In this study, we introduce an integrated framework for managing the complex interdependence between urban infrastructures and the socioeconomic environment within which it evolves, in pursuit of sustainable and environmentally cleaner urban living. The framework addresses the nature of individual preferences for more sustainable urban infrastructures, and how we can use this knowledge to improve urban form in ways that reduce environmental impacts. Using metropolitan Atlanta as a case study, we developed a survey that focuses on the preferences of Atlanta residents for low-impact development (LID) and transit-oriented development (TOD), with responses collected on the Mechanical Turk crowed-source platform. Using these responses we developed a latent-class residential community choice model for four distinctive classes of respondents that revealed heterogeneous preferences for community amenities. Next, we integrated the results of these individual choices into an agent-based market diffusion model, to predict land-use pattern, and to explore policies that drive greater adoption of more compact development. Finally, we used the results of this data collection and modelling to estimate the carbon emission reduction potentials from more compact development driven by LID and TOD. In the future, we will continuously refine the steps and address the issues including survey sample bias to make the framework more reliable and useful for sustainable urban infrastructure planning, design and implementation.