University of Southern California Dissertations and Theses

Façade retrofit performance evaluation: Predicting energy conservation potential from renovations to the building envelope

(USC Thesis Other)

Façade retrofit performance evaluation: Predicting energy conservation potential from renovations to the building envelope

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Façade Retrofit Performance Evaluation
Predicting Energy Conservation Potential from Renovations to the Building Envelope

by
Kenya I. Collins
_______________________________________________________
A Thesis Presented to the
FACULTY OF THE
SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA

In Partial Fulfillment of the
Requirements for the Degree
MASTER OF BUILDING SCIENCE

August 2018
2

COMMITTEE
Douglas Noble, FAIA, Ph.D.
Director, Master of Building Science
School of Architecture
dnoble@usc.edu
213 740 2723

Anders Carlson, S.E., Ph.D.
Adjunct Assistant Professor
School of Architecture
andersca@usc.edu
213 740 1054

Lucio Soibelman, Ph.D.
Department Chair, Sonny Astani Department of Civil and Environmental Engineering
Viterbi School of Engineering
soibelman@usc.edu
213 740 0609

3

ACKNOWLEDGEMENTS

I would like to express my deepest appreciation to the University of Southern California, School
of Architecture Master of Building Science faculty for supporting this research; namely,
committee chairs Douglas Noble and Anders Carlson. However, there are a number of professors
and volunteer review board members who offered input, knowledge, and support. Thank you to
Karen Kensek, Joon-Ho Choi, and Kyle Konis. Additionally, special thanks to Lucio Soibelman
of the Viterbi School of Engineering.
Furthermore, I’d like to express my deepest gratitude to the family and friends who supported me
through my academic career. This encouragement inspired me to continue working throughout
challenging times. Because of this community, I am able to follow the sparks in my curiosity that
drive design and innovation of spaces and environments for the world.

4

CONTENTS
COMMITTEE ............................................................................................................................... 2
ACKNOWLEDGEMENTS ......................................................................................................... 3
CONTENTS................................................................................................................................... 4
LIST OF FIGURES ...................................................................................................................... 7
LIST OF TABLES ........................................................................................................................ 9
ABSTRACT ................................................................................................................................. 10
Chapter 1: Introduction ............................................................................................................. 12
1.1 A Contemporary Approach to Façade Retrofit .............................................................. 12
1.2 Significance of the Building Envelope........................................................................... 13
1.3 Evaluating Enclosure Performance ................................................................................ 14
1.4 Terms .............................................................................................................................. 20
1.5 Digital Technology to Support Façade Retrofit ............................................................. 20
1.6 Scope of Work ................................................................................................................ 21
1.7 Chapter Structure............................................................................................................ 22
Chapter 2: Previous Work and Document Review .................................................................. 24
2.1 Chapter Overview .......................................................................................................... 24
2.2 Relevant Organizations, Codes, and Standards .............................................................. 24
2.3 Contemporary Envelope Renovation Strategies ............................................................ 24
2.4 Predicting Retrofit Potential using Digital Tools ........................................................... 27
2.5 Documentation of Physical Conditions using Digital Methods ..................................... 29
2.6 Precedents....................................................................................................................... 29
2.7 Conclusions and Takeaways .......................................................................................... 30

5

Chapter 3: Methodology............................................................................................................. 31
3.1 Chapter Overview .......................................................................................................... 31
3.2 Initial Analysis ............................................................................................................... 32
3.3 Mathematical Model ...................................................................................................... 35
3.4 Simulation ...................................................................................................................... 38
3.5 Trials............................................................................................................................... 44
3.6 Intended Use of Results.................................................................................................. 45
3.7 Chapter Summary ........................................................................................................... 46
Chapter 4: Results ..................................................................................................................... 47
4.1 Chapter Overview .......................................................................................................... 47
4.2 Mathematical Analysis ................................................................................................... 47
4.3 Results of Simulation ..................................................................................................... 60
4.4 Chapter Summary ........................................................................................................... 82
Chapter 5: Interpretation ........................................................................................................ 83
5.1 Chapter Overview .......................................................................................................... 83
5.2 Interpretation of Mathematical Analysis ........................................................................ 83
5.3 Interpretation of Simulation ........................................................................................... 88
5.4 Alignment with Industry Objectives .............................................................................. 92
5.5 Facilitators and Obstacles............................................................................................... 93
5.6 Chapter Summary ........................................................................................................... 95
Chapter 6: Conclusions and Future Work ............................................................................ 96
6.1 Conclusions .................................................................................................................... 96
6.2 Improvements and Future Work .................................................................................... 98
BIBLIOGRAPHIC REFERENCES........................................................................................ 100
APPENDIX A: CBECS Data Descriptions ............................................................................. 103
6

APPENDIX B: MATLAB Script ............................................................................................. 105
APPENDIX C: Results ............................................................................................................. 108

7

LIST OF FIGURES
Figure 1.6-1 ................................................................................................................................... 22
Figure 3.1-1 ................................................................................................................................... 31
Figure 3.2-1 ................................................................................................................................... 32
Figure 3.3-1 ................................................................................................................................... 35
Figure 3.4-1 ................................................................................................................................... 38
Figure 3.6-1 ................................................................................................................................... 45
Figure 4.2-1: Census Division Histogram .................................................................................... 48
Figure 4.2-2: Exterior Wall Construction Histogram ................................................................... 49
Figure 4.2-3: Roof Construction Histogram ................................................................................. 49
Figure 4.2-4: Roof Tilt Histogram ................................................................................................ 50
Figure 4.2-5: Building Shape Histogram ...................................................................................... 50
Figure 4.2-6: Glass Percentage Histogram ................................................................................... 51
Figure 4.2-7: Characteristics Decision Tree ................................................................................. 52
Figure 4.2-8: Building EUI Histogram ......................................................................................... 54
Figure 4.2-9: Annual Electricity Consumption Histogram ........................................................... 54
Figure 4.2-10: Annual Major Fuel Consumption Histogram ....................................................... 55
Figure 4.2-11: Electricity Use for HVAC Histogram ................................................................... 55
Figure 4.2-12: Major Fuel Use for HVAC Histogram .................................................................. 56
Figure 4.2-13: HVAC Electricity Use vs Wall Construction Material ......................................... 57
Figure 4.2-14: HVAC Electricity Use vs Roof Construction Material ......................................... 57
Figure 4.2-15: HVAC Fuel Use vs Wall Construction Material .................................................. 58
Figure 4.2-16: HVAC Fuel Use vs Roof Construction Material .................................................. 59
Figure 4.3-1: Test Building Plan ................................................................................................... 61
8

Figure 4.3-2: Test Building Axonometric View ........................................................................... 61
Figure 4.3-3: Test Building Elevations ......................................................................................... 62
Figure 4.3-4: Test Building Energy Model ................................................................................... 63
Figure 4.3-5: Wall Type 1 and Wall Type 2 ................................................................................. 64
Figure 4.3-6: EUI for Sample C .................................................................................................... 66
Figure 4.3-7: EUI for HVAC for Sample C .................................................................................. 67
Figure 4.3-8: Energy Cost per Square Foot for Sample C ............................................................ 67
Figure 4.3-9: New Orleans Total Building EUI............................................................................ 68
Figure 4.3-10: New Orleans Lifecycle Energy Cost ..................................................................... 69
Figure 4.3-11: New Orleans Annual Fuel Use for HVAC............................................................ 69
Figure 4.3-12: New Orleans Annual Electricity Use for HVAC .................................................. 70
Figure 4.3-13: New Orleans Annual HVAC Operating Cost ....................................................... 70
Figure 4.3-14: New Orleans Percent Difference........................................................................... 71
Figure 4.3-15: Los Angeles Total Building EUI .......................................................................... 72
Figure 4.3-16: Los Angeles Lifecycle Energy Cost...................................................................... 72
Figure 4.3-17: Los Angeles Annual Electricity Use for HVAC ................................................... 73
Figure 4.3-18: Los Angeles Annual Fuel Use for HVAC ............................................................ 73
Figure 4.3-19: Los Angeles Annual HVAC Operating Cost ........................................................ 74
Figure 4.3-20: Los Angeles Percent Difference ........................................................................... 74
Figure 4.3-21: Louisville Total Building EUI .............................................................................. 75
Figure 4.3-22: Louisville Lifecycle Energy Cost ......................................................................... 76
Figure 4.3-23: Louisville Annual Electricity Use for HVAC ....................................................... 76
Figure 4.3-24: Louisville Annual Fuel Use for HVAC ................................................................ 77
Figure 4.3-25: Louisville Annual HVAC Operating Cost ............................................................ 77
Figure 4.3-26: Louisville Percent Difference ............................................................................... 78
9

Figure 4.3-27: Chicago Total Building EUI ................................................................................. 79
Figure 4.3-28: Chicago Lifecycle Energy Use and Cost .............................................................. 79
Figure 4.3-29: Chicago Annual Electricity Use for HVAC .......................................................... 80
Figure 4.3-30: Chicago Annual Fuel Use For HVAC .................................................................. 80
Figure 4.3-31: Chicago Annual HVAC Operating Cost ............................................................... 81
Figure 4.3-32: Chicago Percent Difference .................................................................................. 81
Figure 5.2-1: CDF of Major Fuel Use .......................................................................................... 85
Figure 5.2-2: CDF of Fuel Expenditure ........................................................................................ 86
Figure 5.4-1: HVAC Energy Use.................................................................................................. 92

LIST OF TABLES
Table 3.4-1 .................................................................................................................................... 42
Table 3.4-2 .................................................................................................................................... 42
Table 4.2-1: Test Building Characteristics ................................................................................... 53
Table 4.2-3: Sample B Correlation Coefficients........................................................................... 59
Table 4.2-4: Hypothesis Test Results ........................................................................................... 60
Table 4.3-1: Wall Assembly Typologies ...................................................................................... 63
Table 4.3-2: Glazing Assembly .................................................................................................... 64
Table 4.3-3: Roof Assembly Typologies ...................................................................................... 64
Table 4.3-4: Trial Descriptions ..................................................................................................... 65
Table 4.3-5: Sample C Characteristics.......................................................................................... 66

10

ABSTRACT
Thermal properties of the building enclosure affect building energy consumption due to the
relationship to external heating and cooling loads. Once a building is constructed, these
properties are assumed to be fixed. If renovation of the building enclosure takes place, it is
possible to improve thermal properties of the building enclosure and consequently reduce energy
consumption.
Many organizations within the building design and construction industry have set up goals to
reduce energy consumption and carbon emissions. Most current efforts focus on implementing
technology in the design of new buildings; however, renovating the existing building façade is
another potential method to enhance energy performance. It is possible to use statistical analysis
and digital simulation to explore what types of façade renovations can produce energy savings in
existing commercial buildings.
The domain included one story commercial office spaces that are less than 5,000 square feet. The
following building enclosure characteristics were selected: exterior wall material, roof material,
roof tilt, and window-to-wall ratio. It was necessary to examine existing building audit data from
the Commercial Building Energy Consumption Survey (CBECS). Statistical methods were used
to create a profile for how buildings within the domain perform in terms of electricity and major
fuel consumption. In addition, linear correlation and curve-fitting/goodness of fit tests predicted
the likelihood that the consumption data followed a known pattern.
A 1 story, 5,000 square foot structure was modeled in Revit as the sample building. Façade
renovation was simulated on the test building by incorporating at least one of three proposed
energy conservation methods. The energy conservation methods were alterations to the windows
(ECM-I), walls (ECM-II) or roofing (ECM-III) materials that could be done on an existing
building.
These energy conservation methods were implemented over a series of 5 trials. Trial 1 simulated
the current performance of the test building and established a baseline. Trial 2 replaced single
pane windows with double pane insulated windows. Trial 3 added continuous exterior insulation
onto the existing brick wall and replaced windows. Trial 4 maintained pre-existing wall
conditions but modified insulation in the roof and installed a cool roof. Trial 5 was an all-
inclusive simulation of the three energy conservation methods.
The selected locations for simulation, New Orleans, Los Angeles, Louisville, and Chicago,
represent ASHRAE climate zones 2 through 5, respectively. Geometry was modeled in Revit.
Green Building Studio and EnergyPlus estimated annual electricity and fuel consumption and
expenditure for space heating and cooling in the test building pre and post façade renovation.
11

Generally, values for energy performance calculated by Revit were significantly less than values
reported from CBECS data. An increase in performance was measured by a reduction in total
EUI, electricity consumption, fuel consumption, or equivalent carbon emissions.
The percent difference in energy performance between the baseline trial and the trial showing the
lowest building EUI was reported for each test city. Overall, the greatest reduction in electricity
use came from ECM-I and ECM-II. The effect of ECM-III was negligible in its ability to reduce
electricity and fuel consumption.
It was expected that simulation would produce lower values for electricity and fuel consumption.
However, primary interest was placed on the percent difference in consumption and expenditure
values before and after the façade renovation. According to the simulation, the most valuable
façade-specific energy conservation methods involved alterations to the windows and wall
assemblies.

Key terms: building envelope, retrofit, façade renovation, energy efficiency, energy conservation
measures, façade engineering.

12

Chapter 1: Introduction
1.1 A Contemporary Approach to Façade Retrofit
The building enclosure is the façade, or face, and envelope, or skin, of a structure. As a building
ages, the enclosure may require maintenance or renovation to meet performance criteria.
Whether the criteria are for aesthetic, structural, or energy performance, an investigation is
necessary to understand the current condition of the enclosure. Traditional methods for
investigation involve visual surveys and destructive testing to begin to understand current
conditions. With the increasing potential of technology, it is possible to support investigation of
the building enclosure performance through the use data from simulations. Furthermore, it may
be possible to develop strategies for improved building enclosure performance without
conducting a deep energy audit of the structure. Digital tools in use take into account physical
characteristics of a building to approximate building enclosure performance. Simply put, a
contemporary approach to façade retrofit can be a process in which simulation tools aid in the
early stage of investigation to provide designers with a summary of existing conditions and
potential options to explore.
Goals for energy performance and reduction in carbon emissions set up by government agencies
and industry organizations are not currently within reach. According to the U.S. Energy
Information Administration, roughly half of the building stock that will be in use in 2050 is
composed of already existing structures (U.S. Environmental Protection Agency 2012). In order
to meet demands for increased building performance, existing buildings will need to be
renovated. To begin the renovation process, designers need to know how a building is currently
performing.
A methodology was established to assess the thermal performance of an existing building
enclosure. It details what parameters to consider, potential failure mechanisms, and targets for
the performance of the façade. While it does not give prescriptive measures on how to conduct
an investigation of a building façade, it offers suggestions for what parameters may be beneficial
to explore. Focus is placed on building energy performance as a result of characteristics of the
building enclosure.
1.1.1 Hypothesis Statement
A simplified estimation of building thermal performance can be used to establish benchmark
standards for existing building enclosures. Buildings with similar physical characteristics can be
grouped in order to develop a representative model. Data from energy audits can be used to
create a profile of the model groups. This profile is expected to reflect the performance of other
buildings that meet the same criteria as the model group.
13

Additionally, it is possible to measure improved performance of a building by making alterations
to the exterior walls, windows, or roof. This improvement can be measured from energy audit
data or approximated by simulation. It is expected that energy audit data will produce a lower
percent difference in energy performance than simulation data because simulations often cannot
capture many other variables, such as thermal leaks and infiltration, that may affect enclosure
performance.
1.2 Significance of the Building Envelope
1.2.1 Aesthetic
The exterior façade of the building is an occupant’s first introduction to the structure. It is
essentially the building’s face or first impression. Architects uphold the aesthetic appearance of
the building; it is a reflection of design intent. The value of aesthetic appearance was considered
in the façade retrofit process because it is a potentially limiting factor to the retrofit design
options.
1.2.2 Performance: The Building Skin
Building skin is another term for the exterior envelope of the building. Using skin as a reference,
the building envelope can be thought of as the outermost layer of defense against the exterior
elements. The skin is expected to protect occupants and internal components and perform in the
following ways (T. J. Kesik 2014):
• Strength and rigidity of structure
• Stability and durability of materials
• Control of thermal energy transmissions (heat flow)
• Control of air flow
• Control of water vapor flow
• Control of liquid water movement
• Control of acoustic transmissions
• Regulation and distribution of light
• Aesthetics
• Ease of construction
• Cost
Overall, the expectation of the façade is more than a face to look at; rather, it is a functional
organ in the building anatomy.
14

1.3 Evaluating Enclosure Performance
1.3.1 Principles of Enclosure Performance
Designers evaluate the performance of a building enclosure through durability, redundancy,
functionality, continuity, and practicality.
Durability refers to the resilience of an enclosure; its ability to resist damage from external
phenomena such as weather, heat, or impact. Redundancy is a characteristic that supports the
durability of the enclosure. It describes the location and frequency of repetitive measures that
provide protection (T. J. Kesik 2014).
Functionality refers to the purpose of each material used in an enclosure assembly. A material
may have various functions. For example, a material used to restrict air transport may also serve
as a moisture barrier. While the function of each material defines its purpose, continuity
throughout the system guarantees that this purpose is carried out across changes in geometry, at
fenestrations, and other transitions along the building façade. An abrupt end to a material can
allow for gaps or bridges where there is no line of defense, such as water leakage at window
seals. For this reason, continuity should be considered in the design and construction of the
building enclosure.
Practicality is the combination of ease of construction and economy. If a particular enclosure
system works to effectively prevent damage from impact but requires a material thickness or
quantity that is unreasonable for the design, then the system fails in performance because it is
impractical. Similarly, systems that do not have feasible methods for construction or exceed the
project budget also do not satisfy practicality as a performance requirement (T. J. Kesik 2014).
1.3.2 Building Envelope Deterioration and the Demand for Retrofit
Three main demands for retrofit related to the condition of materials in the enclosure assembly
were considered: (1) materials have a defined life expectancy; (2) materials can undergo damage
or failure from external phenomena; (3) increased performance requirements prove some
materials or material assemblies to be insufficient or dangerous.
Some materials are expected to be sufficient for the entire lifetime of the structure, while others
have limited usage time and require maintenance or replacement. In addition to the natural
degradation of materials, materials can undergo damage that reduces their lifetime. Damage can
be the result of various failure mechanisms, which will be defined in Chapter 3. A façade
investigation can help to determine that cause of damage.
Finally, increased performance requirements outlined by engineering agencies, such as
ASHRAE, and sustainability agencies, such as Architecture 2030 and Green Build, can make the
standard performance of some commonly used materials obsolete. In addition, new discoveries
15

in science have proven some materials to be hazardous. The most common example of this is the
use of spray on “popcorn” ceiling finishes for acoustic performance that contain asbestos, a
carcinogen (U.S. Environmental Protection Agency 2016).
According to the U.S. Energy Information Administration, roughly 50 percent of the building
stock that will be utilized in 2050 has already been built. This increases the demand for retrofit if
the building industry aims to meet energy consumption reduction goals outlined by city, state,
and national officials.
1.3.3 Risks and Consequences
There are multiple factors that may contribute as good causes to undergo a retrofit to the building
enclosure. While each of these are not always directly linked to energy performance, they may
affect the enclosure and present the opportunity to consider upgrades to the enclosure.
Structural Fatigue
There are several risks associated with façade elements that are under-maintained, improperly
installed, or become fatigued as a result of stress from external forces. Structural fatigue of
cladding components is a large concern. Structural fatigue can occur as a result of external
forces, such as seismic force or wind pressure, or simply due to improper installation. In the case
of seismic activity, abrupt forces cause the building to accelerate. Each floor or roof plate can
drift a relatively small lateral distance causing stress at anchorage points (Arnold 2009). Wind
pressure is not a constant and continuous force; rather, there are periods of high and low winds
that vary with time. As a result, lateral force resistance members in building experience cycles of
loading and unloading. Over the lifetime of the building, both internal structural components and
external façade elements may become fatigued. In an event where seismic or wind forces exceed
the capacity of anchorage elements for cladding, sections of the cladding can disengage from the
structure and create hazardous scenario of falling or flying objects for surrounding people. An
investigation of the façade can help to determine the potential for or cause of structural failure.
Indoor Air Quality
Indoor air quality is understood as the condition of the air within the building and in close
proximity to building fenestrations and its relationship to occupant health and comfort (U.S.
Environmental Protection Agency 2016). Sick Building Syndrome and moisture-induced
airborne contaminants are two of the most common façade-dependent hazards to IAQ.
Enclosure-related Sick Building Syndrome can be the result of an air-tight façade in conjunction
with little to no fresh air introduced from ventilation. When fuel consumption became a national
concern in 1970, the design new practice was to seal buildings against air infiltration in order to
maximize the efficiency and effect of air conditioning systems (Joshi 2008). These airtight
buildings had low ventilation and, as a result, low air exchange rates; occupants did not receive
16

enough fresh outdoor air. When a building is effectively sealed from air infiltration, there is a
need to modify the ventilation strategy in order to achieve adequate air exchanges to satisfy
occupant health and comfort. To meet this demand, ASHRAE modified standards for the
minimum outdoor air flow rate and required air exchanges (ASHRAE 2013).
Moisture infiltration can affect air quality through the introduction of bacteria and fungus (mold).
Moisture can enter an enclosure system as water vapor or bulk water; exit strategies are based on
the storage, drainage, and drying properties of the assembly. Under certain thermal conditions,
the presence of moisture in an enclosure assembly can be an ideal habitat for bacteria or mold
(Morse and Acker 2014). Populations of such microorganisms can be harmful to occupant
health. In the United States, a certified inspector is required to verify the presence of bacteria or
mold in a building. Investigative openings to the façade can determine the presence and extend
of microorganisms in the building.
Energy Efficiency
Targets established by the Department of Energy, International Energy Conservation Code,
International Green Construction Code, United States Green Building Council, and other
agencies aim to decrease energy consumption, reduce greenhouse gas emissions, and consider
opportunities for passive design strategies in building design. The building façade has potential
to influence energy efficiency through its thermal properties, ability to regulate sunlight, and
ventilation strategies. A retrofit of the façade can enable an existing building to reach
sustainability targets.
1.3.4 Parameters of Interest
Parameters were based on the 2015 U.S. Environmental Protection Agency Commercial
Building Energy Consumption Survey (CBECS). Parameters that are relevant to the building
enclosure were selected. These parameters were intended to be quantified and used as
preliminary conditions and variables of the simulation.
1.3.5 Building Envelope Performance
Performance Categories
The first step is to assess conditions of the building envelope. This allows the building owner to
define the expectations of the building envelope. As previously mentioned, effective enclosure of
a building incorporates principles of durability, redundancy, functionality, continuity, and
practicality. These principles can be explored in a number of ways when defining envelope
performance expectations: (I) structural stability; (II) control of thermal energy flow; (III) control
of moisture/water vapor flow; (IV) control of liquid water movement; (V) regulation of sunlight;
(VI) exterior aesthetic. Ultimately, focus was placed on three main areas: (I) conductive heat
transfer, specifically heat flow through solid wall material; (II) solar radiation, heat flow through
17

glass; and (III) infiltration, heat flow through air via gaps in the enclosure. In addition, cost, ease
of construction, and control of acoustic transmissions are considered valuable to envelope
performance but were not incorporated in analysis.
Each of the performance categories was assigned a metric in order to quantify performance. In
addition, the performance categories can be understood by their relationship to the purpose of
retrofit.
Purpose of Retrofit
The purpose of retrofit describes which performance categories a change to the envelope aims to
address. Three main purposes for retrofit were defined, but it is understood that other strategies
can be implemented to address performance categories. The pre-defined purposes are: (1) reduce
energy consumption from space conditioning such as heating and cooling; (2) reduce infiltration;
(3) increase sunlight distribution into occupied space.
A reduction in energy consumption from space conditioning can be achieved by interventions at
the building envelope in many ways. The main idea is to reduce the heating or cooling loads in
conditioned space by regulating heat transfer. First, it is possible, to a certain degree, to regulate
the transfer of heat through exterior walls (or into any non-conditioned space) by insulation. The
U-value of wall assemblies was used as a metric to quantify the ability to reduce heat transfer by
transmission. The second strategy to reduce thermal loads is to reduce thermal energy lost or
gained by infiltration and ventilation. Air-change rate per hour was used to quantify how
frequently air inside conditioned space was recycled or replaced by outside air. The final method
to reduce thermal loads is to reduce or increase solar heat gain through the building envelope.
Energy consumption relates to performance category (II) control of thermal energy flow.
Regulation of light is addressed when the purpose of retrofit is to increase sunlight distribution
into occupied space. This can be measured by the window-to-wall ratio (WWR) and the use of
electric lighting in spaces contained by at least one exterior wall. The purpose would express
daylighting potential for the building. In addition to supporting category (V), there is potential to
address category (VI) exterior aesthetic.
Level of Intervention
Retrofit to a building envelope can take place all at once or in increments over a longer period of
time. There are benefits and setbacks for either process depending on the building function,
occupancy, typical operation hours, size, age, and the budget of the owner (Patterson, Vaglio and
Noble 2014).
One way to begin to understand which timeline is practical for the building is to define the level
of intervention that a suggested retrofit strategy will have on the building. The level of
intervention describes how invasive a given retrofit strategy can be on normal building
18

operations. Some forms of façade retrofit can take place without having a large effect on daily
operations while others require no occupancy during construction. Some renovations will require
destructive testing as a part of the design phase. Whether it is for investigation or construction,
façade retrofit can put a building out of service for a period of time.
1.3.6 Condition Assessment
Existing strategies for evaluation of buildings include Property Condition Assessment (PCA) and
Facility Performance Evaluation (FPE). Property Condition Assessment is an evaluation
approach geared towards determining real estate value. An inspector observes and rates various
elements of the building based on their current condition. Facility Performance Evaluation is a
type of post-occupancy evaluation that monitors and assesses aspects such as functionality,
accessibility, safety, and cost-effectiveness to support the building commissioning process
(Zimring, Rashid and Kampschroer 2014). An FPE produces more quantitative data than a PCA
because it involves physical measurements, space utilization surveys, operating conditions,
energy use, and surveys to quantify occupant satisfaction. The U.S. General Services
Administration offers guidelines for conducting such investigations. An FPE is a system-based
evaluation strategy while a PCA is a component-based evaluation strategy. Other forms of
condition assessment can be used to determine potential risks associated with existing conditions
in order to provide a explanation for the need to retrofit.
Evaluation methodology that is specific to the façade is necessary for façade retrofit in order to
collect information about the building enclosure and determine which elements of enclosure
influence building performance. For the purpose of façade retrofit, emphasis should be placed on
façade-structure interaction and façade-energy consumption relationships. Condition Assessment
in façade retrofit should resemble some aspects of both evaluation strategies in order to
determine the relationship between components of the façade and mechanical systems that serve
the building.
Property Condition Assessment often focuses on the real estate value of the property according
to the condition of individual elements. This is being used as a starting point for application
development because it is important to understand how individual elements contribute to the
overall performance of the building. It is necessary to improve this system to incorporate
comparisons to baseline standards and references to targets in order to have a relative scale by
which performance can be measured.
1.3.7 Contemporary Investigation Methods
In order to perform a renovation, it is necessary to collect information about the current condition
of the building façade. Conventional methods involve investigators who measure, photograph,
and test specific portions of the building façade. Such investigation may involve nondestructive
19

testing or destructive investigative openings. Many contemporary investigation methods aim to
minimize damage to the existing enclosure.
Photogrammetry
Photogrammetry is the use of photography for measuring the distance between two objects. In
the building industry, photogrammetry is often used to document existing or as-built conditions
in a building (Geodetic Systems 2016).
Thermographic Inspection
Thermographic inspections involve the measurement of surface temperatures of a building
envelope using infrared video or photography (U.S. Department of Energy 2016). The inspection
can help detect air or moisture infiltration in a façade assembly and missing insulation.
Geographical Information Systems (GIS)
The National Geographic encyclopedia defines GIS as “capturing, storing, checking, and
displaying data related to positions on Earth’s surface” (National Geographic 1996).
Building Energy Management Systems
Building Energy Management Systems are currently used to monitor, regulate, and collect data
on power, lighting, and HVAC systems. Many programs offer historic data for users to observe
building energy use over time and report in energy audits, when necessary. Such systems could
be beneficial in a building undergoing façade retrofit to measure energy consumption prior to
and after the façade retrofit takes place.
1.3.8 Digital Support
Digital methods of understanding building thermal performance often include simulations that
require the building to be modeling in computer software. While this technique is beneficial at
producing an approximate understanding of building performance, it can be time consuming and
costly because it requires an experienced designer or consultant.
Post-occupancy data collection using sensors and supporting software is another method of
understanding building energy performance using the support of digital tools. These surveys
often extend long term (often a minimum of one year) and provide data that represents energy
use over the course of the heating and cooling seasons.
20

1.4 Terms
1.4.1 Terminology used in this study
Industry terms used frequently in this study are defined here. Although these terms may have
many implications, it is assumed, for the purpose of this study, that the author is using the term
in the definition prescribed below. The definitions were selected for their broad ability to
summarize the topic, method, or idea in a context relevant to this study.
In addition, see Chapter 2.2 for relevant Codes and Standards.
Façade Renovation
For the purpose of this study, a façade renovation includes any major modification to the roof,
exterior walls, or windows of an existing building.
Energy Conservation Measures (ECM)
An Energy Conservation Measure (ECM) is a strategy for renovation of the building envelope
that is intended to reduce energy consumption for space heating and cooling.
Application
This study used an application program written and operated in MATLAB as the method to
evaluate data and make predictions. The application is intended to be a draft or representative
model to demonstrate how a future stand-alone application or software program can use building
characteristics and historic data to make predictions for a building under consideration for façade
renovation.
1.5 Digital Technology to Support Façade Retrofit
1.5.1 Motivations for this study
There is a demand for retrofit of buildings in the United States. The AEC industry is adapting to
use of digital tools for design, documentation, and analyzing buildings. Façade retrofit opens the
possibility to detect structural fatigue, improve exterior aesthetic, identify locations of
infiltration, and reduce thermal loading conditions in buildings. A study using digital tools to
support renovation design can provide insight on the accuracy of simulation tools and how they
compare to energy audit data. A workflow can be created to initiate investigation for façade
renovation and prepare for design using known performance of similar buildings.
1.5.2 Study Domain
The study aims to support investigations of low to mid rise commercial buildings constructed
between 1950 and 2000 in the United States whose target service life is 30 to 50 years.
21

Commercial and residential buildings which fall into this category are those identified by
International Building Code 2015 Risk Category II.
The study is intended to produce results relevant to commercial office buildings that are 1-story
with roughly 5,000 square feet of occupied space that is thermally regulated by conventional
HVAC systems.
1.5.3 Limitations
Simulation is used for an idealized rough approximation of façade performance. This relies on a
simplified building model and a profile generated from CBECS data. Due to the regulations on
CBECS data, exact building identities are unknown. In addition, the CBECS dataset is not
perfect; some characteristics or consumption values are not reported. As a result of these factors,
the study domain focuses on relatively small-scale structures with common characteristics so that
the representative profile can apply to a large number of existing structures in the United States.
High-rise and super-tall structures were not considered in the study because the size of the
structure challenges the practicality of manual inspection and data entry.
1.5.4 Relevant Topics Not Addressed
Some topics were identified as relevant but were not addressed in the façade investigation
methodology development or application.
Balconies and terraces were not explicitly studied. Therefore, the methodology does not have
specific protocol to address challenges posed at these areas.
1.6 Scope of Work
1.6.1 Purpose and Objectives
The objective is to develop a strategy for reviewing conditions of an existing building in order to
predict if the building is a good candidate for a renovation to the building facade. The second
objective is to identify the benefits of simulation and surveys. This can aid in determining
whether a whole building analysis strategy provides any relevant context for façade performance.
The aim was not to optimize parameters and produce a detailed solution, but to explore many
assemblies that could be implemented and offer potential benefits and limitations of each. In
other words, the aim was not to produce a modular assembly that can be applied to a variety of
buildings; rather, the aim was to take into account information relevant to a specific building and
develop potential solutions tailored to the building parameters and owner-defined goals of the
retrofit.
22

1.6.2 Deliverables
The end product was a comparison of façade renovation methodologies across different climate
zones. This is intended to inform designers what parameters should be considered in each
climate zones, whether or not detailed simulation may be beneficial based on building age, size,
or other factors, and what types of renovation strategies to consider. The intent is for architects,
façade consultants, and experienced building designers to have information available that assists
in derives a set of possible façade assemblies that are appropriate for the building façade retrofit.
1.6.3 Methodology
Figure 1.6-1

1.7 Chapter Structure
Chapter 1:
Offers an introduction, purpose, objective, and scope of the study; defines terms necessary to
understand the methodology; predicts limitations of the study
Chapter 2
Summarizes existing research on façade renovation and the methods and tools which support it;
assesses the benefits and limitations of existing studies and compares them to this research;
evaluates relevant national codes and standards from organizations in the United States and
defines how these were used to establish classification standards, baseline scales, and criteria for
evaluation; reviews similar studies using existing software available to do such work.
23

Chapter 3
Provides detailed methodology for how characteristics were selected; explains the selection of
parameters considered in investigation.
Chapter 4
Describes review of available data and details what values were selected; establishes baseline
and target models for various building types; provides a step by step description of the
methodology used to estimate building performance pre and post renovation; summarizes data
without interpretation of its significance.
Chapter 5
Interprets results of the methodology; speculates justification of the results; explains variances
and unexpected results.
Chapter 6
Offers conclusions to the study; provides summary of the research; outlines future research that
can enhance knowledge; describes future work that can improve on this research.
24

Chapter 2: Previous Work and Document Review
2.1 Chapter Overview
Chapter 2 summarizes existing research on building envelope renovation for the purpose of
energy conservation. Methods and tools used by others to conduct research are assessed
according to benefits and limitations. Purpose and methodology from existing studies are
compared. The potential to improve upon existing studies is introduced.
First, typical renovation strategies are investigated. Then, research which uses digital application
to estimate or predict energy savings is explored. Methods of on-site data collection are
reviewed. Finally, a list of existing property condition assessment tools and building energy
modeling tools is presented.
2.2 Relevant Organizations, Codes, and Standards
The following United States codes and standards were reviewed and are referenced by acronym.
• International Building Code 2015 (IBC)
• International Energy Conservation Code (ICEE)
• International Green Construction Code (IgCC)
• American Society of Testing and Materials (ASTM)
• American Society of Heating, Refrigerating, and Air-conditioning Engineers (ASHRAE)
• Chartered Institution of Building Services Engineers (CIBSE)
• ASTM E2270 – 14 Standard Practice for Periodic Inspection of Building Facades for
Unsafe Conditions
• ASTM E2841 – 11 Standard Guide for Conducting Inspections of Building Facades for
Unsafe Conditions
• ASHRAE Standard 90.1-2016: Energy Standard for Buildings, Except Low-Rise
Residential Buildings
• ASHRAE Standard 90.2 -2007: Energy-Efficient Design of Low-Rise Residential
Buildings
• Standards for the Treatment of Historic Properties, presented by the U.S. Department of
the Interior, National Parks Services
• National BIM Standard, organized by the National Institute of Building Sciences
2.3 Contemporary Envelope Renovation Strategies
Many studies have considered methods to estimate improved energy performance following a
renovation to the building envelope. Some studies in the United States have assessed CBECS
data to determine which categories are relevant to the building enclosure and could have an
impact on energy performance, if modified. Other studies have used benchmarks defined by
25

CIBSE in order to compare estimated energy performance prior to and after a façade renovation.
Despite differences in benchmarks, methodology, and software, each of these works have shown
potential benefits to conducting façade renovation.
2.3.1 Fundamentals of Façade Retrofit
Martinez uses findings from two surveys given to building façade professionals to draw
conclusions about the typical façade retrofit practices and expected outcome of façade retrofit
(Martinez, et al. 2015). A retrofit to the building façade can serve various purposes. According to
this study, aesthetics was the main driver for renovation. This was followed by energy
performance and renovation. The most popular expected outcomes were improved durability,
reduced energy consumption, and increased occupant health and comfort. Many projects
reported that the whole existing façade was replaced in the renovation. The next most popular
renovation type was a modification or replacement of windows. In addition, professionals
reported that it was common to upgrade or replace HVAC systems in conjunction with the
façade renovation.
This provides a basis as to what types of façade renovations have taken place in recent years in
the United States. It may be practical to simulate renovations that follow those most frequently
reported in the study.
2.3.2 Façade Retrofit: Enhancing Energy Performance in Existing Buildings
Martinez analyzes façade retrofit’s ability to enhance energy performance through the use of
passive design strategies (Martinez Arias 2013). The study focused on mid-20
th
century buildings
using mechanically conditioned spaces and the types of alternations that could be made to the
glazing and cladding to improve performance. A review of commercial office constructed
between 1900 and 2006 found building constructed between 1997-2006 to have the highest
energy use intensity (thousand BTU/ square foot).
As a case study, Martinez estimated improved energy performance in a 12-story commercial
office in a mild climate using eQuest and Design Builder. Five different façade renovation
schemes were tested and compared to the baseline performance: single skin, sunshades, over-
cladding, recladding, and double skins. Reductions in energy consumption were achieved in the
first four schemes.
The intent of Martinez’s study matches that of this thesis: multiple façade renovation strategies
were simulated to determine their potential to improve performance. Accordingly, the findings of
the study were considered in detail in the selection of building characteristics for this study.
26

2.3.3 Incremental Façade Retrofit with Curtainwall Technology
Mic Patterson suggests that building codes, inadequate technology, and perceived risk of new
technology are the largest hindrances to achieving buildings with net zero energy consumption
(Patterson, Vaglio and Noble 2014). Since the industry began to focus on energy conservation
and sustainable practices, much of the focus has been on producing energy by integrating
mechanisms such as solar photovoltaic arrays on rooftops. According to Patterson, in order to
reach net zero energy designers much first reduce energy demands in buildings and then follow
up by determining sustainable methods to supply needed energy. The building envelope has the
potential to affect heating loads, cooling loads, ventilation, and lighting. These are four of the
eleven categories of energy consumption considered by the CBECS. Patterson argues that
making incremental adjustments to existing building envelopes is one method to address high
energy consumption. Renovation differs from new design because the current owner of the
building is likely responsible for the utility bills related to energy consumption. This presents a
large incentive for owners of existing buildings to perform renovations that could reduce energy
use and potentially cut down utility bills. By comparison, developers and property owners in new
design may not value implementing sustainable practices because they are not the end user or
owner of the building. Patterson suggests curtainwall technology as a retrofit application for
existing buildings and praises emerging curtainwall technology for its adaptability.
While curtainwall technology is not the only façade retrofit application that offers both aesthetic
appeal and adaptability, it is a popular choice in low and high rise commercial buildings. Glazed
and opaque wall assemblies will be integrated into the thermal performance study. Patterson’s
study offers the valuable notions that building envelope renovations can take place incrementally
and will require investigation and validation of new technology in order to gain industry support.
2.3.4 Investigation of Conventional Energy-Savings Interventions
Pomponi et al. used a single building in London as a case study in an effort to verify the
generally accepted hypothesis that energy savings can be produced by façade refurbishment
(Pomponi, et al. 2015). Pomponi’s case study relied on simulation using IES Virtual
Environment and data collection in the building. Pre-retrofit and post-retrofit energy
consumption was compared against three benchmarks developed by CIBSE. Typical energy
conservation measures that are classified as façade renovation include improving wall insulation,
replacing windows and frames, adding shading devices, and employing natural ventilation.
Some argue that improved insulation of the exterior wall assembly is a more important factor
than control of solar radiation in existing office buildings. Depending on climate type, the
potential savings from aforementioned strategies ranges from 20-50%. However, some studies,
including that of Pomponi et al. have shown that not all ECM truly produce energy savings.
Pomponi’s research estimated a 16.4% reduction in space heating as a result of increased
insulation and a 24% increase in lighting electricity due to shading devices. In addition, some
27

studies have shown that interventions can be beneficial in one season and have negative effects
during other seasons.
2.4 Predicting Retrofit Potential using Digital Tools
It is possible to achieve information to support façade retrofit from data-driven analysis. When
provided sufficient data, machine-learning programs can evaluate existing conditions and
determine which of many façade renovation strategies will produce the greatest reduction in
energy consumption. In this way, complex 3D modeling and simulation can be reduced in the
early stages of design; the program uses quantitative analysis that can be performed without
knowing the exact geometry of the structure.
2.4.1 Decision Methodology for Retrofit Strategies
Carlos Ochoa and I. Guedi Capeluto aimed to develop methodology for selection of building
envelope retrofit using a computational tool (Ochoa and Capeluto 2015). The study focused on
retrofits in Europe aiming to improve energy performance. The computational tool used a variety
of inputs such as location, climate data, and orientation to determine a range of possible façade
assemblies that could be installed on the exterior enclosure of the existing building. The new
system to be attached would be based on a modular unit that could be arranged in a tiled array
and mounted onto the exterior façade.
Ochoa presented a method of organizing and ranking input parameters to the application by
assigning each parameter a family, a range of values, and an importance rating. Ochoa’s
approach to offering assembly types was made of single and combination layers of building
enclosure technologies such as color, glazing, insulation, shading, and ventilation. Envelope
technologies and corresponding combinations were classified as traditional or modern/high-tech.
The application filtered input from the user by evaluating the enclosure elements according to
climate, façade condition, orientation, and design intent. Ochoa’s study assessed the assembly
combinations by percentage of energy savings, ICI, the initial cost index relative to insulation
only, and IRR, investment recovery rate. Results were presented as a range of options organized
by priority, combination, ICI, IRR, and percentage of energy savings (Ochoa and Capeluto
2015).
Ochoa’s decision methodology is beneficial because it provides a precedent for how input data
describing the building envelope can be filtered and selected according to climate, façade
condition, wall orientation, and design intent. The intent of the research was to offer and rank
potential solutions that increase energy efficiency using both traditional design and so-called
“high-tech” design.
Ochoa’s research is based on the assumption that the building energy performance can be
improved by attaching an additional thermal barrier without investigating or modifying the
28

existing envelope assembly. This assumption is limiting because it means that designers will
have no knowledge of the current condition of envelope assembly elements before the
construction process begins. Additionally, while the application estimated potential improved
thermal performance, it did not take vapor drive or the flow of moisture into account. Thermal
properties of an exterior wall assembly have a significant effect on the ability of moisture to flow
through or out of the wall. While this does not necessarily affect the thermal performance of the
building overall, it is important to consider for the aspect of weather-resistance.
As part of a doctoral dissertation at the Izmir Institute of Technology, Gucyteter conducted a
study in which building audit and post-occupancy data was used to calibrate a model of a
commercial building in Izmir, Turkey (Gucyeter and Gunaydin 2012). Energy simulation was
used to replicate building performance under initial conditions and after a proposed envelope
retrofit strategy was installed. Various energy conservation strategies for retrofit of the envelope
were tested and the proposed solution was optimized according to cost, energy consumption, and
social factors. Error was estimated using linear correlation analysis between simulated and
measured temperatures and by checking deviation of simulated temperatures from monitoring
data with root mean square error and mean bias error (Gucyeter and Gunaydin 2012).
Calibration of the model was necessary in order to generate information specific to the building
and increase the level of accuracy in simulation. A strategy similar to this was used to establish
current conditions of envelope performance as a benchmark reference. An error estimation was
used which accounted for the level of certainty for which each parameter was known because
both design values and measured values were used to create a benchmark. The use of dynamic
energy simulation and hourly post-occupancy data to calibrate the application is an improvement
for future research.
2.4.2 Predicting Building Retrofit Opportunities
Marasco and Kontokosta used existing data from New York City energy audits to inform a
machine-learning tool that predicts building retrofit opportunities (Marasco and Kontokosta
2016). While this study was not focused on retrofit of the building envelope, it used existing
building information to estimate the potential for retrofit strategies. The objective was to produce
a data-driven machine-learning tool that could assess energy efficiency potential of an existing
building. The end product would offer building owners an understanding of retrofit options that
have the potential to increase energy efficiency. Specifically, the tool classified known
information about the building and predicted the likelihood that a building would be
recommended for a specific energy conservation method.
The study used statistical analysis to calculate the probability that a particular energy
conservation method could be implemented in a building. The machine first performed text-
based data mining on New York City energy audit proceedings in order to find necessary input
data. The machine used 80% of the audit data to learn statistical behavior. In other words, data
29

was fit to a statistical curve, such as Poisson distribution or Gamma distribution, to instruct the
machine how to analyze future input data. The machine conducted analysis on the remaining
20% of the data. The machine was user-facing and used the Falling Rule List algorithm logic to
classify data and determine the resulting probability that a given energy conservation method
would apply to the selected building (Marasco and Kontokosta 2016).
This strategy for retrofit prediction shows how computational tool can develop a level of
intelligence when provided with sample data. Using this intelligence, the tool can perform un-
biased analysis on future projects.
2.5 Documentation of Physical Conditions using Digital Methods
Larsen et. al. detailed the TES Energy Façade method of using pre-fabricated wall assemblies for
façade retrofit. This method depends on precise as-built and existing conditions of the structure.
This precise documentation was pursued through the use of 3D laser scanning technology in
collaboration with existing BIM tools. The paper discusses surveying and documentation of
existing building facades using a “digital chain” (Larsen, et al. 2011).
Gocer et. al. researched digital methods to collect and integrate building data needed for the
renovation of a building. Specifically, the 3D modeling capabilities of BIM were partnered with
the measurement methods of GIS to collect real geometric information about a historic building
and access it in a 3D digital model (Gocer, Hua and Gocer 2016).
Aydin used of calibrated digital cameras, close-range photogrammetry, and GIS to document
existing conditions of buildings in the city center of Ankara, Turkey. This documentation
supported efforts to renovate existing building facades (Aydin 2014).
While GIS and laser-scanning technology were not implemented into the design of the envelope
condition assessment application, they were considered as potential sources as documenting as-
built information and can be considered for future work.
2.6 Precedents
Property Condition Assessment (PCA) applications and data management tools were reviewed.
2.6.1 Property Condition Assessment Applications
Existing PCA apps include:
• Kykloud by kykloud Ltd
• cStat (Collateral Status) Property Condition Assessment by Advertek, Inc.
• BCAS by Indus Systems, Inc.
• CAMS – Mobile by Integrate Australia Pty Ltd
30

• Mobile Property Evaluation Device (MOPED) by Zehnder Communications, Inc;
2.6.2 Building Energy Modeling Tools
Many building simulation tools exist to support performance modeling. A few include:
• EnergyPlus
• BES IES VE
• eQUEST
• TRNSYS
• DoE2
• Revit Energy Analysis
These tools were explored for potential use and Revit Energy Analysis was selected.
2.7 Conclusions and Takeaways
The document review presented several items as important to take into consideration: the
schedule of renovation; the integration of real and simulated data; the value of renovation to a
building owner; the ability of a renovation strategy to increase or decrease energy consumption.
Renovation to the building enclosure can take place incrementally. This could be potentially
beneficial in order to reduce disturbance to ongoing business, operation, and function of the
building.
Studies which incorporate both simulation data and on-site test data are limited. It may be
beneficial to determine a method to incorporate data from both in order to understand the relative
error.
Energy conservation measures can be more valuable to owners of existing buildings than to
developers of new buildings because owners can expect to see the payback during the time that
they possess the property.
Renovations to the building enclosure do not always produce energy savings. Some measures
can produce savings for one function but create expense in another. In addition, a measure that is
beneficial during one season can be detrimental during the opposing or less mild seasons.
Measures may be more appropriate for climates with extreme heating or cooling seasons rather
than year-round mild conditions.

31

Chapter 3: Methodology
3.1 Chapter Overview
This chapter details methods for parameter selection and discusses the analytical methods used to
generate a profile of building performance using statistical models and energy simulations. The
trajectory is divided into analysis, simulation, and results.
Figure 3.1-1

The initial analysis was conducted in MATLAB. The full CBECS dataset was inputted into a
program that plotted histograms of select variables. Values for variables could then be
determined from the most frequently occurring building types and characteristics occurring in the
dataset. The program also generated scatter plots of building characteristics compared to building
EUI and electricity and fuel consumption data. Potential candidate for curve-fitting were
identified. The data was filtered according to the selected values for each variable. The program
was run again to observe any changes in confirm any candidate for curve-fitting. During the
statistical analysis portion, data was tested against various known distributions. The sample
building was modeled in Revit based on values from the initial analysis. The Energy
Conservation Methods were applied in each trial. Results of the trials were compared to the
baseline state. Energy consumption from the simulation was compared to consumption values
from the initial analysis. Potential benefits of each ECM were identified and presented in a
summary.

32

3.2 Initial Analysis
Figure 3.2-1

The initial analysis involved selected which characteristics of a building enclosure and variables
of the CBECS dataset to consider in the study. This was done by observing statistical behavior of
the data and referencing characteristics considered in work done by others.
3.2.1 Energy Consumption Data
Data describing current energy consumption for the United States building stock was necessary.
While many states now require energy audit data be released for public use, it was determined
that CBECS data would provide a comprehensive summary of the United States building stock
that is not biased towards any local or state code requirements. In this way, trends in
consumption in a given climate zone from the CBECS dataset could possibly be used to predict
consumption for a similar building in that climate zone.
In the CBECS variable list, there are four parameters directly affected by the building enclosure
which relate to building envelope performance: space heating, space cooling, lighting, and
ventilation. These account for 50% of building energy consumption (Patterson, Vaglio and Noble
2014): space heating, space cooling, lighting, and ventilation. Focus was placed on energy
consumption for space heating and space cooling.
CBECS data provides a rough approximation of building performance that was used to generate
benchmark performance profiles. The building models used for computer simulation were set up
to perform in alignment with benchmark performance profiles in their initial state (pre-
33

renovation state). The intent of the simulation was to generate a new dataset that estimates
energy performance after an energy conservation method (ECM) is applied in a renovation.
3.2.2 Building Characteristics
Building characteristics for the study domain and simulations were selected based on the 2012
CBECS commercial building stock (U.S. Environmental Protection Agency 2012).
Characteristics of interest include: census division, year constructed, size, exterior wall
construction material, percent of exterior glass, building shape, number of floors, primary
cooling source, and primary heating source. Other information from 2012 CBECS dataset
include total energy consumption, total energy expenditure, consumption for space heating,
consumption for cooling, building energy use intensity. Consumption for space heating and
cooling was examined by each energy source.
CBECS reported that about half of the buildings included in the survey were 5,000 sq. ft. or
smaller and three-fours of buildings were 10,000 square feet or smaller. Accordingly, the
representative building for analysis was designed to be approximately 5,000 sq. ft. If needed,
later simulations can be conducted on a building representing the next two most popular building
sizes: the 5,000 – 10,000 sq. ft. and 10,000 – 25,000 sq. ft. ranges.
Roughly half of buildings in the survey were constructed before 1980. The median age of a
building is 32 years. While a large majority of buildings were constructed before 1980, more
buildings in the survey were built after 2000 than before 1946. Based on this information, it was
assumed that a building that is a good candidate for façade renovation would have been built
between 1950 and 2000. This is because profiles generated from CBECS data are not likely to
reflect buildings constructed before 1950. In addition, buildings built after 2000 have not yet
reached a service life of at least 30 to 50 years.
According to CBECS, the building activity of most surveyed buildings was listed as office.
Therefore, the simulated building is expected to function as an office under ASHRAE
categorization for energy calculation.
Observing additional trends in the CBECS data will provide exterior wall construction material,
percent of exterior glass, building shape, number of floors, primary cooling source, and primary
heating source.
3.2.3 Benchmark Standard
The first component of analysis was to define a benchmark standard for how a building envelope
is expected to perform in each of the performance categories. Each climate zone was expected to
have a different benchmark based on annual energy demand in each climate.
34

Energy use was measured by consumption intensity, or energy consumption per net floor area, in
order to compare buildings of various sizes. The following energy sources for space heating and
cooling were considered: electricity, natural gas, and fuel oil. Other comparisons were made
using operation hours, occupancy, building function, and geographic location according to the
U.S. census region and division.
Many organizations focusing on reduction in energy use in the U.S. use energy consumption and
carbon emissions at 1990 levels as a benchmark. Due to its wide use, this standard is beneficial
for comparison. Energy consumption for the baseline building performance for each climate will
be compared to 1990 levels as presented by the U.S. D.O.E. and E.P.A. However, when
estimating improved performance of a building enclosure after façade renovation, the baseline
building performance for the specified region and building characteristics was used. This was
expected to more accurately display potential for improvement because the baseline reference
uses characteristics specific to the selected building.
3.2.4 Code Requirements
International Building Code 2015 was reviewed in order to develop more constraints to bound
analysis. Some value considerations include the fire rating of the wall assembly and the
minimum thickness of weather coverings.
3.2.5 Establishing Targets
Initial analysis provided a baseline standard describing how a large segment of commercial
buildings in the U.S. are currently performing. In addition, simulation provided estimates for
how various ECM can potentially alter energy demand and as a result, energy consumption.
These estimates establish a target value for energy consumption that is based on modifications to
an existing building. This allows for a comparison which focuses on a building’s performance
over time; rather than comparing newly constructed buildings to older, existing buildings.
In future work, it may be beneficial to align target performance with Architecture 2030 goals for
energy use and carbon emission reduction.
3.2.6 Software
Mathematical modeling and statistical analysis of CBECS data was conducted in MATLAB.
MATLAB is software for computational analysis which uses matrix-based programming
language to offer matrix manipulations, data plotting and analysis, algorithms execution, and
creation of user interfaces. MATLAB is capable of managing large datasets.
Autodesk Insight was used for energy simulation to offer a simplified modeling environment to
observe changes in energy use intensity as a result in changes to building enclosure elements.
35

Insight is a plug-in to Autodesk Revit that operates in the cloud after receiving building
information from Revit created and designated by the user.
The application resulting from the script for analysis was written in MATLAB and is intended to
be developed into an independent computer program.
3.3 Mathematical Model
Figure 3.3-1

The initial analysis was done to refine and filter the data. Once the data subset was defined and
variables of value were established, hypotheses were made as to potential distributions the data
followed. The data was tested against Gaussian and exponential distributions in order to help
make reasonable predictions about the relationships between façade elements and energy
consumption.
3.3.1 Epistemic Uncertainty
Data and analysis of such figures used in the development of benchmark standards and
behavioral conditions for the application, Renvelope, were assumed to follow epistemic
uncertainty, or uncertainty associated with imperfect knowledge. Mathematical modeling and
statistical analysis were used to generate profiles which represent the expected performance of a
randomly selected commercial building. However, the performance a selected building cannot
be expected to meet the profile with 100 percent accuracy. The level of certainty is expressed by
statistical values such as the confidence interval. Regardless, an estimated degree of error exists
in statistical profiling.
36

The technique of statistical inference for decision-making and design relies on estimation of
some parameters and selection of a probability distribution to model data behavior. Efforts were
made to ensure that estimations were unbiased and methods minimize error.
The benchmark performance profile provided by Envelope is a baseline estimate used for
approximating potential for improvement. The improved performance profile after a façade
renovation is an idealized model. An unknown degree of error exists in the improved
performance profile as it is derived from both statistical analysis and computer simulation.
3.3.2 Statistical Summary of Building Stock
Prior to analysis, it was necessary to first generate a performance profile for the building stock
provided by CBECS sampling. Statistical summary values were generated. These included the
minimum, maximum, mean, variance, standard deviation, and skewness of data. In addition,
histograms and probability distribution function graphs were generated in order to visualize data
behavior. Characteristics were tested for goodness of fit to determine if behavior could be
explained by a known pattern or expression. Finally, the statistical correlation and potential
statistical independence of variable pairs was explored.
Characteristics of interest for the statistical summary of the 2012 building stock include: census
division, year constructed, size, exterior wall construction material, percent of exterior glass,
building shape, number of floors, primary cooling source, and primary heating source.
The statistical summary aided in evaluating if the study domain was appropriate. In addition, it
provided a reference for what kinds of buildings Envelope can provide reasonable estimates for
improved performance.
3.3.3 Statistical Summary of Energy Consumption Data
A statistical summary of building energy consumption data was generated. These included the
minimum, maximum, mean, variance, standard deviation, and skewness of data. In addition,
histograms and probability distribution function graphs were generated in order to visualize data
behavior.
Values of interest include total energy consumption, total energy expenditure, consumption for
space heating, consumption for cooling, building energy use intensity. Consumption for space
heating and cooling was examined by each energy source.
3.3.4 Filtering Data
The initial analysis was conducted on the full CBECS dataset. A profile of the most frequenctly
occurring building in the United States building stock was created. Then, data was filtered such
that all included buildings fall within the building characteristics of the defined profile.
37

3.3.5 Statistical Relationships
Correlation between Parameters
In order to conclude if predetermined ECM may be effective at altering energy performance, it
was necessary to determine if any statistical relationships existed between parameters.
Correlations of interest include comparisons between energy consumption and the following
parameters: census division, year constructed, size, wall construction material, percent of
exterior glass, building shape, and number of floors.
Additional tests will be conducted to determine if selected façade components are statistically
significant as they relate to energy consumption for heating and cooling occupied space. If
correlations were found, it was intended to use them in the logical process for the development
of the computer application, Envelope.
3.3.6 Goodness of Fit
After observing data on histogram plots, a probability distribution was assumed to model the
behavior of selected enclosure characteristics. A goodness of fit test was used to validate whether
or not the assumed probability distribution was a plausible model to represent variable behavior.
Kolmogorov-Smirnov Test (K-S Test)
The Kolmogorov-Smirnov Test (K-S Test) was one method used to determine if a probability
distribution can model experimental data. The discrepancy between the cumulative frequency of
data and the CDF of the distribution signifies goodness of fit. While the critical value varies with
sample size, if the discrepancy is less than the expected value, the distribution models to the
predetermined significance level, α.
Chi-squared Test
The Chi-squared test was another method used to determine if data fit a known probability
distribution to a specified significance level, α. The Chi-squared test compares observed
frequencies from the dataset to theoretical or expected frequencies originating from a given
distribution.
3.3.7 Deliverables
The deliverable from the mathematical modeling and statistical analysis is a set of statistical
profiles which describe how a commercial building performs. Specifically, the profiles aim to
represent the relationship between components of the façade and energy consumption from
heating and cooling of occupied space in commercial buildings. These profiles, whether they can
38

be expressed by statistical distributions and/or approximated probability distribution functions,
serve as the basis for approximation in the computer application.
3.4 Simulation
Figure 3.4-1

A sample building was modeled in Autodesk Revit. Using Green Building Studio and the
EnergyPlus plug-in, simulations estimated the baseline electricity and fuel consumption and
changes in these values as energy conservation methods are applied.
3.4.1 Software
Autodesk Revit with Green Building Studio and EnergyPlus plug-ins were used to conduct
computer simulation.
Green Building Studio uses the EnergyPlus methodology to perform whole building performance
analysis based on location, building function, building materials, and geometric characteristics to
conduct a transient heat flow analysis. Collectively, the software offers approximations for
energy use intensity and life cycle energy use and costs. The analysis produces estimates for
heating and cooling loads considering the various factors. Factors of interest include window
solar transmittance, window conductance, infiltration, and exterior wall assembly. The analysis
report also includes in estimates of monthly fuel and electricity consumption.
Green Building Studio evaluates energy performance of a building based on given characteristics
and location and outputs consumption in terms of annual expenditure per square foot. Since the
software is intended to support designers in the early conceptual phase, Autodesk also offers an
cloud-based plug-in, Insight, that has capability to make slight theoretical modifications as
39

different concepts and compare performance. While Insight was not used for analysis, the end
result of this study aims to provide similar functions as the Insight tool, with specific
concentration on improvements to façade performance.
3.4.2 Baseline Building: “Typical Commercial Office”
Simulation began by modeling the representative building based on characteristics selected from
CBECS data. This building is considered the baseline for performance in the analysis and is
expected to produce similar consumption estimates to the values reported in CBECS for
buildings of similar size, shape, height, and wall construction material in the corresponding
ASHRAE climate zone.
3.4.3 Limitations and Constraints
Fixed Values
Some attributes of a building cannot be altered for renovation in a way that is practical in terms
of cost or means of construction. These items were defined as fixed constraints. Fixed values
include structural loading capacity, framing type and spacing, building orientation, total floor
space, and building height. While it is understood that these items may be changeable, for the
purpose of building envelope retrofit they were considered fixed. There are boundary conditions
of the test subject.
3.4.4 Known Values
Test Subjects
The test building was modeled in Revit and the Green Building Studio and EnergyPlus plug-in
estimated the energy consumption of a mid-sized commercial building in four of the eight
ASHRAE climate zones. In the initial round, there was one test subject whose characteristics
were determined by those in CBECS the dataset. This is considered the statistically-determined
building because characteristics for this building were selected based on mode values from the
statistical summary of the 2012 CBECS data for the U.S. commercial building stock. Additional
simulations can be performed on buildings with different characteristics if needed. The
representative building remains the same for each location.
Locations
The test subject was evaluated in five cities representing the four of the eight ASHRAE climate
zones. For each climate zone, one representative city was designated as the location to perform
simulation. Climate zones 1, 7, and 8 were considered extreme climate conditions and were not
considered for the study.
40

CBECS uses 3 different categories to classify the location and climate of a building: Census
Region, Census Division, Climate Region. Each of these categories has an associated number.
CBCES descriptions
Census Region
''1' = 'Northeast'
'2' = 'Midwest'
'3' = 'South'
'4' = 'West'
Census Division
''1' = 'New England'
'2' = 'Middle Atlantic'
'3' = 'East North Central'
'4' = 'West North Central'
'5' = 'South Atlantic'
'6' = 'East South Central'
'7' = 'West South Central'
'8' = 'Mountain'
'9' = 'Pacific'
Building America Climate Region
'1' = 'Very cold/Cold'
'2 '= 'Mixed-humid'
'3' = 'Hot-dry/Mixed-dry/Hot-humid'
'5' = 'Marine'
'7' = 'Withheld to protect confidentiality'

ASHRAE uses seven climate zones. The ASHRAE climate classification is preferred for
consistency with other studies. In order to make the study comparable across both systems, four
cities were selected that meet four different classifications for CBECS and ASHRAE. Climate
regions according to CBECS description considered include cold, mixed-humid, mixed dry/hot-
dry, and hot-humid. ASHRAE climate zones selected include zones two through five (See
41

Table 3.4-1 below).

42

Table 3.4-1
ASHRAE Climate
Zone
CBECS Climate
Region
Climate Description Potential Representative City
1
4
very hot
Maui, HI
Miami, FL
2
4
hot-humid
New Orleans, LA
Orlando, FL
3
3
warm
Los Angeles, CA
Las Vegas, NV
4
2
mixed-humid
Nashville, TN
Baltimore, MD
Louisville, KY
5
1
cool
Pittsburgh, PA
Chicago, IL
6
1
cold
Minneapolis, MN
Sioux Falls, SD
7
1
very cold
Aspen, CO
8
1
artic
Yukon-Koyukuk, AK
The locations selected include New Orleans, LA, Los Angeles, CA, Louisville, KY, and
Chicago, IL to represent ASHRAE climate zones two through five, respectively.
Table 3.4-2
ASHRAE Climate Zone Selected City Weather station code
5 Chicago, IL 59474
4 Louisville, KY 39988
3 Los Angeles, CA 59381
2 New Orleans, LA 1030701

3.4.5 Energy Conservation Methods
In order to identify and track changes in each Performance Category, variables were defined and
assigned to correlate to a specific Performance Category.
Each variable correlates to a building enclosure characteristic that has potential to alter energy
consumption in one of the three selected Performance Categories. The building enclosure
43

characteristic is non-permanent, meaning it could be modified if a renovation to the building
façade were to take place. The element modified during a façade renovation is called an Energy
Conservation Method (ECM) as it is intended to improve energy performance.
The first Performance Category, thermal performance of the exterior wall assembly, considered
the R-value of wall insulation as the primary variable factor. The thermal resistance, R-value,
and coefficient of heat transfer, U-factor, describe heat transfer by type of material. The
coefficient of heat transfer was selected as metric for evaluating thermal performance of an
exterior wall assembly. Effective U-value of the assembly is considered rather than nominal U-
value of materials. Guidelines for U-factor are in compliance with ASHRAE 90.1-2013. After
reviewing the comparison of ASHRAE Standard 90.1 to International Energy Conservation Code
provided by the U.S. Department of Energy, ASHRAE 90.1 was selected over IECC and IgCC
under the belief that these values are generally more conservative (U.S. Department of Energy, et
al. 2011). Testing was conducted to validate whether a renovation which alters the type and
amount of insulation can improve thermal performance of the exterior wall assembly. The
designated ECM was wall assembly rehabilitation.
The second Performance Category, infiltration, was evaluated by air changes per hour (ACH). In
building design, infiltration is assumed to be as a result of the tightness of construction (U.S.
Environmental Protection Agency 2016). Buildings with looser construction are expected to have
higher infiltration and as a result, more air changes per hour (ACH). Buildings with tighter
construction are expected to have the opposite outcome. In a façade renovation, the ECM to
solve for infiltration was to re-seal fenestrations.
The final Performance Category, thermal performance of exterior glazing, was regulated by
glazing assembly, coatings, and shading. Since solar heat gain and/or heat loss through glazing
have shown to significantly affect energy performance, altering glazing type was assumed to be a
practical ECM as a candidate for a façade renovation.
3.4.6 Simulation of Improved Performance
Simulation was performed in order to estimate a change in energy consumption. For each
performance category, an ECM was applied which was expected to alter the flow of thermal
energy through the façade and impact the annual heating and cooling demand for the building.
The purpose of this is to estimate the energy consumed after a change has been made to the
exterior envelope that is intended to improve energy performance. The difference in energy
consumption between before and after ECM is considered the energy savings. These savings
values were presented in the form of EUI and annual energy expenditure.
Examples of ECM include:
1) Category I: Performance of exterior Glazing Assembly
44

a) Modification: Change the glazing assembly to double pane
2) Category II: Thermal Performance of Exterior Wall Assemblies
a) Modification: Change the wall insulation type and exterior finish
b) Modification: Change the construction type to medium or tight with ACH value
3) Category III: Thermal Performance of Roof Assemblies
a) Modification: Change the roof insulation type
b) Modification: Change roof finish material and/or color
3.4.7 Error and Uncertainty
The purpose of the simplified simulation was to estimate energy consumption for space heating
and cooling. The simulation values were used as baseline values for thermal performance. It is
assumed that actual building performance is less than that of simulations.
3.5 Trials
Trials in simulation were not cumulative; each modification did not include previous
modifications unless specified. Analysis was performed using specific building elements and
their thermal properties rather than conceptual masses.
3.5.1 Trial 1.0: Baseline
This simulation was conducted on the modeled “typical commercial office” building according
to parameters defined from the statistical analysis.
3.5.2 Trial 2.0: Window Rehabilitation
The first trial represents a renovation that rehabilitates or replaces exterior windows. This
includes both improving the thermal performance of the glazing assembly through the use of
coatings, thicker glazing, the installation of double paned windows with an insulated cavity. This
simulation also aims to capture the effects of re-sealing windows and doors to increase tightness
of construction and minimize infiltration.
3.5.3 Trial 3.0: Exterior Wall Modification
The second trial aimed to represent a renovation to the exterior walls. The equivalent renovation
would be the installation of an exterior finish system applied onto the existing wall. The goal of
this form of renovation is to increase isolative properties of the walls. Window replacement was
also included in this trial as exterior wall thickness increases with the addition of the exterior
finish system. Replacing or rehabilitating windows is a practical and typical practice during a
façade renovation.
45

3.5.4 Trial 4.0: Roof Renovation
The third trial simulated changes to the roof only. Energy dissipation through the roof can be
considered an important factor in a single story building with a relatively large surface area to
height ratio.
3.5.5 Trial 5.0: Cumulative Enclosure Renovation
This simulation is a cumulative representation of the first three trials. It includes exterior wall
modification, window rehabilitation or replacement, and roof renovation. This is considered an
enclosure renovation because no changes were made to the HVAC system.
3.6 Intended Use of Results
Data gathered from both statistical analysis and simulation were influential in the development
of the computer application.
Figure 3.6-1

3.6.1 Improved Envelope Assemblies
A set of envelope assemblies were defined to have improved performance according to climate,
building shape, number of stories, and building size (occupied square footage). The high
performing characteristics were to be used in the computer application to suggest modifications
to existing assemblies.
46

3.6.2 Deliverables
Statistical analysis and theoretical simulation were expected to provide information describing
the relationship between enclosure characteristics and energy consumption. The outcome of this
is a comparison between energy audit data and simulation data and the estimated impact of a
façade renovation on energy consumption.
3.7 Chapter Summary
The methodology for development of assessment standards and baseline criteria was established.
Performance categories were defined to provide context for areas of potential improvement.
Energy conservation methods were presented to represent modifications that can be made to an
existing building enclosure. Building characteristics of interest were specified and the study
domain was refined. Parameters to consider were taken from other studies, code, surveys, and
building design professionals. Selected parameters were in alignment with those used in the U.S.
CBECS in order to offer comparison to other existing studies. The need for simulation was
introduced in order to develop expectations for envelope performance and improvement in
specific climate zones. Guidelines to make reasonable predictions about performance according
to the results of the data were established.

47

Chapter 4: Results
4.1 Chapter Overview
Chapter 4 describes results of the mathematical analysis and simulation. First, selected
characteristics for the test building were determined based on observations of characteristics for
buildings within the study domain. Values for energy consumption of buildings in study domain
were presented in a series of figures. Building characteristic variables were plotted against
consumption values and predictions for known distributions were made. These predictions were
tested using two methods for goodness of fit. Additionally, results of the simulation for each city
are presented.
4.2 Mathematical Analysis
4.2.1 Characteristic Selection
The CBECS survey stores responses to each question as a variable with a number or category
values. Of the 1120 variables, roughly 260 were considered relevant. This was refined to 85
variables that were inputted into a vector in MATLAB. This vector was used to form a matrix
which included all 6720 buildings available in the CBECS dataset. A new matrix would later be
created based on the definition of the developed profile.
CBECS variables were separated into categories: building characteristics, renovation data, and
consumption values. MATLAB generated histograms for 18 variables that represented building
characteristics; region, census division, principal building activity, square footage, square
footage category, wall construction material, roof construction material, cool roof, roof tilt,
building shape, percentage of exterior glass, equal glass on all sides, sun exposure to glass,
number of floors, floor to ceiling height, year constructed, year construction category, and
principal building activity. Most building characteristic variables were assumed to be fixed to
help define constraints of the study. However, exterior wall construction material and roof
construction material were considered flexible as these aspects of a building could be changed in
during a façade renovation.
The CBECS variable definitions can be found in Appendix A.
In order to create a profile, it was necessary to look at the data to determine what types of
buildings appeared most frequently in the dataset. The full CBECS dataset describing the U.S.
building stock is considered Sample A. In 2015, the EIA conducted a study on the full 2012
dataset and released a report describing the U.S. Building Stock in terms of mode, or most
frequently occurring, values (U.S. Environmental Protection Agency 2012). According to this
study, roughly half of the building stock is composed of buildings between 1,000 and 5,000 sq.
ft.
48

A smaller sample set, Sample B, was created which included only commercial office buildings
falling within the 1,000 sq. ft. to 5,000 sq. ft. range. In addition, the Sample B was filtered to
remove buildings that had reported a façade renovation. This narrows the dataset to 164
buildings. Using MATLAB, histograms were generated that showed the frequency of each
response for a given variable in Sample B. The mode values were selected as the characteristics
for the representative building that would be simulated.
Figure 4.2-1: Census Division Histogram

Figure 4.2-1 shows the quantity of buildings within the sample that fall within the CBECS Census Divisions: '1' = 'New England',
'2' = 'Middle Atlantic', '3' = 'East North Central', '4' = 'West North Central', '5' = 'South Atlantic', '6' = 'East South Central', '7'
= 'West South Central', '8' = 'Mountain', and '9' = 'Pacific'
49

Figure 4.2-2: Exterior Wall Construction Histogram

Figure 4.2-2 shows the quantity of buildings within the sample that fall within the CBECS categories for exterior wall
construction material: '1' = 'Brick, stone, or stucco', '2' = 'Pre-cast concrete panels', '3' = 'Concrete block or poured concrete
(above grade)', '4' = 'Aluminum, asbestos, plastic, or wood materials (siding, shingles, tiles, or shakes)', '5' = 'Sheet metal
panels', '6' = 'Window or vision glass (glass that can be seen through)', '7' = 'Decorative or construction glass', '8' = 'No one
major type', '9' = 'Other'
Figure 4.2-3: Roof Construction Histogram

50

Figure 4.2-3 shows the quantity of buildings within the sample that fall within the CBECS categories for roof construction
material: '1' = 'Built-up (tar, felts, or fiberglass and a ballast, such as stone)', '2' = 'Slate or tile shingles', '3' = 'Wood shingles,
shakes, or other wooden materials’, '4' = 'Asphalt, fiberglass, or other shingles', '5' = 'Metal surfacing', '6' = 'Plastic, rubber, or
synthetic sheeting (single or multiple ply)', '7' = 'Concrete', '8' = 'No one major type', '9' = 'Other'
Figure 4.2-4: Roof Tilt Histogram

Figure 4.2-4 shows the quantity of buildings within the sample that fall within the CBECS categories for roof tilt: '1' = 'Flat', '2'
= 'Shallow pitch', '3' = 'Steeper pitch'
Figure 4.2-5: Building Shape Histogram

51

Figure 4.2-5 shows the quantity of buildings within the sample that fall within the CBECS categories for building shape: '01' =
'Square', '02' = 'Wide rectangle', '03' = 'Narrow rectangle', '04' = 'Rectangle or square with an interior courtyard', '05' = '"H"
shaped', '06' = '"U" shaped', '07' = '"E" shaped', '08' = '"T" shaped', '09' = '"L" shaped', '10' = '"+" or cross shaped', '11' =
'Other shape', Missing = Not applicable
Figure 4.2-6: Glass Percentage Histogram

Figure 4.2-6 shows the quantity of buildings within the sample that fall within the CBECS categories for percentage of exterior
glass. This is the closest comparison to window-to-wall ratio: '1' = '1 percent or less', '2' = '2 to 10 percent', '3' = '11 to 25
percent', '4' = '26 to 50 percent', '5' = '51 to 75 percent', '6' = '76 to 100 percent', Missing = Not applicable
After observing statistical behavior of the full dataset, Sample A, and the filtered dataset, Sample
B, final characteristics were selected. The main characteristics of interest where building
geometry traits, which would remain constant throughout the study, and enclosure assembly
traits, which would be varied in the iterations of simulated façade renovation.
Figure 4.2-7 provides an overview of the decisions made. Characteristics of each category are
arrange, top to bottom, by their frequency of occurrence in Sample B. However, some values
were selected based on their high frequency in Sample A, or conflict with other selected trait. For
example, a flat roof was selected in order to minimize the effect of orientation of the test building
on the results (a sloped roof could potentially receive more or less direct solar heat gain than a
flat roof). Since a flat roof was selected, asphalt shingles and metal panel roofing were
impractical choices for roofing material.
52

Figure 4.2-7: Characteristics Decision Tree

53

The final traits of the representative test building are summarized in Table 4.2-1.
Table 4.2-1: Test Building Characteristics
CBECS Characteristic Mode Value
Principal building activity Office
Square footage 4900
Square footage category 1
Wall construction material brick
Roof construction material Polymer, plastic
Cool roof materials no
Roof tilt flat
Building shape wide rectangle
Percent exterior glass 20
Equal glass on all sides yes
Glass sides most sunlight more sun
Number of floors 1
Number of underground floors 0
Floor to ceiling height 10

While the aforementioned variables were used to generate the test building, the remainder of the
mathematical analysis focused on energy consumption specific to the following variables: square
footage, exterior wall construction material, roof construction material, roof tilt, building shape,
percent of exterior glass, and floor to ceiling height.
4.2.2 Reported Consumption Values
In addition to considering the number of buildings reporting to each characteristic, it was
beneficial to observe the amount of energy buildings 1,000 sq. ft. – 5,000 sq. ft. use annually.
54

Figure 4.2-8: Building EUI Histogram

Figure 4.2-9: Annual Electricity Consumption Histogram

55

Figure 4.2-10: Annual Major Fuel Consumption Histogram

Figure 4.2-11: Electricity Use for HVAC Histogram

56

Figure 4.2-12: Major Fuel Use for HVAC Histogram

4.2.3 Predicting Relationships & Significance
In order to speculate potential relationships between building characteristics and energy
consumption, a series of scatter plots were generated in MATLAB using the matrix of buildings
measuring between 1,000 and 5,000 sq. ft. with no reported façade renovation. Values for
exterior wall material and roof material were individually plotted against fuel consumption,
electricity consumption, HVAC fuel use, and HVAC electricity use.
57

Figure 4.2-13: HVAC Electricity Use vs Wall Construction Material

'1' = 'Brick, stone, or stucco', '2' = 'Pre-cast concrete panels', '3' = 'Concrete block or poured concrete (above grade)', '4' =
'Aluminum, asbestos, plastic, or wood materials (siding, shingles, tiles, or shakes)', '5' = 'Sheet metal panels', '6' = 'Window or
vision glass (glass that can be seen through)', '7' = 'Decorative or construction glass', '8' = 'No one major type', '9' = 'Other'
Figure 4.2-14: HVAC Electricity Use vs Roof Construction Material

'1' = 'Built-up (tar, felts, or fiberglass and a ballast, such as stone)', '2' = 'Slate or tile shingles', '3' = 'Wood
shingles, shakes, or other wooden materials’, '4' = 'Asphalt, fiberglass, or other shingles', '5' = 'Metal surfacing', '6'
58

= 'Plastic, rubber, or synthetic sheeting (single or multiple ply)', '7' = 'Concrete', '8' = 'No one major type', '9' =
'Other'
Figure 4.2-15: HVAC Fuel Use vs Wall Construction Material

'1' = 'Brick, stone, or stucco', '2' = 'Pre-cast concrete panels', '3' = 'Concrete block or poured concrete (above grade)', '4' =
'Aluminum, asbestos, plastic, or wood materials (siding, shingles, tiles, or shakes)', '5' = 'Sheet metal panels', '6' = 'Window or
vision glass (glass that can be seen through)', '7' = 'Decorative or construction glass', '8' = 'No one major type', '9' = 'Other'
59

Figure 4.2-16: HVAC Fuel Use vs Roof Construction Material

'1' = 'Built-up (tar, felts, or fiberglass and a ballast, such as stone)', '2' = 'Slate or tile shingles', '3' = 'Wood
shingles, shakes, or other wooden materials’, '4' = 'Asphalt, fiberglass, or other shingles', '5' = 'Metal surfacing', '6'
= 'Plastic, rubber, or synthetic sheeting (single or multiple ply)', '7' = 'Concrete', '8' = 'No one major type', '9' =
'Other'
4.2.4 Evaluating Linear Correlation
The correlation coefficients between square footage and energy consumption were determined.
Additionally, the correlation coefficients between the percentage of exterior glass and energy
consumption were determined. Very low correlation coefficients were calculated so it was
assumed that there was no linear correlation between the variables considered.
The calculations were repeated with buildings in Sample B. The values again indicated no linear
correlation between square footage and consumption or percentage of glass and energy
consumption.
Table 4.2-2: Sample B Correlation Coefficients

EUI ELCNS MFBTU ELHVAC MFHVAC
SQFT -0.19764 0.32763 0.364677 0.324731 0.330734
GLSSPC 0.06939 0.135812 0.166463 0.200461 0.1907

60

4.2.5 Hypothesis Testing
Chi Squared test for Gaussian (Normal) Distribution
The Chi Squared goodness of fit test was used to determine if the consumption data could verify
the hypothesis that variables within Sample B follow a Gaussian distribution within a 5%
significance level. Out of all tests conducted, only annual fuel expenditure and annual electricity
expenditure rejected the hypothesis. However, other results, which failed to reject the hypothesis,
reported low p-values.
Lilliefors test for Exponential Distribution
The Lilliefors test was used to determine if the consumption data could verify the hypothesis that
variable within Sample B follow an exponential distribution within a 5% significance level. All
tests failed to reject the hypothesis and reported low p-values. Due to the significantly low p-
values, the curve-fitting tests were deamed inconclusive.
Table 4.2-3: Hypothesis Test Results

Guassian? p-val Exp? p-val
MFBTU 1 0.003398 1 0.001
MFEXP 0 NaN 1 0.001
ELCNS 1 0.000204 1 0.001
ELBTU 1 0.000204 1 0.001
ELEXP 0 NaN 1 0.001
EUIBTU 1 0.002029 1 0.001
ExI 1 0.000167 1 0.001
ELHVAC 1 1.60E-06 1 0.001
MFHVAC 1 0.008568 1 0.001
Table 4.2-3 depicts results of hypothesis tests. A value of 0 indicated that the data rejects the
hypothesis and a value of 1 indicated that the data fails to reject the hypothesis. The p values
were reported to show very weak support for the result of the test.
4.3 Results of Simulation
4.3.1 Building Characteristics
The test building was a 1-story, 4,900 sq. ft. commercial office with an open office floor plan.
The exterior walls were composed of brick with plywood sheathing, a vapor barrier, 4 in. of R-
13 fiberglass batt insulation, and gypsum board. Windows were single pane clear glass in in
aluminum frames. The roof was composed of 6 in. reinforced concrete, a vapor retarder, 4 in. of
rigid foam insulation, and finished with an EPDM coating.
61

Figure 4.3-1: Test Building Plan

Figure 4.3-2: Test Building Axonometric View

N
62

Figure 4.3-3: Test Building Elevations

CBECS reports that fuel is the most common energy source used for space heating and
electricity is the most common energy source used for space cooling. The test building used a 2-
pipe fan coil system with distribution via a constant volume duct system. A water-cooled
centrifugal chiller with COP of 5.26 supplied the chilled water coil. A gas-fired boiler with 84.5
combustion efficiency supplied the hot water coil. Variable volume hot water, chilled water, and
condenser pumps were used. The building occupancy schedule followed a typical commercial
office according to ASHRAE 90.1. Additional information on the HVAC system can be found in
Appendix C.
63

Figure 4.3-4: Test Building Energy Model

4.3.2 Trial Characteristics
The Initial characteristics of the test building were modified in each trial by and energy
conservation method (ECM). The ECMs were selected from precedent studies and by
professional recommendation for façade renovation techniques currently used in practice.
Descriptions of the details of each trial are summarized in the tables below. For all tables below,
dimensions are reported in inches.
Table 4.3-1: Wall Assembly Typologies
Wall Assembly
WT1

WT2

Brick, int. insulation Thin brick with ext. insulation over existing
Dim. Material

Dim. Material
4 Brick

3/4 Thin brick
1/2
Plywood

2 Polyurethane foam Insulation
0 Vapor barrier

0 Vapor barrier
4 fiberglass batt insulation 1/2 Plywood
5/8
Gypsum board

4 Brick

1/2
Plywood

4 Insulation

5/8
Gypsum board

64

Figure 4.3-5: Wall Type 1 and Wall Type 2

Table 4.3-2: Glazing Assembly
Glazing Assembly
GL1

GL2

Single pane, aluminum framed

IGU, aluminum framed
Dim. Material

Dim. Material
1/4 single pane glass

1/4 Glass Pane

1/4 Air Cavity

1/4 Glass Pane

Table 4.3-3: Roof Assembly Typologies
Roof Assembly
RT1

RT2

EPDM over rigid insulation

Cool roof over rigid insulation
Dim. Material

Dim. Material
1/8 EPDM

1/8 cool roof
4 rigid foam insulation

4 rigid foam insulation
1/5 vapor retarder

1/5 vapor retarder
6 reinforced concrete

6 reinforced concrete

65

Table 4.3-4: Trial Descriptions
Test Description
Wall
Type
Roof
Type
Glazing
Type Infiltration
HVAC
Type
1.0 Baseline WT1 RT1 GL1 L AC1
2.0 Window Rehabilitation: Re-glaze WT1 RT1 GL2 M AC1
3.0 Exterior Wall Modification WT2 RT1 GL2 M AC1
4.0 Re-roof WT1 RT2 GL1 L AC1
5.0 Cumulative Enclosure Renovation WT2 RT2 GL2 M AC1

4.3.3 Simulation Estimations and Conversions
Revit with Green Building Studio can estimate the total energy use intensity, electricity EUI, fuel
EUI and lifecycle energy use and costs. Lifecycle energy use is calculated using ASHRAE
standards for building schedule according to the defined function; in this case this function is a
typical commercial office on a 40 hr. weekday work schedule. Green Building Studio uses
electricity and fuel costs specific to the defined city to estimate annual expenditures.
Additionally, Green Building Studio estimates the annual equivalent carbon emissions in tons/yr.
Values for fuel consumption are reported in Therms; these values were converted to MBTU for
consistency with CBECS reporting style.
4.3.4 Model Validation
In order to determine if the characteristics selected for the test building were reasonable for the
study, the test building was compared to 15 buildings from the CBECS dataset. The 15 buildings
created Sample C, which was considered to be the buildings whose characteristics best matched
those selected for the test building. Some buildings within Sample C reported a window
replacement, wall renovation, or re-roof. A baseline energy performance trial was run on the test
building for the four selected locations. The baseline electricity and fuel consumption values
were compared to reported consumption values for buildings in Sample C.

66

Sample C had the following characteristics:
Table 4.3-5: Sample C Characteristics

Mean Stdv
Build Area (sqft.) 3043 1232
Building EUI 47.4 32
HVAC EUI 27.1 20
Energy Cost/sqft. $1.23 1

For all four locations, the estimated building EUI was comparable to values in Sample C. The
estimated fuel EUI values for the test building in Los Angeles, CA and New Orleans, LA were
relatively low when compared to Sample C. However, the estimated EUI values in Chicago, IL
and Louisville, KY were within reasonable range to mean value for Sample C.
Figure 4.3-6: EUI for Sample C

All four locations reported energy cost per square foot values in consistency with the mean of
Sample C. However, these values for sample C showed a relatively large standard deviation (1.0)
because the data is quite dispersed.
0
20
40
60
80
100
120
140
EUI [kBTU/sqft]
Building Energy Use Intensity
67

Figure 4.3-7: EUI for HVAC for Sample C

Figure 4.3-8: Energy Cost per Square Foot for Sample C

4.3.5 Results: New Orleans, LA
New Orleans was selected for analysis to represent ASHRAE climate zone 2 and CBECS climate
code 4 for hot-humid climates. The design low temperature is 30 degrees Fahrenheit and the
design high temperature is 96 degrees Fahrenheit. Electricity costs are $0.08/kWh and fuel costs
are $0.91/therm. New Orleans has the highest fuel costs out of all selected cities.
0
10
20
30
40
50
60
70
80
EUI [kBTU/sqft]
Energy Use Intensity for HVAC
$0.00
$0.50
$1.00
$1.50
$2.00
$2.50
$3.00
Total Energy Cost per Square Foot
68

Baseline Performance
The total building EUI was 55 for the test building in New Orleans, LA. The electricity EUI was
15 and the fuel EUI was 5. The lifecycle electricity use over a 30 year period was the greatest of
the four tested cities at 2,015,831 kWh. By comparison, the lifecycle fuel use was 653,400
MBTU, the least of all tested locations.
Annual electricity use was estimated at 67,194 kWh which is 91% of the building’s energy use.
Approximately 43% of electricity use powers HVAC in the building (29,163 kWh). Annual fuel
use was estimated at 21,700 MBTU. About half (51%) of that amount, equivalent to 11,200
MBTU, is consumed by HVAC.
Annual carbon emissions were estimated at 33 CO2e tons/yr. for electricity and 1 CO2e tons/yr.
for fuel.
Energy Use Intensity
The façade renovations did not have any effect on the estimated total building EUI. Both
electricity and fuel EUI remained unchanged over the four trials.
Figure 4.3-9: New Orleans Total Building EUI

Lifecycle Energy Use and Cost
The lifecycle energy cost decreased when renovations were made to the exterior wall (Trial 3)
and when the windows were replaced (Trial 2).
T1 T2 T3 T4 T5
Total EUI 55 55 55 55 55
0
10
20
30
40
50
60
MBTU/sq. ft./yr
Total EUI
69

Figure 4.3-10: New Orleans Lifecycle Energy Cost

Annual Energy Use
Generally, annual fuel and electricity use decreased in Trials 2 and 3 which represented the
window and wall renovations, respectively. According to the study 800 MBTU of fuel and 1,316
kWh of electricity could be saved. This equated to about $113 in annual savings.
Figure 4.3-11: New Orleans Annual Fuel Use for HVAC

T1 T2 T3 T4 T5
Energy Cost $75,729 $74,738 $74,196 $75,729 $74,196
$73,000
$73,500
$74,000
$74,500
$75,000
$75,500
$76,000
USD
Lifecycle Energy Cost
T1 T2 T3 T4 T5
HVAC Fuel 11,200 10,800 10,400 11,200 10,400
10,000
10,200
10,400
10,600
10,800
11,000
11,200
11,400
MBtu
Annual Fuel Use for HVAC
70

Figure 4.3-12: New Orleans Annual Electricity Use for HVAC

Figure 4.3-13: New Orleans Annual HVAC Operating Cost

Overall Performance
In New Orleans, LA, no measurable difference in EUI or carbon emissions was detected across
all five trials. However, the Trial 3 and Trial 5 showed a reduction in energy and fuel use after
the simulated renovation. The following figures show the difference between the baseline and the
cumulative façade renovation performance.
T1 T2 T3 T4 T5
HVAC Elec 20,064 19,051 18,908 20,064 18,908
18,200
18,400
18,600
18,800
19,000
19,200
19,400
19,600
19,800
20,000
20,200
kWh
Annual Electricity Use for HVAC
T1 T2 T3 T4 T5
HVAC Elec Cost $2,327 $2,258 $2,242 $2,327 $2,242
HVAC Fuel Cost $102 $98 $95 $102 $95
$0
$500
$1,000
$1,500
$2,000
$2,500
$3,000
Cost, USD
Annual HVAC Operating Cost
71

Figure 4.3-14: New Orleans Percent Difference

4.3.6 Results: Los Angeles, CA
Los Angeles, CA was selected for analysis to represent ASHRAE climate zone 3 and CBECS
climate code 3 for hot-dry/mixed-dry climates. The design low temperature is 40 degrees
Fahrenheit and the design high temperature is 95 degrees Fahrenheit. Electricity costs are
$0.12/kWh and Fuel costs are $0.80/Therm. Los Angeles has the highest electricity costs out of
all selected cities.
Baseline Performance
Los Angeles reported the lowest total building EUI at 54 MBTU/sq. ft./yr. for baseline
conditions in the test building. The electricity EUI was 13 and the fuel EUI was 10. Over a 30-
year period, the lifecycle electricity use was 1,742,862 kWh and the lifecycle fuel use was
1,391,500 MBTU.
Annual electricity use was estimated at 58,095 kWh, roughly 81% of the building’s energy use.
About 73% of electricity use powers HVAC in the building (20,064 kWh). Fuel accounted for
19% of annual energy use at an estimated 46,300 MBTU. Of that, 73% is used towards HVAC,
an equivalent to 34,100 MBTU.
0% 0% 0%
Electricity EUI Fuel EUI Total EUI
Percent Difference:
Energy Use Intensity
2%
4%
2%
Electricity Use Fuel Use Energy Cost
Percent Difference:
Life Cycle Analysis
0% 0%
Electricity Fuel
Percent Difference:
Annual Carbon
Emissions
4%
7%
Elec-HVAC Fuel-HVAC
Percent Difference:
Annual HVAC Energy
Use
72

Annual carbon emissions were estimated at 19 CO2e tons/yr. for electricity and 2 CO2e tons/yr.
for fuel.
Energy Use Intensity
The façade renovations had a very small effect on the estimated total building EUI. Both
electricity and fuel EUI dropped from 54 to 53 in Trials 2, 3, and, 5.
Figure 4.3-15: Los Angeles Total Building EUI

Lifecycle Energy Use and Cost
Figure 4.3-16: Los Angeles Lifecycle Energy Cost

T1 T2 T3 T4 T5
Total EUI 54 53 53 54 53
52.4
52.6
52.8
53
53.2
53.4
53.6
53.8
54
54.2
MBTU/sq. ft./yr
Total EUI
T1 T2 T3 T4 T5
Energy Cost $98,362 $96,718 $95,960 $98,362 $95,960
$94,500
$95,000
$95,500
$96,000
$96,500
$97,000
$97,500
$98,000
$98,500
$99,000
USD
Lifecycle Energy Cost
73

Annual Energy Use
Annual fuel and electricity use decreased in Trials 2 and 3 which represented the window and
wall renovations, respectively. According to the study 1,200 MBTU of fuel and 1,156 kWh of
electricity could be saved. This equated to about $146 in annual savings.
Figure 4.3-17: Los Angeles Annual Electricity Use for HVAC

Figure 4.3-18: Los Angeles Annual Fuel Use for HVAC

T1 T2 T3 T4 T5
HVAC Elec 20,064 19,051 18,908 20,064 18,908
18,200
18,400
18,600
18,800
19,000
19,200
19,400
19,600
19,800
20,000
20,200
kWh
Annual Electricity Use for HVAC
T1 T2 T3 T4 T5
HVAC Fuel 34,100 39,900 32,900 34,100 32,900
-
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
MBtu
Annual Fuel Use for HVAC
74

Figure 4.3-19: Los Angeles Annual HVAC Operating Cost

Overall Performance
The following figures show the difference between the baseline and the cumulative façade
renovation performance.
Figure 4.3-20: Los Angeles Percent Difference

T1 T2 T3 T4 T5
HVAC Elec Cost $2,365 $2,246 $2,229 $2,365 $2,229
HVAC Fuel Cost $274 $272 $264 $274 $264
$0
$500
$1,000
$1,500
$2,000
$2,500
$3,000
Cost, USD
Annual HVAC Operating Cost
0% 0%
2%
Electricity EUI Fuel EUI Total EUI
Percent Difference:
Energy Use Intensity
2%
8%
2%
Electricity Use Fuel Use Energy Cost
Percent Difference:
Life Cycle Analysis
2%
0%
Electricity Fuel
Percent Difference:
Annual Carbon
Emissions
6%
4%
Elec-HVAC Fuel-HVAC
Percent Difference:
Annual HVAC Energy
Use
75

4.3.7 Results: Louisville, KY
Louisville, KY was selected for analysis to represent ASHRAE climate zone 4 and CBECS
climate code 2 for mixed-humid climates. The design low temperature is -1 degrees Fahrenheit
and the design high temperature is 91 degrees Fahrenheit. Electricity costs are $0.08/kWh and
Fuel costs are $0.85/thm.
Baseline Performance
The total building EUI was 76 for the test building in Louisville, KY. The electricity EUI was 13
and the fuel EUI was 33. The lifecycle electricity use over a 30 year period was 1,717,283 kWh
and the lifecycle fuel use was 4,415,300 MBTU.
Annual electricity use was estimated at 57,242 kWh, roughly 57% of the building’s energy use.
About 34% of electricity use powers HVAC in the building (19,212 kWh). Annual fuel use was
estimated at 147,100 MBTU. Roughly 90% of that amount, equivalent to 133,700 MBTU, is
consumed by HVAC.
Annual carbon emissions were estimated at 44 CO2e tons/yr. for electricity and 8 CO2e tons/yr.
for fuel.
Energy Use Intensity
Across the five trials, there was very little change in total building EUI. The initial value was 76,
and trials 2, 3, and 5 each reduced that value to 75.
Figure 4.3-21: Louisville Total Building EUI

1 2 3 4 5
Series1 76 75 75 76 75
74.4
74.6
74.8
75
75.2
75.4
75.6
75.8
76
76.2
EUI
Total EUI
76

Lifecycle Energy Use and Cost
The lifecycle energy cost was reduced from $82,193 to $80,511 in Trials 3 and 5.
Figure 4.3-22: Louisville Lifecycle Energy Cost

Annual Energy Use
Annual fuel and electricity use decreased in Trials 2 and 3 which represented the window and
wall renovations, respectively. According to the study 2,100 MBTU of fuel and 1,002 kWh of
electricity could be saved. This equated to about $102 in annual savings.
Figure 4.3-23: Louisville Annual Electricity Use for HVAC

1 2 3 4 5
Energy Cost $82,193 $81,163 $80,511 $82,193 $80,511
$79,500
$80,000
$80,500
$81,000
$81,500
$82,000
$82,500
Cost, USD
Lifecycle Energy Cost
1 2 3 4 5
HVAC Elec 19,212 18,338 18,210 19,212 18,210
17,600
17,800
18,000
18,200
18,400
18,600
18,800
19,000
19,200
19,400
kWh
Annual Electricity Use for HVAC
77

Figure 4.3-24: Louisville Annual Fuel Use for HVAC

Figure 4.3-25: Louisville Annual HVAC Operating Cost

Overall Performance
The following figures show the difference between the baseline and the cumulative façade
renovation performance.
1 2 3 4 5
HVAC Fuel 1,337 1,334 1,316 1,337 1,316
1,305
1,310
1,315
1,320
1,325
1,330
1,335
1,340
Therm
Annual Fuel Use for HVAC
1 2 3 4 5
HVAC Elec Cost $1,606 $1,533 $1,522 $1,606 $1,522
HVAC Fuel Cost $1,135 $1,133 $1,117 $1,135 $1,117
$0
$500
$1,000
$1,500
$2,000
$2,500
$3,000
Cost, USD
Annual HVAC Operating Cost
78

Figure 4.3-26: Louisville Percent Difference

4.3.8 Results: Chicago, IL
Chicago, IL was selected for analysis to represent ASHRAE climate zone 5 and CBECS climate
code 1 for very cold/cold climates. The design low temperature is -9 degrees Fahrenheit and the
design high temperature is 96 degrees Fahrenheit. Electricity costs are $0.08/kWh and Fuel costs
are $0.79/Therm.
Baseline Performance
The total building EUI was 96 for the test building in Chicago, IL, the greatest value reported as
the baseline in all four tested cities. Much of this was energy from electricity (EUI of 12) and
fuel (EUI of 54). The lifecycle electricity use over a 30-year period was 1,651,420 kWh and the
lifecycle fuel use was 7,369,400 MBTU.
Annual electricity use was estimated at 55,047 kWh, roughly 43% of the building’s energy use.
About 31% of electricity use powers HVAC in the building (17,0176 kWh). Chicago is
characterized by very cold winters; this is reflected in the high reported estimates for fuel
consumption. Annual fuel use was estimated at 245,600 MBTU. Roughly 93% of that amount,
equivalent to 230,600 MBTU, is consumed by HVAC.
8%
3%
1%
Electricity EUI Fuel EUI Total EUI
Percent Difference:
Energy Use Intensity
2%
1%
2%
Electricity Use Fuel Use Energy Cost
Percent Difference:
Life Cycle Analysis
2%
0%
Electricity Fuel
Percent Difference:
Annual Carbon
Emissions
5%
2%
Elec-HVAC Fuel-HVAC
Percent Difference:
Annual HVAC Energy
Use
79

Annual carbon emissions were estimated at 42 CO2e tons/yr. for electricity and 14 CO2e tons/yr.
for fuel.

Energy Use Intensity
The façade renovations had a very small effect on the estimated total building EUI. Both
electricity and fuel EUI dropped from 54 to 53 in Trials 2, 3, and, 5.
Figure 4.3-27: Chicago Total Building EUI

Lifecycle Energy Use and Cost
Figure 4.3-28: Chicago Lifecycle Energy Use and Cost

T1 T2 T3 T4 T5
Total EUI 96 95 95 96 95
94.4
94.6
94.8
95
95.2
95.4
95.6
95.8
96
96.2
MBTU/sq. ft./yr
Total EUI
T1 T2 T3 T4 T5
Energy Cost $87,459 $86,726 $85,986 $87,459 $85,986
$85,000
$85,500
$86,000
$86,500
$87,000
$87,500
$88,000
USD
Lifecycle Energy Cost
80

Annual Energy Use
Annual fuel and electricity use decreased in Trials 2 and 3 which represented the window and
wall renovations, respectively. According to the study 2,600 MBTU of fuel and 812 kWh of
electricity could be saved. This equated to about $87 in annual savings.
Figure 4.3-29: Chicago Annual Electricity Use for HVAC

Figure 4.3-30: Chicago Annual Fuel Use For HVAC

T1 T2 T3 T4 T5
HVAC Elec 17,016 16,365 16,204 17,016 16,204
15,600
15,800
16,000
16,200
16,400
16,600
16,800
17,000
17,200
kWh
Annual Electricity Use for HVAC
T1 T2 T3 T4 T5
HVAC Fuel 230,600 230,500 228,000 230,600 228,000
226,500
227,000
227,500
228,000
228,500
229,000
229,500
230,000
230,500
231,000
MBtu
Annual Fuel Use for HVAC
81

Figure 4.3-31: Chicago Annual HVAC Operating Cost

Overall Performance
The following figures show the difference between the baseline and the cumulative façade
renovation performance.
Figure 4.3-32: Chicago Percent Difference

T1 T2 T3 T4 T5
HVAC Elec Cost $1,383 $1,330 $1,806 $1,383 $1,317
HVAC Fuel Cost $1,827 $1,826 $1,806 $2,306 $1,806
$0
$500
$1,000
$1,500
$2,000
$2,500
$3,000
$3,500
$4,000
Cost, USD
Annual HVAC Operating Cost
0% 0%
1%
Electricity EUI Fuel EUI Total EUI
Percent Difference:
Energy Use Intensity
2%
1%
2%
Electricity Use Fuel Use Energy Cost
Percent Difference:
Life Cycle Analysis
2%
0%
Electricity Fuel
Percent Difference:
Annual Carbon
Emissions
5%
1%
Elec-HVAC Fuel-HVAC
Percent Difference:
Annual HVAC Energy
Use
82

4.4 Chapter Summary
Mathematical analysis and geometric simulation were used to generate an understanding of
energy use in 1-story, commercial offices. Mathematical analysis provided context for current
performance of buildings in the domain. This directed the type of building that could serve as a
representative model. Some conflicts existed between frequently occurring characteristics and
practical characteristics to model for simulation. These were resolved and the test building
showed performance values comparable to buildings in Sample C. Simulation was conducted in
four climate zones. Louisville showed the greatest percent difference for Building EUI post-
façade renovation. Los Angeles showed the greatest post-renovation percent difference for total
fuel and electricity use. Chicago and Louisville estimated a 2 percent reduction in annual carbon
emissions. In all four locations, annual electricity savings for HVAC ranged from 3 to 6 percent.
Annual fuel use for HVAC ranged from 1 to 7 percent with Chicago showing the least and New
Orleans showing the greatest potential for reduction post-renovation.

83

Chapter 5: Interpretation
5.1 Chapter Overview
The CBECS dataset was useful in providing general and basic geometric information about
commercial offices in the U.S. building stock and their estimated energy consumption. In order
to improve a mathematical analysis of façade performance of commercial offices in the U.S.,
more detailed information is needed on the thickness and materials used in wall, window, and
roofing assemblies. Despite these limitations, a theoretical test building was created based on
suggestions from Sample B of the CBECS dataset.
A baseline simulation for energy consumption was conducted on the test building. Results were
compared to buildings in Sample C. Since consumption estimates of the test building fell within
statistical range of Sample C, it was accepted as an acceptable representative model. This
indicated that while limited information was provided on façade assemblies, it is possible to
produce a representation when supporting knowledge of industry practices are taken into
account.
Façade renovations on the test building were simulated in four climate zones. Generally, small
decreases in electricity and fuel use for HVAC were measured when modifications were made to
the windows and wall assembly. No impact was measured for renovations to the roof finish
material. The simulation offers the ability to observe the difference the impact of the renovations
across four climates and speculate what variables are of most importance in each climate.
Finally, the results of mathematical analysis and simulation allow for forecasting the impact of
façade renovation in existing buildings. For a more accurate forecast, additional testing is needed
with detail given to material properties and thicknesses.
5.2 Interpretation of Mathematical Analysis
5.2.1 Facilitators and Limitations of Sample B
Population and Sample Size
The CBECS dataset forming the population of buildings representing the U.S. building stock
included a variety of buildings with different functions, geometries, and characteristics. The
population needed to be filtered to a smaller sample size, Sample B, in order to use it to make
educated predictions about building energy performance. Reducing the size of the population by
collecting a sample is often not favorable in mathematical analysis. However, since façade
performance is largely dependent on geometric properties, such as shape and surface area, it was
necessary to use a sample size with such characteristics held as control variables. Similarly,
energy performance of a building is related to the size and function of the building. Accordingly,
square footage and principle building activity were held as control variables as well. While these
84

control variables filtered the original population of 6700 to a sample size of roughly 160, the
buildings within Sample B had more in common which increases the likelihood that they will
have similar building energy and facade performance.
Façade Characteristics
Most buildings within Sample B reported having and exterior wall construction material of brick,
stone, or stucco (category 1) (Figure 4.2-2). This seems reasonable considering that Sample B
included relatively small, 1-sotry office buildings. Categories 3, 4, and 5 captured the majority of
the remaining buildings in Sample B. These categories included typical construction materials of
concrete, siding, and metal panels, which are used quite frequently in low rise buildings in the
U.S. If the study included high rise commercial offices more buildings may have reported
having all glazed facades.
There was less variation in reporting for roof tilt (Figure 4.2-4). The majority of buildings in
Sample B reported a shallow pitch. This was closely followed by steeper pitch and flat roof. For
roof materials, a large number of buildings reported using asphalt or fiberglass shingles (Figure
4.2-3). This was followed by category 5, metal surfacing, and category 2, slate or tile shingles.
This differs greatly from buildings in Sample A, which reported flat roofs with synthetic
sheeting. The decision was made to select a flat roof for the simulation model in order to reduce
the impact of roof tilt and orientation. If the test building was modeled with a sloped roof, the
orientation of the building would become more significant. While flat roofs are not preferred in
some climates, such as those that receive large amounts of snow, it offered the most general roof
type for simulations across four ASHRAE climate zones.
The building shape and orientation have an effect on the amount of sun exposure the façade
receives. CBECS does not provide much information as to how an individual building is
oriented. However, the survey offers 11 categories to report building shape. Most buildings in
the whole population and in Sample B reported having category 2, wide rectangle (Figure 4.2-5).
This was interpreted as a building with a ration closer to 2:1 than 4:1. It is not surprising that
most buildings reported rectangular or square shapes as this is a typical form in U.S. architecture.
Older buildings would expect to have rectangular or “alphabet” shapes to allow for more
sunlight.
Most buildings in Sample B reported having 2 to 10 percent of the façade made up of glass. The
next most popular categories were less than 1 percent and 11 to 25 percent. Energy code suggests
less than 40% exterior glass is ideal for satisfactory energy performance (ICC 2012). Since the
sample set reported lower values in exterior glazing, it was expected that modifications to the
windows would not produce a large difference in energy consumption. However, the Revit
Energy Analysis report on the baseline test building reported that window conductive and
window solar were significant factors affecting heating and cooling loads in some climates. The
85

test building had approximately 20% window coverage across the exterior façade. Therefore,
window to wall ratio is important to consider even in buildings with relatively low values.
5.2.2 Predictions on Consumption
The intent of the mathematical analysis was to use data from existing buildings to make
reasonable predictions about the effect of façade characteristics on energy consumption.
Specifically, the goal was to show that strategically changing the façade characteristics in a
renovation can reduce energy consumption. Accordingly, the reported values for energy
consumption were reviewed with intent to identify relationships and make an educated guess as
to the expected values if strategic changes are made to the façade.
The Empirical CDF of Consumption
Histograms of annual electricity and fuel consumption showed greater number of buildings
reporting low consumption values and decreasing number of building reporting higher values for
consumption. The cumulative distribution functions of these vectors were plotted. After
reviewing the CDF plot, the hypothesis that fuel consumption follows an exponential distribution
was tested using the Lilliefors test. The test returned a rejection of the hypothesis. This is not
unexpected since the hypothesis was based purely on observation.
Figure 5.2-1: CDF of Major Fuel Use

There were cases where a hypothesis was made as to a possible distribution but testing proved
inconclusive. For example, fuel expenditure was tested against exponential distribution using the
Lilliefors test a returned a failure to reject the hypothesis. The reported p value was 0.001, which
doubts the validity of the hypothesis. The assumption is that the data does not truly follow an
86

exponential distribution within a 5% significance level. The test was repeated at a 10%
significance level and presented similar behavior. Therefore, the data does not follow an
exponential distribution within any reasonable significance level.
Figure 5.2-2: CDF of Fuel Expenditure

A similar process occurred with the analysis of electricity consumption.
Energy Use vs. Façade Characteristics
Based on the statistical description of Sample B, buildings using an exterior wall material of
brick, stucco, or stone consume the most electricity and fuel for HVAC (Figure 4.2-13 and
Figure 4.2-15). Due to this, it was expected that the test building would generate estimates of
relatively high energy consumption when compared to other buildings of similar size and
occupancy schedule. This was reflected in the results.
When looking at the relationship between roof constructional material and energy consumption
in 1-story buildings, there was no clear correlation (Figure 4.2-14 and Figure 4.2-16). It is
assumed that this is because the heating and cooling loads are more likely impacted by the
thermal properties of the total roof assembly than by the characteristics of the external roof finish
material. While it is known that cool roofs may impact cooling load in some climates, a more
detailed study of roof assemblies is necessary to determine the performance of each roof finish
material.
The exterior wall material and roof material were used to make assumptions about the enclosure
assembly. The intent was to use this as a baseline condition and then modify it and detect any
changes in performance. Overall, it is understood that the enclosure material characteristics
provided in the CBECS dataset do not provide enough information to make reasonable
87

conclusions about overall thermal enclosure performance. This showed that a more detailed
simulation was necessary and that some parameters of the enclosure assembly would need to be
inferred.
The Possibility for Correlation
It was not expected that any façade parameters would show strong linear correlation to energy
consumption. This is because, as mentioned previously, the CBECS dataset provides information
on finishes rather than full assemblies. Despite this knowledge, a few variables were tested using
the correlation coefficient strategy.
Square footage and percentage of exterior glass were tested against total building EUI, electricity
consumption, fuel consumption, electricity used for HVAC, and fuel used for HVAC. Both
Sample A ad Sample B were tested for linear correlation. It was assumed that Sample A would
show stronger correlation, if any, due to the larger quantity of data included in the sample set.
However, neither dataset produced coefficients of absolute value higher than 0.36. This indicates
that neither a linear nor inverse relationship can be assumed from the variables.
5.2.3 Producing a Test Building
Although the mathematical analysis made it clear that the CBECS data has limitations, it did
provide satisfactory information to design a test building. Whether the test building truly is
representative of a large portion of commercial buildings in the U.S. building stock is still in
question. The main matter of importance is whether the test building represents the façade
characteristics of many buildings; this is necessary in order to have a baseline building whose
façade performs as that of a “typical” building would.
Firstly, the majority of buildings in the U.S. building stock filed their primary building activity as
“commercial office”. While this does not necessarily provide information about façade
characteristics, it determines the anticipated occupancy schedule of building, as defined by
ASHRAE Standard 90.1. This is necessary for analysis because the occupancy will determine the
quantity of hours where space is conditioned annually. Therefore, this was considered a
characteristic of high impact and it was beneficial in the definition of a typical commercial
building.
The ability to match the square footage category to that most frequently occurring in the CBECS
dataset was necessary for two reasons. First, the square footage of heated and cooled space is
directly correlated to the amount of energy the building consumes. Secondly, the square footage
of a building is related to the surface area of the building, a variable related to the building
façade. In addition, the building shape, number of floors, and floor to ceiling height all played
into the geometric properties of the building that relate to the façade. Therefore, choosing a 1-
88

story, 5,000 square foot, rectangular-shaped building as the test model seemed to be a logical
decision in a façade analysis study.
CBECS provides variables that describe the exterior finish material for walls and roofs.
Additionally, it questions whether or not a cool roof finish material was used. The most common
wall construction material was the category for brick, stucco, or siding. Brick was selected for
the test building. The most common roof tilt was a shallow-pitched roof. In order to alleviate the
issue of roof tilt and building orientation, a flat roof was selected for the test building. This
meant that the roof construction material and finish would need to be a material compatible with
flat roofs. Although asphalt shingles was the most commonly-reported roof material, a
plastic/rubber roof material was used because it was the most frequently occurring material that
is logical for application on a flat roof.
The glazing related variables that CBECS records are: percent of exterior glass, whether there is
equal glass on all sides, and whether or not most glass faces the sun path. These are façade-
specific parameters that can help indicate the potential for solar heat gain. Since solar heat gain is
a factor that is beneficial in some seasons in some climates and harmful in orders, it was
necessary to include these parameters in the study. The percent of exterior glass in Sample A was
20% but this value was only 1-10% in Sample B. The test building had a wind to wall ratio
between 10% and 20%. It was an east-west oriented rectangle that did not have equal glass on all
sides and had most glass facing south. These decisions were made to be account for the lack of
information on shading by surrounding buildings, shading on glazing surfaces, and variations in
building orientations. It would have been helpful to know more details about glazing
characteristics such as frame types, glazing thickness, and number of glass panes. These
parameters were determined based on standard of practice for buildings constructed before 1990.
The mathematical analysis used statistical methods to determine what characteristics were most
frequently occurring in 1-story, commercial office buildings. The characteristics selected for the
test building did not match these mode values exactly, but used them when feasible. While more
detailed information could have been useful to help define façade-specific variables, sufficient
data was provided by CBECS to roughly represent a typical office building. The parameters
provided by CBECS were meant to be a guideline for the definition of the test building. In this
case, they proved to be so.
5.3 Interpretation of Simulation
5.3.1 Theoretical Test Building
The baseline test building used aluminum-framed, ¼ in. single pane windows. In Trial 2, these
windows were replaced with aluminum-framed IGU composed of two ¼ glass panes with an air
cavity. This window assembly was selected because it is known to yield increased thermal
89

properties. It was tested to determine the difference in performance of the window across the
four climate zones represented.
The façade renovation selected for the exterior walls was to add an exterior layer onto the
existing façade. This was because there was interest in determining the effect of additional
insulation on wall performance. This can be done, without demolishing the existing façade, by
adding an exterior layer of continuous exterior insulation and a relatively thin finish material.
The window modification was included in the wall renovation because adding thickness to the
wall would affect the windows. Additionally, it is practical to replace windows during an exterior
wall renovation.
It was assumed that air tightness could be improved during an exterior façade renovation;
therefor, the ACH was changed from loose in the baseline case to medium in Trials 2 and 3. The
true tightness of construction is not often known with confidence or measured for verification.
However, indoor air quality studies have shown improvements in weather-sealing assemblies and
installation methods in the years following the introduction of HVAC in commercial buildings
(U.S. Environmental Protection Agency 2016).
The simulated re-roof modified the EPDM roofing used in the baseline test building to a cool
roof material in Trial 4. The intention was to reduce black body radiation on the roof surface and
in term reduce heat gain through the roof in the cooling season. This had not measurable impact
on the test building in any of the four climate zones. This indicates that a more detailed
simulation method is needed to analyze the impact of cool roofing materials across different
climate zones.
Results of the baseline trial show that estimated energy consumption in the test building fell
within the range of buildings in Sample C. Accordingly, the theoretical test building was
considered sufficient for the study. The ability to create a building model that accurately
represents a range of existing buildings is limited by the level of detail the model can hold and
the amount of detailed data available on the existing building stock.
5.3.2 Comparison by Renovation Trial
New Orleans, LA
New Orleans represented ASHRAE climate zone 2 and CBECS climate code 4 for hot-humid
climates. ASHRAE Standard 90.1 records 1513 HDD and 6910 CDD for New Orleans.
Therefore, a façade renovation that reduces cooling load is ideal for this climate.
Trial 2, which added additional insulation to exterior walls reduced fuel use for HVAC by 400
MBTU annually and electricity use for HVAC by 1013 kWh annually. Trial 3, which replaced
single pane windows with IGU, reduced fuel by an additional 400 MBTU but had little impact
on electricity use. Trial 4, the re-roof, had no impact on fuel or electricity use. Overall, this tells
90

us that buildings in hot-humid climates that use both fuel and electricity about equally as energy
sources can see energy savings by increasing insulative properties of wall and window
assemblies. If less heat is conducted through the building enclosure during the cooling season,
the cooling load is reduced.
Los Angeles, CA
Los Angeles, CA represented ASHRAE climate zone 3 and CBECS climate code 3 for hot-
dry/mixed-dry climates. Los Angeles has 1458 HDD and 4777 CDD. While Los Angeles is
cooling-dominated climate, it differs from other test locations because it is a relatively dry
climate. This reduces the amount of energy consumed for dehumidifying outside air.
In Los Angeles, Trial 2 reduced annual electricity consumption for HVAC by 1,013 kWh and
annual fuel use for HVAC by 200 MBTU. Trial 3 produced a greater reduction (1000 MBTU) in
fuel use for HVAC. Increasing thermal performance of the walls had a greater impact on
electricity use, while increasing performance of the windows had a greater impact on fuel use.
This indicates an impact from solar heat gain (direct or via radiation) in addition to typical
considerations for heat conduction from outside the air temperature.
Trial 4 had no impact on energy performance. Accordingly, the results of the cumulative
renovation for Trial 5 replicate those of Trial 3.
Louisville, KY
Louisville, KY represented ASHRAE climate zone 4 and CBECS climate code 2 for mixed-
humid climates. Louisville has 4514 HDD and 4000 CDD. Since buildings in this climate shares
and relatively equal amount of time in heating and cooling mode, façade performance should not
be optimized for one season.
The test building was estimated to produce a 5% reduction in electricity use for HVAC and a 2%
reduction in fuel use for HVAC. According to the energy report, window solar heat gain was the
largest façade-specific factor contributing to the cooling load. In the heating season, window
conductive heat loss was the largest façade-specific factor contributing to heat loss. By replacing
windows with IGU, annual fuel use for HVAC was reduced by 1,800 MBTU. This is greater
than reported reductions in New Orleans and Los Angeles.
Chicago, IL
Chicago, IL represented ASHRAE climate zone 5 and CBECS climate code 1 for cold climates.
Chicago has 6536 HDD and 2941 CDD.
During the cooling season, window solar heat gain is the largest façade related factor
contributing to the cooling load. Thermal performance of the wall and roof assemblies is critical
91

during July and August for cooling and December through March for heating. Electricity
consumption for HVAC was reduced by 651 kWh from Trial 1 to Trial 2. An additional 161
kWh was saved by replacing windows in Trial 3. The total annual electricity savings is estimated
at 812 kWh. Very little fuel savings was produced from Trial 1 to Trial 2. However, fuel
consumption was reduced by 2,600 MBTH in Trial 3. This was the largest savings produced by
the window modification in any climate.
5.3.3 Comparison across Test Locations
Out of all four locations tested, the greatest energy savings in terms of energy use intensity was
produced in Louisville, KY. In Louisville, electricity EUI was reduced by 7.69% in Trial 5, the
cumulative façade renovation. Fuel EUI was reduced by 3.03 % and the total building EUI was
reduced by 1.37%. For comparison, the test building produced a 1.03% savings in Chicago, IL
and a 1.85% savings in Los Angeles, CA in total building EUI.
It was expected that each climate zone would produce different results for energy use for HVAC.
A surprising low difference in energy use for HVAC occurred in Chicago, IL. When compared
to the baseline, the cumulative façade renovation had a lower impact than expected in fuel use
for HVAC. It was anticipated that adding additional insulation to the exterior walls would reduce
heat loss in the building during the heating season and improving the SHGC of windows would
reduce the heat gain during the cooling season. In Chicago, the test building showed a 1.13%
decrease in fuel use and a 4.77% decrease in electricity use for HVAC.
By comparison, Los Angeles, which is a relatively mild climate, showed a 3.52% decrease in
fuel use and 5.76% decrease in electricity use fir HVAC. This is likely due to the number of days
requiring heating and cooling in each location. According to ASHRAE Standard 90.1, Los
Angeles has 1458 HDD and 4777 CDD while Chicago has 6536 HDD and 2941 CDD. In both
locations, solar and conductive heat gains through windows were reduced. However, the impact
on heating and cooling load differed due to the intensity in each climate.
In Louisville, KY, the test building showed lower reductions in fuel use (1.57%) and higher
reductions in electricity use (5.22%) for HVAC. This performance is comparable to that in
Chicago. However, Louisville has almost equal number of heating and cooling degree days at
4514 HDD and 4000 CDD, respectively.
The effect of the cumulative renovation on energy use for HVAC was most unique in New
Orleans, LA. Results showed a 7.14% decrease in fuel use and a 3.63% decrease in electricity
use. This is significant because New Orleans is a cooling-dominated climate with 1513 HDD and
6910 CDD. Since only 50% of fuel use goes towards HVAC, it was not expected to see a great
difference in fuel use. Fuel use for HVAC was 70-93% in other tested locations.

92

5.4 Alignment with Industry Objectives
The U.S. Department of Energy reports that the median site EUI for a commercial office is 67.3
BTU/ sq. ft. Sample C reported a median site EUI of 56.7. The site EUI of the test building
varied from 20 to 70, depending on the location. While the EUI values for sample set and test
building did not fall far from the national average, more aggressive façade renovation strategies
will need to be researched in order to significantly reduce total building EUI.
Façade renovation is often used to address issues of aesthetic degradation and weather-resistance
that may exist in aging buildings. As shown, façade renovation may also offer the added benefit
of improved energy performance. Collectively, these benefits can work to extend the service life
of an existing building.
An additional objective would be to determine methods in which façade renovation can bring
older buildings up to current energy code requirements or better. This would mean that the
façade renovation would need to have a significant impact on the total building energy use.
Architecture 2030 aims for all new buildings, developments, and major renovations to be carbon
neutral by 2030. A façade renovation in which more than 25% of the building envelope
undergoes renovation is considered a major renovation by the Architecture 2030 standard.
Façade renovation can be paired with deep energy retrofits in order to achieve aggressive energy
reduction goals.
Figure 5.4-1: HVAC Energy Use

IL KY CA LA
Pre 288,659 199,251 102,558 110,704
Post 283,288 193,733 97,414 106,291
Arch 2030 202,061 139,476 71,791 77,493
-
50,000
100,000
150,000
200,000
250,000
300,000
350,000
Energy [kBTU]
HVAC Energy Use
Pre- and Post-renovation
93

5.5 Facilitators and Obstacles
5.5.1 Key Facilitators
The key facilitators of the study were (1) the ability to design a representative building based on
existing building data, and (2) the ability to detect changes in energy performance by making
alterations to the building façade.
The ability to define a representative building was key to carry out this study. Some researchers
use an existing building as a precedent. For example, Pomponi et. al. used a single building in
London as a case study and simulated various energy conservation methods on this building in
IES VE using local climate data (Pomponi, et al. 2015). Martinez, whose study focused
specifically on façade energy conservation methods, also used an existing building (Martinez
Arias 2013). The benefit to this strategy is that there is enough available information about the
building that the model can be calibrated to match known consumption values specific to the
building in the climate. While this provides detailed, location-specific results, the goal of this
study was to determine if a generalized case would be effective at representing a larger group.
The large variation in the total building EUI of the baseline building in each location indicates
that the general approach may require some refinement to produce acceptable estimations.
However, the test building produces consumption values comparable to buildings within Sample
C without detailed information on façade assemblies of building within the sample set.
Therefore, it was possible to generate a test building from characteristics of Sample B that
produced consumption values comparable to buildings in Sample C. This suggests that, in a
given climate zone, it is possible to use a knowledge of facade typologies to design a building
that represents a larger population. For such climate, this representative building can be used to
estimate energy performance as a result of the characteristics of the façade.
Since the building enclosure contributes to a portion of the heating and cooling load, it is
valuable to detect changes in these loads. The Revit analysis is a simplified method to estimate
building energy performance. EnergyPlus, IES VE, and eQuest are popular options for whole
building energy modeling. These software packages offer more detailed definition of HVAC
equipment and similar ability to modify enclosure assembly specifications. Therm, a program
made available by the Department of Energy, offers detailed thermal energy transfer through
façade assemblies, building components, and glazing elements but is not intended for whole
building energy models. Revit, with the help of Green Building Studio run by the EnergyPlus
enginer, was able to estimate heating and cooling loads based on climate data and detect what
factors of the building and environment contribution to such loads. In future development, this
study could be repeated using multiple software packages to compare the variation in estimates.
Overall, Revit offered enough information about loading conditions and energy consumption to
make reasonable assumptions about façade performance in different climates.
94

5.5.2 Limitations and Obstacles
Several obstacles were encountered over the course of the study. These varied from limitations in
available data to details overlooked in early analysis. These obstacles serve as potential areas of
improvement in future research on façade renovation.
Energy Audit Data
In early development, there were a few options of available data on the performance of
buildings. Some regional and state jurisdictions now require that building owners release energy
consumption data. While this was an option, the study was intended to be broad or general, and
apply to a variety of climate zones. Using a local city or state’s data would limit the data to a few
distinct climate zones. Ultimately, the decision was made to use the CBECS dataset because it is
publicly available and could allow for research to be duplicated or compared to existing studies
that use information from CBECS. As mentioned, the full CBECS dataset offered information on
roughly 6700 buildings. This was reduced to 164 buildings when filters were applied to create
Sample B, which defined the study domain. The limited number of data points reduces the
validity of the mathematical analysis.
Roof Material Analysis
CBECS provides information on the roof finish material, roof tilt, and whether or not a cool roof
is used. This was not sufficient information to make predicts on the effect of exterior roof
material on thermal performance. The total thermal properties of the roofing assembly determine
the rate of heat transfer through the roof into conditioned spaces. This was reflected in the
simulation as the roof material had no impact on overall electricity or fuel consumption in the
test building.
Heat Loss Estimation
Energy simulation in Revit approximates heat transfer through the wall enclosure in as a
generalized 3D heat flow approximation. In practice, thermal imaging studies have revealed that
many inconsistencies exist along the wall due to structural members, installation errors, and
mechanical and electrical fixtures and components (U.S. Department of Energy 2016). These
items are often a source of infiltration or unexpected thermal leaks. In order to produce a more
accurate simulation of façade performance, these factors will need to be included in the heat loss
estimation. In a future iteration of research, software geared towards 2D steady-state or transient
heat flow, such as Therm, could be used to determine more detailed information about heat flow
through the building enclosure.
95

Limited Test Cases
The study was limited to three types of façade renovation, namely walls, glazing, and roof
modifications, tested in four different climate zones. Each type of renovation tested only one
assembly that differed from the baseline case. This was done to simplify the number of trials and
diversity of changes. Adding additional test cases for each type of façade renovation is one
method to increase the impact of the study. For example, for the exterior wall renovation, five
different walls types can be tested in each city. This would result in 20 different cases and
ultimately add to the depth of knowledge about the performance of exterior wall assemblies. If
the approach for five assemblies was expanded to glazing and roof types, the total number of
tests would increase to 60. Additionally, the tests can be expanded to include seven ASHRAE
climate zones resulting in 105 tests. While this is a great number of tests, it may be possible to
automate a great deal of the simulation through a plug-in.
Accuracy of Simulation
Accuracy of simulation is another limitation of the study. Since it is known that simulation is an
estimation, it may be beneficial to understand how far off the simulated performance is from a
real building’s performance. While this study compared simulation results to building energy
consumption data from CBECS, an error approximation is another method to understanding the
differences between real and simulated performance. The error estimation could signal the
possibility of inaccurate results. One effective method to conduct this error analysis could be to
model a building with known parameters and energy consumption and compare the simulation to
the actual energy consumption. Another method would be to work with simulation software
developers to understand the algorithms used in analysis and identify error approximations that
correspond to the algorithms used.
5.6 Chapter Summary
The mathematical analysis was able to provide enough information to define a test building that
could represent a larger population of buildings. This was consistent with the hypothesis that
known data could help create a model. However, the mathematical analysis did not have enough
façade-specific information to make predictions about the performance of a building based on
characteristics of its façade. Hence, simulation was necessary. Simulation proved to be helpful in
estimating façade performance and estimating changes in consumption based on improvements
to the building façade. While the improvements differed across climate zones, the results
generally indicate that façade renovations tailored to the demands of the climate can reduce
electricity and fuel use for space heating and cooling. Overall, a more detailed study is necessary
to be able to make predictions about what façade renovations produce more savings in each
climate.

96

Chapter 6: Conclusions and Future Work
6.1 Conclusions
The contemporary approach to façade retrofit involves simulation and modeling in the early
conceptual design phase. This can help to produce a rough approximation of building envelope
performance prior to investing in construction. The focus of this research was to produce a
general model for the theoretical investigation of existing building enclosures and estimate the
potential for a renovation to the façade to reduce energy consumption use for heating and cooling
occupied spaces. Ultimately, existing buildings may be able to undergo a façade renovation to
bring them to current energy code standards or better. This could potentially extend the service
life of the building.
Buildings undergo façade retrofit for various reasons: aesthetic, thermal, structural, weather-
resistance, or more. A thermal performance-based renovation aims to reduce energy
consumption but can also address other aspects of the façade that may be of concern to a
building owner or designer.
Conventional methods for design of a façade retrofit generally involve some form of
investigation prior to design of the new system. This investigation can involve a detailed analysis
of energy use data or a physical investigation of the existing enclosure that can involve
investigative openings or non-destructive testing. Introducing digital modeling and simulation is
not intended to replace these existing methods of gathering and analyzing relevant data. Rather,
mathematical and geometric modeling provide additional information on thermal behavior.
Document review revealed that there are many existing studies of façade thermal performance
and potential for retrofit to offer improved thermal performance. Many of which focus on
improving performance of curtain wall systems. Despite the differences in façade types
considered, these studies offered guidance for what software to use, how to set up geometric
models, and what unaddressed challenges still exist in this field of study. This study differed
because simulation was conducted in Revit using the EnergyPlus engine.
Research on deep energy audits and retrofits were also useful. These studies often looked deeper
into HVAC equipment and focused specifically on the insulation used in the walls than the who
façade system. They are significant, however, due to their contribution to the understanding of
how to interpret energy audit data.
Research in machine learning and the potential for data-driven decision methodology was
relevant as it presented an automated strategy for façade retrofit decisions based on trends in
data.
97

The CBECS collects information on energy consumption that may relate to the thermal
performance of commercial buildings in various climate zones. An attempt was made to
marginalize this data such that it represented a key, frequently occurring group of buildings in
the U.S. building stock. Thus, the domain of the study was defined: 1-story, commercial office
buildings whose occupied square footage falls between 1,000 – 5,000 sq. ft. When additional
restricted parameters where applied to the dataset, it was reduced to only 164 representative
buildings. This subset, called Sample B, represented 3% of the total dataset. Sample B was used
to calibrate the geometric model.
One aspect of research that can be improved is the definition of targets for energy performance.
A target for performance in alignment with an existing system should be developed for the
façade retrofit. This target could be energy code requirements or more rigorous systems aiming
for large reductions in carbon emissions and net energy consumption.
The mathematical analysis involved reviewing statistical behavior of the data and subsequent
samples of data. The goal was to determine a statistical summary of the building stock and the
annual energy consumption of such buildings. Results verify anticipated relationships such as
those between building size and energy use or building age and energy use. However, results
showed little to no correlation between façade parameters and energy consumption. The
explanation for this is that CBECS does not collect information on the overall assembly or wall
R-value, only exterior construction materials of the walls and roof. This was not detailed enough
to make reasonable assumptions about the thermal behavior of the overall wall or roof assembly.
Some information on glazing on the façade is collected. CBECS reports the percentage of
exterior glass and the ratio across various elevation but the type and thickness of glazing is not
collected. Ultimately, such details not provided by CBECS were decided by what is commonly
used in practice in buildings constructed prior to 1990.
Since the geometric modeling was done a theoretical structure, rather than an existing building,
the model could not be calibrated using existing energy use data. To overcome this, the
theoretical model in four climate zones was compared to Sample C, 15 buildings with similar
characteristics from various climate zones. In Los Angeles and New Orleans, the electricity EUI
was close to the mean of Sample C, but the fuel EUI was significantly lower than the mean value
of the sample. In Chicago and Louisville, both the fuel and electricity EUI of the theoretical
building were comparable to the mean values of Sample C. It should be noted, that due to
differing climate zones represented in Sample C, there was a very large standard deviation in fuel
use.
Of the four locations tested, Louisville, KY showed the greatest reduction in electricity and fuel
building EUI. In this location, simulation estimated a 5.22 percent reduction in electricity use for
HVAC and a 1.57 percent reduction in fuel use for HVAC. However, in New Orleans, LA, the
test building estimated the largest reduction in fuel use for HVAC (7.14%) and the lowest
98

reduction in electricity use for HVAC (3.63%). These differences in energy use across climate
zones indicates that more attention should be given to which type of façade renovation should
take place in each climate.
Generally, the same pattern for changes in total energy consumption were produced over the
trials of façade renovation regardless of the location. The main difference was in the magnitude
of savings in each location and the type of fuel for which savings was produced. This is
consistent with known information about façade performance: fuel most often used for heating,
so places with a large number of heating degree days the façade should increase insulation to
minimize heat loss.
After conducting both mathematical analysis and geometric simulation, a few items were
revealed. The quality and efficacy of theoretical modeling for façade performance is affected by
the level of detail that simulation software can interpret and the scale of the building itself. Revit
allows for families to be defined to include materials with their thermal properties. It uses the
EnergyPlus approximation method for estimating infiltration through walls, floors, and the roof.
These methods for approximate are useful but must be considered relative to the scale of the
building. This study modeled a rectangular building with a 2:1 ratio in plan. The high facade
surface area to building volume ratio implies that the performance of the façade is significant
because much of the building’s conditioned space is within considerable proximity to exterior
walls. Multi-story buildings and buildings with low surface area to building volume ratios may
be affected by façade performance on a smaller scale. Overall, buildings considered good
candidates for façade renovation would be those whose thermal loads are directly impacted by
the performance of the façade.
The façade renovation was estimated to have relatively low impact on the total building EUI.
Since it is known that the façade is estimated to make up roughly 30 to 45 percent of factors
affecting energy performance, this speaks to the notion that space heating and cooling are not the
only factors that should be considered in the design of a façade renovation. Daylighting potential,
for example, could be a method to reduce energy use for interior lighting.

6.2 Improvements and Future Work
This study can be improved in many ways. The first concept would be to test various types of
energy conservation methods for each component of the façade renovation. For example, four
different wall types can be tested with assemblies of various R-values. This will generate more
information about the effect of insulation in different climates. A similar approach can be
considered for roofing assemblies. Various types of window assemblies can be tested as well.
Glazing assembly had a large effect on electricity consumption in all climates. In order to
99

optimize glazing assembly selection according to climate type, different glazing assemblies
should be simulated.
Another potential improvement would be to run the same simulations in various types of
software. Each software package has different offerings for enclosure set up and analysis type. A
study on the similarities and differences in estimated façade performance in different software
(or by different codes) could be beneficial.
To improve the workflow of analysis, macros should be integrated into the simulation process.
These macros would be pre-defined and would eliminate the need to manually set each
simulation. This automates a portion of the analysis which increases efficiency, and can
potentially allow for more simulations to take place.
The next step would be to write a program that uses the results of the energy simulations to make
predictions on what type of façade renovation should be implemented in each climate. The
program would either need to be fed prescriptive measures to make decisions or be given a
portion of the data for learning and use the remainder of the data to make predictions.

100

BIBLIOGRAPHIC REFERENCES
Arnold, Chris. 2009. Seismic Safety of the Building Envelope. National Institute of Building
Sciences. November 18. Accessed July 2016.
ASHRAE. 2013. Standard 62.2 - Ventilation and Acceptable Indoor Air Quality in Low-Rise
Residential Buildings. American Society of Heating, Refrigerating, and Air-Conditiong
Engineers.
ASHRAE. 2016. Standard 90.1 Energy Standard for Buildings, Except Low-Rise Residential
Buildings. American Society of Heating, Refrigeration, and Air-Conditioning Engineers.
ASHRAE. 2007. Standard 90.2 Energy-Efficient Design of Low-Rise Residential Buildings.
American Society of Heating, Refrigeration, and Air-Conditioning Engineers.
Aydin, Cevdet Coskun. 2014. "Designing Building Facades for the Urban Rebuilt Environment
with Integration of Digital Close-range Photogrammetry an Geographical Information
Systems." Automation in Construction 43: 38-48.
Geodetic Systems. 2016. What is Photogrammetry. Geodetic Systems, Inc. Accessed September
2016. http://www.geodetic.com/v-stars/what-is-photogrammetry.aspx.
Gocer, Ozgur, Ying Hua, and Kenan Gocer. 2016. "A BIM-GIS Integrated Pre-retrofit Model for
Building Data Mapping (." Building Simulation 513-527.
Gocer, Ozgur, Ying Hua, and Kenan Gocer. 2016. "A BIM-GIS Integrated Pre-retrofit Model for
Building Data Mapping." Building Simulation 513-527.
Gucyeter, Basak, and H. Murat Gunaydin. 2012. "Optimization of an Envelope Retrofit Strategy
for an Existing Office Building." Energy amd Buildings 55: 647-659.
Hakkinen, Tarja. 2012. "Systematic Method for the Sustainability Analysis of Refurbishment
Concepts of Exterior Walls." Construction and Building Materials 37: 783-790.
ICC. 2012. International Energy Conservation Code. International Code Council.
International Code Council. 2015. International Building Code. ICC, Inc.
Joshi, Sumedha M. 2008. "The Sick Building Syndrome." Indian Journal of Occupational &
Environmental Medicine 12 (2): 61-64. Accessed October 2016. doi:10.4103/0019-
5278.43262.
Kesik, Ted J. 2014. Building Envelope Design Principles and Strategies. National Institute of
Building Sciences. October 23. Accessed July 2016.
101

Kesik, Ted J. 2014. Building Envelope Design Guide. National Institute of Building Sciences.
October. Accessed April 2016.
Larsen, Knut Einar, Frank Lattke, Stephan Ott, and Stefan Winter. 2011. "Surveying and Digital
Workflow in Energy Performance Retrofit Projects using Prefabricated Elements."
Automation in Construction 20: 999-1011.
Lucid Design Group, Inc. 2015. "Building Dashboard." University of Southern California
Campus. Los Angeles, California. http://buildingdashboard.net/usc/#/usc.
Marasco, Daniel E., and Constantine E. Kontokosta. 2016. "Applictions of MAchine Learning
Methods to Identifying and Predicting Building Retrofit Opportunities." Energy and
Buildings 128: 431-441.
Martinez Arias, Andrea Soledad. 2013. Facade Retrofit: Enhancing Energy Performance in
Existing Buildings. Masters Thesis, Los Angeles: University of Southern California.
Martinez, Andrea, Michael Patterson, Anders Carlson, and Douglas Noble. 2015. "Fundamentals
in Facade Retrofit." Procedia Engineering (118): 934-941.
McArthur, J.J., and C.G.H. Jofeh. 2016. "Portfolio Retrofit Evaluation: A Methodology for
Optimizing a Large Number of Building Retrofits to Achieve Triple-bottom-line
Objectives." Sustainable Cities and Societies 27: 263-274.
Morse, Roger, and Don Acker. 2014. "Indoor Air Quality and Mold Prevention of the Building
Envelope." Whole Building Deisgn Guide. National Institute of Building Sciences. April
11. Accessed September 2016.
National Geographic. 1996. Geographic Information System. National Geographic Encyclopedia.
Accessed September 2016.
Ochoa, Carlos, and I. Guedi Capeluto. 2015. "Decision Methodology for the Development of an
Expert System Applid in an Adaptive Energy Retrofit Facade System for Residential
Buildings." Renewable Energy 78: 498-508.
Pareja, T Fernadez, A Garcia Pablos, and J de Vicente. 2013. "Terrestrial Laser Scanner
Equipment Calibration." The manufacturing Engineering Society International
Conference. Procedia Engineering. 278-286. Accessed September 2016.
doi:10.1016/j.proeng.2013.08.191.
Patterson, Mic, Jeffrey Vaglio, and Douglas Noble. 2014. "Incremental Façade Retrofits:
Curtainwall Technology as a Strategy to Step Existing Buildings Towards Net Zero
Energy." Energy Procedia (Elsevier Ltd.) 3150-3159.
102

Pomponi, Francesco, Eric R.P. Farr, Poorang Piroozfar, and Jonathan R. Gates. 2015. "Facade
Refurbishment of Existing Office Buildings: Do Conventional Energy-Saving
Interventions Always Work?" Journal of Building Engineering 3: 135-143.
Prowler, Don. 2010. "Mold and Moisture Dynamics." Whole Building Design Guide. National
Institute of Building Sciences. June 12. Accessed September 2016.
U.S. Department of Energy. 2016. Thermographic Inspections. Office of Energy Efficiency &
Renewable Energy. Accessed August 2016.
http://www.energy.gov/energysaver/thermographic-inspections.
U.S. Department of Energy, Eric Makela, Jennifer Williamson, and Erin Makela. 2011.
"Comparison of Standard 90.1-2012 and the 2012 IECC with Respect to Commercial
Buildings." Energy Efficiency and Renewable Energy, September.
U.S. Environmental Protection Agency. 2012. Commercial Buildings Energy Consumption
Survey. U.S. Department of Energy.
—. 2016. Indoor Air Quiality. July 21. Accessed September 2016.
Wang, Fulton, and Cynthia Rudin. 2015. "Falling Rule Lists." Journal of Machine Learning.
Zimring, Craig, Mahbub Rashid, and Kevin Kampschroer. 2014. "Facility Performance
Evaluation." Whole Building Design Guide. National Institute of Building Sciences.
October 27. Accessed September 2016.

103

APPENDIX A: CBECS Data Descriptions
Variable
name
Variable
type
Format Label Values/Format codes
PUBID Char Building
identifier
00001 - 06720
REGION Char $REGION. Census
region
'1' = 'Northeast'
'2' = 'Midwest'
'3' = 'South'
'4' = 'West'
CENDIV Char $CENDIV. Census
division
1' = 'New England'
'2' = 'Middle Atlantic'
'3' = 'East North Central'
'4' = 'West North Central'
'5' = 'South Atlantic'
'6' = 'East South Central'
'7' = 'West South Central'
'8' = 'Mountain'
'9' = 'Pacific'
SQFT Num COMMA10. Square
footage
1,001 - 1,500,000
SQFTC Char $SQFTC. Square
footage
category
'01' = '1,000 square feet or less'
'02' = '1,001 to 5,000 square feet'
'03' = '5,001 to 10,000 square feet'
'04' = '10,001 to 25,000 square feet'
'05' = '25,001 to 50,000 square feet'
'06' = '50,001 to 100,000 square feet'
'07' = '100,001 to 200,000 square feet'
'08' = '200,001 to 500,000 square feet'
'09' = '500,001 to 1 million square feet'
'10' = 'Over 1 million square feet'
WLCNS Char $WLCNS. Wall
construction
material
1' = 'Brick, stone, or stucco'
'2' = 'Pre-cast concrete panels'
'3' = 'Concrete block or poured concrete
(above grade)'
'4' = 'Aluminum, asbestos, plastic, or
wood materials (siding, shingles, tiles, or
shakes)'
'5' = 'Sheet metal panels'
'6' = 'Window or vision glass (glass that
can be seen through)'
'7' = 'Decorative or construction glass'
'8' = 'No one major type'
'9' = 'Other'
104

RFCNS Char $RFCNS. Roof
construction
material
1' = 'Built-up (tar, felts, or fiberglass and a
ballast, such as stone)'
'2' = 'Slate or tile shingles'
'3' = 'Wood shingles, shakes, or other
wooden materials'
'4' = 'Asphalt, fiberglass, or other
shingles'
'5' = 'Metal surfacing'
'6' = 'Plastic, rubber, or synthetic sheeting
(single or multiple ply)'
'7' = 'Concrete'
'8' = 'No one major type'
'9' = 'Other'
RFTILT Char $PITCH. Roof tilt 1' = 'Flat'
'2' = 'Shallow pitch'
'3' = 'Steeper pitch'
BLDSHP Char $SHAPE. Building
shape
01' = 'Square'
'02' = 'Wide rectangle'
'03' = 'Narrow rectangle'
'04' = 'Rectangle or square with an
interior courtyard'
'05' = '"H" shaped'
'06' = '"U" shaped'
'07' = '"E" shaped'
'08' = '"T" shaped'
'09' = '"L" shaped'
'10' = '"+" or cross shaped'
'11' = 'Other shape'
Missing = Not applicable
GLSSPC Char $GLSPCT. Percent
exterior glass
1' = '1 percent or less'
'2' = '2 to 10 percent'
'3' = '11 to 25 percent'
'4' = '26 to 50 percent'
'5' = '51 to 75 percent'
'6' = '76 to 100 percent'
Missing = Not applicable
NFLOOR Num Number of
floors
1 - 14
994 = 15 to 25
995 = More than 25
FLCEILHT Num

Floor to
ceiling height
6 - 50
995 = More than 50
YRCON Num Year of
construction
995 = Before 1946
1946 - 2012
YRCONC Char $YRCONC. Year of
construction
category
'01' = 'Before 1920'
'02' = '1920 to 1945'
'03' = '1946 to 1959'
'04' = '1960 to 1969'
'05' = '1970 to 1979'
'06' = '1980 to 1989'
'07' = '1990 to 1999'
'08' = '2000 to 2003'
'09' = '2004 to 2007'
'10' = '2008 to 2012'
105

APPENDIX B: MATLAB Script
% Envelope
% Plot Data (statistics)
% File origin: 17-09-09
% Last updated: 17-09-16
% kicollin@usc.edu

% The goal of this program is to plot the CBECS data in histograms and
% scatter plots. Histograms will be observed to hypothesize if data fits a
% known distribution. Scatter plots will be observed to hypothesize
% potential correlation between varuables.

% Characteristics Histograms
histplot_region = histogram(REGION_r);
histplot_cendiv = histogram(CENDIV_r);
histplot_PBA = histogram(PBA_r);
histplot_sqft = histogram(SQFT_r);
histplot_sqftc = histogram(SQFTC_r);
histplot_wlcns = histogram(WLCNS_r);
histplot_rfcns = histogram(RFCNS_r);
histplot_rfcool = histogram(RFCOOL_r); %% Y/N response
histplot_rftilt = histogram(RFTILT_r);
histplot_bldshp = histogram(BLDSHP_r);
histplot_glasspc = histogram(GLSSPC_r);
histplot_eqglss = histogram(EQGLSS_r);
histplot_sungls = histogram(SUNGLS_r);
histplot_nfloor = histogram(NFLOOR_r);
histplot_flceilht = histogram(FLCEILHT_r);
histplot_yrcon = histogram(YRCON_r); %need to remove values <1900
histplot_yrconc = histogram(YRCONC_r);
histplot_act1 = histogram(ACT1_r);

% Renovation Histograms
histplot_renov = histogram(RENOV_r);
histplot_renadd = histogram(RENADD_r);
histplot_renrdc = histogram(RENRDC_r);
histplot_renwll = histogram(RENWLL_r);
histplot_renins = histogram(RENINS_r);
histplot_renwin = histogram(RENWIN_r);
histplot_renrff = histogram(RENRFF_r);
histplot_renhvc = histogram(RENHVC_r);

% Consumption Histograms
histplot_mfbtu = histogram(MFBTU_r);
histplot_mfexp = histogram(MFEXP_r);
histplot_elcns = histogram(ELCNS_r);
histplot_elbtu = histogram(ELBTU_r);
histplot_elexp = histogram(ELEXP_r);
histplot_EUIBTU = histogram(EUIBTU_r); %may need to remove 0 values
histplot_ExI_r = histogram(ExI_r); %need to remove 0 values
% specific to HVAC
histplot_mfhtbtu = histogram(MFHTBTU_r);
histplot_mfclbtu = histogram(MFCLBTU_r);
histplot_mfvnbtu = histogram(MFVNBTU_r);
histplot_elhtbtu = histogram(ELHTBTU_r);
histplot_elclbtu = histogram(ELCLBTU_r);
histplot_elvnbtu = histogram(ELVNBTU_r);
histplot_elhvac = histogram(ELHVAC);
histplot_mfhvac = histogram(MFHVAC);

% Consumption Scatter Plots
scat_MFCvSQFT = scatter(SQFT_r, MFBTU_r); %CANDIDATE for curvefitting, exponential
scat_MFCvWLCNS = scatter(WLCNS_r, MFBTU_r);
scat_MFCvRFCNS = scatter(RFCNS_r, MFBTU_r);
scat_MFCvGLSSPC = scatter(GLSSPC_r, MFBTU_r); %CANDIDATE for curvefitting, gaussian
scat_ELBTUvSQFT = scatter(SQFT_r, ELBTU_r); %CANDIDATE
scat_ELBTUvWLCNS = scatter(WLCNS_r, ELBTU_r);
scat_ELBTUvRFCNS = scatter(RFCNS_r, ELBTU_r);
scat_ELBTUvGLSSPC = scatter(GLSSPC_r, ELBTU_r); %CANDIDATE
106

scat_MFEXPvSQFT = scatter(SQFT_r, MFEXP_r); %CANDIDATE
scat_MFEXPvWLCNS = scatter(WLCNS_r, MFEXP_r);
scat_MFEXPvRFCNS = scatter(RFCNS_r, MFEXP_r);
scat_MFEXPvGLSSPC = scatter(GLSSPC_r, MFEXP_r); %CANDIDATE
scat_ELEXPvSQFT = scatter(SQFT_r, ELEXP_r); %CANDIDATE
scat_ELEXPvWLCNS = scatter(WLCNS_r, ELEXP_r);
scat_ELEXPvRFCNS = scatter(RFCNS_r, ELEXP_r);
scat_ELEXPvGLSSPC = scatter(GLSSPC_r, ELEXP_r); %CANDIDATE
scat_EUIBTUvSQFT = scatter(SQFT_r, EUIBTU_r); %CANDIDATE wth is going on???
scat_EUIBTUvWLCNS = scatter(WLCNS_r, EUIBTU_r);
scat_EUIBTUvRFCNS = scatter(RFCNS_r, EUIBTU_r);
scat_EUIBTUvGLSSPC = scatter(GLSSPC_r, EUIBTU_r); %CANDIDATE
%these are better
scat_MFHVACvSQFT = scatter(SQFT_r, MFHVAC);
scat_MFHVACvWLCNS = scatter(WLCNS_r, MFHVAC);
scat_MFHVACvRFCNS = scatter(RFCNS_r, MFHVAC);
scat_MFHVACvGLSSPC = scatter(GLSSPC_r, MFHVAC);
scat_ELHVACvSQFT = scatter(SQFT_r, ELHVAC);
scat_ELHVACvWLCNS = scatter(WLCNS_r, ELHVAC);
scat_ELHVACvRFCNS = scatter(RFCNS_r, ELHVAC);
scat_ELHVACvGLSSPC = scatter(GLSSPC_r, ELHVAC);

% Envelope
% Correlations Coefficients (statistics)
% File origin: 18-01-17
% Last updated: 18-01-19
% kicollin@usc.edu

% This script defines the correlation coefficients between selected
% variables

%combine consumption vaues for heating, cooling, and ventilation:
ELHVAC = ELHTBTU_r + ELCLBTU_r + ELVNBTU_r;
MFHVAC = MFHTBTU_r + MFCLBTU_r + MFVNBTU_r;

% find the correlation coefficient for the following:
CCR_SQFTvEUIBTU = corrcoef(SQFT_r, EUIBTU_r);
CCR_SQFTvELCNS = corrcoef(SQFT_r, ELCNS_r);
CCR_SQFTvMFBTU = corrcoef(SQFT_r, MFBTU_r);
CCR_SQFTvELHVAC = corrcoef(SQFT_r, ELHVAC);
CCR_SQFTvMFHVAC = corrcoef(SQFT_r, MFHVAC);

% Since these are not linear values, we cannot evaulate linear
% correlation between these values and the corresponding numbers.
% CCR_WLCNSvEUIBTU = corrcoef(WLCNS_r, EUIBTU_r);
% CCR_WLCNSvELCNS = corrcoef(WLCNS_r, ELCNS_r);
% CCR_WLCNSvMFBTU = corrcoef(WLCNS_r, MFBTU_r);
% CCR_WLCNSvELHVAC = corrcoef(WLCNS_r, ELHVAC);
% CCR_WLCNSvMFHVAC = corrcoef(WLCNS_r, MFHVAC);
%
% CCR_RFCNSvEUIBTU = corrcoef(RFCNS_r, EUIBTU_r);
% CCR_RFCNSvELCNS = corrcoef(RFCNS_r, ELCNS_r);
% CCR_RFCNSvMFBTU = corrcoef(RFCNS_r, MFBTU_r);
% CCR_RFCNSvELHVAC = corrcoef(RFCNS_r, ELHVAC);
% CCR_RFCNSvMFHVAC = corrcoef(RFCNS_r, MFHVAC);

CCR_GLSSPCvEUIBTU = corrcoef(GLSSPC_r, EUIBTU_r);
CCR_GLSSPCvELCNS = corrcoef(GLSSPC_r, ELCNS_r);
CCR_GLSSPCvMFBTU = corrcoef(GLSSPC_r, MFBTU_r);
CCR_GLSSPCvELHVAC = corrcoef(GLSSPC_r, ELHVAC);
CCR_GLSSPCvMFHVAC = corrcoef(GLSSPC_r, MFHVAC);

% combine into a row vector
CCR_SQFT = [CCR_SQFTvEUIBTU(1,2) CCR_SQFTvELCNS(1,2) CCR_SQFTvMFBTU(1,2)...
CCR_SQFTvELHVAC(1,2) CCR_SQFTvMFHVAC(1,2)];
% CCR_WLCNS = [CCR_WLCNSvEUIBTU(1,2) CCR_WLCNSvELCNS(1,2) CCR_WLCNSvMFBTU(1,2)...
% CCR_WLCNSvELHVAC(1,2) CCR_WLCNSvMFHVAC(1,2)];
% CCR_RFCNS = [CCR_RFCNSvEUIBTU(1,2) CCR_RFCNSvELCNS(1,2) CCR_RFCNSvMFBTU(1,2)...
% CCR_RFCNSvELHVAC(1,2) CCR_RFCNSvMFHVAC(1,2)];
CCR_GLSSPC = [CCR_GLSSPCvEUIBTU(1,2) CCR_GLSSPCvELCNS(1,2) CCR_GLSSPCvMFBTU(1,2)...
107

CCR_GLSSPCvELHVAC(1,2) CCR_GLSSPCvMFHVAC(1,2)];

CCRvals = [CCR_SQFT; CCR_GLSSPC]

% Envelope
% Hypothesis Testing (statistics)
% File origin: 18-01-19
% Last updated: 18-01-19
% kicollin@usc.edu

% This script tests the hypothesis that data fits a Normal or
% Exponential distribution

%combine consumption vaues for heating, cooling, and ventilation:
ELHVAC = ELHTBTU_r + ELCLBTU_r + ELVNBTU_r;
MFHVAC = MFHTBTU_r + MFCLBTU_r + MFVNBTU_r;

% to test for normal distribution
% h = chi2gof(x)
% 0 = Normal (does not reject the hypothesis)
% 1 = Non Normal (rejects the hypothesis)

[G_mfbtu, pG_mfbtu] = chi2gof(MFBTU_r);
[G_mfexp, pG_mfexp] = chi2gof(MFEXP_r);
[G_elcns, pG_elcns] = chi2gof(ELCNS_r);
[G_elbtu, pG_elbtu] = chi2gof(ELBTU_r);
[G_elexp, pG_elexp] = chi2gof(ELEXP_r);
[G_euibtu, pG_euibtu] = chi2gof(EUIBTU_r); %may need to remove 0 values
[G_ExI_r, pG_ExI_r] = chi2gof(ExI_r); %need to remove 0 values
% specific to HVAC
[G_elhvac, pG_elhvac] = chi2gof(ELHVAC); %use the compiled data for simplification
% G_elhtbtu = chi2gof(ELHTBTU_r);
% G_elclbtu = chi2gof(ELCLBTU_r);
% G_elvnbtu = chi2gof(ELVNBTU_r);
[G_mfhvac, pG_mfhvac] = chi2gof(MFHVAC); %use the compiled data for simplification
% G_mfhtbtu = chi2gof(MFHTBTU_r);
% G_mfclbtu = chi2gof(MFCLBTU_r);
% G_mfvnbtu = chi2gof(MFVNBTU_r);

Gtests = [G_mfbtu; G_mfexp; G_elcns; G_elbtu; G_elexp; G_euibtu; G_ExI_r;...
G_elhvac; G_mfhvac];
pvals_Gtests = [pG_mfbtu; pG_mfexp; pG_elcns; pG_elbtu; pG_elexp;...
pG_euibtu; pG_ExI_r; pG_elhvac; pG_mfhvac];

% to test for exponential distribution
% h = lillietest(x,'Distr','exp')
% 0 = Normal (does not reject the hypothesis)
% 1 = Non Normal (rejects the hypothesis)

[E_mfbtu, pE_mfbtu] = lillietest(MFBTU_r);
[E_mfexp, pE_mfexp] = lillietest(MFEXP_r);
[E_elcns, pE_elcns] = lillietest(ELCNS_r);
[E_elbtu, pE_elbtu] = lillietest(ELBTU_r);
[E_elexp, pE_elexp] = lillietest(ELEXP_r);
[E_euibtu, pE_euibtu] = lillietest(EUIBTU_r); %may need to remove 0 values
[E_ExI_r, pE_ExI_r] = lillietest(ExI_r); %need to remove 0 values
[E_elhvac, pE_elhvac] = lillietest(ELHVAC);
[E_mfhvac, pE_mfhvac] = lillietest(MFHVAC);

Etests = [E_mfbtu; E_mfexp; E_elcns; E_elbtu; E_elexp; E_euibtu; E_ExI_r;...
E_elhvac; E_mfhvac];
pvals_Etests = [pE_mfbtu; pE_mfexp; pE_elcns; pE_elbtu; pE_elexp;...
pE_euibtu; pE_ExI_r; pE_elhvac; pE_mfhvac];

% HYPtests = [Gtests Etests];

HYPtests = [Gtests pvals_Gtests Etests pvals_Etests]

108

APPENDIX C: Results
Chicago, IL

Louisville, KY

Los Angeles, CA

Trial Data Location Data Energy Use Intensity Life Cycle Energy Use & Cost
F F $/kWh $/Therm kWh/sf/yr kBTU/sf/yr kBTU/sf/yr kWh kBTU $
Trial Type Temp min Temp max Electrical Cost Fuel Cost Electricity EUI Fuel EUI Total EUI Electricity Use Fuel Use Energy Cost
T1 baseline -9 96 $0.08 $0.79 12 54 96 1,651,420 7,369,400 $87,459
T2 window -9 96 $0.08 $0.79 12 54 95 1,631,882 7,366,200 $86,726
T3 wall -9 96 $0.08 $0.79 12 54 95 1,619,344 7,289,100 $85,986
T4 roof -9 96 $0.08 $0.79 12 54 96 1,651,420 7,369,400 $87,459
T5 renov -9 96 $0.08 $0.79 12 54 95 1,619,344 7,289,100 $85,986
Annual Carbon Emissions Annual Energy Use
tons/yr tons/yr kWh kBTU kBTU kWh
Trial Electricity-CO2e Fuel-CO2e Electricity Use2 %Elec Electricity Cost Fuel Use2 %Fuel Fuel Costs HVAC Fuel %Fuel-HVAC HVAC Fuel Cost HVAC Elec %Elec-HVACHVAC Elec Cost
T1 42 14 55,047 43% $4,475 245,600 57% $1,946 230,600 93% $1,827 17,016 31% $1,383
T2 42 14 54,396 43% $4,422 245,500 57% $1,945 230,500 93% $1,826 16,365 30% $1,330
T3 41 14 53,978 43% $4,388 242,900 57% $1,925 228,000 93% $1,806 16,204 30% $1,806
T4 42 14 55,047 43% $4,475 245,600 57% $1,946 230,600 93% $2,306 17,016 31% $1,383
T5 41 14 53,978 43% $4,388 242,900 57% $1,925 228,000 93% $1,806 16,204 30% $1,317
Trial Data Location Data Energy Use Intensity Life Cycle Energy Use & Cost
F F $/kWh $/Therm kWh/sf/yr kBTU/sf/yr kBTU/sf/yr kWh kBTU $
Trial Type Temp min Temp max Electrical Cost Fuel Cost Electricity EUI Fuel EUI Total EUI Electricity Use Fuel Use Energy Cost
T1 baseline -1 91 $0.08 $0.85 13 33 76 1,717,283 4,415,300 $82,193
T2 window -1 91 $0.08 $0.85 12 32 75 1,691,074 4,406,300 $81,163
T3 wall -1 91 $0.08 $0.85 12 32 75 1,679,530 4,350,700 $80,511
T4 roof -1 91 $0.08 $0.85 13 33 76 1,717,283 4,415,300 $82,193
T5 renov -1 91 $0.08 $0.85 12 32 75 1,679,530 4,350,700 $80,511
Trial Data Annual Carbon Emissions Annual Energy Use
tons/yr tons/yr kWh kBTU kBTU kWh
Trial Type Electricity Fuel Electricity Use2 %Elec Electricity Cost Fuel Use2 %Fuel Fuel Costs HVAC Fuel %Fuel-HVAC HVAC Fuel Cost HVAC Elec %Elec-HVAC HVAC Elec Cost
T1 baseline 44 8 57,242 57% $4,785 147,100 43% $1,249 133,700 90% $1,135 19,212 34% $1,606
T2 window 43 8 56,369 57% $4,712 146,800 43% $1,247 133,400 90% $1,133 18,338 33% $1,533
T3 wall 43 8 55,984 57% $4,680 145,000 43% $1,231 131,600 90% $1,117 18,210 33% $1,522
T4 roof 44 8 57,242 57% $4,785 147,100 43% $1,249 133,700 90% $1,135 19,212 34% $1,606
T5 renov 43 8 55,984 57% $4,680 145,000 43% $1,231 131,600 90% $1,117 18,210 33% $1,522
Trial Data Location Data Energy Use Intensity Life Cycle Energy Use & Cost
F F $/kWh $/Therm kWh/sf/yr kBTU/sf/yr kBTU/sf/yr kWh kBTU $
Trial Type Temp min Temp max Electrical Cost Fuel Cost Electricity EUI Fuel EUI Total EUI Electricity Use Fuel Use Energy Cost
T1 baseline 40 95 $0.12 $0.80 13 10 54 1,742,862 1,391,500 $98,362
T2 window 40 95 $0.12 $0.80 13 10 53 1,712,470 1,286,800 $96,718
T3 wall 40 95 $0.12 $0.80 13 10 53 1,700,448 1,355,500 $95,960
T4 roof 40 95 $0.12 $0.80 13 10 54 1,742,862 1,391,500 $98,362
T5 renov 40 95 $0.12 $0.80 13 10 53 1,700,448 1,355,500 $95,960
Trial Data Annual Carbon Emissions Annual Energy Use
tons/yr tons/yr kWh kBTU kBTU kWh
Trial Type Electricity-CO2e Fuel-CO2e Electricity Use2 %Elec Electricity Cost Fuel Use2 %Fuel Fuel Costs HVAC Fuel %Fuel-HVAC HVAC Fuel Cost HVAC Elec %Elec-HVAC HVAC Elec Cost
T1 baseline 19 2 58,095 81% $6,849 46,300 19% $372 34,100 73% $274 20,064 35% $2,365
T2 window 19 2 57,082 81% $6,730 46,200 19% $371 33,900 73% $272 19,051 33% $2,246
T3 wall 19 2 56,681 81% $6,683 45,100 19% $363 32,900 72% $264 18,908 33% $2,229
T4 roof 19 2 58,095 81% $6,849 46,300 19% $372 34,100 73% $274 20,064 35% $2,365
T5 renov 19 2 56,681 81% $6,683 45,100 19% $363 32,900 72% $264 18,908 33% $2,229
109

New Orleans, LA

Trial Data Location Data Energy Use Intensity Life Cycle Energy Use & Cost
F F $/kWh $/Therm kWh/sf/yr kBTU/sf/yr kBTU/sf/yr kWh kBTU $
Trial Type Temp min Temp max WWR Electrical Cost Fuel Cost Electricity EUI Fuel EUI Total EUI Electricity Use Fuel Use Energy Cost
T1 baseline 30 96 $0.08 $0.91 15 5 55 2,015,831 653,400 $75,729
T2 window 30 96 $0.08 $0.91 15 5 55 1,989,815 641,700 $74,738
T3 wall 30 96 $0.08 $0.91 15 5 55 1,976,349 628,500 $74,196
T4 roof 30 96 $0.08 $0.91 15 5 55 2,015,831 653,400 $75,729
T5 renov 30 96 $0.08 $0.91 15 5 55 1,976,349 628,500 $74,196
Trial Data Annual Carbon Emissions Annual Energy Use
tons/yr tons/yr kWh kBTU kBTU kWh
Trial Type Electricity-CO2e Fuel-CO2e Electricity Use2 %Elec Electricity Cost Fuel Use2 %Fuel Fuel Costs HVAC Fuel %Fuel-HVAC HVAC Fuel Cost HVAC Elec %Elec-HVAC HVAC Elec Cost
T1 baseline 33 1 67,194 91% $5,362 21,700 9% $198 11,200 51% $102 29,163 43% $2,327
T2 window 33 1 66,327 91% $5,293 21,300 9% $194 10,800 50% $98 28,296 43% $2,258
T3 wall 33 1 65,878 91% $5,257 20,900 9% $190 10,400 49% $95 28,104 43% $2,242
T4 roof 33 1 67,194 91% $5,362 21,700 9% $198 11,200 51% $102 29,163 43% $2,327
T5 renov 33 1 65,878 91% $5,257 20900 9% $190 10,400 49 $95 28104 43% $2,242

Abstract (if available)

Abstract Thermal properties of the building enclosure affect building energy consumption due to the relationship to external heating and cooling loads. Once a building is constructed, these properties are assumed to be fixed. If a renovation of the building enclosure takes place, it is possible to improve thermal properties of the building enclosure and consequently reduce energy consumption. ❧ Many organizations within the building design and construction industry have set up goals to reduce energy consumption and carbon emissions. Most current efforts focus on implementing technology in the design of new buildings

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Mitigating thermal bridging in ventilated rainscreen envelope construction: Methods to reduce thermal transfer in net-zero envelope optimization

PDF

A proposal for building envelope retrofit on the Bonaventure Hotel: a case study examining energy and carbon

PDF

Double skin façade in hot arid climates: computer simulations to find optimized energy and thermal performance of double skin façades

PDF

Facade retrofit: enhancing energy performance in existing buildings

PDF

Using architecture in the fight against global warming: presenting viable energy-saving renovation design strategies to homeowners via an interactive web learning tool

PDF

Energy use intensity estimation method based on building façade features by using regression models

PDF

Real-time simulation-based feedback on carbon impacts for user-engaged temperature management

PDF

Kinetic facades as environmental control systems: using kinetic facades to increase energy efficiency and building performance in office buildings

PDF

Building retrofitting evaluation: Energy savings and cost effectiveness of building retrofits on Graduate Art Studios at the University of California, Los Angeles

PDF

The effectiveness of enviro-materially actuated kinetic facades: evaluating the thermal performance of thermo-bimetal shading component geometries

PDF

Skin fit and retrofit: challenging the sustainability of curtainwall practice in tall buildings

PDF

District energy systems: Studying building types at an urban scale to understand building energy consumption and waste energy generation

PDF

Zero net energy institutional building

PDF

Building energy performance estimation approach: facade visual information-driven benchmark performance model

PDF

Design of double skin (envelope) as a solar chimney: adapting natural ventilation in double envelope for mild or warm climates

PDF

ctrl+z: exploring the effects of undoing retrofits to pre-war buildings in Los Angeles

PDF

Energy performance of different building forms: HEED simulations of equivalent massing models in diverse building surface aspect ratios and locations in the US

PDF

Energy saving integrated facade: design and analysis using computer simulation

PDF

Night flushing and thermal mass: maximizing natural ventilation for energy conservation through architectural features

PDF

Intelligent adaptive automation: activity-driven and user-centered building automation

Asset Metadata

Creator Collins, Kenya Inay (author)

Core Title Façade retrofit performance evaluation: Predicting energy conservation potential from renovations to the building envelope

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Publication Date 07/16/2018

Defense Date 07/15/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag building envelope,energy conservation measures,energy efficiency,façade engineering,façade renovation,OAI-PMH Harvest,retrofit

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Noble, Douglas (committee chair), Carlson, Anders (committee member), Soibelman, Lucio (committee member)

Creator Email kenyainaycollins@gmail.com,kicollin@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-16420

Unique identifier UC11669028

Identifier etd-CollinsKen-6402.pdf (filename),usctheses-c89-16420 (legacy record id)

Legacy Identifier etd-CollinsKen-6402.pdf

Dmrecord 16420

Document Type Thesis

Format application/pdf (imt)

Rights Collins, Kenya Inay

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA