基于收益管理的邮轮客舱分配与定价模型

基于收益管理的思想对邮轮客舱分配与定价问题进行了研究。结合邮轮运营中的个性化特点，例如消费者团体人员构成多样、较长的预售周期以及救生位和儿童看护人员的数量限制等。在不失一般性的前提下，建立了整数规划模型用以确定在预售周期内的不同预售阶段中各种客舱类型的待售数量及其价格，以达到使邮轮公司收益最大化的目的。实验分析表明，该模型在实际应用中是有效的且呈现出显著的年增长趋势，可明显提高邮轮公司的收益。此外，设计了一种基于韦伯分布的EM算法用以解决模型中涉及到的需求量的无约束估计问题。数值算例研究表明，该算法收敛速度快且无约束估计过程可靠有效。

A Model of Cruise Cabin Allocation and Pricing Based on Revenue Management

The world cruise tourism industry has been developing rapidly in the past 50 years. As a kind of leisure tourism with high fixed cost, how to make cruise companies obtain higher profits in limited voyages has attracted only a small number of scholars to study. Cruise industry has the same typical revenue management characteristics as hotel and aviation industry. Whether the cruise industry can achieve the same success as hotel and aviation industry by using revenue management theory to improve the income of cruise companies remains to be studied.Based on the idea of revenue management, the issue of cruise cabin allocation and pricing is studied. Combined with the personalized characteristics of the cruise operations, such as the diverse composition of consumer groups, a longer pre-sale cycle and the limitation of the number of life-saving seats and child care staffs.(1)Without loss of generality, an integer programming model was established to determine the quantity and price of various cabin types for sale in different pre-sale stages in the pre-sale cycle, in order to maximize the profit of the cruise company. (2) Because the traditional pre-sale system only records the actual pre-sale data, it does not consider the demand that has arrived but has not been met, that is, the part whose demand is greater than the number of cabin booking limits are subject toType-ΙCensoring. In order to obtain the deterministic demand data and apply it to the model proposed in this paper, an EM algorithm based on Weber distribution is designed to solve the unconstrained estimation of demand involved in the model. The accuracy and convergence of the algorithm are proved by numerical examples.(3) This paper compares the integer programming model and the traditional unified pricing model, and compares their ability to improve revenue. The two models use the same demand data, and the results show that the proposed integer programming model can obtain higher revenue and show a significant annual growth trend. In summary, the research in this paper can provide a scientific decision-making tool for the operation and management of cruise companies.