NSF Award
2309668
This collaborative research project will advance national equity and welfare by developing diverse fairness measures to support policy-making for emerging infrastructure systems with disadvantaged populations involved. Emerging infrastructure systems, such as shared e-scooter systems, are increasingly impacting existing civil infrastructures like public transit, but it remains unclear how to systematically and reliably quantify fairness in such systems due to high operational dynamics. This award will contribute to a deeper and more comprehensive understanding of the role of fairness and the associated tradeoffs that arise when considering fairness in the design and operations of these emerging infrastructure systems. The research findings will be integrated into undergraduate and graduate courses to educate students on the challenges of measuring and promoting fairness in civil infrastructures amid the impact of emerging technologies. Moreover, the project team will prepare educational material based on the findings for the Successful Transition and Enhanced Preparation for Undergraduates Program at the University of Florida and the High School Virtual Summer Research Program at the Stevens Institute of Technology to attract high school graduates, particularly women and minorities.<br/><br/>This project aims to create a novel framework for designing diverse fairness measures that can effectively evaluate fairness in emerging infrastructure systems where conflicting interests from stakeholders and inherent complex dynamics pose unique challenges. The framework will use multi-level utility functions to qualify the positive and negative impacts of an emerging system on various population groups. These utility evaluations will be mapped to a fairness value by a social welfare function that combines multiple fairness perspectives. Additionally, the project will develop new optimization models for addressing operational dynamics to preserve fairness and effective exact solution methods, such as a customized Benders decomposition method based on theoretical analysis and a specialized generation and selection procedure for high-quality Benders cuts. A data-driven approach will be employed to construct an uncertainty set to compute a robust estimation of the price of fairness for any given fairness level under a specific fairness measure. A case study on shared micro-mobility systems and extensive experiments using publicly available real-world data will be conducted to analyze the tradeoffs and provide insights for policy-making.<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.
