Improving the representation of consumers' choice in transport within energy system models

Abstract

Transport is a fundamental driver of economic development and a key supporter of welfare. Nonetheless, it is responsible for approximately 28% of global final energy use, 23% of the global energyrelated CO2 emissions and is regarded as the most complicated sector to decarbonise. In order to reduce the carbon intensity and energy consumption of the transportation sector, both technological and behavioural changes are required. Energy system models are a valuable tool for long-term energy planning. Decision makers have been using these models for more than three decades to explore alternative pathways towards greenhouse-gas (GHG) emissions free energy systems and to test the potential impact of policy measures. Bottom-up (BU) energy-economy-environment-engineering (E4) models in particular can provide a detailed technological representation of the energy system. However, these models are generally weak at representing human behaviour, despite it is a fundamental aspect of decision making in the transportation sector. This PhD thesis fills this gap by proposing several methodologies that improve the representation of consumers’ choice in passenger transport within energy system models, thus paving the way for the possibility to carry out novel transport analyses and to consider a wider array of decarbonisation policies. The first part of this thesis reviews the current scientific literature regarding integrated energy and transport models. It highlights the failing in representing consumers’ decision making and identifies modal choice and vehicle choice as key behavioral features to be integrated in the modeling framework to overcome the existing limitation. Following these two findings, this thesis presents the methodologies developed within the scope of this PhD research to incorporate modal choice and vehicle choice in TIMES-DK, the TIMES model of the Danish energy system. The methodologies developed can be classified in two categories: those that extend the structure of the TIMES model to accommodate novel transport-specific variables and those that link the TIMES model with an external transport model. Thanks to the broad spectrum of approaches developed and tested within the scope of this PhD research, this thesis ultimately aims at acting as a guide for fellow researchers interested in including behavioural realism of transport users’ choice in E4 models. The thesis describes how traditional limits in the representation of behaviour within BU optimization E4 models can be addressed with the different approaches developed. Then, it compares the various methodologies with respect to the capability to capture key behavioural features, to answer different policy questions and to the modeling efforts required to reproduce the models. The novel methodologies proposed inaugurate the possibility to perform more comprehensive analyses of decarbonisation pathways, which include both the behavioural and technological dimension. The results of the PhD study indicate that modal shift potentially has a positive contribution to the decarbonisation of the energy system, helping to reach carbon-neutral energy system in Denmark in 2050 at faster pace and with lower cumulative emissions. The analyses carried out within the scope of this PhD research find that car transport is likely to maintain the highest modal share also in the future, suggesting that modal shift should be accompanied by the electrification of the car sector to comply with the Danish environmental targets and overarching climate targets. The analyses are intended to inform Danish policy makers dealing with energy and transport planning on the beneficial contribution of modal shift and of the electrification of the car stock to reduce GHG emissions. The Nordic experience and the findings of the modeling analyses are used to give policy recommendations on the measures that the authorities should put in practice to encourage modal shift away from car to more sustainable modes of transport and to promote the deployment of electric cars. Finally, this PhD thesis discusses future research to address the remaining gaps concerning the representation of consumers’ choice within BU optimization E4 models and suggests interesting energy and transport analyses that should be performed using the novel models hereby proposed.