮助零售商把握市场需求，获得更高的利润。此外，本文还发现，产品市场需求均值的不确定性程度对需求预测更新的价值有着正向影响，但产品市场需求本身的不确定性程度对预测更新的价值却是负面影响；同时，只有当第一次的采购成本与库存成本的和与第二次的采购成本相差不大时，需求预测更新才有价值。

关键词 双重不确定 信息双采购源 两周期需 求贝叶斯信息更新

毕业论文外文摘要

Title   Dual Sourcing Policy under Bi-level Uncertain Demand Information              

Abstract：With the rapid development of science and technology, the customer’s demand is becoming increasingly personalized and persified, and the market demand is showing increasing uncertainty. Nowadays, under the unpredictable market environment, not only the market demand is a random variable, but also the parameter is stochastic. It forces the retailer to make quick response to the customer’s demand. Two ordering opportunities policy emerges after the quick response theory, which can adjust the market variety better and more practical.

This paper studies the dual sourcing policy of multi-stage demand. Different from the one-time demand, multi-stage demand must be taken by considering pre-period inventory level and actual demand. In this paper, a dual sourcing policy model based on Bayesian demand information updating is established by using the inverse method, and the optimal ordering quantity is found. In order to compare the impact of Bayesian information updating, this paper also analyses the purchasing policy without Bayesian information updating, and finds the optimal ordering quantity. Then, through analyzing a relevant example proves that Bayesian information updating can help retailers grasp the changes of market and gain more profits. This paper also found that the higher uncertainty of the parameter, the value of the demand forecast updating is higher. However, the uncertainty of the market demand is a negative impact on the value of the demand forecast updating. In addition, when the purchasing cost in the second period is not much higher than the sum of the purchasing cost and inventory cost in the first period, the demand forecast updating is valuable.

Keywords  Bi-level uncertain demand information; Two ordering opportunities; Multi-stage demand; Bayesian information updating

目次

1绪论 1

1.1研究背景和意义 1

1.2相关文献综述 2

1.2.1两次或多次订购机会，一次性需求的相关文献 2

1.2.2两次或多次订购机会、两阶段或多阶段需求的相关文献 4

1.3研究内容 6

2相关理论概述 8