AI and the World's Hard Problems · Economics & social sciences · arXiv · 2026-10-19
On the possibilities and limitations of LLM-based multi-agent systems for multi-criteria decision-making in healthcare operations management
Multi-criteria models for healthcare operations need weights for trade-offs among patient waiting time, provider workload, and operating cost, and these weights are hard to quantify. The authors propose a framework that assesses LLM-agent-generated preferences as optimization inputs, testing explicit ratings in outpatient scheduling and pairwise choices among emergency department configurations. Deliberation among agents erases role differences that simple averaging preserves, pairwise choices depend largely on presentation order, and robust optimization often selects different schedules than average weights do.