论文部分内容阅读
相较于单舱位无约束估计方法,多舱位方法能有效避免收益管理系统对同航班舱位间需求的高估问题.同时,无约束需求数据的多分布假设更加符合收益管理实践中不同情况下的需求特征.为此,首先建立了基于正态、对数正态和伽玛分布的多舱位Spill模型;然后,使用数值算例说明了所提方法的可行性;最后,通过与单舱位方法比较,验证了所提多舱位Spill模型的有效性,表明了数据的离散程度是影响各分布假设下Spill模型无约束估计效果的主要因素.
Compared with the single-class unconstrained estimation method, the multi-class method can effectively avoid the problem of over-estimation of demand between the revenue management system and the same flight class space.Meanwhile, the multi-distribution hypothesis of unconstrained demand data is more in line with the different situations in the revenue management practice Demand characteristics.Therefore, a multi-cabin Spill model based on normal, logarithm normal and gamma distribution is established firstly. Then, numerical examples are given to illustrate the feasibility of the proposed method. Finally, by comparing with the single-class method , Which verifies the validity of the proposed spill model. It shows that the degree of data dispersion is the main factor that affects the effect of unconstrained Spill model under various distributional assumptions.