Optimizing Service Operations via LLM-Powered Multi-Agent Simulation
This work addresses the problem of optimizing service operations for researchers and practitioners by providing a novel simulation tool, though it is incremental in combining existing LLM and optimization techniques.
The paper tackles the challenge of modeling human behavior in service systems by introducing an LLM-powered multi-agent simulation framework for optimizing service operations, achieving superior performance over benchmarks in a sustainable supply chain application and uncovering strong designs in a contest design case study.
Service system performance depends on how participants respond to design choices, but modeling these responses is hard due to the complexity of human behavior. We introduce an LLM-powered multi-agent simulation (LLM-MAS) framework for optimizing service operations. We pose the problem as stochastic optimization with decision-dependent uncertainty: design choices are embedded in prompts and shape the distribution of outcomes from interacting LLM-powered agents. By embedding key numerical information in prompts and extracting it from LLM-generated text, we model this uncertainty as a controlled Markov chain. We develop an on-trajectory learning algorithm that, on a single simulation run, simultaneously constructs zeroth-order gradient estimates and updates design parameters to optimize steady-state performance. We also incorporate variance reduction techniques. In a sustainable supply chain application, our method outperforms benchmarks, including blackbox optimization and using LLMs as numerical solvers or as role-playing system designers. A case study on optimal contest design with real behavioral data shows that LLM-MAS is both as a cost-effective evaluator of known designs and an exploratory tool that can uncover strong designs overlooked by traditional approaches.