NSF Award
2245506
Large-scale infrastructure systems (e.g. gas, power, and water networks) often cross local, regional, and national jurisdiction boundaries. When this is the case, each jurisdiction oversees a certain part of the entire infrastructure system and can elect to pursue its own socio-economic objectives using different policy instruments. Such a jurisdictional divide naturally creates a fertile environment for regulatory competition. This project will develop mathematical models that study implications of this competition on the optimal infrastructure design. As an example, it will focus on electric power networks that are currently undergoing a rapid transition toward smart grids. If successful, this project will provide tools for infrastructure operators and planners (e.g. power utilities and state/federal regulators), as well as citizens, to evaluate the effects of different regulatory structures on achieving policy objectives. Project outcomes will be conveyed by means of direct outreach and via professional workshops, conferences, and peer-reviewed publications. Furthermore, broader impacts include training of future STEM researchers at graduate and undergraduate levels via project mentorship, independent research studies, and research-informed improvements to existing courses, as well as early-stage research opportunities for local high-school students from underrepresented groups in STEM (via NYU's Applied Research Innovations in Science and Engineering program).<br/><br/>This project will develop formal mathematical models for optimizing policy decisions for large-scale, hierarchical, network-constrained infrastructure systems. This model will endogenously optimize policy decisions that are currently considered as exogenous parameters in infrastructure planning tasks and allow for studying different forms of regulatory competition between multiple jurisdictions. To represent this competition and account for all aspects of infrastructure operating and planning tasks, the project will leverage the multi-leader-common-follower game-theoretic framework. This framework is notoriously known to be computationally demanding and, therefore, this project will develop scalable algorithmic solutions, based on the Progressive Hedging algorithm, to implement the framework for realistically large networks. The algorithmic component of this project will investigate solution quality, e.g. its optimality and uniqueness properties. Broader scientific impacts are anticipated in models and algorithms to solve multi-stage optimization problems under uncertainty. This project will directly benefit researchers and developers working in the area of operations research and smart grids/cities, with extensions relevant to public policy and network economics communities.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
