Interpretable and Actionable Vehicular Greenhouse Gas Emission Prediction at Road link-level

To help systematically lower anthropogenic Greenhouse gas (GHG) emissions, accurate and precise GHG emission prediction models have become a key focus of the climate research. The appeal is that the predictive models will inform policymakers, and hopefully, in turn, they will bring about systematic changes. Since the transportation sector is constantly among the top GHG emission contributors, especially in populated urban areas, substantial effort has been going into building more accurate and informative GHG prediction models to help create more sustainable urban environments. In this work, we seek to establish a predictive framework of GHG emissions at the urban road segment or link level of transportation networks. The key theme of the framework centers around model interpretability and actionability for high-level decision-makers using econometric Discrete Choice Modelling (DCM). We illustrate that DCM is capable of predicting link-level GHG emission levels on urban road networks in a parsimonious and effective manner. Our results show up to 85.4% prediction accuracy in the DCM models' performances. We also argue that since the goal of most GHG emission prediction models focuses on involving high-level decision-makers to make changes and curb emissions, the DCM-based GHG emission prediction framework is the most suitable framework.