Answers are grounded in this paper and cite their sources.
Field implementation studies spanning six data-driven building operation and maintenance (DBOM) application domains (model-based predictive control (MPC), occupant-centric control (OCC), automated demand response (ADR), reinforcement learning control (RLC), fault detection and diagnostics (FDD), and virtual metering (VM)) were reviewed. The review aimed to make inter- and intra-application domain comparisons in terms of implementation processes and outcomes. The studies were classified by different levels of integration effort and sensing infrastructure complexity according to the quantity and kind of their inputs and outputs. An analysis is conducted on how energy-related success metrics (e.g., energy use, cost, demand) correlate with the integration effort and the complexity of the infrastructure. In general, MPC and ADR studies were found to require a higher level of integration effort than OCC studies; yet, OCC studies tend to require upgrades in the sensing infrastructure, unlike MPC and ADR studies. 95 % of those reporting post-implementation energy metrics achieved more than 10 % improvement in their energy-related objectives. The analysis results also highlight that a large fraction of the FDD and VM studies did not follow a measurement and verification procedure to quantify energy-related performance improvements upon field implementation. 90 % of the papers did not include feedback from building operators or occupants after the implementation.
No takes yet. Share an insight, caveat, or question.
Gunay et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: