基于顾客选择模型的演唱会售票策略研究

发布时间：2018-02-09 17:53

本文关键词： 多项式罗吉特选择模型消费者偏好动态规划销售策略　出处：《西南财经大学》2013年硕士论文　论文类型：学位论文

【摘要】：目前,我国的演唱会市场正在飞速发展,演唱会的数量和收益都在不断提高,但是由于该行业的发展时间较短,在市场运作和服务方面还有一些不足,学者们大多针对演唱会本身的环境和定价等进行研究(崔寒,2002;陈瑜妍,2008,夏梦琦,2012)。本文将消费者的购买行为和销售策略结合起来,在一个垄断的单一演唱会市场的背景下,对基于消费者选择的动态收益进行模拟研究。下面介绍本文的研究思路,首选提出了研究背景和意义,描述并比较在经济学和市场研究领域中应用较多的两个基于消费者选择的模型。接下来根据演唱会的座位特点和消费者的不同偏好选择,决定演唱会门票的三个不同属性：票价高低,座位离舞台远近和座位的角度。并依据Logit模型量化出顾客对不同产品的选择概率。在进行演唱会门票销售模拟时,本文根据文献资料和作者的经验合理设定了一些参数,将演唱会的门票根据上述的门票属性分为五个等级,并依据动态规划的思想,将零售商的售票过程分为一个相关联的三阶段决策问题,对于每一阶段,不同类别的消费者到达比例不同,而零售商对五类门票有不同的组合策略。在收益管理的研究中,由于现实生活中变量的因素很多而且很复杂,要得到最优的解有一定的难度,作者在研究中受到了Qian Liu和Garrett van Ryzin (2008)的收益方程的启发来进行模拟。然而他们的动态规划模型由于将时间分得过细,而状态空间的维数过大,在大多数现实生活中是不可解的。再加上动态规划过程没有准确的公式或算法来适合所有的情况,都要具体问题具体分析,因此,本文提出了一个近似的决策问题,用近似的方法将收益函数变为可计算的形式,这在某种程度上降低了运算的复杂度,更容易解决问题。通过对现实生活中几种情况的模拟：一,三个阶段不同消费者到达比例的变化(分别是商务消费者先来,敏感消费者先来和消费者到达情况差不多的情形)；二,市场需求量的变化：三,产品种类的变化。在以上的情况中根据消费者对不同产品的选择概率,比较不同情况下相对较好的收益和产品组合,作者希望得到一个直观的认识,即要如何安排才能使门票尽可能的卖完,并让有能力买高价票的消费者尽可能的买到此类产品,从而提高演唱会的收益。最后,本文对提高演唱会门票收益管理提出了三个建议,分别是：提供折扣或优惠卷；将市场分的更细；不断更新收益管理的方案。本文研究发现基于消费者偏好的离散选择模型是一个具有经济效益的方式,对研究消费者的偏好和支付意愿有一定的价值,可以为演唱会供应商进行更好的市场细分,制定更有效的运营策略提供帮助。由于本文的模拟只是复杂的商业活动中一种或几种情况,考虑的参数不一定全面和精确,在对消费者的类别(效用)和数量上的模拟仍然有一些局限性。在今后的研究中,可以在一个合适的范围进行市场调研使数据更加接近真实。而针对模拟自身的复杂性,随着计算机技术的飞速进步,基于动态规划的收益管理问题也会得到不断的提高和完善,学者们通过严谨科学的运筹规划,将不断完善动态收益管理的性能。
[Abstract]:At present, China's market is the rapid development of the concert, the concert volume and revenue are constantly improving, but because the development time of the industry is relatively short, there are some shortcomings in the operation of the market and services, most of the scholars in the concert itself and environment Pricing Research (Cui Han, 2002; Chen Yuyan 2008, summer, Mengqi, 2012). This will be the purchase behavior of consumers and sales strategies together, will be in a single market singing a monopoly in the background, the simulation research of the dynamic earnings based on consumer choice.
The following describes the research ideas of this paper, first introduced the research background and significance, and describes the application more comparison in economics and market research in the field of two consumer choice model based on. Then according to the different characteristics and consumer preference seat concert choice, three different properties determine concert tickets: tickets or seats from the stage the distance and the angle of the seat. And on the basis of Logit model to quantify the probability of selection of customers of different products.
The concert ticket sales simulation, this paper based on the literature and the author's experience in setting some parameters, the tickets for the concert tickets according to the above mentioned characteristics are divided into five levels, and on the basis of the theory of dynamic programming, the ticket retailer is divided into an associated three stage decision-making problem, for in each stage, different categories of consumers at different proportion, and the retailer has a combination of different strategies for five kinds of tickets.
In the study of revenue management, because the variables in real life are many and very complex, to get the optimal solution has a certain degree of difficulty, the author is Qian Liu and Garrett van Ryzin in the study (2008) inspired the yield equation to simulate. However, their dynamic programming model because of the time allocated to however, the dimension of the state space is too large, in most real life is not solvable. Coupled with the dynamic programming process without accurate formula or algorithm suitable for all situations, have to analyze specific issues, therefore, this paper proposes an approximate decision problem with approximate methods for calculate the variable profit function form, which to some extent reduce the computational complexity, easier to solve the problem. Through the simulation of several real life situations: first, three different stages of consumer arrival ratio Change (are consumers first, consumers first and consumers to reach the sensitive situation about the situation); two, changes in market demand: three, changes in product variety. In these cases according to the probability of consumer choice of different products in different conditions, a relatively good income and product portfolio, the author hope to get an intuitive understanding, that is how to arrange the tickets sold out to make as much as possible, and to have the ability to buy expensive tickets to consumers as much as possible to buy such products, so as to improve the concert revenue. Finally, this article is to improve the ticket revenue management has put forward three suggestions are: to provide discount or coupon; the market is divided more detailed; constantly updated revenue management solutions.
This study found that the discrete choice model based on consumer preferences is a cost-effective way, have a certain value for the study of consumer preferences and willingness to pay for the concert, can the supplier better market segments, make the operation more effective strategies to help.
Because the simulation is complex business activities in one or more of the parameters, not necessarily comprehensive and accurate, in the consumer category (utility) and the number of simulation still has some limitations. In the future, can carry out market research in a suitable range of the data is more close to true.
According to the complexity of simulation, along with the rapid progress of computer technology, the problem of revenue management based on dynamic programming will continue to improve and improve. Scholars will continuously improve the performance of dynamic revenue management through rigorous scientific operation planning.

【学位授予单位】：西南财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F719;F224

【参考文献】