Location: Rm 2018, 1880 Pratt Drive

Title: Predicting “heavyweight” decisions: An agent based model for predicting residential solar PV purchasing & a software platform to conduct online social experiments

In this talk I will go over two projects being conducted at Sandia National labs.

Increasing use of clean, renewable energy can help reduce U.S. carbon emissions and oil dependence.  Given that residential energy use is 21% of energy consumption in the US, there has been increased interest in understanding, and predicting, residential consumer behavior towards purchasing solar photovoltaic (PV) panels. Better prediction of consumer behavior can reduce customer acquisition "soft costs" which will reduce solar PV prices. 

I will provide an overview of efforts at Sandia National Labs to develop a computational model of residential consumer behavior. In particular, we have created an agent-based model, trained on economic, demographic and household data from San Diego County, to predict solar PV purchasing trends. 

“Heavyweight” decisions – those that have large economic impact and long-term consequences – are often influenced by many factors, including economic, social, and cognitive. Increasingly, decision making is influenced by online information gathering and influence as well.

To understand decision making, and how online interaction influences this process, we are developing the “Controlled, Large Online Social Experimentation” (CLOSE) platform -- to conduct online, longitudinal, social experiments. I will describe the problems we faced in development of the system and the solutions we came up with to address these issues.

Speaker Bio:
Dr. Kiran Lakkaraju is a Senior Member of the Technical Staff at Sandia National Laboratories, New Mexico in the Cognitive Science and Applications group. He has extensive experience in agent based modeling and simulation of social systems. His interests focus on developing agent models that capture “heavyweight” actions, like adopting solar PV, and understanding human behavior online (both observationally and experimentally) for validation of social models. He is currently building a social experimentation platform (the Controlled, Large, Online Social Experimentation platform). In addition, he has led an effort to study player behavior in Massively Multiplayer Online Games at Sandia National Labs. He is Co-PI of a SunShot funded grant to develop data driven models of residential PV adoption.