Design and Evaluation of a Constraint-Based Energy Saving and Scheduling Recommender System

Development of low-cost and inter-operable home sensing products in recent years has motivated the development of consumerlevel energy and home monitoring software solutions to exploit these new streams of data available to end-users. In particular, this opens up the home energy space as an area of high potential for the use of consumerlevel energy optimisation with home-owners actively engaged with data about their energy use behaviour. We describe the development of a tablet-based home energy cost saving and appliance scheduling system which calculates behaviour change suggestions that save occupants on their energy bills while avoiding disruption to their regular routines. This system uses a Constraint Satisfaction Problem Solver to compute savings based on real-world sensor data, and to generate revised schedules in a user-friendly format, operating within a limited computing environment and achieving fast computation times.