基于供应商易化零售商调拨的供应链优化与双向协调机制——理论模型与行为实验检验

As business competition becomes increasingly fierce, the problem of stock-out and leftover in inventory control goes hand in hand, and becomes prevalent in reality. Inventory sharing through transshipment serves as an important tool performs well in solving this problem. However, significant hurdles usually exist in successfully implementing the transshipment programs due to the lack of incentive mechanism etc., in the process of transshipment. To cope with this problem, many leading suppliers try to participate in the transshipment and facilitate it. However, although the supplier plays an important role in the process of transshipment, the decisions the retailers make on their order and transshipment policy from their own perspective usually leads to lower supply chain performance. So, what policy should the supplier adopt to facilitate the transshipment program’s performance is the main problem in this project. This project studies Classical theory, behavioral test and behavioral decision by theoretical model and behavioral experiment, respectively. Firstly, the classical theory (complete rationality) is used to design a mechanism which about four kinds of supply chain optimization and bidirectional coordination mechanism (horizontal coordination within the retailers and vertical coordination between the supplier and retailer system). The mechanism will be used as a benchmark model for future research. Then the behavioral experiment will be used to estimate the benchmark model and the behavioral factors refined from test are used to describe the difference between the benchmark model and the experimental data. At last, the project will construct the behavioral decision model by incorporating the behavioral factors in the framework of the classical theory. In this project, not only the benchmark theoretical framework of supply chain optimization and bidirectional coordination mechanism has been established, but also broken the limitation of the classical theory by the introduction of behavioral factors, which makes the model have good precision. Theoretical method and decision reference for inventory management will also be presented systematically in this project. The benchmark theoretical framework of supply chain optimization and bidirectional coordination mechanism has been established, and by the introduction of behavioral factors which makes the model have good precision, the limitation of the classical theory has been broken.

市场竞争使产品缺货与滞销成为一种常态，调拨是解决该矛盾的有效手段。但由于零售商之间可能由于缺乏激励机制等因素导致调拨失败，因此很多供应商参与并且易化零售商之间调拨。尽管供应商在调拨中扮演重要角色，但零售商独立决策订货与调拨导致调拨效率较低。因此，供应商如何设计机制提高供应链绩效是本项目研究的问题。本项目采用理论模型和行为实验检验相结合的方法从经典理论(完全理性)模型、行为实验实证和行为决策模型三方面对该问题进行研究。首先采用经典理论设计出四种供应链优化与双向协调机制作为研究的基准模型；然后采用行为实验对基准模型进行检验，提炼出用以刻画基准模型与实验数据存在差异的行为因素；最后将行为因素纳入经典理论框架构建行为决策模型。该项目不但建立供应链优化与双向协调机制的基准理论框架，而且还引入行为因素，实现对经典理论局限性的突破，使行为决策模型具有更好的预测能力，从而为库存管理提供理论方法和决策依据。

随着市场竞争的加剧与技术更新速度的加快，产品的生命周期越来越短，市场需求由此变得越来越不确定，这使得企业缺货与滞销现象成为了一种常态。这种供需的不匹配性，常常给企业的生产运作带来严重损失，调拨是解决该矛盾的有效手段。因此，如何设计机制提高供应链绩效是本项目研究的问题。首先，本项目对基于调拨价格和讨价还价能力分配机制下供应链优化与协调的经典理论进行了研究。在调拨价格分配机制下，研究发现调拨价格在满足一定的条件下，不管是否存在顾客的转移行为，其订货纳什均衡均总是存在；在纳什讨价还价框架下，仅当调拨价格为调拨量的隐函数时，才存在协调价格使得双方可以协调，并且证明了在联盟内部谈判顺序对利润分配并不会产生影响。其次，基于上述模型设计库存行为实验，对改进的新模型进行验证。通过VFP软件构建实验平台，并且依据控制自变量信息的完整性设计三个对照实验，构造基于QRE的零售商订货行为有限理性模型的似然函数，将实验数据与模型进行拟合。结果显示，纳什均衡模型对实验数据的解释能力较为有限。实验数据与纳什均衡模型的对数似然值低于随机反应均衡模型。再次，为了研究存在调拨行为时零售商订货数量的系统偏差，引入行为博弈论中的随机最优反应均衡，并通过引入干扰博弈概念构造干扰订货博弈的随机最优反应均衡，从而证明了随机最优均衡的存在性。引入完美均衡概念并构造出完美均衡为找到一条从任意订货量通向完美均衡订货量的光滑路径，由此可以通过合理的算法并根据证明的光滑路径计算出完美均衡点。最终，通过本项目建立了调拨行为的经典理论(完全理性)模型、行为实验实证和行为决策模型三方面的理论体系、方法体系和技术体系。