Construct an Artificial Population with Urban and Rural Population Differences Considered - To Support Long-Term Care System Evaluation by Agent-Based Simulation.

Bottom-up simulation approaches such as agent-based simulation are attracting attention in Chinese Long-Term Care (LTC) studies. To enable the deployment of agent-based modelling approaches in evaluating LTC systems, a computational base which entails individual and household details is necessary. In this work, we propose a method to construct such an artificial population with the urban and rural population differences considered. Given the situation that nationwide disaggregated records on health issues of households and individuals are not available for the Chinese case, we first propose a method to generate such records based on the first wave of China Health and Retirement Longitudinal Study. Drawn upon above records containing health-related attributes, we then propose a revised combinatorial optimization method to construct an artificial population mirroring a Chinese city including both urban and rural residents. It will be among the first methods in constructing such artificial populations to support agent-based simulation approaches in LTC studies.