Designing zones for school choice systems requires eliciting complex preferences and balancing multiple stakeholder objectives. We propose a stakeholder-in-the-loop framework for school zone generation that iterates between using optimization to generate zone boundaries for given preferences, and allowing stakeholders to participate and learn their preferences by reacting to zones. To facilitate stakeholder participation, we use LLMs to translate between natural language preferences and optimization constraints. To enable real-time use of our framework, we develop faster computational approaches for the multi-school zoning problem using both math programming and sampling-based methods. Our framework produces zones with substantially improved diversity and proximity metrics relative to existing benchmarks, while also generating individual-level preference representations that can be aggregated using standard social choice methods. Our approach has supported preliminary discussions about school zone boundaries in San Francisco and is generalizable to other redistricting contexts.

Dr. Irene Lo is an assistant professor in the Department of Management Science & Engineering at Stanford University. Her research sits at the intersection of operations research, computer science theory, and economic theory. She designs markets and allocation systems that improve both efficiency and equity, with applications in education, the environment, and the developing world. She leads a Stanford Impact Lab on Equitable Access to Education, co-launched the ACM Conference series on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO), and is a William T. Grant Scholar.