An energy model is a computer-generated representation of a building used to estimate the impact of proposed changes to the building on energy use. It is a comparative tool most commonly used:

New construction

• To evaluate the impact of different design options (e.g., a chilled-beam system as opposed to a conventional overhead single-duct variable-air-volume system).

• To compare the performance of a proposed design to local energy-code requirements.

• To determine the impact of change orders (e.g., less exterior-wall insulation) during construction.

Existing buildings

• To understand how a building is consuming energy.

• To quantify deficiencies in HVAC controls (e.g., how much energy and money non-functioning air-handling-unit [AHU] supply-air-temperature reset is wasting).

• To identify/estimate savings from facility improvements.

• As part of a measurement-and-verification (M&V) program.

Modeling Methods

There are three approaches to creating an energy model: forward, inverse, and hybrid (Table 1). To determine which is most appropriate for a project, consider:

• The project type (new construction or existing building).

• The type of data available (e.g., construction documents, building-automation-system [BAS] trend data).

• Who will be available to answer questions concerning the building’s operations (e.g., the engineer of record, the facility manager/engineer).

• The level of accuracy or resolution (i.e., system or equipment) needed.

• The purpose of the model (e.g., to diagnose HVAC-equipment-operation deficiencies, to estimate whole-building energy use).

• How quickly the building owner wants to see results.

• Whether the model outputs will be used to make capital-investment decisions.

• Whether the model will be used as part of a M&V program or ongoing commissioning process.

More often than not, for new-construction projects, detailed forward models best meet the needs of design teams. Forward models offer the flexibility to evaluate different design options and compare overall energy efficiency to the local building energy-efficiency code.

For existing-building projects, forward or inverse modeling can be used, depending on the needs of the building owner and the availability of building documentation/data. Typically, though, hybrid modeling is used to estimate measure-level energy savings for facility-improvement measures and capital-intensive projects. The hybrid approach offers a few time-saving advantages:

• Inverse models can be created from BAS or portable-logger trend data to estimate the baseline operation of a system.

• Forward modeling can be used to estimate energy use (e.g., resulting from the proposed implementation of supply-air-temperature reset on an AHU with a constant supply-air-temperature setpoint).

Modeling Throughout the Life of a Building

Energy models can be used throughout the life of a building for comparative reasons (e.g., whether HVAC systems are operating based on designed and commissioned controls sequences of operations).

A whole-building-simulation model can be used prior to or during schematic design and design development to compare different design options. Once a design is finalized, construction documents can be compared to the local energy code to estimate the design’s efficiency. During construction, an energy model can be used to estimate the impact of change orders on energy use and building performance. At the end of construction, an energy model can be used to compare a building as it was built to how it was designed to determine the effect of any change orders or operational deficiencies.

Throughout the remainder of a newly constructed building’s life, an as-built whole-building-simulation model can be used to evaluate operations-and-maintenance (O&M) procedures and identify opportunities to improve them. A model also can be used to estimate the impact of interior-space changes (e.g., a tenant-space remodel, the conversion of a conference room to a data center) on energy use and as part of an ongoing commissioning program to verify a building is operating as designed and commissioned. If a M&V program is implemented, an energy model can be used to measure energy savings.

For older buildings and buildings for which a whole-building-simulation model was not created during design and construction, an energy model can be used for a variety of reasons. Depending on the modeling approach, an energy model can be used to estimate energy savings, diagnose O&M deficiencies, and measure energy savings in relation to a pre-established energy-use baseline.

M&V, Existing-Building Commissioning, and Energy Models

The mechanisms for using an energy model to diagnose and correct system deficiencies and project and measure energy savings are M&V and existing-building commissioning.

M&V. In M&V, an energy model is used to determine energy savings relative to an energy-use baseline. The International Performance Measurement & Verification Protocol (IPMVP) offers two options for performing M&V:

Option C: Whole Building. An energy-use baseline is established using an inverse model of at least 24 months of utility bills. The baseline is compared with post-implementation utility bills—up to 12 months’ worth, depending on the type of building and the measure(s) implemented—to determine annual energy savings. Great care should be taken when using fewer than 12 months of post-implementation utility bills. To minimize risk, an analysis can be performed each month to determine if annualized energy savings are converging toward an energy-savings value. Figure 1 shows how the duration of post-implementation monitoring affects final energy savings.

Option D: Calibrated Simulation. Whole-building-simulation-modeling software simulates building systems and interactive effects hourly. The resulting model is calibrated to at least 24 months of utility bills to within a specified tolerance (Table 2). This is the baseline, a copy of which is calibrated to post-implementation data—at least 12 months’ worth, depending on the measure(s) implemented in the building. The post-implementation model then is compared with the baseline model to determine energy savings.

While both options measure energy savings, only Option D provides a model capable of identifying differences between how measures are operating and how they were specified and commissioned to operate.

Existing-building commissioning. Existing-building commissioning is a process for systematically improving the controllability of building systems. Existing-building commissioning often leads to ongoing commissioning, the focus of which is to ensure measures persist (i.e., there is no degradation in performance) and to identify and implement additional measures.

A whole-building-simulation model created using IPMVP Option D can be a useful tool for testing the persistence of measures during the post-implementation monitoring period or ongoing-commissioning cycle. BAS trend data are compared with whole-building-simulation-model hourly outputs. If a measure is found to have degraded or to be operating unexpectedly, the commissioning provider investigates, and the whole-building-simulation model can be used to estimate the impact on energy savings. Once issues have been identified, information can be communicated to the building owner.

Final Thoughts

More often than not, highly detailed whole-building-simulation modeling does not fit within the constraints of an existing-building-commissioning-project budget. In such a case, it is the project manager’s responsibility to manage the expectations of the client. It also is the project manager’s responsibility to manage the energy analyst’s time and effort to meet the needs of the project while staying on budget. Following are tips for limiting the modeling effort:

• Allow extra time, especially if a calibrated whole-building-simulation model is required. This will help manage client expectations.

• Prioritize needs. How much detail is needed? What measures will be analyzed? Will the model be used only to estimate energy savings, or will it be used as part of M&V work?

• When calibrating, categorize the model’s input data as “known” or “unknown.” Focus on the known items first.

• Simplify systems or equipment that will not be investigated or is in the “unknown” category.

Following these rules, one can create an energy model that provides sufficient flexibility and reliable results and meets the needs and budget of a project.


1) Effinger, M., Anthony, J., & Webster, L. (2009, November). Case studies in using whole building interval data to determine annualized electrical savings. Paper presented at Ninth International Conference for Enhanced Building Operations, Austin, Texas. Available at

A project manager for SSRCx, the facilities-commissioning division of engineering-design-and-facility-consulting-firm Smith Seckman Reid Inc., Christopher M. Morales, LEED AP O+M, provides new-construction and existing-building commissioning, energy- and water-auditing, and LEED-project-management services to building owners. He has five years of experience in the commercial-buildings energy-conservation industry, during which he has gained extensive experience in energy modeling and trending data analysis.

Did you find this article useful? Send comments and suggestions to Executive Editor Scott Arnold at