Behavioral Study of Personalized Automated Demand Response in Workplaces

This study conducts a quasi-randomized controlled trial in a workplace with personalized lighting control and investigates the impact of automated demand responses on employees. To meet this objective, we set up four treatment groups: opt-in performance incentive, opt-in fixed incentive, opt-out performance incentive, and opt-out fixed incentive. Then, we estimate the average treatment effects (intent-to-treat and treatment-on-treated) of demand response. The results reveal a significant reduction in electricity consumption during peak hours for all four treatment groups.