SCC-IRG Track1: Smart energy assistants for affordable housing communities

Purdue University, West Lafayette IN

Investigators

Abstract

Almost 25% of U.S. households experience high energy burdens, referring to the percentage of gross household income spent on energy costs. As a result, more than 20% fall behind on their energy bills and suffer from utility service disconnection. To empower affordable housing communities with long-term energy access, a drastically different paradigm is required—one driven by new Smart and Connected technology that achieves energy security and improved well-being to all US families in need. Our goal is to realize a new paradigm for smart energy assistants that monitor energy use and energy needs and create customized community-based interventions that have the largest potential to mitigate energy-related disparities experienced by low to moderate income households. To achieve this goal, we will bring together community residents and diverse stakeholders (state and regional housing authorities, utility service providers, municipal administrations, community action groups, housing developers/landlords) to co-design an interactive cross-platform for smart energy assistants that: 1) Collect and organize community sociotechnical data and create a community-shared knowledge platform; 2) Integrate a counterfactual engine developed using a theory-informed, hierarchical Bayesian approach that reveals causal links between socioeconomic disparities and system determinants; 3) Generate customized policies for interventions that are tailored to the unique needs of each community; 4) Leverage an AI layer empowered by Large Language Models (LLMs) to interface with residents and stakeholders and the developed tools and databases. Also, the AI layer facilitates effective communication among residents in need and enables stronger connections between community residents and stakeholders, lowering the barriers to residents understanding and accessing energy programs. Our sociotechnical research will have direct positive impacts in five pilot communities in Indiana with ultimate goal to extend these benefits to 4.5 million affordable housing units nationwide. Our foundational contributions of Smart and Connected Communities sociotechnical research will utilize AI powered by LLM as the basis for smart energy assistants that will integrate energy use and energy needs and create customized community-based interventions that have significant potential to reduce energy-related disparities experienced by low to moderate income households. Our research approach will impact diverse application domains, including energy, transportation, water, urban planning. This will also lead into ground-breaking insights on AI-supported decision-making pertaining to measures of accuracy, trust, and social influence across dimensions of the technical and social system. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.