Beyond Energy Modeling: A New Paradigm for Performance Analysis

An earlier version of this essay appeared on www.sefaira.com and on the SketchUp Blog.

The traditional energy modeling paradigm is fundamentally misaligned with the design process, and incapable of delivering on the promise of high performance. The alternative is a completely different way of thinking about performance — what I call "Performance-Based Design."

Energy Modeling Isn't Working for Design

When architects think of “energy modeling,” they think of a complex, arcane activity involving energy modeling specialists (in-house experts or consultants), number-laden reports, and turnaround times on the order of days or weeks. Energy modeling is a slow, one-design-at-a-time, inputs-first paradigm. Every part of that description is problematic — and it's worth diving deeper to understand why.

1. Slow: Traditional energy modeling tools are complex and require many inputs, which means that even experts need significant time to set up and run a model. Design, on the other hand, moves at breakneck speed, particularly in the early stages when the big, impactful decisions are being made. The architect simply has no time to stop designing and do energy modeling as a separate process.

2. One design at a time: Design is about exploring a design space to find the best solution. Energy modeling is about simulating a singular design. As a result, it tends to lead to refinements on a design rather than informing fundamental design moves.

3. Inputs-first paradigm: Energy modeling tools require numerous inputs before they produce any results. Not only does this demand significant time and expertise, it requires that the architect has a fully-formed design before the process can even begin. It comes after design — and by that time it’s too late.

The result is a mismatch between the process of design and the process of energy modeling. This is largely because energy modeling tools were not intended to be architectural design tools. They are useful for designing mechanical systems and performing validation analysis — and for these uses, the energy modeling paradigm will continue to be a necessary part of the overall design process.

But the energy modeling paradigm is about as far removed from the creative, iterative, real-time nature of design as you can get. If we want a tool that can meaningfully inform design — that can have significant impact on energy consumption, operating cost, and the capital costs of the HVAC systems; that is useful from Schematic Design through Construction Documents, when architects are actually making design decisions — then we need a new approach.

How Architects Work

When I set out to solve this problem at Sefaira, I spent significant time with the product team talking with architects to understand their processes and workflows. We found that design moves incredibly quickly; that architects spend most of their time in their 3D modeling or BIM environment; and that their overhead of switching between applications is high.

In addition, we discovered that the performance-related questions architects ask during design are very different from the questions that energy modeling is instrumented to answer. Architects wanted to know:

1. What’s important? What elements of the design should they focus on? Where are the biggest opportunities to have an impact on performance? Is it a tighter envelope, preventing solar gains, or providing good daylight?

2. What are the constraints? Designers want to know the constraints and boundaries of the problem. If they know early on that south shading is critical, or that a 55% glazing ratio is ideal, or that a 50-to-60-foot floorplate provides a great balance of daylighting and energy performance, they can use these as meaningful constraints to inform the design.

3. Is it possible to do X? Can the design be naturally ventilated? Could passive solar address most of the heating needs? Could we meet all energy needs from on-site renewables?

4. How does my design compare? — to other design options, to typical buildings in my area, to an ASHRAE 90.1 baseline, to 2030 Challenge goals? When exploring the design space, comparisons are critical.

These are the questions that the new approach must answer.

A New Approach: Feedback, Not Analysis

The alternative to energy modeling is a real-time, multiple-options-at-a-time, results-first paradigm in which seamless performance feedback informs design decisions throughout the design process. It can't depend on a finished design; it needs to help designers explore the design space and understand the field of possibilities. At Sefaira, we termed this approach "Performance-Based Design."

If energy modeling is a slow, one-design-at-a-time, input-heavy approach, Performance-Based Design is the opposite: a real-time, multiple-options, results-first paradigm. Each of these capabilities is critical for aligning performance feedback with the design process.

1. Real-time. Sefaira provdes immediate feedback on energy, daylight, and other metrics, with a strong preference to delivering these in the architect’s native design environment. We want feedback to be always available to the designer — a fundamental departure from the idea of “analysis” as a separate, after-the-fact activity.

2. Multiple options. Sefaira is built on the idea of exploration — of testing out different ideas, whether through real-time analysis within the 3D modeling environment, via comparison of options or iterations, or by quickly investigating the impact of many different design strategies. Exploration is key for answering design-oriented questions like “What’s important in my design?” and “Which option is best?”

3. Results-first. With only two inputs — space use and location — architects can start seeing feedback on design decisions. The feedback is designed to be actionable — helping designers understand how the building is operating and what they can do to improve. This allows designers to shape the design based upon results, rather than simply analyzing a design that’s already complete.

What’s revolutionary about this approach is that it makes achieving performance a creative endeavor — no different from meeting the constraints of program, context, and site that have inspired architects for generations. By making performance accessible, it opens up a new design space, and empowers architects to address performance as a standard part of the design process. This is the future of performance: a seamless, essential part of design.

Using Performance to Inform Design

What does it mean for analysis to inform design? Here are a few ways of going about it:

Exploratory modeling: Explore building form, floorplate dimensions, glazing ratios, etc. with simple models before moving on to specific designs (i.e. prior to sketching & ideation). Understand what’s driving energy use and daylight potential, and what elements of the design are important for moving these in the right direction. These elements can become design constraints — inputs for creating an actual design.

Design comparisons: Quickly compare different design options to understand the type and relative magnitude of performance impacts. Comparisons can help to inform design decisions large and small — from massing shape to façade design to shading strategies.

Sensitivity analysis: Use parametric analysis to see which design parameters and passive strategies have the biggest impact on performance. This early knowledge makes it possible to integrate these strategies as core aspects of the design. (Read more about how to do this analysis in my post on Sensitivity Analysis in Concept Design, or in this post on Predictive Modeling.)

The Performance Based Design approach is already aligned with how architects work — but like any new practice or technology, it requires effort (and strong leadership) to transition from “business as usual.” For more on getting started with a performance-based design approach, see my post on Integrating Performance Analysis Into Your Design Process, which provides five recommendations for starting to do analysis in a fast, lightweight way.