LIVE image

On this Department of Veterans Affairs Omaha VA Medical Center, energy modeling was used from the outset to analyze nine different massing schemes down to three more schematic schemes, then throughout the design of the selected scheme to optimize building massing, mechanical systems, daylighting, and onsite renewable energy usage.On this Department of Veterans Affairs Omaha VA Medical Center, energy modeling was used from the outset to analyze nine different massing schemes down to three more schematic schemes, then throughout the design of the selected scheme to optimize building massing, mechanical systems, daylighting, and onsite renewable energy usage. Image – Leo A Daly

The green building industry focuses far too much on energy modeling to predict performance, not to make early design decisions.

Let me put my headline question another way: Is prediction of building performance the highest use of energy modeling during building design?

This question came to mind after I sent an email to our members this morning about our upcoming (July 9th) webcast on Energy Modeling for Early Design Decisions. One individual responded to me and questioned the credentials of the panel of experts we assembled for this roundtable discussion.

He wrote: “Until you have an energy model that is a good prediction of the way a building operates and the amount of energy used (and is verified by a independent 3rd party) then one needs to be wary” about claims of expertise in modeling.

There is a lot of value in an energy model that reflects the building “as built”: validating LEED points, validating code compliance, and setting the stage for measurement and verification.

But, as I wrote recently in the EBN article, Energy Modeling: Early and Often, “after the massing, orientation, envelope and glazing design, and mechanical systems in a building are already specified, and hundreds of hours of work have already been put into those designs—the modeling might have little value beyond keeping score.” With a quote from Marcus Sheffer, who will be on our webcast panel (and who  wrote the book on energy modeling in an integrative design process), I went on to say:

“It blows me away that that’s where we are,” says Marcus Sheffer, an energy consultant with 7group. As critics have pointed out, a 'green' building modeled to save a certain amount of energy doesn’t necessarily end up doing that. Given accurate inputs, models are accurate at forecasting energy use, says Sheffer, but “models can’t accurately predict the future”: actual operating conditions will always differ from modeled conditions. This typically happens because equipment and controls are installed differently than modeled, or because weather patterns or occupancy are different than expected. The real value of modeling is not predicting energy use but making relative comparisons among design options, says Sheffer.

I agree with Sheffer. There is way too much focus put into the energy model and how predictive it is of performance, and not enough put on iterative modeling during early design to choose between designs A, B, C, D, and so on.

Our most stringent energy codes and prescriptive guidelines today get us no more than 40% savings over common practice.  What we need are 50%, 60%, and greater savings, and we need early modeling to get there.

Do you agree? Disagree? Why? I hope you’ll join me and Marcus Sheffer, Troy Hoggard, Amanda Bogner, and Prasad Vaidya for a lively discucssion at our July 9th, 1:30 p.m. ET webcast on Energy Modeling for Early Design Decisions. It’s free to sign up.

If you enjoyed this article, sign up for BuildingGreen email updates:



1 unfortunately, you are right posted by Brent Eubanks on 06/20/2013 at 09:52 pm

What you say - that energy models are used for scorekeeping, rather than for making design decisions - is absolutely in line with my experience.  Worse, gaming of these models seems to be rampant.  There are far too many ways - many of them legitimate - in which the model can diverge from reality; the result is that it's very easy to get any required result, while the manipulations required to achieve this are hidden deep in the details of the model, where no one will see them without an extensive (and expensive) model audit.

This is an unsurprising outcome of how these models are used - as a proxy for future performance, and a basis for awarding accolades.  Given what's at stake, the fact that these models can be abused means that they necessarily will be.

As far as I can see, there is no real way to fix this.  Standards bodies can come up with increasingly sophsiticated modeling rules in an effort to stymie gaming, but this is just an arms race with the less-ethically-constrained members of our industry, and it is one that the good guys must necessarily lose: every time you increase the complexity of the modeling process by adding yet another set of rules, it creates yet another set of potential loopholes or tricks that can be exploited.  Complexity in this case is the opposite of transparency.

The only solution that I can see is to get away from using models as the basis for code approval or awards.  Base the assesement of building performance on actual measured, recorded data, and let models go back to what they are actually good for - being design tools.

2 Parametric Solutions posted by Hernando Miranda on 06/24/2013 at 05:03 pm

Energy models should be used as parametric analysis tools. They are good at comparing different ECMs. Mr. Sheffer notes this, and people long involved know this, but the concept is lost on people developing LEED requirements.

The USGBC is amassing energy use data for all LEED v3 projects. Evidently to analyze and confirm whether LEED approved energy analyses match actual use data. That will never happen, except by luck. Results that match are as suspect as results that do not.

The real problem with predictiing energy use are process loads and schedules of equipment use. Existing energy modeling software doesn't work well with the former. Running parametrics on schedules would not be difficult. Running parametrics on process loads is very difficult. Process loads are the gray zone of energy analyis.

3 Huzzah! posted by Graham Irwin on 06/24/2013 at 04:11 pm

I thank you for your insightful article and couldn't agree more. Likely beyond the scope of this discussion (but closely related) is the fact that many building rating criteria (California's Title 24, HERS, LEED, etc) are likewise geared toward "compliance" rather than "performance." As such, they compare the design to itself rather than an objective performance level, yielding little useful information on routes to performance improvement.

In our firm, we utilize Passive House energy modeling software (PHPP) and have enhanced it to enable extensive parametric analysis of our projects. We do this as standard practice and consider it the only way to get a good sense of what the building "wants" for cost-optimized performance. More info here I shall endeavor to attend your upcoming discussion.


Graham Irwin, principal

Essential Habitat

4 Models posted by Larry Spielvogel, PE, FASHRAE on 06/26/2013 at 09:59 pm


All Energy Models are wrong.  Some are useful.


Larry Spielvogel, PE, FASHRAE

5 As-built energy models hard to sell posted by Renee Azerbegi on 06/26/2013 at 10:52 pm

I agree with Marcus that modeling should be used to make early design decisions only. And that someday when the US adopts European standards we will be just doing energy models to predict EUIs and not comparing one ficticious model to the next.  We call an end of design LEED model and an as-built energy model "apples to oranges". The New Building Institute should not have compared the end of design model to utility bills for example as they were not meant for comparison. The problem is that the USGBC never required an end of construction model so no one updates the model post construction unless there is a major change. Trying to sell an "as-built" energy model to clients has been challenging since LEED doesnt require it and the owners dont see value in it once there are utility bills to review instead, particularly since you can't really do an "as-built" until three to six months post occupancy once plug loads and actual occupancy schedules and HVAC schedules are determined. The M&V credit doesnt go far enough in selling the "as-built" model or making results useful for owners.

Renee Azerbegi, President

Ambient Energy

6 So what is the value to an as-built energy model? posted by Brent Eubanks on 06/27/2013 at 02:41 pm

I find that I agree with those who question the value of an as-built energy model.  While it would be nice to have, I can't see a justification for spending the time/money required in most cases.  Once you have enough data to calibrate the model, you could just proceed with a M&V process based on real data and ignore the model.  Why rely on a computer model when you have the highest-fidelity model possible (i.e. the real world) available?

It's only value would seem to be scorekeeping.  I think we do far too much scorekeeping as it is.  I'd rather spend that effort in the design process.

7 As a Design Tool posted by Waleed AlGhamdi on 07/01/2013 at 11:05 am

Much has been said about using energy models as design tools, but I doubt that will show any accurate results either. I've been trained as an architect, and have worked for sometime as one. I've switched to energy modeling a couple of years ago, but intend to return to practicing architecture down the road. When I model buildings, I often think about how can I make use of this tool had I been designing the building (architectural design, that is).

I still need a long way to go in order to fully grasp energy modeling, but when it comes to energy consumption, all architectural design strategies that can be used in a building to affect energy consumption are "passive"; i.e. making use of the available energy and other factors in the environment (sunlight for increased illumination and decreased heating and cooling loads, proper choice of massing, optimized fenestration selection, etc). However, these decisions don't need a sophisticated energy model, like eQUEST. The variables are so many, that it is not practical to explore them all. 

In our work, we sometimes compare different alternatives for HVAC systems in the early stages of the design to present to a client. However, even this exercise is so early in the process that many of the architectural features aren't finalized. In addition, the HVAC design isn't finalized either, and there's a lot of guessing and assuming that could -in extreme situations- render the initial model worthless, had the assumption been all different than what the design team decided to do at the end. 

In some of the projects we handled, be it LEED or code compliance models, we sometimes get involved early. We have had luck in some cases where we advise the design team to make a change in an area where we can see they're hurting, and we are often successful in convincing the design team to make the change. For LEED projects, we are not as successful, as the model is often done later after the design has been finished, and in some cases it's down half way through construction. In those cases, the whole exercise seems like a big loss with a lot of opportunities missed.

However, when you get involved early on with schematic design or even design development, keeping track of all changes and revisions of the design can prove difficult, and sometimes impossible, unless you start from scratch every time a change has been made to the design. 

As an architect, I often find only a few optimal alternatives for massing of a building or group of buildings (unless it's a major development or a campus). So there's not a lot to begin with. Creating an energy model may help choose one alternative over the other, but I don't see much use of a "whole building model" beyond that, unless we are talking about multiple levels of modeling on different platforms to study different aspects of the building, which makes the whole issue even more laborious and complicated (e.g. using COMFEN for fenestration vs. eQUEST for the whole building). I question the feasibility of taking this route. 

I hope to see some of these issues touched upon in the upcoming webinar. 



8 Just like MPG... posted by Christopher Schaffner on 07/05/2013 at 10:10 am

...your actual mileage may vary.

Why should we expect building energy models to be different? Absolute energy numbers depend on occupant behavior, not physics. 

9 Hubris posted by Russell Higgins on 12/05/2013 at 11:12 pm

A building is created, built, run, impacted, used and abused by dozens, hundreds, and even tens of thousands of the least predictable and controlable forces on our planet.  HUMAN BEINGS.  Yet, I"ve NEVER, in 30 yrs in the industry seen an modeling input accounting for this extreme variability, well, there is the odd program you pick your best guess as to infiltration.  Not sure what we need to address this, database of buildings indicating variance to modeling along with analysis of the human factors that caused the model to fail.  In depth studies of our clients building users, occupants, etc effect modeling results? 

For me, basic sustainable design isn't something that needs to be modeled, modeling sub systems to fine tune design is useful.

Add comment

Login or register to post comments

— Share This Posting!

About the Authors


Recent Discussions

posted by atwilson
on Feb 3, 2016

I believe the Viridian decking we installed is 15/16" thick by 2-1/2" wide; I'm not seeing those dimensions on the Viridian...

posted by Meredith_1
on Feb 3, 2016

Hi there! I've been sourcing reclaimed exterior deck material for a client, and am interested to know:

The thickness of the material you...

posted by tristan
on Jan 29, 2016

Erica, 3-way LED bulbs are sold in stores and online. Just do a search or ask at your local hardware store.

Recent Comments

Saving Wood from the Landfill, Without the Supply Issues

Alex Wilson says, "Meredith, I believe the Viridian decking we installed is 15/16" thick by 2-1/2" wide; I'm not seeing those dimensions on the Viridian website now,..." More...

Meredith Williams says, "Hi there! I've been sourcing reclaimed exterior deck material for a client, and am interested to know: The thickness of the material you used..." More...

Beat the Bulb "Ban": LED Replacement Lamps in a New Light

Tristan Roberts says, "Erica, 3-way LED bulbs are sold in stores and online. Just do a search or ask at your local hardware store." More...

Erica E. Tactical says, "Anyone know a way to replace incandescent 3-way light bulbs with LEDs? My table lamps are all 3-way and I like the ability to have ambient or..." More...

Autoclaved Aerated Concrete (AAC): Will the U.S. Ever Lighten Up?

Simon Hatton says, "Merhis Projects are an Australian Developer and Construction company and have been using their patented Merhis Building System for the last 4 years...." More...