非平稳需求多周期自适应库存控制研究

发布时间：2018-06-30 07:51

本文选题：自适应库存管理 + 指数平滑法　；参考：《青岛大学》2014年硕士论文

【摘要】：库存在生产系统乃至供应链中扮演者重要角色。高效的库存管理可以在降低成本的前提下保证生产销售的正常运行,为产品的流通提供可靠保障；反之,库存管理不当将会占用大量流动资金并且产生高额的库存管理费用。传统的EOQ模型及其扩展模型有效解决了需求确定或者需求分布信息已知时的库存管理问题。随着社会的发展,顾客消费水平的提高,产品生命周期不断缩短,更新换代速度加快,许多产品的需求呈现非平稳特征,而且需求分布信息难以获取,在这种背景下,如何科学有效地管理库存问题成为库存领域研究的焦点之一。自适应库存控制方法是解决这类问题的有效方法。制定有效的自适应库存控制策略具有重要意义。所谓自适应库存控制,就是将自适应控制理论与传统库存控制相结合。具体方法为：在面对多周期库存控制问题时且客户需求信息不确定的情况下,每到库存盘点时刻都会检测库存控制的效果,并根据库存效果与库存控制目标(比如：顾客服务水平、库存成本等)的差距自动调整参数,以确保库存控制能及时作出调整,进而使得需求出现变化时可以减小不确定性对库存的影响,进而达到预设的库存控制目标。未来顾客需求的预测在库存控制研究中往往起着至关重要的作用,学者通常运用指数平滑法作为对未来需求预测的模型。然而,指数平滑法中系数在选取时往往采取的利用人工经验或是反复测试来完成,并没有一种可靠而明确的方法,使得预测值与实际值偏差较大,造成指数平滑法的预测精度偏低,阻碍了指数平滑法的广泛应用。而将自适应控制思想与传统指数平滑法结合产生的自适应指数平滑算法能够在需求波动时较传统指数预测法更准确地预测顾客需求,从而提高了预测精度,提高了库存控制的效果。本文目标是探讨非平稳随机需求环境多周期随机库存自适应库存控制方法,主要内容包括：在探讨自适应库存控制方法的同时,结合自适应指数预测算法并以此建立一个满足预设服务水平的库存模型。前三章主要介绍自适应库存控制基本内容以及库存控制的基本理论,同时介绍指数平滑法及其改进算法；四、五两章建立了自适应库存控制模型。第六章用计算机对顾客需求进行仿真,并通过动态地调整平滑因子和安全因子,使库存控制能够满足预设的客户服务水平。仿真结果表明,本文提出的库存控制模型在非平稳需求环境下能有效稳定在预设顾客服务水平。
[Abstract]:Inventory plays an important role in the production system and even in the supply chain. Efficient inventory management can ensure the normal operation of production and sales under the premise of reducing costs, and provide reliable guarantee for the circulation of products. Conversely, improper inventory management will occupy a large number of liquidity and lead to high inventory management costs. The traditional EOQ model and its extended model effectively solve the inventory management problem when the requirement is determined or the requirement distribution information is known. With the development of society, the improvement of customer consumption level, the product life cycle is shortened, the speed of renewal is accelerated, the demand of many products presents the non-stationary characteristic, and the information of demand distribution is difficult to obtain, under this background, How to manage inventory scientifically and effectively becomes one of the focuses in inventory field. Adaptive inventory control is an effective method to solve this kind of problem. It is of great significance to establish an effective adaptive inventory control strategy. Adaptive inventory control is a combination of adaptive control theory and traditional inventory control. The specific methods are as follows: in the face of multi-cycle inventory control problem and customer demand information is uncertain, every inventory count time will check the effectiveness of inventory control, And automatically adjust the parameters according to the gap between inventory effect and inventory control objectives (such as customer service level, inventory cost, etc.) to ensure that inventory control can be adjusted in time. When the demand changes, it can reduce the impact of uncertainty on inventory, and then achieve the pre-set inventory control goal. The prediction of future customer demand often plays an important role in the research of inventory control. Scholars usually use exponential smoothing method as the model of forecasting future demand. However, in the exponential smoothing method, the coefficients are often chosen by artificial experience or repeated testing, and there is no reliable and definite method to make the predicted value deviate greatly from the actual value. The prediction accuracy of exponential smoothing method is on the low side, which hinders the wide application of exponential smoothing method. The adaptive exponential smoothing algorithm combined with the traditional exponential smoothing method can predict the customer demand more accurately than the traditional exponential forecasting method when the demand fluctuates, thus improving the prediction accuracy. Improve the effect of inventory control. The purpose of this paper is to discuss the adaptive inventory control method of multi-period stochastic inventory in the non-stationary stochastic demand environment. The main contents include: while discussing the adaptive inventory control method, Combined with adaptive exponential prediction algorithm, an inventory model satisfying the preset service level is established. The first three chapters mainly introduce the basic contents of adaptive inventory control and the basic theory of inventory control. At the same time, the exponential smoothing method and its improved algorithm are introduced. In chapter 6, the customer demand is simulated by computer, and by adjusting the smoothing factor and security factor dynamically, the inventory control can meet the preset customer service level. The simulation results show that the inventory control model proposed in this paper can effectively stabilize the customer service level in the non-stationary demand environment.
【学位授予单位】：青岛大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：O227

【参考文献】