Sensitivity Analysis of a large-scale system dynamics immigration model

摘要

Sensitivity Analysis (SA) is studying the impact of input changes (nature and magnitude) on outputs. Specifically, analysts and decision makers are typically interested in understanding how much output variation is produced, by varying the inputs of a system. The system in this case is a system dynamics computer model and the objective is to perform SA on the Immigration Simulation model that has been developed to support the Secure Border Initiative (SBI). The ultimate objective is to enhance the validity of the Immigration Simulation Model (and possibly other future models) and minimize the risk taken in making decisions based on models' output. In this paper, a framework is provided for the SA process that was used in analyzing the SBI Immigration Simulation model. This framework consists of the following processes: (1) Input Variables Analysis, (2) Output Variables Analysis, (3) Experimental Design, (4) Model(s) Execution, and (5) Statistical Analysis. The main focus of this paper, however, is the experimental design used to carry out the sensitivity analysis.