Understanding School Energy Use in the Mountain West: Breaking down the latest commercial building energy consumption survey for educational facilities
EmailTemplate-Fluid
(Optional) This text will appear in the inbox preview, but not the email body.
CBECS 2018 Update: We have compiled and oragnized the CBECS micro-data to be released. You can download the Excel files here:Click Here
The rest of this blog is an in-depth analysis of the 2018 CBECS education building data.
Understanding School Energy Use in the Mountain West: Breaking down the latest commercial building energy consumption survey for educational facilities.
The purpose of identifying Commercial Building Energy Consumption Survey (CBECS) trends is so that users can more accurately forecast energy consumption, set energy goal targets, design workflow or policies, and make informed decisions about energy trends. This blog is an in-depth look at educational facilities within the Mountain west region. We will use the data available to explore the energy trends of Elementary Schools, High Schools, Preschools/Daycare, Other classroom education buildings, Middle/Junior Highs, and Multi-Grade schools (any K-12).
Data sets and samples are collected from the Commercial Building Energy Consumption Survey or CBECS. CBECS is a survey conducted periodically to assess the national building energy use and infrastructure. It should be noted that data provided by the EIA on their website about CBECS trends is based on a weighted sample size that uses population estimates to infer how many buildings in the area exist with a similar profile. The Energy Information Administration, which administers CBECS defines educational facilities as “buildings used for academic or technical classroom instruction, such as elementary, middle, or high schools, and classroom buildings on college or university campuses. Buildings on education campuses for which the main use is not educational instruction, are included in the category relating to their use. For example, administration buildings are considered office, dormitories are considered lodging, and libraries are considered public assembly.”
The subcategory of education is organized into the following building uses:
Elementary or Middle School
High School
Multi-grade K-12
College or University
Preschool or Daycare
Adult Education
Career or Vocational Training
Religious Education
The sample size of the data provided by EIA on the national level is 936 while the Idaho Power (IPC) Climate Region sample size is 306. This sample of 306 does not necessarily mean all (or any) of these buildings are in Idaho. Each building’s identity is masked and anonymous.
For our analysis we created a new boundary to filter data by climate zones 1 and 2. The boundary is called the Idaho Power Climate Region or IPC CR. Climate zones are defined by CBECS using a threshold for heating and cooling degree days to create boundaries which are slightly different from ASHRAE climate zones. After splicing together various sets of data made available and formatting in Excel the IDL performed analysis similar to our 2012 CBECS study.
Idaho Power Climate Region - Climate Zones 1 & 2
The purpose of applying the climate region is to allow users to better understand building performance as it relates to climate conditions similar to where their building is located.
Shown in figure 1 is the breakdown of the total 306 building sample size. Data and analysis for pre-school/daycare as well as ‘other classroom education’ buildings should be considered but not be relied upon because each category has a very small sample size, 9 and 13 respectively.
Figure 1 - Sub-categories sample size.
However, we can apply an additional filter to our data set to narrow down the geographic location of the buildings. Currently, with the IPC CR we are considering 32 states, but CBECS collects information about geographic regions such as census region West which contains the divisions Pacific and Mountain. With the census region West filter, we can reduce the number of states being considered to 10. Then if we apply the census division filter for Mountain, we will only be considering 8 states.
Because of these filters we reduced our sample size from 306 to 46 to 31 which results in some subcategories being unable to provide trends as shown in the next two figures below. We will not be going into that data set with this blog post but if you would like more information about it, please let us know.
Sample size of the census region west filter
Percent of census region west filter that is mountain and pacific division.
The following figure, 2, shows the average EUI by building type and end use category. We can compare it to the overall average of the entire sample. 3 building types (High School, Other, and Middle/Junior High) have a higher average EUI than the overall average while the following building types come in under the average, Elementary, Preschool/Daycare, and Multi-grade schools. However, we are not comparing the building types by size which will give us a more level playing field to assess performance. One trend is clear regardless and that is heating. Heating is the largest category for energy consumption and consumes the most energy in schools for two reasons. The first is due to the climate where we have significantly more heating days than cooling days and secondly most schools are not in session for summer when the cooling days occur for our climate. With plug loads (Misc. & Computers) increasing due to the ‘smart’ device era and replacement of traditional teaching methods with digital or hybrid methods, it is crucial for schools to become more energy efficient where possible.
Figure 2 - Shows the EUI by category and building type for the IPC CR, 306 buildings.
Since heating, cooling, and ventilation make up most of the energy use, 54%, regardless of building subcategory we will look at HVAC trends first to identify viable energy efficiency measures. When considering fuel consumption, we looked at buildings that only use natural gas and electricity as a fuel source which resulted in a sample size of 203.
Fuel consumption data that was not considered include oil/diesel/kerosene, propane or bottled gas, district steam, and district hot water. The sample size of this dataset is 103 and the IDL will be revisiting it with a focus on propane used for heating to identify schools that could be in rural areas. In our experience schools in rural areas have to rely on other fuel sources since many don’t have access to a natural gas pipeline in Idaho, but this is not universally true.
Of all buildings being considered, figure 3 shows the natural gas consumption compared to electricity for each education building subcategory. All building types are within two percentage points of the average apart from pre-school/daycare and other building types.
Figure 3 - Electric vs Gas consumption for all buildings, 306.
Figure 4 shows annual total consumption in kBTUs. Other, middle/junior high, and multi-grade are within 2 million kBTUs of the average while elementary and preschool/daycare consume less than 5 million kBTUs annually. High schools consume the most kBTUs annually at 20.33 million in part because of the 56 buildings considered, 54 have a square footage greater than 100,000. The average square footage for high schools is 289,000. Once again, this is true for education facilities in these climate zones, not necessarily for Idaho.
Figure 4 - Shows the average annual total conumption of kBTUs by building type.
The importance of this can be realized when considering the associated cost of consuming kBTUs which can be seen in table 1 below.
Table 1 - Annual Utility Cost Breakdown.
While the information displayed in Table 1 is useful, as mentioned earlier, we are not evaluating energy on a level playing field. To have a better understanding of annual kBTUs consumed and its associated cost we must consider building size in addition to type and climate. Table 2 breaks down annual total consumption into four square footage categories (0 to 9,999, 10,000 to 49,999, to 50,000 to 99,999, and greater than 100,000). If a subcategory value is ‘N/A’ then it means, there were no sample buildings available for that category. Sample sizes for each square footage range and sub-categories are shown below, Figure 5.
Figure 5 - Sample size by square footage for only electric and natural gas fuel sources.
Table 2 - Annual kBTUs with associated cost by building sub-category and square footage category.
Figure 6 - Comparing the average kBTUs and cost for all builidng as well as their square footage category.
With the data organized by square footage and associated cost analyzed we can now apply a factor to create a unit that may be more applicable for early design estimates. One question to consider is how many kBTUs will I be spending per student and how much will it cost annually. While also taking into account building type and size. CBECS does not collect student attendance, however, it does collect the total number of seats available in a classroom. Specifically, the question text states “How many students can be seating in all of the classrooms in this building at one time? If you’re not sure, please provide your best estimate.” Therefore, we can estimate the maximum student attendance. The average number of students for each building subcategory is shown in figure 7.
Figure 7 - Average number of students by building sub-category and square footage category.
Since we have the estimated number of students for our sub-categories as well as organizing by square footage, we can estimate how many kBTUs will be used per student, shown in table 3. Architects and Engineers can use this ratio to estimate energy consumption of a school by knowing the square footage and approximate number of students that will be attending. This trend is intended to be applied in the early design process as a rule of thumb. Depending on design decision and equipment specification the end energy use may be greater than or less than the estimate. The highest kBTUs per student can be found in the 100,000 square footage category under High Schools, 13,790.04 kBTU/Student, and Other, 21,917.07 kBTU/Student.
Table 3 - Annual kBTUs per average number of students.
The Other sub-category should be taken with a grain of salt because the sample size is 7 while High School’s sample size is 56. The kBTUs we are considering are derived from natural gas and electricity usage. Figure 3 from earlier shows the ratio of natural gas to electricity consumption for our building types but we can filter that as well by square footage to have better understanding of how energy is being consumed but also at what cost. Figures 8 to 14 demonstrate the ratio of natural gas to electricity by building subcategories and square footage.
Figure 8 - Annual percentage of electric and gas consumption for all education buildings.
Figure 9 - Annual percentage of electric and gas consumption for Elementary buildings.
Figure 10 - Annual percentage of electric and gas consumption for High Schools buildings.
Figure 11 - Annual percentage of electric and gas consumption for Pre-school/Daycare buildings.
Figure 12 - Annual percentage of electric and gas consumption for Other education buildings.
Figure 13 - Annual percentage of electric and gas consumption for Middle/Junior High buildings.
Figure 14 - Annual percentage of electric and gas consumption for Multi-grade (K-12) buildings.
CBECS collects the annual consumption of various fuel types as well as the annual expense associated with the fuel type. The annual consumption and cost for natural gas and electricity was used to determine averages. The average square footage from buildings in each square footage category was used to calculate the cost of electricity and natural gas per square foot. In the future it would be beneficial to separate electricity usage by category. Doing so would allow us to separate the energy usage into two halves, building usage and occupancy demand. However, this metric would have issues as well that would need to be interpreted. For instance, if there is only 1 teacher and 3 students it would show the same base load usage if there was 1 teacher and 30 students. The context being that we turn on all classroom lights regardless of attendance, 3 vs 30. Therefore, a more ideal building typology to use such a metric would be for commercial office buildings. Until such data is made available, we can apply a student factor to the cost of electricity in addition to the cost per square foot to use as a rule of thumb estimate. Still the cost of electricity per student may be more meaningful to a school administrator than electricity cost per square foot.
Table 4 - Annual electric cost by building sub-category and square footage category.
Next, we apply the same filters and factor we used for kBTUs and electricity to natural gas usage, shown in table 5.
Table 5 - Annual gas cost by building sub-category and square footage category.
HVAC Upgrades
Tables 6 through 9 show the energy use for central HVAC system types for all buildings and their subcategories. The total column is based on all energy use categories while the tables only show heating, cooling, and ventilation. We’ve included the median total EUI, sample size, and average square footage to provide context for understanding the performance of different systems. Please keep in mind that the data presented in tables 6 through 9 do not consider the age or maintenance of HVAC systems which can impact operational energy efficiency. Also, buildings who consider upgrading their CAVs to an energy efficient HVAC such as DOAS with VRF or Heatpump would see more energy savings than those who upgrade from a VAV or DCV HVAC system. Again, context is key to determining savings, base load and operational energy. One issue that presents itself is the capital cost which can be significant for full HVAC system overhaul, however, as shown heating is the largest consumption of energy usage for schools. Therefore, schools that invest in their HVAC systems can expect to see significant operational savings which results in a reduction of utility cost thereby reducing the operating costs of the school. In addition to the energy savings schools can expect to improve indoor air quality and occupant comfort which should also increase occupant productivity. Indoor air quality is a critical non-EEM (Energy Efficiency Measure) that is often overlooked when considering school buildings regardless of grade level. This is because children are the primary occupants of the building and therefore the exchange rate of air for developing children or young adults is much higher than adults (office workers) which means that the HVAC system needs to have more frequent air exchanges in classroom to, for example, avoid building up levels of CO2 (Carbon Dioxide) to an unhealthy level. For more information on the benefits of a high performance classroom please visit our YouTube channel here (https://www.youtube.com/watch?v=pZHKxySQbbM&ab_channel=UIIDLChannel) for a lecture Dr. Woods has developed and continues to update.
Table 6 - Heating, Cooling and Ventilation for CAV EUI.
Table 7 - Heating, Cooling and Ventilation for VAV EUI.
Table 8 - Heating, Cooling and Ventilation for DCV EUI.
Table 9 - Heating, Cooling and Ventilation for DOAS EUI.
Figure 15 - EUI for High School HVAC system types over 100,000 square feet.
Figure 16 - EUI for Other education HVAC system types over 100,000 square feet.
HVAC Savings
Using a NEEA report from better bricks to estimate the cost of upgrading an HVAC system to DOAS – VRF would cost approximately $13 to $20 per square foot. The twenty dollars has been adjusted from $17 for inflation and rural locations. Table 10 below shows the estimated cost by building subcategories using their average square footage to upgrade their HVAC systems. Table 11 below considers the same situation but instead of using average square footage we used a range of square footages from 5,000 to 200,000. Both table 10 and 11 do not take into account incentive programs offered by utility companies such as Idaho Power. For those in Idaho Power service territory you can view the list of HVAC retrofit incentives offered here (https://www.idahopower.com/energy-environment/ways-to-save/savings-for-y...) and if you need help understanding the incentives you can find your Idaho Power Energy Advisor here (https://tools.idahopower.com/forms/EnergyAdvisorContact/Submit).
Table 10 - Estimated cost to upgrade HVAC from CAV to DOAS/VRF
Table 11 - Estimated cost to upgrade by square footage range.
Commission HVAC Controls
CBECS does not collect information about buildings being commissioned but instead focuses on if a building has had regular HVAC maintenance or an energy management and control system which can be used for heating, cooling, or other building mechanical systems. Of those who responded to the survey 94.77% responded yes to “is there any regularly scheduled maintenance and repair for the (heating/cooling) system?”. An issue with this question is that it doesn’t consider any of the control settings for the system but rather refers only to the hardware of the HVAC system. Often, the right equipment can be in place, but the controls are fighting each other or just not allowing the systems to work properly – like thermostats with scattered settings so that the same room is being heated and cooled at the same time. Commissioning all controls can save 13% of total energy but also an 8 to 20% reduction in equipment maintenance. We can estimate the cost for upgrading controls at $1.18/SQFT (Tilgner).
Table 12 - Estimated cost to add thermostats by square footage range.
Installing HVAC controls for a building that currently has none is also a significant capital cost, but again, this is an area where schools should be making investments in their building infrastructure due to HVAC making up the largest consumption, 65%, of energy for schools. Installing thermostats can cost $75 to $300 per thermostat but can result in an estimated 10% to 25% of HVAC energy savings. For bigger schools installing full DDC the cost range is $1,000 to $1,500 per control point or $5.00 to $8.00 per square foot.
Table 13 - Estimated cost to add controls by square footage range.
Building Shape & Envelope Sealing
Figure 17 - Building shape sample size by Education sub-category.
Figure 18 - Average total EUI by building shape and sub-category.
Figure 19 - Wall construction by building sub-category.
Figure 20 - Insulation upgrades by building sub-category.
Figure 21 - Glazing type by building sub-category.
Envelope Sealing is one of the most cost-effective measures. This can include re-sealing windows and doorways and patching any gaps in construction. This is especially helpful for older buildings. “Annual energy savings from air barrier improvement resulting from testing due to the measure ranges from $5.07 to $71.88 per thousand square feet of floor area” (Hart 4). This EEM is critical for schools to implement since heating accounts for the largest consumption of energy therefore schools would see a significant return on investment. By having a high-performance envelope facility managers will not have to heat or cool the building as often, but also could see a reduction in loads for the system size. A potential reduction in load sizing would make replacing or adding additional equipment cheaper.
Lighting Upgrades or Retrofits
Figure 22 - Lighting system types present in education building by sub-category.
Fluorescents are the dominant lighting type, 64%, for educational buildings regardless of the subcategories. LEDs are the second largest at 32% and last is compact fluorescent at 4%. Approximately 30% of the lighting systems examined are of only one type while 70% were a hybrid of lighting systems with fluorescent and LEDs being the most common hybrid.
There are a variety of options available to upgrade the school’s lighting system, but the two most common options are replacing the bulbs and bypassing the ballast or conversion/retrofit kits. One issue with just replacing the bulbs of a fixture is that the luminous efficacy and distribution of light is vastly different. Therefore, additional lighting controls may be needed to meet occupant satisfaction for doing tasks in the space which will add to the cost. LED retrofit kits are designed to upgrade existing fluorescent fixtures by replacing the fixture’s bulb and ballast to be compatible with LED technology. The upfront cost is significantly higher than only replacing the bulbs of fluorescent lighting and payback is calculated in two ways. The first is the base load reduction of kW which is calculated from the reduction of Watts being consumed from fluorescent to LEDs. Second is the control strategies savings such as high-end trim, occupancy/vacancy sensors, scheduling, daylight harvesting, and dimming. Control strategies savings are much harder to calculate because they depend on how the building is being used. We can predict or infer how a building will be used but collecting data on how the building is being used is invaluable to create trends. Building use trends allow facility managers to calibrate and tune the building for optimal performance to meet energy efficiency needs, but also, satisfy occupant comfort requirements. This is true for all building control strategies, not just lighting and that’s why the estimated savings are presented in ranges or weighted averages.
Energy consumption for lighting has been on a downward trend since the introduction of control strategies for fluorescent lighting systems. In 2003 lighting accounted for 36% of total electricity consumption for commercial buildings but that fell to 17% in 2012. The reduction is being continued by the adoption of LED lighting systems. LEDs share of the market has increased 35% from 9% in 2012 to 44% in 2018 and is the only lighting type to gain market share in the 2018 CBECS. Early estimates for the 2018 CBECS survey suggest lighting will account for 15.7% of electricity consumption in buildings. Lighting control strategy savings are dependent entirely on the kWh pool available and generally are expressed as a percentage. The greater the kWh pool then the better the savings from control strategies. There is change occurring in the market where more control strategies are required by code, hardware cost is increasing, but the rate of return is diminishing. This is because LEDs have become so efficient that there is a significantly smaller kWh pool of lighting power to apply percentage savings. In response to this market disruption the industry is responding by introducing ‘smart’ lights or Luminaire Level Lighting Controls. The future of lighting controls will be in data collection by serving as the backbone of a data highway that runs throughout the building for energy management systems. At least in the AEC industry but other industries will find value through other methods of reaching building occupants and establishing trends from data collection. For those interested in upgrading lights to LEDs the rule of thumb from the State and Local Climate and Energy Program estimates the upgrades cost at $0.90 to $1.20 per square foot. In addition, Idaho Power offers incentives for lighting systems based on kWh saved. For more information about the commercial lighting incentives please visit: https://www.idahopower.com/energy-environment/ways-to-save/savings-for-y...
Conclusion
Analysis or interpretation of data sets must be done strategically so that a user can isolate effects and derive what questions should be asked from trends. Moreso, context is crucial in understanding data for removing miscellaneous questions, such as, why is heating the largest consumption of energy across the board for all buildings. Context tells us that because we applied a climate filter then we will have more heating days than cooling days. Okay, but why is cooling non-existent because there are cooling days in the climate filter we applied. Looking at the building typology we can determine that the building’s operating schedule is different from a typical commercial building by which schools do not operate during summer (mid-May to mid-August). What cooling we are seeing is from the handful of weeks where schools need to cool but also from schools that are on a year-round schedule.
It is not enough to install new hardware or upgrade existing hardware and expect energy savings. Yes, you can receive a base load of savings from changing hardware but being able to continually save energy is where the real payback for building systems can be found. This is achieved through operational savings by ensuring, one, that the equipment is operating as designed and two, that we are maximizing our efficiency to use the least amount of energy possible while maintaining occupant comfort. To ensure that equipment is operating as designed it is verified uses two methods, commissioning and energy audits or assessments. Commissioning verifies that installation and basic operating standards/procedures have been programmed correctly. Energy audits or assessments verify that the equipment continues to perform as designed but also is responding to programming changes over time. Energy audits do not need to be conducted every six months or a year, but a schedule should be created to collect valuable verifiable data about the building’s mechanical operation. So, a two- or three-year audit timetable is acceptable depending on the building type and use but the takeaway is not to wait until something goes wrong to assess a building’s performance.
One of, if not the biggest, issues with optimizing building system performance is that the purpose focuses on only saving energy. If a building’s control strategies conflict with the occupants’ use, i.e. comfort, of the building then they will be overridden, and a building could end up using more energy than its design intent. Performing an energy audit or assessment can also include occupant comfort surveys to collect feedback for how occupants want/are using the building’s spaces. If you have any questions, comments, or would like to access the data we used please contact Dylan Agnes at dagnes@uidaho.edu. The IDL will be revisiting this data set to apply the weighted population factor to verify trends and identify any key differences from the IPC CR boundary against the national data set.
Dylan Agnes
After earning a Bachelor of Science in Architecture from the University of Idaho, Moscow, Dylan studied the science and engineering of building design, completing a Master's in Architecture with an emphasis in urban planning and net-zero/energy efficiency building design. As a student he worked at the Integrated Design Lab and gained hands-on experience in the practice of Integrated Design. As an IDL Research assistant, Dylan worked with both the architectural and engineering side of integrated design, providing a broader opportunity to cross over fields of study. Since graduation, Dylan has been working as a Research Scientist at the IDL and has been working on a wide range of projects from Energy Modeling to Daylighting Design.
Works Cited
“2018 Commercial Buildings Energy Consumption Survey - Consumption and Expenditures Highlights.” U.S. Energy Information Administration, Dec. 2022.
“Commercial Building Energy Consumption Survey (CBECS) - User’s Guide to the Public Use Microdata File.” U.S. Energy Information Administration, Jan. 2023.
“Commercial Buildings Energy Consumption Survey (CBECS) - Trends in Lighting in Commercial Buildings.” U.S. Energy Information Administration, 17 May 2017.
Hart, R, et al. “Envelope Air Tightness For Commercial Buildings.” U.S. Department of Energy, Dec. 2018. https://www.osti.gov/biblio/1489004
Hines, Emma, and Sara Ross. “HVAC Choices for Student Health and Learning: What Policymakers, School Leaders, and Advocates Need to Know.” RMI and UndauntedK12, Jan. 2023.
Liu, L, et al. “Life Cycle Cost Analysis of LED Retrofit and Luminaire Replacements for Four-Foot T8 Troffers.” The Society of Light and Lightings, June 2023.
Idaho Power Climate Region - Climate Zones 1 & 2
Figure 1 - Sub-categories sample size.
Sample size of the census region west filter
Percent of census region west filter that is mountain and pacific division.
Figure 2 - Shows the EUI by category and building type for the IPC CR, 306 buildings.
Figure 3 - Electric vs Gas consumption for all buildings, 306.
Figure 4 - Shows the average annual total conumption of kBTUs by building type.
Table 1 - Annual Utility Cost Breakdown.
Figure 5 - Sample size by square footage for only electric and natural gas fuel sources.
Table 2 - Annual kBTUs with associated cost by building sub-category and square footage category.
Figure 6 - Comparing the average kBTUs and cost for all builidng as well as their square footage category.
Figure 7 - Average number of students by building sub-category and square footage category.
Table 3 - Annual kBTUs per average number of students.
Figure 8 - Annual percentage of electric and gas consumption for all education buildings.
Figure 9 - Annual percentage of electric and gas consumption for Elementary buildings.
Figure 10 - Annual percentage of electric and gas consumption for High Schools buildings.
Figure 11 - Annual percentage of electric and gas consumption for Pre-school/Daycare buildings.
Figure 12 - Annual percentage of electric and gas consumption for Other education buildings.
Figure 13 - Annual percentage of electric and gas consumption for Middle/Junior High buildings.
Figure 14 - Annual percentage of electric and gas consumption for Multi-grade (K-12) buildings.
Table 4 - Annual electric cost by building sub-category and square footage category.
Table 5 - Annual gas cost by building sub-category and square footage category.
Table 6 - Heating, Cooling and Ventilation for CAV EUI.
Table 7 - Heating, Cooling and Ventilation for VAV EUI.
Table 8 - Heating, Cooling and Ventilation for DCV EUI.
Table 9 - Heating, Cooling and Ventilation for DOAS EUI.
Figure 15 - EUI for High School HVAC system types over 100,000 square feet.
Figure 16 - EUI for Other education HVAC system types over 100,000 square feet.
Table 10 - Estimated cost to upgrade HVAC from CAV to DOAS/VRF
Table 11 - Estimated cost to upgrade by square footage range.
Table 12 - Estimated cost to add thermostats by square footage range.
Table 13 - Estimated cost to add controls by square footage range.
Figure 17 - Building shape sample size by Education sub-category.
Figure 18 - Average total EUI by building shape and sub-category.
Figure 19 - Wall construction by building sub-category.
Figure 20 - Insulation upgrades by building sub-category.
Figure 21 - Glazing type by building sub-category.
Figure 22 - Lighting system types present in education building by sub-category.