I was green, fresh out of college, and had just planned my first building. My boss simply asked, "What is the efficiency of this building"? I had no idea. "Let me check...Ill get that to you as soon as possible"

12 grueling hours, a sloppy excel sheet, and a CAD file with far too many polylines later, the answer was 79.5% efficient. "Why is it so **inefficient**?" I never stopped to consider what the metric actually meant.

**Building efficiency is the ratio of usable area to non-usable area**. In multifamily its generally a ratio of net-rentable area to all other areas. Today I will go into the weeds of multifamily efficiency--one of the building optimization metrics.

**Part One: Consider the Building Components**

Time to roll out the building planning legos...

For this test I built an interactive spreadsheet for the local efficiency of theoretical building components. Using two variables, unit depth and corridor width, we can compute a huge amount of data for a dozen building components. Dead-end corridors are the most efficient building component, quickly followed by double-loaded corridors and 90 degree turns. None of this is huge news, until we look at how inefficient it is to wrap and line a parking structure. Splitting corridors from the corner of a garage is the least efficient component.

That plan I drew years ago? It had four **Wrap C (76%)** conditions (see diagram above) and one **Liner D (73%)** conditions per level. This is why the building was inefficient.

**Part Two: Getting More Efficient**

Let us assume I am planning a project with 27' deep units and 6' wide corridors. I currently am under pro-forma by about 1,000 NRSF. I should consider using a 5' corridor. This boost in 1.66% efficiency for a 100,000 NRSF project could gain up to 1600 NRSF.

**Part Three: Diminishing Returns**

I ran the math from a 2' deep unit (would never happen) to a 40' deep unit (does this happen?). Boosting the unit depth has diminishing returns, but meaningful gains if boosting the unit depth a couple of feet is possible.

**Part Four: Using Residential Engine to Test Building Efficiency**

I did 120 site plan iterations on the same property in Residential Engine. This generated a massive amount of data to use for backing up the "building component" theory above.

Here is the spread of efficiencies. As low as **79.78%** and as high as **85.16%**

Site Plan Scheme 36 is the **least efficient design at 79.78%:**

Site Plan Scheme 231 is the **most efficient design at 85.16%:**

**Part Five: Using TestFit for "Advanced Metrics"**

We built out an export function to tell us several atypical metrics from the 120 plans generated above:

- Percent of Double Loaded Corridors
- Number of 3 and 4 way intersections
- Linear feet of corridor

No shock here, but designing with double-loaded corridors improves efficiency.

This also was not much of a shock, but more intersections generally means a more inefficient building.

Similar to the one above, but more a pronounced trend:

Any thoughts? Test Fit is an instant multifamily site planning application. Send me an email: clifton@testfit.io