Electrifying light-duty passenger transport for CO2 emissions reduction: A stochastic-robust input–output linear programming model

Electrification of light-duty passenger vehicles has been widely considered as an important strategy for decarbonizing the transport sector. This study proposes a novel input–output linear programming model with stochastic-robust optimization methodology to identify the optimal pathways to reduce direct and embodied emissions from light-duty passenger transport under technology cost and emission intensity uncertainties. The model introduces detailed transport technologies into a well-known economic input–output model so that the carbon dioxide (CO2) emissions embodied in production inputs of vehicles in the entire economic system can be fully accounted for. We apply the model to a case study in China, which shows that tightening cumulative CO2 emissions from light-duty passenger transport by 30% of the base case requires large deployment of electric vehicles, especially plug-in hybrids and battery electric vehicles. The emission reductions mainly come from vehicle operation-related emissions. Nevertheless, the total reduction in operation-related emissions is offset by the increase in capital-related CO2 emissions by 13%–18%, depending on the variability in future emission intensity of electricity. Finally, we compare the data uncertainty handling performance of stochastic-robust model with a stochastic programming model and two deterministic optimization models under optimistic and pessimistic parameter settings. The comparison shows that the optimal technology portfolio generated from the proposed model is not only reliable in achieving the predefined emissions target, but is also able to reduce the total investment and operation costs in the light-duty passenger transport sector under data uncertainty.