客户关系动态优化模型与实证研究

运筹与管理 ›› 2014, Vol. 23 ›› Issue (4): 117-123.

客户关系动态优化模型与实证研究

马少辉, 谭慧, 代逸生

江苏科技大学经济管理学院,江苏镇江 212003

收稿日期:2013-05-01 出版日期:2014-04-25
作者简介:马少辉(1972-),男,河北承德人,博士,副教授,主要研究领域为决策支持系统;谭慧(1990-),女,江苏宿迁人,硕士生,主要研究领域为客户关系管理;代逸生(1964-),男,重庆人,博士,教授,主要研究领域为管理信息系统。
基金资助:
国家自然科学基金资助项目(71171100,71273121)

Customer Relationship Dynamic Optimization Model and Empirical Study

MA Shao-Hui, TAN Hui, DAI Yi-Sheng

School of Economy and Management, Jiangsu University of Science and Technology, Zhenjiang 213002, China

Received:2013-05-01 Online:2014-04-25

摘要/Abstract

摘要： 提出了客户关系与营销活动的动态交互模型,以长期收益最大化为目标,优化企业的营销活动。模型假设客户关系可离散为几个层级状态,并设客户关系所处状态受营销活动的影响而动态的变化,服从马尔可夫决策过程。客户关系状态所处层级不可直接观测,但其与客户购买水平有概率相关关系。提出模型参数估计的最大似然估计方法。以国内某企业的客户关系管理数据为例,说明了模型变量的定义方法,通过客户交互历史数据估计模型参数,并对客户管理策略进行优化。结果表明,最优策略管理下期望提升客户价值61％～82％。

关键词: 客户关系管理, 动态优化, 马尔可夫过程, 客户终生价值

Abstract: In this paper, we propose a dynamic model to capture the interactivity between marketing actions and customer relationship. The model uses long term rewards as the objective to optimize marketing actions. We assume that customer relationship can be discretized to several levels, and the relationship levels customer stays in is affected by marketing actions. This dynamic process is assumed as a Markov process. Which levels customer stays in is not observable directly, but it is probabilistically correlated with customer purchasing levels. To obtain the model parameters, we further propose a maximum likelihood estimation approach. In empirical study, we apply the model to a dataset from a company in China. We show how to define the model variables, how to estimate the model parameters, and finally how to optimize the customer relationship management policies. The results show that we can increase expected customer lifetime value by 61％～82％ when using the optimized customer management policy.

Key words: customer relationship management, dynamic optimization, markov process, customer lifetime value

中图分类号:

F272.3

马少辉, 谭慧, 代逸生. 客户关系动态优化模型与实证研究[J]. 运筹与管理, 2014, 23(4): 117-123.

MA Shao-Hui, TAN Hui, DAI Yi-Sheng. Customer Relationship Dynamic Optimization Model and Empirical Study[J]. Operations Research and Management Science, 2014, 23(4): 117-123.

参考文献

[1] Fournier S, Yao J L. Reviving brand Loyalty: a reconceptualization within the framework of consumer-brand relationships[J]. International Journal of Research in Marketing, 1997, 14(6): 451-472.
[2] Fournier S. Consumers and their brands: developing relationship theory in consumer research[J]. Journal of Consumer Research, 1998, 24(4): 343-373.
[3] Bolton R, Lemon K N, Verhoef P C. The theoretical underpinnings of customer assets management: a framework and propositions for future research[J]. Journal of the Academy of Marketing Science, 2004, 32(3): 271 -292.
[4] Bolton R N. A dynamic model of the duration of the customer's relationship with a continuous service provider: the role of satisfaction[J]. Marketing Science, 1998, 17 (1): 45-65.
[5] Thomas J. The importance of linking customer acquisition to customer retention[J]. Journal of Marketing Research, 2001, 39 (2): 262-268.
[6] Dwyer F R. Customer lifetime valuation to support marketing decision making[J]. Journal of Direct Marketing, 1997, 11(4): 6-13.
[7] Aaker J, Susan F, Brasel A S. When good brands do bad[J]. Journal of Consumer Research, 2004, 31 (3): 1-18.
[8] Rossi M, Robert E M, Allenby G M. The value of purchase history data in target marketing[J]. Marketing Science, 1996, 15 (4): 321-340.
[9] G nül F F, Kim D, Meng Z S. Mailing smarter to catalog customers[J]. Journal of Interactive Marketing, 2000, 14 (2): 2-16.
[10] G nül F F, Hofstede F. How to compute optimal catalog mailing decisions, marketing science[J]. 2006, 25(1): 65-74.
[11] Chen Y X, Iyer G. Consumer addressibility and customized pricing[J]. Marketing Science, 2002, 21(2): 197-208.
[12] Feinberg F M, Krishna A, Zhang Z J. Do we care what others get? a behaviorist approach to targeted promotions[J]. Journal of Marketing Research, 2002, 39(4): 277-291.
[13] Bitran G R, Mondschein S V. Mailing decisions in the catalog sales industry[J]. Management science, 1996, 42(9): 1364-1381.
[14] G nül F F, Shi M E. Optimal mailing of catalogs: a new methodology using estimable structural dynamic programming models[J]. Management Science, 1998, 44 (9): 1249-1262.
[15] Elsner R, Krafft M, Huchzermeier A. The 2003 ISMS practice prize winner-optimizing rhenania's direct marketing business through dynamic multilevel modeling(DMLM) in a multicatalog-brand environment[J]. Marketing Science, 2004, 23(2): 192-206.
[16] Simester D I, Sun P, Tsitsiklis J N. Dynamic catalog mailing policies[J]. Management science, 2006, 52(5): 683-696.
[17] Fader P S, Hardie B G S, Lee K L. Counting your customers the easy way: an alternative to the pareto/NBD model[J]. Marketing Science, 2005, 24(2): 275-284.
[18] Cassandra A, Littman M, Zhang N. Incremental pruning: a simple, fast, exact method for partially observable Markov decision processes[C]. Proceedings of the 13th Annual Conference on Uncertainty in Artificial Intelligence, San Francisco: Morgan Kaufmann Publishers, 1997. 54-61.