ICP For Adaptive Reuse The Investor Confidence Project's (ICP) Investor Ready Energy Efficiency (IREE) Adaptive Re-Use Protocol Addendum aims to enhance the ICP’s IREE certification protocols to address the unique challenges of certifying energy efficiency projects in buildings without available baseline data due to factors such as adaptive re-use, prolonged periods of vacancy, or recent property acquisition that limit access to historical operational data.
A baseline makes predicting and validating the potential energy savings resulting from energy efficiency improvements easier and can provide calibration of energy savings estimates. The absence of a baseline can pose a significant challenge for energy efficiency projects, particularly in scenarios of adaptive reuse, long-unoccupied buildings, or recently purchased buildings with no accessible historical data.
The ICP, part of the Green Building Certification Inc (GBCI), is a system to standardize the development of energy efficiency projects to increase investor confidence through predictable and reasonable energy savings outcomes. Typically, this involves establishing a baseline normalized for weather and other non-routine events against which the post-retrofit performance is compared. However, scenarios like adaptive reuse (e.g., converting commercial spaces into residential use), unoccupied buildings for extended periods, or newly acquired buildings without historical data present unique challenges. This lack of actual historical utility data or verifiable operational assumptions means that utilizing a whole building simulation model may be required, and all significant assumptions must be agreed to as reasonable by all parties and then verified in the performance period based on actual building performance and usage patterns.
The current ICP standards primarily rely on the International Performance Measurement and Verification Protocol (IPMVP) Option C for measurement and verification (M&V), which uses historical utility data to establish a baseline. This is problematic for buildings without such data because it precludes using directly measured and calibrated data to validate energy savings. This necessitates the application of Option D from the IPMVP, which relies on simulated baselines (dynamic building simulation or “energy modeling”). However, without historical operational data, these simulations are inherently uncalibrated and require numerous assumptions that cannot be verified against ground truth before the retrofit. The ICP Adaptive Re-Use approach provides a system to validate upfront assumptions, which inherently requires additional engineering and owner representation and a system to validate models in the performance period. Because a complete change in building end use is comprehensive, projects that follow this approach should utilize this addendum for the baseline, savings calculations, and M&V sections of all protocols, and a simulation model is required.
Situations Where Lack of Baseline Data Makes Current ICP Protocols Ineffective:
Adaptive Re-use: In adaptive reuse scenarios, commercial buildings transformed into residential units or other uses do not have relevant historical energy usage data that reflect the property’s new function. This transformation fundamentally changes energy consumption patterns, making previous data, if available, largely irrelevant.
Extended Unoccupied Periods: Buildings that have remained unoccupied for lengthy durations exhibit no historical energy usage data during these periods. Any pre-vacancy data fails to reflect the current state of the building’s systems and envelope, which may have deteriorated or become less efficient over time.
Recently Purchased: Newly acquired buildings often come without accessible historical utility data due to privacy concerns, lost records, or previous owners' unwillingness to share information. This leaves new owners needing a reference point for past energy consumption patterns.
Approach to Address These Challenges The challenges associated with the lack of baseline data in energy efficiency projects are significant, affecting the accuracy and reliability of savings predictions. By adopting a robust, transparent, and continuously evolving modeling approach, stakeholders can mitigate these issues, improving investor confidence and the effectiveness of energy efficiency measures. Given these issues, a detailed approach involves:
Developing a Baseline Without Reliable Historical Data In scenarios where reliable historical data are unavailable—such as adaptive reuse projects, buildings with prolonged vacancies, or newly acquired properties—a baseline is critical to validate energy savings projections. This section outlines two viable approaches based on project complexity and resources: advanced simulation modeling for robust, complex retrofits and standardized, code-based assumptions for more straightforward projects.
1. Simulation Model Development for Robust and Complex Retrofits For projects where precise baseline accuracy is essential, a detailed simulation model can be used to project energy use. Advanced building simulation tools allow for assumptions about future building characteristics, occupancy patterns, and energy usage behavior, providing a dynamic and comprehensive baseline.
Task: Develop a simulation model using dynamic building simulation software, ensuring it meets industry standards for accuracy and includes an 8760-hour annual simulation. Document all assumptions, from HVAC and lighting efficiency to operational schedules.
Objectives: To simulate a robust, representative baseline that reflects anticipated usage, offering stakeholders a transparent and credible energy savings projection for high-impact retrofit projects.
2. Standards Assumptions and Code Requirements for Simpler Projects For projects unable to support extensive modeling, a baseline (total annual energy consumption and end-use disaggregation) can be constructed using public datasets, such as the U.S. Energy Information Administration’s Commercial Building Energy Consumption Survey (CBECS) for commercial buildings, the Natural Resources Canada's National Energy Use Database (NEUD) or the U.S. Department of Energy’s (DOE) Residential Energy Consumption Survey (RECS) for multifamily residential buildings. These surveys provide energy consumption benchmarks by building type, location, age, and size, allowing more straightforward projects to establish a reasonable baseline reference.
Task: Establish standards to utilize CBECS, NEUD, or RECS data to align baseline assumptions with building demographics or assumptions based on relevant codes or standards applicable during construction, including typical equipment efficiencies (e.g., non-condensing heating, pneumatic HVAC controls).
Objectives: To provide a practical and data-supported baseline using comparable buildings’ median energy use, offering a credible starting point for simple projects where custom modeling is not feasible.
Documentation and Transparency Maintaining detailed records of all assumptions, methodologies, and adjustments made to the model. This transparency is crucial for stakeholders to understand the basis of the predictions and for building trust in the modeled outcomes.
Task: Require comprehensive documentation of all processes—from model development or assumptions made using and actual performance metrics—should be maintained and accessible for audit purposes.
Objective: To maintain transparency and allow stakeholders, including quality assurance assessors, investors, and certification bodies, to review the methodology and outcomes.
Appendix A: Challenges with Simulation Modeling as a Baseline
Energy savings are calculated by comparing current consumption against a historical standard or baseline. Without this comparative measure, any model must rely heavily on assumptions regarding how the building would have operated under normal conditions.
For the low-data retrofit use cases, a model or estimate of baseline may be the only viable path to measuring performance outcomes. Still, all parties must understand the challenges and risks associated with this approach. The following example compares utility bill-calibrated third-party simulation models to the results of a baseline generated through regression modeling of actual pre-post consumption. It is important to note that the variance below is from calibrated models. One should expect that a model that cannot be calibrated because of a lack of data or adaptive reuse will have even more significant variance.
The Pacific Gas and Electric (PG&E) whole-building study highlights specific challenges in applying Option D-modeled baselines. The image below shows how the IPMVP Option-D modeled baselines vary compared to various IPMVP Option-C models. These Option-D models were calibrated to energy consumption, which is impossible for the adaptive reuse use case we are addressing in this document.
Bias and Variance: The study notes a consistent bias toward overestimating energy savings when using modeled baselines compared to measured baselines. This is primarily because models must assume specific operational settings, like HVAC runtimes, lighting usage, outside air percentages, or other variables such as occupancy patterns, internal loads, or outside air infiltration, which may vary. The variance is also significant, indicating a wide discrepancy between predicted and actual energy usage.
Assumptions in Model: Hundreds of assumptions are necessary to simulate how a building uses energy. These can include occupancy levels, equipment efficiencies, and behavioral patterns of the building occupants. The more assumptions made, the greater the potential for error if these assumptions do not align closely with actual conditions.
Verification Challenges: Verifying these assumptions or calibrating the model against actual data is only possible with an existing condition baseline, leading to questions about the reliability of the model predictions. The example above is for models that have been utility bill-calibrated with assumptions based on actual usage of the building being modeled, which should generally provide a higher quality model than what is possible in this adaptive reuse approach. Because of these challenges, a more detailed quality assurance process and ensuring all parties are comfortable with baseline assumptions are necessary.