On the surplus accuracy of data-driven energy quantification methods in the residential sector

Abstract Increasing trust in energy performance certificates (EPCs) and drawing meaningful conclusions requires a robust accurate determination of building (BEP). However, existing by law prescribed engineering methods, relying on physical principles, are under debate for being error-prone practice ultimately inaccurate. Research has heralded data-driven mostly machine learning algorithms, to be promising alternatives: various studies compare methods with clear advantage terms prediction accuracy BEP. While previous only investigated the BEP, it yet remains unclear which reasons cause–effect relationships lead surplus methods. In this study, we develop discuss theory how data collection, type auditor, quantification method, its relate one another. First, introduce quantifying BEP method-agnostically investigate influence several design parameters, such as expertise auditor issuing EPC, our theory. Second, evaluate literature. We find that positively relationships, compensating deficits due auditors’ lack expertise, leading high accuracy. provide recommendations future research enable informed use