A meta-modeling framework in public health emergency management

Bin Chen,Peng Zhang

DOI: https://doi.org/10.1177/0037549717733064

2017-10-04

SIMULATION

Abstract:Epidemic transmission is a common type of public health emergency that is difficult to forecast and often causes substantial harm. Artificial societal models provide a novel approach to the study of public health problems. However, public health emergency management (PHEM) always involves multi-disciplinary and multi-hierarchical models that complicate the work of modeling. Models are also made more complex by the consideration of new requirements and interactions. Therefore, we propose a domain-specific methodology to guide the modeling process in PHEM. By analyzing domain characteristics and modeling requirements, a meta-modeling framework can be constructed, containing the basic elements with which to construct an artificial society to study epidemic transmission. In this paper, the designs of meta-models are discussed in detail, and domain models are implemented by code generation, which enables the support of large-scale, agent-based computational experiments on the KD-ACP platform. Case studies of Ebola are outlined, emergency scenarios are reconstructed based on pre-designed meta-models, and “scenario-response” experiments are presented. This study provides a valuable framework and methodology with which to study complex social problems in PHEM. The proposed method has been verified effectively and efficiently.