【作者】 李江峰；

【导师】 张顺明；

【作者基本信息】 厦门大学 ， 应用数学， 2012， 博士

【摘要】 有效管理库存一直以来都是运筹学和管理学研究主要议题,也是整个供应链有效运行的重要环节。面对快速变化的随机需求,每个制造企业如何决定自己的最优产量?每个零售企业如何决定自己的最优库存量?一些重要的服务业如何设计最优的运作空间?这些都是世界上各大企业所面对的共同难题。而作为对这类问题研究的理论基石,经典的报贩模型由于其结构简单,并能够刻画大量现实中商业和生产运作模式,从而具有广泛的应用前景,因此长期以来都是学者和管理者强烈关注的焦点。近30年来,在随机环境下,基于期望效用理论的传统经济和金融理论由于无法解释现实世界中一些异象,因而受到越来越多学者的质疑,一些学者通过把人类的行为因素纳入到决策模型中发展了行为经济学和行为金融学。目前,这一领域是经济学和金融学研究最为热点的主题.管理决策同所有经济和金融决策一样,都是同人类行为有着密不可分的关系。因而,近10年来,也有少数学者试图把行为经济学和金融学的理论引入到管理学的研究之中,提出了行为运作管理的思想。相对于行为经济学和行为金融学的发展,行为运作管理(BOM)仍然是一个刚刚起步领域.此外,在有关行为报贩模型的研究中,现有文献大多数考虑报贩的风险厌恶、损失厌恶、缺货厌恶、有限理性和过度自信等行为偏差。最近,大量的实验证据表明,情绪在人们的决策过程中具有重要的影响,来自心理学的研究表明悲观总是和人类的一切消极情绪密切相关,而乐观总是一切积极情绪密不可分。因此,悲观和乐观偏差在人类决策过程中扮演着重要角色.所以,无论从学术还是从实践的角度来考虑,研究悲观和乐观偏差对于报贩决策行为有怎样的影响是一个很有价值的研究问题。在这样的动机下,本文从悲观和乐观偏差的视角下对经典的报贩模型进行讨论,进行了以下三方面的研究：1.研究了悲观乐观偏差对于报贩订购行为的影响,并同期望效用和风险中性框架下的结果进行了比较。证明了悲观预期效用报贩的订购量小于风险厌恶的期望效用和风险中性报贩的订购量,这说明概率相关的风险态度对供应链的效率造成更大的损失,对于乐观的报贩在适当条件下有相反的结果,因此悲观和乐观偏差加剧了供应链的不稳定性。通过构造例子说明Schweitzer和Cachon(2000)实验的“中心聚集效应”在AU框架下能够给予合理的解释。不同于现有的比较静态文献,考察了随机需求期望和方差的改变对于悲观和乐观报贩决策行为的影响。2.研究了价格相关需求环境的报贩问题。在乘和加需求模型下讨论了悲观和乐观报贩的定价订购决策.在乘需求模型和适当的条件下,证明了悲观预期效用报贩制定的价格高于相应的对偶效用报贩和风险中性报贩,而订购量小于后面两者；乐观报贩的的最优价格小于对应的对偶效用报贩和风险中性报贩,而订购量大于后面两者。在加需求模型下,悲观预期效用报贩的最优价格小于对应的对偶效用和风险中性报贩的最优价格,但无法确定三者之间最优订购量之间的关系:乐观报贩有相反的定价行为.3.研究了悲观和乐观偏差对于资本限制报贩决策问题的影响。考察了悲观和乐观偏差对于报贩订购和借贷行为的影响,以及悲观和乐观偏差对于银行利润和设定贷款上限决策的影响.证明了悲观(乐观)的报贩和银行的均衡订购量少于(大于)二者都是风险中性情形下的对应数量.而悲观(乐观)的银行设定的贷款上限总是小于(大于)风险中性银行所设定贷款上限,从而降低(增加)了报贩由于借贷而可能破产的概率.更多还原

【Abstract】 How to manage inventory effectively has been being the focus of research in operation and management science and the important step in effective operation of the whole supply chain. In face of fast changing stochastic demand, how does each manufacturer decides its optimal production? how does every retailer makes its inventory decision? how do those service enterprises design their optimal op-eration space? All these problems are common obstacles encountered by various types of enterprises around the world. As the cornerstone of theory about this kinds of issues, the classic newsvendor model has been a highlight to the scholars and managers in the fields for a long time, due to its elegant structure and excel-lent ability to capture a great number of situations of businesses and manufacture operation in practice and its extensive application prospects.Over the past three decades, traditional economic and financial theories based on the expected utility have been challenged by many scholars, owing to their failure to explain some anomalies in reality. Some researchers devel-oped behavioral economics and behavioral finance by incorporating human be-ing’s behavioral biases into decision making process, which are the hottest topics in modern economics and finance. Management decision-making are very much tied to the human’s behaviors, as do economic and financial decision. Therefore, a few scholars have made attempt to draw on some tools of behavioral economics and finance to investigate operation management in recent ten years, and ad-vanced theories of behavioral operation management(BOM). Compared with the development of behavioral economics and finance, BOM is just a beginning. Fur-thermore, most of the extant works in the BOM field just consider few behavioral bias such as risk aversion, loss aversion, stock-out aversion, boundary rationality and overconfidence. A great number of empirical evidences show that human be-ing’s emotions have a significant role in their decision-making processes recently. Some researches on psychology have demonstrated that was always related to all of the human’s passive feeling, and optimism bias mingled with any positive emo-tions. Consequently, the pessimism and optimism biases will have great impact on people’s decision-making. Thus, it is obviously a valuable research subject, not only academically but also practically, to study what roles do the pessimism and optimism biases play in the newsvendor problem.Motivated by these considerations, this paper reexamines the classic newsven-dor problem from the perspective of the pessimism and optimism biases. The main work comprises three aspects as following.1. We have studied the influence of the pessimism and the optimism biases on newsvendor’s ordering decision, and draw some comparisons among these re-sults derived from the frameworks that based on the expected utility and risk neutral news vendor problem. Our main results show that optimal order quan-tities of a pessimistic anticipated utility newsvendor are always less than those of risk aversion EU newsvendor and risk neutral newsvendor, which illustrates that newsvendor’s probability-dependent attitudes toward risk will result in a greater loss in the efficiencyof the supply chain. Under some adaptive conditions, the opposite conclusion can be drawn for optimistic newsvendor. Therefore, the pessimism and the optimism biases aggravate the instability of the supply chain system. In addition, the put-to-center effect originated from Schweitzer and Ca-chon’s experiment (2000) can also be explained reasonably under AU framework by constructing a specific example. Finally, distinguished from those published literatures of comparative static analysis in newsvendor problem, we examine ef-fects of changes in the mean and variance of the stochastic demand on pessimistic and optimistic newsvendor’s decision-making behaviors.2. We have revised the newsvendor problem under the price-dependent set-ting and studied pessimistic and optimistic newsvendor’s joint pricing and order-ing decision-making under multiplication and addition demand model. In mul-tiplication demand model and some adaptive conditions, we demonstrate that an anticipated utility newsvendor’s optimal price, with pessimistic bias, is higher than that of corresponding dual utility newsvendor and risk neutral newsvendor, while optimal order quantities are less than associated those.As far as optimistic newsvendor is concerned, we can obtain opposite conclusion. When it comes to addition demand model, an anticipated utility newsvendor’s optimal price, with pessimistic bias, is lower than that of corresponding dual utility newsven-dor and risk neutral newsvendor under addition demand model, the situation is completely contradict to the optimism. Unfortunately, there do not exist unam-biguous conclusions about the relationship of order quantities among different decision making models.3. We have examined the effects of the pessimistic and optimistic bias on decision behavior of capital-constrained newsvendor. The influences of the pes-simistic and optimistic bias are investigated on decisions of newsvondor’s order-ing and borrowing loan, and on decisions of bank’s setting loan upper-limits and bank’s profits. We have proved that the equilibrium order quantities between pessimistic (optimistic) newsvendor and bank are less (more) than those when both of them are risk neutral decision makers. Furthermore, the loan upper-limits set by bank with pessimistic (optimistic) bias are less (more) than that set by risk neutral bank, as a result, the newsvendor’s bankruptcy probability, resulted from borrowing loan from bank, will be lower (higher).更多还原