Global sustainable energy and environmental policies have increased the need to understand how new energy innovations diffuse into the market. The transport sector is currently a major source of unsustainable energy use contributing ~20-25% global CO2 emissions. Although the potential benefits of alternative fuel vehicle (AFV) technologies to reduce CO2 emissions and fossil fuel dependency have been demonstrated, many uncertainties exist in their market diffusion. It is also not well understood how policy can influence rapid diffusion of AFVs. To transition to a more sustainable energy-transport system, we need to understand the market conditions and factors necessary for triggering widespread adoption of new energy innovations such as AFVs. Modelling the diffusion of innovations is one way to explain why some ideas and technologies spread through society successfully, while others do not. These diffusion processes are characterized by non-linear interactions between heterogeneous agents in complex networked systems. Diffusion theory has typically been applied to consumer durable goods but has found less application to new energy and environmental innovations. There is much scope for advanced diffusion methods to inform energy policy. This depends upon understanding how consumer behaviour and technologies interact and can influence each other over time. There is also need to understand the underlying mechanisms that influence adoption behaviour among heterogeneous agents. This thesis tackles the above issues using a combination of empirical data analysis, scenarios, and simulation modelling as follows: 1) We first develop the empirical basis for assessing innovation diffusion from a technology-behavioural perspective, where we explicitly account for interactions between consumer preferences and technological performance across different spatial and temporal scales; 2) Scenarios are then used to disaggregate consumer markets and analyze the technological and behavioural factors that might trigger large-scale adoption of AFVs; 3) We then case-analyze the UK transport sector and develop a model of the dynamics between how vehicle technologies and consumer preferences can change and influence the diffusion process; 4) Finally, we develop exploratory simulations to assess how social network effects can influence individual adoption behaviour; 5) We close with policy implications of our findings, contributions and limitations of the thesis, and possible avenues for taking the research forward.