Critical Review of Field Implementation of Data-driven Operation and Maintenance Technologies

Strategic Insights -

Researchers at Carleton University and the National Research Council Canada have conducted a comprehensive review of field implementation studies of data-driven building operation and maintenance (DBOM) applications. The study aimed to compare the implementation processes and outcomes of various DBOM applications, including 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). According to the research, the analysis highlights the correlation between energy-related success metrics and the integration effort and sensing infrastructure complexity of the DBOM applications.

Key Takeaways:

  • 95% of DBOM applications that reported post-implementation energy metrics achieved more than 10% improvement in their energy-related objectives.
  • MPC and ADR studies required a higher level of integration effort than OCC studies, but OCC studies often required upgrades in sensing infrastructure.
  • 90% of FDD and VM studies did not include a measurement and verification procedure to quantify energy-related performance improvements after implementation.
  • A large fraction of FDD and VM studies did not provide feedback from building operators or occupants after implementation.
  • The analysis results suggest that DBOM applications with higher complexity of infrastructure and integration effort tend to have better energy-related outcomes.

Statistics:

  • 95% of DBOM applications achieved more than 10% improvement in energy-related objectives.
  • 90% of FDD and VM studies did not include a measurement and verification procedure.
  • 90% of papers did not include feedback from building operators or occupants after implementation.
  • 10% of MPC and ADR studies required upgrades in sensing infrastructure.

Sources:

  • A Critical Review of Field Implementation of Data-driven Operation and Maintenance Technologies To Reduce Building Energy Use and Ghg Emissions (Energy and Buildings, 2025;346).
  • NewsXr (Information Technology Newsweekly, November 4, 2025).