Utilizing large language models to simulate parking search

This study explores the capability of Large Language Models (LLMs) to mimic general economic decision-making patterns relevant to parking choices. We conducted two series of experiments. In the first, a LLM acted as a survey participant, choosing between parking options based on attributes and sociodemographic profiles, which we term “Personas.” The model’s choices aligned with findings from parking behavior literature, consistently selecting lower-cost parking options while minimizing walking distances and search times, demonstrating utility-maximizing decision-making. Additionally, its responses revealed a willingness to trade off between price and time savings, reflecting diminishing marginal utility and an increased willingness to pay among higher-income Personas. The model also exhibited trends not well-documented, such as older Personas’ heightened sensitivity to reduced walking distances. In the second series of experiments, the model engaged in a dynamic decision-making scenario that emulated a serious game, continuously making decisions under uncertainty regarding expected cruising time. The model exhibited myopia and risk attitudes similar to those of humans in comparable experiments, with heightened risk aversion in older Personas. Our experiments highlight the potential of LLMs to emulate hypothetical scenarios across diverse populations, geographical settings, and policy and technological innovations. We outline a pathway toward leveraging LLMs as tools to support and complement human participants in the design of behavioral studies.