Smart Meter Data-Driven Customizing Price Design for Retailers

Abstract

Designing customizing prices is an effective way to promote consumer interactions and increase the customer stickiness for retailers. Fueled by the increased availability of high-quality smart meter data, this paper proposes a novel data-driven approach for incentive-compatible customizing time-of-use (ToU) price design based on massive historical smart meter data. Consumers' ability to choose freely and consumers' willingness are fully respected in this framework. The Stackelberg relationship between the profit-maximizing retailer (leader) and the strategic consumers (followers) in an incentive-compatible market is modeled as a bilevel optimization problem. Smart meter data are used to estimate consumer satisfaction and predict consumer behaviors and preferences. Load profile clustering is also implemented to cluster consumers with similar preferences. The bilevel problem is integrated and reformulated as a single mixed-integer nonlinear programming (MINLP) problem and then simplified to a mixed-integer linear programming (MILP) problem. To validate the proposed model, the smart meter dataset from the Commission for Energy Regulation (CER) in Ireland is adopted to better illustrate the whole process.