System-Level Key Performance Indicators
Energy efficiency is an underused resource—even though it saves money for both consumers and utilities1 and, on a global level, is a key element of any plan to address climate change.2,3
A primary reason that energy efficiency is not deployed to its full potential is that—unlike solar panels or electric vehicles—energy efficiency is often invisible; it hides within wall insulation or works behind the scenes through the best practices a technician takes to install a piece of equipment.
Even when energy efficiency is visible, it does not carry much social cachet: very few people boast about their new heat-pump water heater as they do about a new Tesla electric vehicle. Further complicating the need to elevate the conversation around energy efficiency is the lack of widely agreed upon language for identifying the efficiency of specific building services. While we label product-specific efficiencies (e.g., seasonal energy efficiency ratio [SEER] measures the efficiency of air-conditioning equipment) and measure efficiency at the building-level (e.g., energy use intensity can measure energy use per home, per person or per unit of floor area), we don’t have clear or consistent measures for the efficiency of energy services delivered to building occupants.
At the June 2019 ASHRAE Conference, Legrand North America and researchers from the Pacific Northwest National Laboratory (PNNL) and Lawrence Berkeley National Laboratory (LBNL) participated in a panel on “Building Energy Efficiency at the System Level.” The panelists explored system-level key performance indicators (KPIs) and proposed ways to implement them within building energy management practices and policies. While performance diagnostics are most effective at the system level, KPIs presently are mostly available for the component- or whole-building-level, leaving a gap in system-level KPIs. Incorporating the development of system-level KPIs into performance benchmarking, analysis, and measurement and verification has enormous potential for identifying and achieving deeper energy savings. The ASHRAE panel introduced a suite of system-level KPIs, including an HVAC system-level performance metric that has been piloted in a building energy code.
System-level KPIs quantify energy performance by measuring the capacity or quality of a service provided by a building system—such as comfort or illuminance—per unit of energy used to provide the service. For instance, an HVAC system-level KPI would go beyond the efficiency aspects captured by the SEER rating of an air-conditioner to more holistically account for additional aspects of the system—e.g., fan, air filter, ductwork length, ductwork insulation, presence or absence of economizers or energy recovery—that have a significant impact on how much energy is used to keep occupants comfortable. Clear, commonly understood system-level KPIs would provide greater transparency into the performance of building systems, and could also help increase our understanding of the interactive effects of different components within a system.
System-level KPIs are just starting to have real-world implications for building design and operations. Washington State recently approved an HVAC-focused KPI called “Total System Performance Ratio” (TSPR) into its energy code, which will take effect in July 2020.4 The TSPR compares a building’s annual heating and cooling loads to the amount of energy required to provide these heating and cooling services. Washington State modified the TSPR metric for their code by assessing carbon emissions rather than energy use. PNNL aims to facilitate the use of TSPR by architects and engineers in the state by developing a module specific to Washington for the U.S. Department of Energy’s Building Energy Asset Score.5 The module is just one example of a system-level tool; as more regions consider the benefits of system-level KPIs, different tools or even different KPIs may need to be developed to serve a state’s needs with consideration for their specific policy objectives, and to allow building designers to optimize HVAC systems using a more comprehensive set of increasingly complex options.* Well-designed system-level KPIs have the potential to support performance-based compliance that offers design flexibility while ensuring that any efficiency tradeoffs made in the design process don’t undermine the ultimate efficiency of the building and its systems.
Building operators can track system-level KPIs to identify performance problems, including building systems that are consuming more energy than expected. Once an underperforming system is identified, system-level KPIs can help with diagnosing the problem as well. For example, high energy consumption in an HVAC system might lead an operator to recognize excessive energy use by a particular fan within that system, further facilitating discovery of a malfunctioning motor, sensor or actuator. By flagging issues within specific building systems, system-level KPIs can thus help diagnose problems and more quickly restore systems to their proper operation.
The recent growth in smart meters and sensors,6 and in turn the growing availability of building data, increases the options for development of system-level KPIs and broadens their potential benefits. System-level KPIs offer the opportunity to capture and focus building data in a useful way to better inform how we measure, benchmark, demonstrate, and manage energy performance. The coordination of sensors and controls to help manage and track system-level KPIs will require significant coordination among building operation and design professionals to ensure that the KPIs enhance and improve the accuracy of both diagnostics and predictive modeling.
The potential benefits of system-level KPIs—including energy, greenhouse gas, and cost savings—continue to be explored and demonstrated, but what is required to implement them? Once you have decided to deploy system-level KPIs to help manage energy use, which KPIs should you use? The KPI that best fits a particular application depends on several factors, including the types of performance issues being targeted, how much building data is already being collected, and available resources to support additional data collection and analysis. A systems approach enables us to center energy efficiency around the human experience and how effectively we deliver the building services that people need and want, including a fresh and comfortable environment and enough light to work and play. This approach is supported by TSPR, which instead of looking only at HVAC energy use looks at the ratio of services provided (cooling and heating loads) to energy use.
A building that sacrifices services, for example more relaxed heating and cooling setpoints or reduced hours of operation, will show reduced HVAC energy use, but not a better TSPR.
System-level KPIs can provide a lens to help designers and operators focus on delivering these services in a way that minimizes energy consumption. LBNL has identified the following key parameters to consider when defining a KPI: energy use, power demand, responsiveness to control, responsiveness to service demand, and aggregation level. In addition to identifying the scope of data necessary for a well-documented KPI, LBNL has already defined 43 KPIs that measure the performance of major building systems, including indoor and outdoor lighting, cooling, heating, ventilation, water distribution, hot-water service, and miscellaneous energy loads.7
Other concrete ways to leverage system-level KPIs and think beyond component-level efficiency—to help architects, engineers, and building operators optimize the efficiency of building services—include:
- Integrating system-level KPIs into existing building databases—for example, system-level KPIs could be added to LBNL’s Building Performance Database8 or to the General Services Administration’s Energy Usage Analysis System—or developing new system-focused databases;
- Using system-level KPIs to denote high performance, including within minimum efficiency requirements (e.g., in ANSI/ASHRAE/IES Standard 90.1 or the model International Energy Conservation Code) or within beyond-code programs and standards (such as the Energy Star certification program or LEED certification program);
- Integrating system-level KPIs within building automation systems, building energy modeling, and building information systems, which could include a report on system-level KPIs for major end use systems.
Just as system-level KPIs can shine a light on the performance of building systems, they can similarly help us understand the performance of our overall energy system. Mitigating climate change is a significant motivator for taking action to make buildings more energy efficient. As noted, system-level KPIs can help highlight and hone in on the performance of building systems to drive deeper efficiency in a cost-effective manner. However, to realize the full environmental and energy-savings benefits of taking a systems approach, we need to consider what goes on beyond the building and how a building relates to the energy system at large.
The quality of energy used has environmental implications that directly scale with the quantity of energy used—in other words, we should consider the carbon intensity as well as the amount of energy being used in a building. The efficiency of distributed power depends on the dynamics within the larger system of the building and the grid, and the interactions between a power system’s peak generation and a building’s peak consumption.9
System-level KPIs can play a role in advancing energy efficiency in both the design and operation of buildings in a way that focuses on energy efficiency improvements without sacrificing the occupant experience. Developing system-level KPIs to measure and help manage the energy used in buildings at all levels—from a single system to whole buildings to large-scale building-to-grid interactions—will unlock greater efficiency gains at a lower cost and will help us realize the full potential of energy efficiency.
- Gilleo, A. 2017. “New data, same results – Saving energy is still cheaper than making energy.” American Council for an Energy-Efficient Economy (ACEEE). https://aceee.org/blog/2017/12/new-data-same-results-saving-energy.
- Creuheras, S. “The Role of Energy Efficiency in Long-Term Climate Change Planning.” World Resources Institute. https://www.wri.org/climate/expert-perspective/role-energy-efficiency-long-term-climate-change-planning; and/or https://www.nrdc.org/experts/sheryl-carter/ramping-energy-efficiency-key-address-climate-change.
- Carter, S. 2016. “Ramping Up Energy Efficiency Key to Address Climate Change.” Natural Resources Defense Council. https://www.nrdc.org/experts/sheryl-carter/ramping-energy-efficiency-key-address-climate-change.
- Goel, S., Jonlin, D., Rosenberg, M. 2018. “TSPR: The Total System Performance Ratio as a Metric for HVAC System Efficiency.” ACEEE. https://aceee.org/files/proceedings/2018/node_modules/pdfjs-dist-viewer-min/build/minified/web/viewer.html?file=../../../../../assets/attachments/0194_0286_000366.pdf#search=%22TSPR%22.
- Building Energy Asset Score. Department of Energy. https://buildingenergyscore.energy.gov/.
- Kutepatil, O. 2017. “Soaring smart electric meter market: Enabling a balanced profitability landscape for the consumers and the energy companies.” Consulting-Specifying Engineer. https://www.csemag.com/articles/soaring-smart-electric-meter-market-enabling-a-balanced-profitability-landscape-for-the-consumers-and-the-energy-companies/.
- Hong, T., Li, H., Sofos, M. 2019. “System-level key performance indicators (KPIs) for building performance evaluation.” ASHRAE Transactions 125(2).
- Building Performance Database. Lawrence Berkeley National Laboratory. https://bpd-info.lbl.gov/.
- Frick, N., Eckman, T., Goldman, C. 2017. “Time-varying value of electric energy efficiency.” Lawrence Berkeley National Laboratory. https://emp.lbl.gov/publications/time-varying-value-electric-energy.•
* For example, Washington state includes a requirement for dedicated outdoor air systems combined with a cycling space conditioning system for some occupancies within its state code. This requirement is reflected in the Asset Score module’s baseline system, which sets the performance threshold. However, this requirement might not be appropriate for other jurisdictions, and different performance thresholds would need to be reflected in any tools or modules developed for them. (Back to text)
About the Author
John Mayernik is an energy analyst at the National Renewable Energy Laboratory and a member of the Alliance to Save Energy’s Systems Efficiency Subcommittee. He is based in Washington, DC.