Policy simulation will enable the creation of a “virtual laboratory” for studying think tank problems. With the full collaboration of experts from multiple fields and the support of computer technology, policy simulation uses multidisciplinary knowledge as its core to translate real-world problems into the digital world. This supports experts, scholars, and policymakers in conducting scenario analysis and policy evaluation in a “virtual laboratory”, and optimizing solutions for think tank problems. This chapter will analyze the relationship between think tank problem research and policy simulation, analyze the process of constructing policy simulations for think tank problems based on the DIIS process convergence method and MIPS logic layer method, and discuss typical policy simulation methods to provide a reference for policy simulation in think tank problem research.