Collaborative Research: Strategic Planning of Internet Services in the Presence of User-Initiated Innovations: Implications for Network Neutrality

互联网提供的信息服务对社会至关重要，并产生巨大的经济效益。互联网爆炸性增长的一个重要因素是没有使用费，这在很大程度上是由于互联网上的历史服务-尽力服务。由于没有使用费，互联网成为新的用户创建的应用程序的温床。然而，随着网络的成熟，已经清楚的是，仅尽力服务不能满足用户的期望以及更新的、更复杂的应用的服务质量要求。这些新的要求使得有必要考虑基于使用费的托管服务，并保证服务性能。允许网络提供商提供这些服务是否会对尽力服务和互联网的特性产生负面影响，是当前网络中立性辩论的核心问题。本研究项目采用运筹学的方法来解决这一重要的政策问题。核心是网络提供商的利润最大化规划的模型和分析，其中用户发起的创新的影响起着核心作用，这是新的分析处理。虽然短期的利润激励可能会诱使提供商降低最佳服务的质量，以促使消费者支付使用费并使用托管服务，但这样做会抑制用户发起的应用程序的创建，而这些应用程序会产生未来的需求，从而产生新的利润机会。通过将后者的影响整合到规划模型中，本研究可以告知网络提供商最大努力服务对其长期繁荣的重要性。 调查人员还将分析监管机构如何创造一个政策环境，以利用这种影响，使私人提供商的利润激励与公共利益保持一致。如果这些目标得以实现，将有助于社会避免在托管服务上的艰难选择，并使互联网在扩大服务范围的同时受益于高质量的尽力而为服务。本研究项目将模拟和分析网络服务提供商的业务战略，并调查由此产生的与网络中立性相关的公共政策的影响。第一步将是为提供商开发一个多期规划模型，通过控制基础设施投资、带宽分配和服务定价来实现利润最大化。 与以前的方法不同，该模型具有一个反馈回路，在这个回路中，供应商的利润最大化决策与用户发起的创新相互作用。这种互动的强度取决于许多因素，如需求的价格弹性，用户对延迟的不容忍，以及积极的网络外部性。通过解决这个扩展的规划模型和研究最优的解决方案，依赖关系将被量化，并得出结论，关于用户发起的创新的存在如何影响网络提供商的激励，以保持在各种情况下的最佳努力服务的质量。然后，该模型和分析将分别应用于赞助数据服务和内容分发网络的两个案例研究。结果将用于实证验证和模型的改进。关于公共政策的讨论将侧重于加强提供从用户发起的创新中获得的最大努力服务的激励。治疗将在多供应商环境中进行。在这样的环境中，提供商可以从作为其他提供商的高质量尽力服务的结果的创新中受益，即使其自己的尽力服务的质量低于标准。为了解决由此产生的搭便车问题，将分析从过去网络系统的经验和其他国家采取的办法中得出的不同政策选择。

The Internet provides information services that are essential to society and generate great economic benefits. An important contributing factor for the explosive growth of the Internet has been the absence of usage fees, which has been made possible in large part by Best Effort Service, the historical service on the Internet. The absence of usage fees has made the Internet a breeding ground for new user-created applications. However, as the network has matured, it has become clear that Best Effort Service alone cannot meet users` expectations and also the Quality of Service requirements of newer, more sophisticated applications. These new requirements makes it necessary to consider usage fee-based Managed Services with guaranteed service performance. Whether allowing network providers to offer these services will negatively affect Best Effort Service and the character of the Internet are issues at the heart of the current Network Neutrality debate. This research project employs the methodology of Operations Research to address this important policy question. In the core are models and analysis of network providers' profit-maximizing planning in which the influence of user-initiated innovations plays a central role, which is new in analytic treatments. While short-term profit incentives may induce a provider to degrade Best Effort Service to push consumers to pay the usage fee and use the Managed Services, doing so suppresses the creation of user-initiated applications, which generate future demands, and thus new profit opportunities. By integrating the latter effect into the planning model, the study can inform network providers about the importance of Best Effort Service for their long-term prosperity. The investigators will also analyze how regulators can create a policy environment to leverage this effect to align private providers' profit incentive with public interest. If these goals are achieved, the research project will help society to avoid the difficult choice on Managed Services, and allow the Internet to benefit from an expanded scope of services while sustaining Best Effort Service at a high quality level.This project will model and analyze business strategies of network service providers and investigate resulting implications for public policies that are pertinent to Network Neutrality. The first step will be to develop a multi-period planning model for providers to maximize profit by controlling infrastructure investment, bandwidth allocation, and service pricing. In a departure from previous approaches, the model features a feedback loop in which the provider's profit-maximizing decisions interact with user-initiated innovations. The strength of such interactions depends on many factors, such as price elasticity of demand, users' intolerance to delay, and positive network externality. By solving this extended planning model and studying optimal solutions, the dependencies will be quantified and conclusions will be drawn regarding how the presence of user-initiated innovations affect network providers' incentive for preserving the quality of Best Effort Service in various scenarios. The model and analysis will then be applied to two case studies on Sponsored Data Service and Content Delivery Networks, respectively. The results will be used for empirical validation and refinements to the model. The discussion on public policies will focus on enhancing the incentive for offering Best Effort Service that is gained from user-initiated innovations. The treatment will be carried out in a multi-provider environment. In such an environment a provider can benefit from innovations that are the results of other providers' high-quality Best Effort Service, even though the quality of its own Best Effort Service is sub- standard. To counter the resulting free-rider problem, different policy alternatives derived from experiences with past network systems and approaches taken by other countries will be analyzed.