Ramesh Johari, Management Science and Engineering and, by courtesy, Computer Science and Electrical Engineering
Demand response (DR) programs offer incentives to individual users to encourage them to shift their energy use to off-peak times. Effective deployment of DR programs requires experimentation by both users and the utility: the utility experiments to learn the preferences of individual users, and tailor the program design to maximize participation; while users themselves may not know whether they want to participate in a DR program without trying it out first. These dual learning processes interact with each other to shape the performance of the DR program. In this project, the researchers will reconsider the design of efficient DR programs from first principles, taking into account learning and experimentation on the part of users and the utility. The goal is to provide design guidance for utilities on how to run, and learn from, demand response programs across a heterogeneous pool of users.
Publications and media:
"Cournot Games with Uncertainty: Coalitions, Competition, and Efficiency" Association for Computing Machinery X (2015): 1-20.