论文部分内容阅读
传统的收益管理多航班无约束估计方法假设顾客到达时的购买决策是一次性的,未能充分考虑策略型顾客的跨期替代行为.在仅能获取产品的历史可观察订购量、历史预售开放状态以及市场份额信息的情况下,基于顾客偏好排名列表建立了考虑顾客策略行为的非参数离散选择模型.针对历史预售数据的不完备性,采用EM算法对顾客到达率和非参数离散选择模型的概率质量函数进行联合估计,并提出了考虑历史顾客策略行为的“初始需求”无约束估计计算方法.使用数值算例说明了所提方法的可行性,通过与现有文献中已有方法比较,验证了所提多航班方法能够反映产品价格变化对顾客选择行为的影响,并能更加有效地避免需求预测对未来顾客“初始需求”的高估.
Traditional Revenue Management Multi-flight Unconstrained Estimation Method Suppose that the customer’s purchase decision at the time of arrival is one-off and fails to take full account of the intertemporal substitution behavior of strategic customers. Open state and market share information, a nonparametric discrete choice model considering customer strategy behavior is established based on the customer preference ranking list.According to the incompleteness of historical presale data, EM algorithm is used to analyze the customer arrival rate and nonparametric discrete choice The probability and mass function of the model are estimated jointly and the calculation method of unconstrained estimate of initial demand considering the historical customer strategy behavior is proposed.The numerical example is given to illustrate the feasibility of the proposed method.According to the existing literature Some methods are compared to verify that the proposed multi-flight method can reflect the impact of product price changes on customer choice behavior, and can more effectively avoid the overestimation of future demand for customers “initial demand ”.