供应链中序列决策优化方法的研究

发布时间：2018-01-15 01:17

本文关键词：供应链中序列决策优化方法的研究　出处：《河北大学》2016年硕士论文　论文类型：学位论文

【摘要】：供应链中序列决策过程是指当决策者在供应过程中制定一项决策后又遇到一些新情况,并需要根据新情况再进行下一项决策的过程.由于信息不断地被揭露,后一事件在前一事件的基础之上发生,决策者不但需要权衡当前的情况,还需要在事件尚未发生时,权衡未来事件的利弊得失.在任何一个决策过程中,都存在诸多不确定性因素,如果决策者忽略不断更新的信息,而盲目制定决策的话,不但得不到最高的收益,还会给决策者带来巨大的风险,序列决策的研究目的就是尽可能降低不确定性因素带来的风险.所以利用市场反馈的信息,制定合适的序列决策,就成为了论文研究的重点.序列决策中的不确定性可能来自于客观因素,也可能来自于主观因素,所以论文的工作从随机不确定性和模糊不确定性两个方面来研究供应链中的序列决策问题.在随机环境中,主要针对三类典型的序列决策问题进行分析,分别建立了双层随机期望值模型.采用先处理下层规划问题,根据下层问题的最优解求解上层规划问题的方法从而得到最优决策的解析解.接着,通过数值实验说明了建模思想与求解方法的可行性.在模糊环境中,首先研究在供应链中同时做决策的情况,得到了原模型的等价子模型.其次讨论两种序列决策问题,用等价值指标度量公司的收益,建立了双层模糊等价值模型,并通过优化软件对模型进行了求解.最后同样通过数值实验说明了建模思想与求解方法的可行性.文章的主要工作可以概括为以下几个方面：(1)针对随机环境中的价格决策先于数量决策的序列决策问题,建立双层期望值模型,给出下层规划问题的通解；(2)在供应和需求服从均匀分布的条件下,得到了最优数量决策的解析解和最优价格决策所满足的方程式；(3)针对两类随机环境中数量决策先于价格决策的序列决策问题,分别建立随机期望值模型；(4)对于随机环境中价格决策先于数量决策问题,进行了数值实验及参数分析；(5)对于模糊环境中的同时决策、数量决策先于价格决策和价格决策先于数量决策问题分别建立等价值模型,并通过软件对模型进行了求解,进行数值实验说明了求解方法的可行性.
[Abstract]:The sequential decision-making process in the supply chain refers to the process in which the decision maker makes a decision in the supply process and then meets some new situations and needs to make the next decision according to the new situation. The latter event occurs on the basis of the previous event. The decision maker not only needs to weigh the current situation, but also needs to weigh the pros and cons of the future event when the event has not happened. There are many uncertain factors, if the decision-makers ignore the constantly updated information, and make decisions blindly, not only can not get the highest return, but also will bring huge risks to the decision-makers. The purpose of sequential decision making is to reduce the risk of uncertainty as much as possible. The uncertainty in sequential decision-making may come from objective or subjective factors. So the work of this paper is to study the sequence decision problem in supply chain from two aspects of stochastic uncertainty and fuzzy uncertainty. In the random environment, we mainly analyze three typical sequence decision making problems. The two-level stochastic expected value model is established, and the analytic solution of the optimal decision is obtained by first dealing with the lower level programming problem and solving the upper level programming problem according to the optimal solution of the lower level problem. The feasibility of modeling idea and solution method is demonstrated by numerical experiments. In fuzzy environment, the case of simultaneous decision making in supply chain is studied first. The equivalent sub-model of the original model is obtained. Secondly, two kinds of sequential decision problems are discussed. The profit of the company is measured with equal value index, and the two-layer fuzzy equal value model is established. Finally, the feasibility of modeling idea and solving method is illustrated by numerical experiments. The main work of this paper can be summarized as follows: 1). In order to solve the sequential decision problem of price decision before quantity decision in random environment. The two-level expected value model is established, and the general solution of the lower level programming problem is given. 2) under the condition of uniform distribution of supply and demand, the analytical solution of optimal quantity decision and the equation of optimal price decision are obtained. (3) to solve the sequential decision problem in which the quantity decision is prior to the price decision in two kinds of random environment, the stochastic expectation value model is established separately. 4) numerical experiment and parameter analysis are carried out for price decision before quantitative decision in random environment. 5) for simultaneous decision in fuzzy environment, the value model of quantitative decision is established before price decision and price decision before quantity decision, and the model is solved by software. Numerical experiments show the feasibility of the method.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：O225

【相似文献】