ENHANCING POLICY ANALYSIS IN SYSTEM DYNAMICS USING MULTIPLE OBJECTIVE OPTIMISATION

System Dynamics is a well-established methodology for modeling complex systems, and involves building models where key variables are represented as stocks and flows. A key challenge in system dynamics is to determine the best combination of parameters that will maximize or minimize a certain payoff. Optimisation has been widely used in System Dyanmics, and many of the commercially available tools contain optimisation modules. However, to date, none of these systems have encorporated multiple objective optimisation, an approach which allows decision makers to trade-off between many conflicting objectives. This paper presents the first system dynamics modeling environment that encorporates multiple objective optimisation, and demonstrates it application through the widely known beer-game simulation model.