智能系统软件的定价模式:数据驱动升级视角  

Pricing smart enterprise software:A data-driven upgrading perspective

作　　者：戴军 窦一凡[1] 黄丽华[1] 张睿诚 DAI Jun;DOU Yi-fan;HUANG Li-hua;ZHANG Rui-cheng(School of Management,Fudan University,Shanghai 200433,China;College of Sciences,Shanghai University,Shanghai 200444,China)

机构地区：[1]复旦大学管理学院,上海200433 [2]上海大学理学院,上海200444

出　　处：《管理科学学报》2022年第9期1-14,共14页Journal of Management Sciences in China

基　　金：国家自然科学基金资助项目(71672042,91746302,71822201)。

摘　　要：智能系统软件是指能够结合客户的使用数据来实现持续升级的软件系统.当前5G技术和工业物联网的飞速发展,大量的系统数据被采集并被用于智能系统软件的算法训练,从而帮助实现质量提升.在这一背景下,以SAP和Oracle等为代表的智能系统软件提供商积极从传统的自有模式转型为订阅服务模式.本文采用博弈论建模的研究方法来探索了这一定价模式变化趋势的合理性.相比文献,本文提出用户数据积累越多则软件升级幅度越大.研究发现,在每一类定价模式下,外部的市场特征和内部的升级能力都对于智能系统软件的价格优化有着直接影响.进一步地,本文发现订阅和自有模式都有可能为最优,这取决于软件提供商是否需要对于未来价格进行承诺.这一结论对于智能互联时代的系统软件提供商在商业模式设计方面提供了理论依据.Smart enterprise software is the software that leverages client data for upgrades and quality improvement. With the rapid development of ongoing 5G and industrial Internet-of-Things technologies, massive data are being accumulated for training the algorithms embedded in the enterprise software. In the meantime, enterprise software providers such as SAP and Oracle have actively transformed the pricing scheme from traditional perpetual license model to subscription-based model. A game-theoretical approach is explored to explore the rationale of this transition. The paper draws from real-world practices to make the assumption that the software quality improvement is more significant when more data are collected for algorithm training, which appears novel in the literature. It is found that both pricing models can be optimal, depending on both external market characteristics and internal upgrades. Furthermore, our results show that the optimal pricing scheme also depends on whether the software provider is able to commit to future prices. Our findings provide a theoretical basis for the business model design of enterprise software providers in the smart and connected era.

关 键 词：系统软件 人工智能 信息产品定价 商业模式选择 

分 类 号：F272.3[经济管理—企业管理]