See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/303531438

Innovative Flexible Structural System Using Prefabricated Modules

Article in Journal of Architectural Engineering · May 2016

DOI: 10.1061/(ASCE)AE.1943-5568.0000214

CITATIONS READS

67 5,921

4 authors:

Tharaka Gunawardena Tuan Duc Ngo

University of Melbourne University of Melbourne
39 PUBLICATIONS 607 CITATIONS 409 PUBLICATIONS 19,006 CITATIONS

SEE PROFILE SEE PROFILE

Priyan Mendis Jose Alfano

University of Melbourne Columbia University
468 PUBLICATIONS 10,950 CITATIONS 4 PUBLICATIONS 93 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Ultra-High Strength Reinforced Concrete (VHSC) (100 - 150 MPa) in Construction View project

Prefabricated Stressed-Skin Engineered Timber Flooring Systems View project

All content following this page was uploaded by Tharaka Gunawardena on 17 January 2019.

The user has requested enhancement of the downloaded file.

Case Study

Innovative Flexible Structural System Using

Prefabricated Modules
1 2 3 4
Tharaka Gunawardena ; Tuan Ngo ; Priyan Mendis ; and Jose Alfano

Abstract: Prefabricated modular structures are increasingly becoming popular as a strategy that can be used to achieve cost-effective and
speedy construction. However, there is an absence of detailed engineering research and case studies dealing with the structural performance of
this technology. Although such prefabricated modules are already used for some buildings, they cannot be categorized as exclusively modular
structures because a core structure is used in most of them to carry the lateral loads. This paper gives a brief overview of prefabricated modular
structures and an introductory preview into an innovative structural system that can be adopted to deliver a purely modular construction. The
introduced structural system was assessed through a case study of a multistory apartment building that was constructed using prefabricated
modules. The static and dynamic behavior of the structure was analyzed using finite-element analysis techniques with the aid of a three-
dimensional (3D) computer model. Torsional or twisting effects are a major problem for the designers of buildings. Results show that with the
introduced system, the elevator shafts can be flexibly shifted around the plan without causing adverse torsional effects to the structure. The
introduced technology is discussed in detail along with the key points that make it a viable option to be used in medium- and high-rise buildings.
DOI: 10.1061/(ASCE)AE.1943-5568.0000214. © 2016 American Society of Civil Engineers.

Author keywords: Prefabrication; Modular structures; Structural systems; Speedy construction.

Introduction to Prefabricated Modules
Background
Because of their fast delivery and convenience on site, prefabricated
modular structures have great potential for changing conventional
construction methods at a rapid rate. Prefabricated building modules
(e.g., apartments, office spaces, staircases) can be fully constructed
with architectural finishes and services inside a quality-controlled
factory environment; these modules are then ready to be delivered
and assembled on site to form a load-bearing structure. Most
manufacturers will accordingly accommodate architectural designs
with innovative modular units.
Modular technology has already been used for low-rise
structures around the world. A great example among many is the
Little Hero low-rise apartment building in Melbourne, Australia
(Figs. 1 and 2), which consists of 58 single-story apartment modules
and 5 double-story apartment modules. The authors were part of the
development team for this project. The eight modular stories were
assembled with finishes within 8 days, and the building was
constructed in a site with a very narrow access road, demonstrating
the many advantages of modular construction.
1
Ph.D. Student, Dept. of Infrastructure Engineering, Univ. of Melbourne,
Parkville, VIC 3010, Australia (corresponding author). E- mail:
tharakag@student.unimelb.edu.au
2
Associate Professor, Dept. of Infrastructure Engineering, Univ. of
Melbourne, Parkville, VIC 3010, Australia.
3Professor, Dept. of Infrastructure Engineering, Univ. of Melbourne,
Parkville, VIC 3010, Australia.
4Director, Alfano Architects Pty Ltd., PO Box 127, VIC 3927, Australia.
Note. This manuscript was submitted on September 15, 2014; approved
on January 28, 2016; published online on May 25, 2016. Discussion period
open until October 25, 2016; separate discussions must be submitted for
individual papers. This paper is part of the Journal of Architectural
Engineering, © ASCE, ISSN 1076-0431.
In most cases, including that of the Little Hero building, a cast-
in- situ concrete core, a steel core, or a number of cores will act as
the primary lateral-load-resisting system. Prefabricated modules are
tied to one another and to the core(s) by means of steel connections
through which all lateral loads will be transferred to the core(s).
Therefore, these systems are not truly completely prefabricated
modular systems.
This paper introduces a new system in which structures can be
built without the need for an in-situ core to resist lateral loads. This
in turn will ensure that the structure will have full access to the
benefits that modular construction has to offer.
The performance of the system was studied in terms of general
structural stability and behavior against wind and dynamic earth-
quake forces to form an initial idea about the structural feasibility of
the new system.
Features of Prefabricated Modular Structures
Because modern architecture develops innovative designs, buildings
cannot rely on fixed modules. With modular structures, a building
de- signer is free to lay out a building in the conventional manner,
suiting a client’s desire and the requirements of the market. The
building is then adjusted and divided into units that are in width and
length suitable for transportation and lifting into position by a crane
on site.
The prefabrication of buildings has proven to reduce
construction waste by up to 52% (Jaillon et al. 2009), mainly
through means of minimized offcuts (Osmani et al. 2006). This in
turn results in significantly improved efficiency in the energy, cost,
and time required for construction (Aye et al. 2012).
The features of such building units (modules) are as follows:
 The modules can incorporate all components of a building,
including stairs, elevator shafts, façades, corridors, and services.
 The modules are constructed in a quality-controlled production
facility. A unit’s length, width, and height can vary from project
to project.
 There is minimal work on site to complete the buildings because
the façade and interiors themselves form part of modules.

© ASCE 05016003-1 J. Archit. Eng.

J. Archit. Eng., 05016003
Fig 1: Multi storey modular building 'Little Hero' in Melbourne,
Australia (reprinted from Gunawardena et al. 2014a)
Fig 2: 'Little Hero' during construction (reprinted from Gunawardena et
al. 2014b, with permission from Open House International)
 The modules can easily be removed from the main structure for
future reuse or relocation.
 Modular construction at present reduces construction time by
over 50% for a site-intensive building (Lawson et. al. 2012).
 Reduced construction time means that the building starts
generating revenue for the client much sooner than it does after a
conventional construction.
New Structural System
Introduction
Most of the new modular constructions are built as a system of
modules that are connected laterally to a cast-in-situ or prefabricated
steel core (Fig. 3) that eventually acts as the primary lateral-load-
resisting element. Further, in many of these, the floor levels are cast
with concrete subsequent to placing the modules. Although these are
innovative technologies themselves, they prevent the structure from
© ASCE 05016003-2
J. Archit. Eng., 05016003
Fig 3: 17 storey modular building stabilized by concrete cores (reprinted
from Lawson et al. 2012, with permission)
being considered a purely modular construction. As a result, these
structures do not enjoy the full benefits of modular construction. This
study aimed to identify ways in which an assembly of pre- fabricated
modules alone could be used to engineer a structurally feasible
construction that would reward the client with full benefits of modular
construction.
New System
A new structural system in which a cast-in-situ or prefabricated core
is not the predominant lateral-load-resisting component is introduced
in this paper. The elevator core in this system is intended to be formed
with steel elements as a part of some of the prefabricated modules
themselves and therefore will not be the central component in the
lateral-load-resisting system. The prefabricated modules are stacked
vertically and connected horizontally through bolted plates.
The lateral congruity for lateral load transfer is provided by these
connections and improved greatly through the introduction of
modules with stiff concrete walls. Because these stiff modules, which
are strategically placed in the main structure, resist the majority of
lateral loads and transfer them down to the foundation, the need for a
stiff central core becomes less critical. The structure can now act as a
purely modular system.
This technology has the potential to reduce construction time
significantly by completely eliminating the time and costs incurred in
building the traditional core of a low-rise building. The elevator shafts
and staircases can be accommodated in the prefabricated modules
themselves. It also gives the opportunity for the stiffer walls to be
constructed using innovative materials, such as composites and high-
J. Archit. Eng.
 Shear centre

Centre of mass

(a) (b)
Fig 4: Placement of the core in conventional buildings; (a) concentric
core where torsional forces are minimized, (b) eccentric core where
higher torsions apply on the structure

Fig 5: Four different conceptual placements of the elevator shaft (on

plan view) made possible with the new system

Fig 7: 3D finite element model for the 20 storey structure (Note: only the
load bearing elements are modelled)

The elevator core can even be staggered vertically. Thus, architects

Fig 6: Two different conceptual arrangements of the lift core made
possible with the new system, on elevation view
strength concrete and steel. Concrete walls can be built by filling after
the modules are placed.
More importantly, the new system provides architects with a lot of
freedom with a structure that is not limited by the placement of a core
that takes critical lateral loads. In conventional structures, the core
would ideally be situated in the center so that the shear center of the
structure coincides with the center of mass, reducing torsional forces.
Fig. 4 illustrates this phenomenon.
Because the stiffer modules take the lateral loads, the elevator core
can be moved around in the building as the architect pleases while
considering the vertical transportation requirement. Fig. 5 shows
different plan arrangements that are possible with the new system.
will have further freedom to change the placement of the elevator core
at different levels of the building (Fig. 6), which is not normally
possible in conventional construction because the elevator core is
required to transfer the loads to the foundation.
With its potential to further reduce construction time and cost
and to be a great addition to the range of innovations that architects
can work with, this new system shows great promise in being
developed as an economical and sustainable solution for modern
construction.
Case Study
Methodology
A 20-story medium-rise building was modeled as a modular structure
(Figs. 7 and 8) using the software program ETABS. ETABS is widely
used by structural engineers around the world for the analysis of
buildings, especially resistance against lateral loads. Two types of

© ASCE 05016003-3 J. Archit. Eng.

J. Archit. Eng., 05016003
 Fig 8: Plan View of a typical floor of the 3D finite element model above. (Note: The 100mm thick wall elements in modules are shown with the thicker
gray lines).

Fig 9: 3D view of a single module with stiffer walls (Type 1) (Note: Fig 10: 3D view of a single module without stiff walls (Type 2)
only the load bearing elements are modelled) (Note: only the load bearing elements are modelled)

modules were used for each story in the model; Type 1 (Fig. 9) had
stiffer reinforced concrete walls of 100 mm thickness, and Type 2
(Fig. 10) had lightweight wall panels without such rein- forced
concrete infills. In current practice, the outer walls of such
prefabricated modules are constructed of various lightweight
materials, including thin steel sheets and sandwich panels. Berman
and Bruneau (2005) have shown how thin steel sheets could enhance
the stiffness of infill wall panels. However, because the aim of this
research was only to ascertain the adequacy of the introduced system,
the modeling was kept simple, and only the supporting frame was
considered in the model, as shown in Fig. 10. The weight of the wall
panels was added to the supporting beam elements.
Columns and beams that constituted the load-transferring system
were part of the prefabricated module, and their continuity was

© ASCE 05016003-4 J. Archit. Eng.

J. Archit. Eng., 05016003
ensured through the stiffness of the bolted connection plates. The
lateral connections were modeled as frame elements with equivalent
stiffness and pinned at the common joint to allow independent
rotation of the connected modules (Annan et al. 2009). These
connections were assumed to be rigid connections for the purpose of
forming the computer model, as proposed by Annan et al. (2009).
To suit the particular structure considered, the stiffer modules (Type
1) were arranged in the three middle sets of modules (Fig. 8). A set
of modules with an opening in the middle that accommodates four
elevator shafts (variation of Type 2) was located at the rear ends of
the building. In practice, this module can be designed with the
staircases and ducts that usually accompany a conventional elevator
core.
The structure was assessed in terms of its general structural
stability, natural period, modes of vibration, and performance against
lateral earthquake and wind loads. It was important to see the
characteristics shown through the mode shapes to understand how the
structure would behave dynamically under different loading
conditions. The participation of these structural modes in the lateral
response of the structure was considered by using the complete
quadratic combination (CQC) method.
The structure was analyzed with a dynamic earthquake response
spectrum (SpecY) according to the Australian Earthquake code, AS
Fig 11: Lift shafts: (a) placed in the rear (Configuration 1, the original configuration); (b) placed in the corner module (Configuration 2); (c) placed
in the center module (Configuration 3)
© ASCE 05016003-5
J. Archit. Eng., 05016003
1170.4 (Standards Australia 2007), and a static wind load (WindY)
according to the Australian Wind code, AS 1170.2 (Standards
Australia 2011). The earthquake response spectrum was given a
hazard factor of 0.08 to match the subsoil conditions in Melbourne
according to AS 1170.4 (Standards Australia 2007), and the wind
loads are calculated with a basic wind speed of 46 ms−1. For
simplicity, the results were analyzed only in the critical y-direction to
obtain an initial idea of how structurally feasible the new system was.
Once the concept is applied to high-rise structures, the lateral loads
will become more critical. Therefore, an understanding of the
structure’s behavior under earthquake and wind forces provides
valuable insight into how the system can be modified for future high-
rise use.
Further, to highlight the key benefit of this particular system, three
different configurations (two in addition to the original one) of the
elevator shafts were analyzed to identify whether the centers of mass
and stiffness varied drastically. These configurations are shown in
Figs. 11(a–c). The new ETABS v (0).13 makes it possible to calculate
centers of rigidity and mass for each floor (CSI Knowledge Base
2014). Because the floors of the model were typical throughout, it was
adequate to calculate the results for one floor.
It is also understood that the system may present both geometric
and material nonlinearities under more severe lateral loads. As a
J. Archit. Eng.
preliminary investigation into the performance of the introduced The story drift in a structure is the difference of displacements of
structural system under more severe earthquake loads, a nonlinear the floors above and below, and is expressed as a percentage when it
earthquake time-history analysis was carried out using ETABS by is given as a ratio to the height of the floor (Chopra 2001, p. 506). The
applying six time histories (Table 1) to the model with the original story drifts resulting from the static wind and response spectrum
configuration. The lateral connections for this analysis were modeled analyses in this building were also within acceptable levels of below
as nonlinear link elements with equivalent stiffness. The maxi- mum 1.5% as specified in AS 1170.4 (Standards Australia 2007).
drift that would result from each record was observed. The analysis The maximum drifts that resulted from the nonlinear earthquake
was carried out with and without P-D effects as two separate analyses. time-history analysis are presented in Table 4. Once the geometric
nonlinearities were considered, the resulting drift values were slightly
higher than the code-recommended 1.5% for two of the time histories
Discussion of Results analyzed, whereas the others are still within acceptable limits.
The results for centers of mass and rigidity of the three different
The results for structural period and modes of vibration are presented elevator shaft configurations are shown in Tables 5–7. It is very
in Table 2. The modal mass participation factors provide an evident that neither the center of mass nor the center of rigidity shifted
understanding of the influence of each mode of vibration on the considerably from the original configuration, although the position of
behavior of the building. Modes with higher participation are more the elevator shaft was changed. This is an indication that such an
easily excited by forces applied to the structure (Irvin 2013). This, it alteration will not cause any additional adverse twisting or torsion on
is important to see which modes are prominent because this will
indicate how the structure will behave when different loads are
applied. Table 3: Summary of Results against Lateral Loads
As summarized in Table 2, the fundamental period was 3.75 s. The Maximum storey
fundamental mode was a translational mode with a rotational mode Load Case Base Shear (kN)
drift
as the second. Results showed that the rotational modes can have Wind Y (Static) 5494 0.13%
considerable influence on the behavior of the building. This result is
Spec Y (Dynamic) 1232 1.20%
important in analyzing the behavior against lateral loading, especially
for high-rise applications.
The lateral loads were also within acceptable ranges for the member Table 4: Summary of Maximum Drift Results from the Nonlinear Earthquake
sizes used in the design of the structure. It was beyond the scope of Time-History Analysis
this particular study to carry out a complete analysis of the structure Maximum Drift Maximum Drift
against lateral loads, but the basic evaluation resulted in a favorable Ground motion
Without Considering Considering P-Δ
conclusion on the structural feasibility of the new system. A summary record
P-Δ Effects Effects
of results against wind and earthquake forces is given in Table 3.
Chichi 0.62% 1.69%
As per AS 1170.4 (Standards Australia 2007), the base shear of a
structure is the “horizontal equivalent static shear force acting at the Tabas 0.63% 1.34%
base of a structure” resulting from horizontal forces such as wind and Dunzce 1.24% 1.55%
earthquake loads acting on the structure. The values of the base shears
were of realistic and acceptable values, as shown in Table 3. Kocaeli 0.85% 0.99%
Imperial Valley 0.75% 1.43%
Table 1: Earthquake Ground Motion Records Applied for the Nonlinear
Time-History Analysis Loma Prieta 1.15% 1.17%

Magnitude (Moment
Record PGA Duration
Magnitude Scale)
Chichi 7.63 0.278 g 90.0 s Table 5: Centre of Mass and Rigidity for Configuration 1 (Original
Configuration)
Tabas 7.40 0.852 g 32.8 s
Parameter X (m) Y (m)
Dunzce 7.14 0.535 g 25.8 s
Centre of Mass 17.10 5.33
Kocaeli 7.40 0.312 g 21.1 s
Imperial Valley 6.90 0.519 g 40.0 s Centre of Rigidity 17.10 5.05

Loma Prieta 7.10 0.371 g 60.0 s

Table 6: Centre of Mass and Rigidity for Configuration 2
Table 2: Modal Frequencies and Mass Participation Factors Change from Original
Y Configuration
Parameter X (m)
Mode Period (s) Frequency (Hz) UX UY RZ (m)
X (m) Y (m)
1 3.75 0.27 73.2% 0.0% 0.0% Centre of Mass 17.35 5.35 +0.25 +0.02
2 2.39 0.42 16.7% 0.2% 67.4% Centre of Rigidity 17.10 5.50 0.00 +0.45

3 1.79 0.56 6.9% 60.9% 0.1% Table 7: Centre of Mass and Rigidity for Configuration 3
UX: Participation of the modal mass to translational movement in X
direction Change from Original
UY: Participation of the modal mass to translational movement in Y Parameter X (m) Y (m) Configuration
direction X (m) Y (m)
RZ: Participation of the modal mass to rotational movement about Z Centre of Mass 17.10 5.35 0.00 +0.02
(vertical axis) Centre of Rigidity 17.10 5.50 0.00 +0.45

© ASCE 05016003-6 J. Archit. Eng.

J. Archit. Eng., 05016003
the structure, thereby giving the full benefit of this new system, as
expected.
Concluding Remarks
Many structures around the world that have been classified as
modular constructions in fact rely heavily on a conventional core
structure as the primary lateral-load-resisting mechanism. This is a
limitation that prevents the structure from enjoying the full benefits
of modular construction.
The introduced system is a step forward in designing low to
medium-rise structures as purely modular constructions. Model
analysis showed that the system results in a structure that behaves
within the parameters set out by design standards for conventional
structures under normal loading conditions. The nonlinear earthquake
time-history analysis considered the possible geometric and material
nonlinearities that may arise in the system when subject to more
severe dynamic lateral loads. Although the resulting drift values were
mostly within the specified limits, further studies are required to
obtain a more detailed understanding on the failure mechanisms and
redundancies of the system.
This study also confirmed that the newly introduced structural
system allows the elevator shaft to be positioned in various
arrangements without causing additional adverse torsional effects.
Although the advantages of the new system were proven to be
desirable, further research is required to confirm the validity,
especially for tall buildings. Further, a detailed future study into the
economic and sustainability aspects of this system will ensure that it
is developed as a solid methodology to reap the full benefits of purely
modular construction.
It is apparent that the basic idea behind the new system is feasible
from a structural perspective. The added flexibility for designing with
less stress on maintaining a core structure at a particular place makes
the system very attractive for architects as well.
The vision is to develop this technology for the use of taller
buildings. For this purpose, the system can be modeled and designed
with improved connections and an appropriate placing of
stiffer modules. Different materials, such as high-strength steels
and concrete and composites with various mechanical properties, can
be evaluated for their potential use in connections and stiff walls in
the stiffer modules.
The modules can be designed in such a way that they can replicate
the use of a conventional core. For example, they can be used in
collaboration with a smart system to support a crane and also sup-
port auto-climbing features.
References
Annan, C. D., Youssef, M. A., and El Naggar, M. H. (2009).
“Seismic vulnerability assessment of modular steel buildings.” J.
Earthquake Eng., 13(8), 1065–1088.
Aye, L., Ngo, T., Crawford, R. H., Gammampila, R., and
Mendis, P. (2012). “Life cycle greenhouse gas emissions and
energy analysis of prefabricated reusable building modules.”
Energy Build., 47(0), 159–168.
Berman, J. W., and Bruneau, M. (2005). “Experimental
investigation of light-gauge steel plate shear walls.” J.
Struct. Div., 10.1061/ (ASCE)0733-9445(2005)131:2(259), 259–
267.
Chopra, A. K. (2001). Dynamics of structures, Prentice Hall, Upper
Saddle River, NJ.
CSI Knowledge Base,
hhttps://wiki.csiamerica.com/display/etabs/Centerþofþrigidityi,
(Sep. 9, 2014).
© ASCE 05016003-7
J. Archit. Eng., 05016003
View publication stats
Annan, C. D., Youssef, M. A. and El Naggar, M. H. (2009).
"Seismic Vulnerability Assessment of Modular Steel Buildings",
Journal of Earthquake Engineering 13(8): 1065-1088.
AS 1170.2: 2011: Structural Design Actions. Part 2: Wind Actions,
Standards Australia, New South Wales.
AS 1170.4: 2007: Structural Design Actions. Part 4: Earthquake
Actions in Australia, Standards Australia, New South Wales.
Aye, L., Ngo, T., Crawford, R. H., Gammampila, R. and Herath, N.
(2012). "Life cycle greenhouse gas emissions and energy analysis
of prefabricated reusable building modules." Energy and
Buildings, 47(0): 159-168
Berman, J. W. and Bruneau, M. (2005). “Experimental Investigation
of Light-Gauge Steel Plate Shear Walls”. Journal of Structural
Engineering, American Society of Civil Engineers, 259-267.
Chopra, A. K., “Dynamics of Structures”, Prentice Hall, New
Jersey, USA. 2001
CSI Knowledge Base,
<https://wiki.csiamerica.com/display/etabs/Center+of+rigidity>,
(Sept 9, 2014).
ETABS 15.2.0 [Computer software]. Computers and
Structures, Inc., Berkeley, CA.
Gunawardena, T., Mendis, P., Ngo, T., Aye, L. and Alfano, J.
(2014a). “Sustainable Prefabricated Modular Buildings”,
Proceedings of 5th International Conference on Sustainable Built
Environment, Kandy, Sri Lanka.
Gunawardena, T., Ngo, T., Mendis, P., Aye, L. and Crawford, R.
(2014b). “Time Efficient Post-disaster Housing Reconstruction
with Prefabricated Modular Structures.” Open House
International, 39(3): 59-69.
Irvin, T. (2013). “Effective Modal Mass and Modal Participation
Factors, Revision H”,
<http://www.vibrationdata.com/tutorials/ModalMass.pdf>, (Dec
10, 2014)
Jaillon, L., Poon, C. S. and Chiang, Y. H. (2009). "Quantifying the
Waste Reduction Potential of Using Prefabrication in Building
Construction in Hong Kong." Waste Management, 29(1): 309-320.
Lawson, R. M., Ogden, R. G. and Bergin, R. (2012). “Application of
Modular Construction in High-rise Buildings”, Journal of
Architectural Engineering, American Society of Civil Engineers,
148-154.
Osmani, M., Glass, J. and Price, A. (2006). “Architect and
Contractor Attitudes to Waste Minimisation”, Waste and Resource
Management, 2(1), 65-72.
J. Archit. Eng.