An Integrated Approach using Data Mining and System Dynamics to Policy Design: Effects of Electric Vehicle Adoption on CO_2 Emissions in Singapore

摘要

This study aims to demonstrate the utility of System dynamics (SD) thinking and data mining techniques as a policy analysis method to help Singapore achieve its greenhouse gases (GHG) emission target as part of the Paris climate agreement. We have developed a system dynamics model called Singapore electric vehicle and transportation (SET) and analyzed the long-term impacts of various emission reduction strategies. Data mining techniques were integrated into SD modelling, to create a more evidence-based decision-making framework as opposed to the prevalent intuitive modelling approach and ad hoc estimation of variables. In this study, data mining was utilized to aid in parameter fitting as well as the formulation of the model. We discovered that the current policies put in place to encourage electric vehicle (EV) adoption are insufficient for Singapore to electrify 50% of its vehicle population by the year 2050. Despite not achieving the electric vehicle target, the projected CO_2 emission still manages to be significantly lower than the year 2005 business as usual scenario, mainly because of switching to a cleaner fossil fuel for power generation as well as curbing the growth of vehicle population through the Certificate of Entitlement (COE). The results highlighted the usefulness of SD modelling not just in policy analysis, but also helping stakeholders to better understand the dynamics complexity of a system.