buildings

Article
Cyclic Evaluation of Exterior Beam-to-Column Connections for
Intelligent Accelerated Construction of Industrial Facilities
Mingming Ji 1,2 , Wei Chen 1 , Shi Zeng 2 and Xinyu Zhao 2, *

1 Guangzhou Engineering Contractor Group Co., Ltd., Guangzhou 510310, China

2 State Key Laboratory of Subtropical Building Science, South China University of Technology,
Guangzhou 501641, China
* Correspondence: ctzhaoxy@scut.edu.cn; Tel.: +86-135-7037-3755

Abstract: The rapid development towards automated construction has been witnessed in recent years
mainly due to the growing shortage of skilled labor. Against that backdrop, an accelerated method,
with the aid of robotic cranes, is emerging in China to speed up the construction of industrial facilities
such as warehouse structures. This method requires that the steel bars in the precast beams do not
extend beyond the beam ends to facilitate the temporary fixation of the robotic crane at the top of
cast-in-place columns. This, nonetheless, brings a series of new problems, one of which is how to
choose a suitable anchoring type for the beam bottom bars. To address this issue, three large-scale
exterior beam-to-column connections were fabricated and tested under lateral load reversals. Two
anchorage forms, namely, mechanical splices and grouted sleeves, were adopted and compared with
the monolithically cast specimen. The test results showed that the specimen using the grouted sleeves
had similar seismic performance to that of the cast-in-place specimen, whereas the specimen using
the mechanical splices presented significant post-peak deterioration under positive beam bending
moments. This happened because the congestion of steel bars within the joint core made it difficult to
fully tighten the beam bottom bars into the threaded couplers; consequently, a “slop” was formed
which could substantially impair the cyclic behavior of the specimen. As such, it is suggested that
grouted sleeves should be more applicable and reliable for the new construction method. This
counter-intuitive finding also indicates that, for intelligent construction, no detail should be taken for
Citation: Ji, M.; Chen, W.; Zeng, S.; granted, but rather needs due consideration.
Zhao, X. Cyclic Evaluation of Exterior
Beam-to-Column Connections for Keywords: precast concrete; rebar anchorage; RC connections; intelligent construction; accelerated
Intelligent Accelerated Construction construction; seismic performance
of Industrial Facilities. Buildings 2023,
13, 62. https://doi.org/10.3390/
buildings13010062

Academic Editor: Marco Di 1. Introduction

Ludovico In stark contrast to the low productivity of the construction sector relative to other
Received: 27 November 2022 sectors, the strong demand for new constructions is one of the major challenges that all civil
Revised: 18 December 2022 engineering departments, firms and engineers have to face. Furthermore, the increasing
Accepted: 23 December 2022 scarcity of natural resources, low level of automation in design and construction, and lack
Published: 27 December 2022 of labor force are putting the sector in an unsustainable predicament [1–4].
Under such circumstances, the industry must undergo fundamental changes. One
solution is to espouse the importance of the adoption of prefabrication in construction so as
to achieve sustainable development. The advantages of prefabrications have long been rec-
Copyright: © 2022 by the authors.
ognized [5–10]: high cost-efficiency, powerful mass production, and replaceability in case
Licensee MDPI, Basel, Switzerland.
of severe damage. Since prefabricated structures are produced under factory-controlled
This article is an open access article
conditions, they benefit from better and standardized material quality, precise geometry,
distributed under the terms and
and lower cost compared to cast-in-place structures. Additionally, precasting reduces
conditions of the Creative Commons
weather-related holdups and is cleaner because it eliminates the need for on-site casting
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
and curing. The use of prefabricated elements also reduces health and safety risks by re-
4.0/).
ducing on-site workmanship. Moreover, shrinkage and creep of prefabricated components

Buildings 2023, 13, 62. https://doi.org/10.3390/buildings13010062 https://www.mdpi.com/journal/buildings

Buildings 2023, 13, 62 2 of 19

occur mainly during factory production. In general, by reducing waste emissions, prefabri-
cated buildings offer greater energy efficiency and sustainability compared to traditional
monolithic systems.
Of a precast structural system, the beam-to-column connections are the most vul-
nerable part, susceptible to extensive damage under seismic events [11]. For this reason,
considerable research studies have been conducted to better understand the complex be-
havior of these connections in precast concrete structures [12]. In recent years, the precast
industry has also attempted to develop imaginative new connections that would perform
well during an earthquake [13].
For example, French et al. [14] have conducted extensive cyclic tests on various precast
connections. They found that, although the plastic hinge was forced to develop within the
connection region, the specimens still exhibited ductility and energy dissipation character-
istics similar to those of ordinary monolithically cast elements. Through experimentation,
Soubra et al. [15] highlighted the superiority of using fiber-reinforced concrete in critical
regions of precast frames; they demonstrated that the alternative measure can dramatically
improve the specimens’ cyclic behavior—displacement ductilities even up to 6.0 were
achieved. Choi et al. [16] developed a new wet precast connection which uses steel plates
to realize joint continuity and consequently obtains improved seismic performance. Lago
et al. [17] demonstrated the viability of an innovative dry-assembled precast structural
system with special mechanical connection devices. Senturk et al. [18] developed a novel
monolithic-like precast connection using bolted plates. The new connection system is
easy to assemble and exhibits considerably higher (up to 34%) ductility and ultimate de-
formability compared with the monolithic connections. Correal et al. [19] found that the
precast connections in one-story industrial facilities designed as per modern ACI codes
have good seismic resistance and can resist a lateral drift ratio of at least 3.5%. Recently,
some innovative hybrid precast connections have also been developed and their adequate
seismic performance has been well illustrated [20–24].
According to [25], the degree of automation in precast constructions is more advanced
than any other part of the construction sector. Indeed, the integration of automation and
robotics into precast constructions holds promise for sustainable performance in terms
of increased construction efficiency, waste reduction, improved workplace safety and
enhanced intelligent living environments [26].
Against this background, an accelerated construction method using robotic cranes
developed for industrial facilities, such as warehouse structures, is emerging in China. As
shown in Figure 1, the structure demonstrated is “semi-precast” in that the columns are cast
in situ while the beams are prefabricated in plant. Most notably, the lifting of the precast
beams can be greatly accelerated with the help of a robotic crane shown in the figure. The
robotic crane can move as needed over the cast-in-place columns. It is temporarily fixed to
the top of four erected columns during operation. After being lifted the precast beams are
placed on the erection corbels (Figure 1).
This construction method substantially increases the hoisting speed, which is especially
suitable for large industrial buildings. However, it requires that the longitudinal steel bars
of the precast beams do not extend out of the beam ends, so as to facilitate the temporary
connection between the crane’s crawling feet and the top surface of the erected columns.
ildings 2023, 12, x FOR
Buildings PEER
2023, REVIEW
13, 62 3 of 21 3 of 19

Concealed Precast
beam bars Robotic
crane beam

Cast-in-situ
Erection column
corbel

Figure 1. AnFigure
example project
1. An using
example a four-leged
project using arobotic crane.
four-leged robotic crane.

After the
This construction crane has
method walked away,
substantially headed
increases thereinforcement
hoisting speed, is used
which within the joint core
is espe-
cially
Buildings 2023, 12, x FOR PEER to anchor
suitable
REVIEW the industrial
for large bottom rebars of the However,
buildings. precast beams as wellthat
it requires as their top rebars in place,
the longitudinal 4 of 21 as
steel bars ofillustrated in Figure
the precast beams do 2a. not
Then the joint,
extend out ofthethe
upper
beamcolumn,
ends, soand thefacilitate
as to remainingthepart of the
temporarybeams are cast
connection at the the
between job crane’s
site, as shown
crawling in feet
Figure
and2b.
the top surface of the erected
columns.
After the crane has walked away, headed reinforcement is usedCast-in-situ within the joint core
to anchor the bottom rebars of the precast beams as well as their upper
top column
rebars in place, as
Beam upper bars
illustrated in Figure(placed the joint, the upper column, and the remaining part Topping
on site)
2a. Then of the
beams are cast at the job site, as shown in Figure 2b.
Headed For the beam bottom bars, they can be connected to the headed reinforcement by
rebars different splicing methods. Despite different types of proposed new precast connection
[16–24], the most practical methods are to use mechanical couplers or grouted sleeves.
However, which connection approach is better is a question. Mechanical couplers are
commonly used to connect reinforcing bars, but it is not clear whether they can work in
Different Beam bot- Corbel
connectors
areas densely tom reinforced
bars with rebars (such as within beam–column joints). Precast Grouted beam
sleeves are the primary forms of rebar connection in precast structures, yet there are often
Cast-in-situ
concerns about their construction quality. Cast-in-situ
lower column lower column

(a) Installation of steel cages on site (b) After completing the in-situ casting

Figure
Figure 2. Schematic
2. Schematic diagram
diagram of the
of the construction
construction process
process of semi-precast
of semi-precast exterior
exterior connections.
connections.

ToFor theout
seek beam
thebottom bars, they
most suitable waycanofbeconnecting
connected to theheaded
the headedreinforcement
reinforcement by different
with the
beam bottom bars, three large-scale exterior beam–column connections were fabricatedthe
splicing methods. Despite different types of proposed new precast connection [16–24],
most
and practical
tested undermethods are to use
cyclic loading. Twomechanical
specimenscouplers or grouted
using different bar sleeves.
splicing However, which
methods (i.e.,
connection approach is better is a question. Mechanical couplers are commonly
mechanical couplers or grouted sleeves) were compared with the control cast-in-place used
specimen. Test results enabled thorough examination of the seismic performance of the
exterior connections in terms of stiffness, strength, and energy dissipation. The potential
defects of the mechanical splicing method were pointed out, thus providing design in-
sights and ensuring safety of the semi-precast constructions.
Buildings 2023, 13, 62 4 of 19

to connect reinforcing bars, but it is not clear whether they can work in areas densely
reinforced with rebars (such as within beam–column joints). Grouted sleeves are the
primary forms of rebar connection in precast structures, yet there are often concerns about
their construction quality.
To seek out the most suitable way of connecting the headed reinforcement with the
beam bottom bars, three large-scale exterior beam–column connections were fabricated
and tested under cyclic loading. Two specimens using different bar splicing methods
(i.e., mechanical couplers or grouted sleeves) were compared with the control cast-in-place
specimen. Test results enabled thorough examination of the seismic performance of the
exterior connections in terms of stiffness, strength, and energy dissipation. The potential
defects of the mechanical splicing method were pointed out, thus providing design insights
and ensuring safety of the semi-precast constructions.

2. Test Program
2.1. Specimen Design
Three large-scale T-shaped exterior beam-to-column connections were constructed
in this study. One of them was a cast-in-place specimen (designated as LZ1), whereas the
other two were semi-precast units, constructed in the way just depicted in Figure 2. The
two semi-precast specimens adopted mechanical couplers or grouted sleeves to connect
beam bottom bars and headed reinforcement in the joint region; they are designated as LZ2
and LZ3, respectively. All the specimens shared the same geometry and dimensions, as
detailed in Figure 3. The beam of each specimen was 2800 mm long (measured from the
beam tip to the beam–joint interface) with a cross-section of 200 mm × 500 mm. The upper
and lower columns were of the identical length, each measuring 1350 mm. They also had
the same cross-sections, being 400 mm square. An erection corbel was set on the top of the
lower column for easing the seating of the precast beam. The corbel’s height was 200 mm;
it extended 100 mm out of the inner edge of the lower column.
The three specimens were reinforced in exactly the same manner. Three D22 lon-
gitudinal bars were placed at the top and bottom of the beam, along with D8 stirrups
set at 100 mm c/c spacing (this spacing was reduced to 50 mm near to the beam loading
point). The stirrups were closed with 135◦ bends and 10 bar-diameter extensions to comply
with the current concrete design code [27]. The longitudinal reinforcement of the column
comprised twelve D22 bars symmetrically distributed around the perimeter of the column
section. To meet the minimum requirement for transverse reinforcement in joint and adja-
cent column regions, D8 hoops at a spacing of 100 mm were placed throughout the column
(Figure 3), which provided four legs acting in either direction.
LZ1, a monolithic specimen constructed without cold joints, was used as a control
specimen (Figure 3a). For LZ2 and LZ3, three mechanical couplers and half-grouted sleeves
were, respectively, used to splice the headed bars and the beam bottom bars (Figure 3b,c).
The embedded length of the headed bars for LZ1 through LZ3 was 300 mm (=13.6db , where
db is the bar diameter), and the area ratio of the rounded head to the bar, Ah /Ab , was 4.8.
These values of the embedded length and the Ah /Ab ratio satisfied the recommendations
of Wallace et al. [28] (the suggested values were 12db and 4.0, respectively). As such, the
pull-out failure of the headed bars was not anticipated. In addition, the anchorage of the
headed bars left a 100 mm back cover behind the outer edge of the heads, being sufficient
to prevent the breakout failure of column concrete cover when the headed bars were in
compression [28,29].
All the three specimens were designed based on the strong column–weak beam
principle [30]; the column to beam moment strength ratio was 2.75 (the required ratio per
ACI 318-19 [31] is 1.4). Consequently, flexural hinging was expected to occur in the beam
adjacent to the column face. This situation enabled comparison of the efficacy of the two
different reinforcement splicing methods.
ldings 2023, 12, x FOR PEER REVIEW 5 of 21
Buildings 2023, 13, 62 5 of 19

400
Steel plate

A-A
12D22

400
D8@100

S ection A -A

300 B-B
3200

Headed
bars

200
200
100

6D22

500
D8@100

S teel plate
S ection B-B

400 2200 600

(a) LZ1

400
Steel plate

A-A
12D22
400

D8@100

S ection A -A
Cas t-in-place
300 concrete B-B
3200

H eaded
bars

M echanical 200
coupler
150

Pre-cas t
concrete 6D22
2D16
500

D8@100

Mechanical coupler
S teel plate
Section B-B

400 2200 600

(b) LZ2

Figure 3. Cont.
ldings 2023, 12, x FOR PEER REVIEW 6 of 21
Buildings 2023, 13, 62 6 of 19

400
Steel plate

A-A
12D22

400
D8@100

S ection A -A
Cas t-in-place
300 concrete B-B
3200

H eaded
bars

Half grouted 200

s leeve

150
Pre-cas t
concrete 6D22
2D16
500 D8@100

S teel plate Half-grouted sleeve

Section B-B

400 2200 600

(c) LZ3

Figure 3. Dimensions
Figure 3. Dimensions and reinforcement
and reinforcement details of thedetails of the three specimens.
three specimens.
2.2. Material Properties and Specimen Fabrication
LZ1, a monolithic specimen constructed without cold joints, was used as a control
specimen (Figure C453a).
grade
Forconcrete
LZ2 and (with
LZ3,the nominal
three cubic strength
mechanical couplersequal
andtohalf-grouted
45 MPa) was used for
the specimens. Specifically, the actual cubic compressive strength of the concrete used for
sleeves were, respectively, used to splice the headed bars and the beam bottom bars (Fig-
the precast part of the specimens was measured as 44.1 MPa at the test period, while the
ure 3b,c). The embedded length of the headed bars for LZ1 through LZ3 was 300 mm
corresponding value for the cast-in-situ parts was 49.9 MPa. High-strength cementitious
(=13.6db, where db is the bar diameter), and the area ratio of the rounded head to the bar,
grouting material (101.2 MPa) was used to fill the half-grouted sleeves.
Ah/Ab, was 4.8. These values of the embedded length and the Ah/Ab ratio satisfied the rec-
The measured tensile strengths of the steel bars used in this study are listed in Table 1.
ommendations of Wallace et al. [28] (the suggested values were 12db and 4.0, respectively).
The tensile capacity of the couplers and sleeves specified by manufacturers was in excess
As such, the pull-out failure of the headed bars was not anticipated. In addition, the an-
of 1.5~2.0 times the yield strength of the D22 steel bars. Therefore, those connectors should
chorage of the headed bars left a 100 mm back cover behind the outer edge of the heads,
not experience failure during the tests.
being sufficient to prevent the breakout failure of column concrete cover when the headed
bars were in compression [28,29].
Table 1. Strengths of steel reinforcement.
All the three specimens were designed based on the strong column–weak beam prin-
ciple [30]; the column Barto beam
Type moment strength ratio Yieldwas 2.75 (the required ratio
Strength per ACI
Ultimate Strength
318-19 [31] is 1.4). Consequently,
D8 flexural hinging was426 expected to occur in the beam 648ad-
jacent to the column face. D22This situation enabled comparison
390 of the efficacy of the578
two
different reinforcement splicing methods.
Each specimen was equipped with two 20 mm thick steel plates, fixed at the column
2.2. Material Properties and Specimen Fabrication
ends by welding them to the longitudinal bars. LZ1 was cast monolithically in one go,
C45 grade
whileconcrete
LZ2 and(with the nominal
LZ3 were cubic strength
manufactured accordingequal to 45
to the MPa)shown
process was used for 2.
in Figure
the specimens. Specifically, the actual cubic compressive strength of the concrete used for
the precast2.3.
partTest
of the specimens
Setup, was measured
Instrumentation, as 44.1
and Loading MPa at the test period, while the
Protocol
corresponding value
Figurefor the cast-in-situ
4 shows parts used
the test setup was 49.9 MPa.
in this High-strength
study. cementitious
Each specimen was loaded in this
grouting material (101.2 MPa) was used to fill the half-grouted sleeves.
statically determinate test rig with the column in a lying position; that is, each specimen
was rotated 90 degrees to accommodate the setup. The column was pin-supported at both
ends, while a MTS actuator was installed at the top of the (vertical) beam to apply (cyclic)
lateral loading. The pin-to-pin distance of the column was 3700 mm.
ildings 2023, 12, x FOR PEER REVIEW 8 of 21
Buildings 2023, 13, 62 7 of 19

Beam-joint
interface
LVDTs

Figure
Figure 4. Test Testinstrumentation
setup4.and setup and instrumentation
(unit: mm). (unit: mm).

A series of linear variable displacement transducers (LVDTs) and strain gauges were
3. Test Results
arranged within and around the joint core to measure the beam and column rotations
3.1. General(including
Observationsthe curvatures at the beam–joint interface; see the inset of Figure 4), joint shear
As expected, all three
deformation, andspecimens
strains in thedeveloped beam hinging failure and behaved in a
reinforcement.
ductile manner,Prior
withto the exception the of LZ2 load,
whena itmanually
was subjected to beam
axial upward loading 0
applying lateral controlled load of 0.3A g f c (=1609 kN)
(a positivewas
beam moment
applied to was
the generated
column. Then, under thethis specimens
condition and were then the beam
tested underbottom
a displacement-
bars and thecontrolled
connectors cyclic
were load history based on the loading protocol defined in ACI 374.1-05 [32],
in tension).
Figurewhich consisted
5 shows of three cycles
the progression of cracksof push and pull
observed for each
in each drift at
specimen amplitude ± %): 0.1, 0.25,
different(drift
0.50, 0.75, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0. The cycles were
ratios, where the drift ratio was defined as the deflection at the beam loading point, Δ, controlled quasi-statically (0.3
to 1.5 mm/s). Testing continued until the lateral load decreased
divided by the distance from the point to the center line of the column, Lc (=2800 mm; c.f. by more than 15% of the
maximum
Figure 4). Based measured force.
on inspections during A the
datacourse
acquisition system
of testing, thewas used to
damage store all the
progression ofdata.
the test specimens can be summarized as follows:
3. Test
For LZ1, Results
evenly distributed hairline flexural cracks initiated in the beam by a drift
3.1. General Observations
ratio of 0.25%. No significant cracking at the beam–joint interface was observed. As the
As expected,
specimen was loaded cyclically, all three
cracksspecimens
on one side developed
of the beambeam(tophinging failure opened
or bottom) and behaved in a
while those ductile
on themanner, withclosed,
other side the exception
and then of LZ2
vice when
versa.itWhen
was subjected
the drift to beam
ratio upward loading
attained
(a positive
0.75%, a small diagonal beam
crackmoment
emanating was from
generated
the top under thisof
surface condition
the corbel and
wasthen the beam bottom
captured.
bars and
Flexural cracks alsothe connectors
occurred at thewere in tension).
junction between the beam top and the joint. With
Figure 5 shows the progression
further loading, no distress was found within the of joint
cracks observed
region, in each
but the damagespecimen
to the at different drift
spec-
imen was concentrated gradually at the beam hinging area, especially at the bottom of the point, ∆,
ratios, where the drift ratio was defined as the deflection at the beam loading
beam out of divided by theFinally,
the corbel. distance thefrom the point
concrete in thisto the center
region wasline of the crushed
severely Lc (=2800 mm; c.f.
column, (Figure
Figure 4). Based on inspections
6a), signifying a complete failure of LZ1. during the course of testing, the damage progression of the
test specimens can be summarized as follows:
For LZ1, evenly distributed hairline flexural cracks initiated in the beam by a drift
ratio of 0.25%. No significant cracking at the beam–joint interface was observed. As the
specimen was loaded cyclically, cracks on one side of the beam (top or bottom) opened
while those on the other side closed, and then vice versa. When the drift ratio attained
0.75%, a small diagonal crack emanating from the top surface of the corbel was captured.
Buildings 2023, 13, 62 8 of 19

Flexural cracks also occurred at the junction between the beam top and the joint. With
further loading, no distress was found within the joint region, but the damage to the
specimen was concentrated gradually at the beam hinging area, especially at the bottom
Buildings 2023, 12, x FOR PEER REVIEW 9 of 21
of the beam out of the corbel. Finally, the concrete in this region was severely crushed
(Figure 6a), signifying a complete failure of LZ1.

LZ 1 LZ 2 LZ 3

Drift ratio = +0.5% Drift ratio = +0.5% Drift ratio = +0.5%

Drift ratio = +1% Drift ratio = +1% Drift ratio = +1%

Drift ratio = +1.5% Drift ratio = +1.5% Drift ratio = +1.5%

Drift ratio = +2.5% Drift ratio = +2.5%

Figure 5. Comparison of cracking patterns before significant concrete crushing.

Figure 5. Comparison of cracking patterns before significant concrete crushing.
ldings 2023,Buildings
12, x FOR2023,
PEER13,REVIEW
62 10 of 21 9 of 19

(a) LZ1

(b) LZ2

(c) LZ3
Figure 6. Photographs after the tests.
Figure 6. Photographs after the tests.

In the case In of the

LZ2case of LZ2
using using mechanical
mechanical couplers, the couplers,
incipientthecracking
incipient cracking
was similarwas
to similar to
that of LZ1that
priorof to
LZ1theprior
drifttoratio
the drift ratio of
of 0.75%. 0.75%. However,
However, small cracks
small flexural flexuralappeared
cracks appeared
at at the
the junctionjunction
betweenbetween the precast
the precast part and part
theand the cast-in-place
cast-in-place partbeam.
part of the of theAfterwards,
beam. Afterwards, the
the upper andupper
lowerandcracks
loweratcracks at the beam–joint
the beam–joint interfaceinterface were coalesced
were coalesced and widened.
and widened. How- However,
ever, the joint core remained intact throughout the testing. Beyond a drift ratio of 1.0%,1.0%,
the joint core remained intact throughout the testing. Beyond a drift ratio of an an abrupt
abrupt loss in strength was witnessed when upward (positive) loading was applied to theto the beam.
loss in strength was witnessed when upward (positive) loading was applied
Yetspecimen
beam. Yet the the specimen wasable
was still still to
able to sustain
sustain successive
successive cycling
cycling and theand the resistance
resistance was was kept
and even increased under beam downward loading. This
kept and even increased under beam downward loading. This continued until a positive continued until a positive drift
ratio
drift ratio of 2.5%of 2.5% was arrived
was arrived at. this
at. At At this point,
point, thethe beam
beam hadhad softenedcompletely,
softened completely,as as evidenced
evidenced by a drop in strength of 90% in the positive loading direction. The cause of such a failure
by a drop in strength of 90% in the positive loading direction. The cause of such
Buildings 2023, 13, 62 10 of 19

was speculated to be related to the pull-out failure of the beam bottom bars or the joint
headed bars under beam upward loading, which will be discussed later.
As to LZ3 with grouted sleeves, its damage progression resembled that of LZ1. Flexural
cracks developed during positive and negative drift excursions. There was no sign of
damage within the joint core. The specimen ended up failing due to the major crushing of
the concrete in the plastic hinge region of the beam.
Figure 6 presents the photographs of the specimens after the completion of testing.
For LZ1 and LZ3, the failure pattern near the beam–joint interface clearly confirmed the
description by Englekirk [13]: “The Achilles heel of a properly conceived concrete ductile
frame beam has always been the toe (no pun intended) of the frame beam.”
Unlike LZ1 and LZ3, no concrete crushing was observed in LZ2 (Figure 6b). Never-
theless, in LZ2 a very wide vertical crack exhibited at the junction between the beam and
the joint. Moreover, the lower part of the crack close to the corbel was obviously wider
than the upper part. This phenomenon was probably induced by the significant slippage
of the reinforcing bars here under the action of beam positive bending. Presumably, the
reinforcing bars had been pulled out of the mechanical couplers. Later explorations will
corroborate this conjecture.
Note that the presence of the corbel arrested the damage in the beam end compression
zone under negative bending. This was why the beam crushed region was either shifted
outside the corbel (LZ1) or more severe on the top of the beam end (LZ3).

3.2. Global Load–Displacement Response

Table 2 lists the main results of the current tests (including the characteristic loads and
displacements). The global load–displacement (V b –∆) relationships measured at the beam
tip are shown in Figure 7; V b here denotes the lateral load acting on the beam.

Table 2. Main test results.

Vy 1 ∆y 1 Vp 1 ∆p 1
Specimen
(+) (−) (+) (−) (+) (−) (+) (−)
LZ1 87.2 86.6 23.2 25.0 98.4 101.1 98.0 98.3
LZ2 85.6 88.2 23.0 25.5 94.7 104.8 27.7 113.1
LZ3 87.1 91.2 23.3 23.4 101.6 104.0 53.4 83.9
1 ‘V y ’ represents the yield load obtained by the Park method; ‘∆y ’ denotes the displacement at V y ; ‘V p ’ represents
the peak load; ‘∆p ’ denotes the displacement at V p .

For LZ1, the hysteresis loops were overall symmetric. No significant pinching was
found, and the specimen represented typical flexural failure mode. A drift ratio of 4.0%
was achieved for LZ1, then the load declined under negative loading, which was associated
with the beam concrete crushing as described previously.
For LZ3, its hysteresis loops matched qualitatively well with those of LZ1. However,
the displacement at the peak load, ∆p , of LZ3 was less than that of LZ1 and when the
specimen was loaded toward +4.0% drift ratio, the resistance abruptly dropped off, due
primarily to the concrete crushing at the beam end.
LZ2, however, presented a completely different hysteretic behavior. When the positive
drift reached 27.7 mm (1.0% drift ratio), the load-bearing capacity of the specimen declined
dramatically and unexpectedly, and only 12% of the capacity remained at 2.5% drift ratio.
The specimen still reserved a considerable strength in the negative loading direction though,
which made the positive and negative responses totally asymmetric, and a strong pinching
was also clearly observed.
Figure 8 compares the skeleton curves of the specimens. Clearly, the skeleton curve of
LZ3 was similar to that of LZ1, because of their similar flexural failure mode. This was also
indicative of the comparable load–displacement capacity of the cast-in-situ specimen and
the semi-precast specimen using grouted sleeves. In addition, according to [18], a beam–
column connection can be considered to have good seismic performance if it can sustain at
 Table 2. Main test results.

Vy 1 Δy 1 Vp 1 Δp 1
Specimen
(+) (-) (+) (-) (+) (-) (+) (-)
Buildings 2023, 13, 62 LZ1 87.2 86.6 23.2 25.0 98.4 101.1 98.0 98.311 of 19
LZ2 85.6 88.2 23.0 25.5 94.7 104.8 27.7 113.1
LZ3 87.1 91.2 23.3 23.4 101.6 104.0 53.4 83.9
1 ‘Vy’arepresents
least 3.5% drifttheratio
yieldwith
load aobtained
maximum by the Park
25% method; ‘Δy’ capacity
load-bearing denotes the displacement
loss. at Vyboth
Apparently, ;
‘Vp’ represents the peak load; ‘Δp’ denotes the displacement at Vp.
LZ1 and LZ3 met this performance criterion.
Drift Ratio（%） Drift Ratio (%)
-5.0 -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0 -5.0 -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0
125 125

100 100

75 75

50 50

25 25
Vb (kN)

Vb (kN)
0 0

-25 -25

-50 -50

-75 -75

-100 -100

-125 -125
-140 -112 -84 -56 -28 0 28 56 84 112 140 -140 -112 -84 -56 -28 0 28 56 84 112 140
Δ (mm) Δ (mm)

(a) LZ1 (b) LZ2

Drift Ratio (%)
-5.0 -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0
120
100
80
60
40
20
Vb (kN)

0
-20
-40
-60
-80
-100
-120
-140 -112 -84 -56 -28 0 28 56 84 112 140
Δ (mm)

Buildings 2023, 12, x FOR PEER REVIEW (c) LZ3 13 of 21

Figure7.7.Beam
Figure Beam tip
tip load–displacement
load–displacement relationships.
relationships.

Drift Ratio（%）
-4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0
125

100

75

50

25
Vb (kN)

-25

-50
LZ1
-75 LZ2
-100 LZ3
-125
-112 -84 -56 -28 0 28 56 84 112
Δ (mm)

Figure 8. Skeleton curves of hysteresis loops.

3.3. Energy Dissipation and Secant Stiffness Degradation

Figure 9 compares the energy dissipation of the specimens at different drift ratios. As
can be seen, LZ3′s energy absorbing capacity agreed well with that of LZ1. This again
demonstrated the viability of using grouted sleeves as a bar splicing solution. LZ2′s en-
 0

Vb (
-25

-50
LZ1
-75 LZ2
Buildings 2023, 13, 62 -100 LZ3 12 of 19

-125
-112 -84 -56 -28 0 28 56 84 112
Δ (mm)
The skeleton curve of LZ2 shows an obvious post-peak strength decay in the positive
Figure 8. Skeleton curves of hysteresis loops.
loading direction. As stated, this was thought to be induced by the pull-out failure of
the Energy
3.3. joint bottom headed
Dissipation bars Stiffness
and Secant or the beam bottom bars. However, in the negative loading
Degradation
direction, the skeleton curve of LZ2 almost coincided with that of LZ1. This suggests that
Figure 9 compares the energy dissipation of the specimens at different drift ratios. As
the be
can mechanical couplers
seen, LZ3′s could stillcapacity
energy absorbing effectively transfer
agreed forces
well with under
that compression.
of LZ1. This again
demonstrated the viability of using grouted sleeves as a bar splicing solution. LZ2′s en-
3.3. Energy Dissipation and Secant Stiffness Degradation
ergy absorbing capacity correlated tightly with those of LZ1 and LZ3 before the drift ratio
Figure
of 1.0%, 9 compares
but beyond theformer
that the energy dissipation
showed of thecapacity
a decreased specimens at different
to dissipate drift
energy, es-ratios. As
can be seen,
pecially LZ30 s energy
in the positive absorbing
direction. capacity
This can be ascribedagreed well with
to the pull-out thatasof
failure LZ1. This again
aforemen-
demonstrated the viability of using grouted sleeves as a bar splicing solution. LZ20 s energy
tioned.
Figure capacity
absorbing 10 showscorrelated
comparisons of thewith
tightly secant stiffness
those of LZ1 degradation between
and LZ3 before thethe speci-
drift ratio of 1.0%,
mens. Obviously,
but beyond that had the pull-out
the former failure
showed not occurred
a decreased in LZ2, to
capacity thedissipate
stiffness degradation
energy, especially in
of
thethe three specimens
positive direction.could
Thishave
can been very similar.
be ascribed to the pull-out failure as aforementioned.
10

9
LZ1
8 LZ2
Energy dissipation (kN·m)

7 LZ3

0
-4 -3 -2 -1 0 1 2 3 4
Drift Ratio (%)
Figure
Figure9.9.Energy
Energydissipation evolutions.
dissipation evolutions.

Buildings 2023, 12, x FOR PEER REVIEW Figure 10 shows comparisons of the secant stiffness degradation between 14 the
of 21speci-
mens. Obviously, had the pull-out failure not occurred in LZ2, the stiffness degradation of
the three specimens could have been very similar.
7

6 LZ1
LZ2
Secant stiffness (kN/mm)

5 LZ3

0
-4 -3 -2 -1 0 1 2 3 4
Drift Ratio (%)

Figure10.
Figure Secantstiffness
10.Secant stiffness degradation.
degradation.

3.4.Strain
3.4. StrainResponse
Response
Figure1111shows
Figure showsthe
thestrain
strain profiles
profiles along
along thethe beam
beam bottom
bottom barsbars
nearnear the beam–joint
the beam–joint
interface under positive loading for the three specimens. These strains were
interface under positive loading for the three specimens. These strains were important important
because they could provide useful information about bar slippage [33]. If the adjacent
strains of a bar recorded between two successive drifts remain the same or decrease, it
signals the occurrence of pull-out of the bar [34].
 -4 -3 -2 -1 0 1 2 3 4
Drift Ratio (%)

Figure 10. Secant stiffness degradation.

3.4. Strain Response

Buildings 2023, 13, 62 13 of 19
Figure 11 shows the strain profiles along the beam bottom bars near the beam–joint
interface under positive loading for the three specimens. These strains were important
because they could provide useful information about bar slippage [33]. If the adjacent
because they could provide useful information about bar slippage [33]. If the adjacent
strains of a bar recorded between two successive drifts remain the same or decrease, it
strains of a bar recorded between two successive drifts remain the same or decrease, it
signals the occurrence of pull-out of the bar [34].
signals the occurrence of pull-out of the bar [34].

1.4 B1~B8

1.2

1.0
Bar strain (ε/εy)

0.8
B6
0.6 B5
DR=0 .5%
B8
0.4 B4 DR=1 %
B3 DR=1 .5%
0.2 B2 DR=2 %
B1 DR=2 .5%

12, x FOR PEER REVIEW 0.0 15 of 21

-300 -200 -100 0 100 200
Distance from beam-joint interface (mm)

(a) LZ1

1.4 B1~B8

1.2

1.0
Bar strain (ε/εy)

B7
0.8
B6 B8
0.6
B5 DR=0.5%
0.4 DR=1%
B4 DR=1.5%
B2
0.2 B3 DR=2%
B1 DR=2.5%
0.0
-300 -200 -100 0 100 200
Distance from beam-joint interface (mm)
(b) LZ2

1.4 B1~B7

1.2

1.0
Bar strain (ε/εy )

0.8 B6
B7
0.6
B5
DR=0.5%
0.4 B4 DR=1%
B3 DR=1.5%
0.2 B2
DR=2%
B1 DR=2.5%
0.0
-300 -200 -100 0 100 200 300
Distance from beam-joint interface (mm)

(c) LZ3
Figure 11. Strain profilesFigure
along 11.
beam bottom
Strain bars.along beam bottom bars.
profiles

As can be seen, the strains of LZ1 and LZ3 could increase steadily along the embed-
ded length, whereas the strains at B1~B4 of LZ2 decreased after a drift ratio of 1.5%. This
implied that slip had taken place in the key regions of LZ2. This also explained why LZ2
exhibited poor seismic performance under positive loading.

3.5. Drift Contributions

 Buildings 2023, 13, 62 14 of 19

As can be seen, the strains of LZ1 and LZ3 could increase steadily along the embedded
length, whereas the strains at B1~B4 of LZ2 decreased after a drift ratio of 1.5%. This
implied that slip had taken place in the key regions of LZ2. This also explained why LZ2
, x FOR PEER REVIEW exhibited poor seismic performance under positive loading. 16 of 21

3.5. Drift Contributions

The LVDT instrumentation, as illustrated in Figure 12, was set to record the major
Remarkably,contributions
even at a drift
fromratio of only
different 1.5%, components
assembly the B-J interface rotation
to the beam contribution
overall lateral drifts. The
recorded 69% forfive
LZ2, which
drift was almost
contributions twice
were: (1) that ofelastic
the column LZ1 flexural
and LZ3. This further
deformation; (2) thecon-
joint shear
firmed a significant B-J interface
deformation; slipbeam
(3) the had occurred in LZ2. outside the beam hinge; (4) the beam
flexural deformation
flexural deformation within the beam hinge; and (5) the beam–joint (B-J) interface rotation.
100% 3% 2% 2% 2% 100% 1% 2% 2% 2% 1%
5% 2% 2% 2%
12% 11% 10% 8% 2% 2%
12% 11% 8% 4%
3% 13%
2% 3% 3% 15%
80% 80% 15%
31% 41%
Contribution to lateral

Contribution to lateral
32% 32% 34% 51% 50% 28% 31% 29%
Displacement (%)

Displacement (%)
56% 53%
60% Column 60% Column
Joint
Joint
Beam (hinge)
Beam (hinge) 25% 30% B-J Interface
40% 24% 25% 26% 39% B-J Interface 40% 26%
Beam (o/s hinge)
Beam (o/s hinge)
39%
55% 69%
30% 36% 40% 42%
20% 20%
30% 29% 27% 32% 29% 28%
20% 20%
13% 9% 10%
6% 5% 4%
0% 0%
0.25 0.5 0.75 1 1.5 2 2.5 3 3.5 0.25 0.5 0.75 1 1.5
Drift Ratio (%) Drift Ratio (%)

(a) LZ1 (b) LZ2

100% 2% 2% 2% 2% 2% 2% 2% 2% 2%
5% 4% 3% 2%
9% 6%
15% 13% 12%
80%
Contribution to lateral

49% 47%
Displacement (%)

40% 48%
36% 41% 38% 58% 57%
60%
Column
Joint
Beam (hinge)
40% 13% B-J Interface
15% 20% 28% Beam (o/s hinge)
29% 40% 45%
24% 28%
20%
34% 29%
27%
21%
14% 11% 9%
6% 5%
0%
0.25 0.5 0.75 1 1.5 2 2.5 3 3.5
Drift Ratio (%)

(c) LZ3

Figure 12. Components of12.

Figure contribution
Componentsto
of beam lateraltodisplacement.
contribution beam lateral displacement.

As can be seen, the drift contributions from the different assembly components of LZ1
4. Post-Test Check and Discussion
and LZ3 were overall similar. The contributions from the column and joint deformations
4.1. A Close-Up View ofminor
were LZ-2 and not relevant, which was consistent with the experimental observation that
The failure ofthe column and joint were in an intact state. The primary deformation contributions came
LZ-2 was unexpected. As an afterthought, it should be attributed to the
from the beam bending deformation and the rotations at the B-J interface. Note that the
use of the mechanical couplers
rotation in interface
at the B-J LZ-2. Inincluded
an attempt to elucidate
the slip the genesis
of reinforcement ofopening
and the the failure
of flexural
of LZ-2, the specimen
crackswas
at thebroken apart after testing to check the embedded reinforcing
interface.
bars and the threaded couplers. Figure 13 shows a close-up view of the disassembled LZ-
2. Apparently, the lengths of the bottom headed bars anchored into the couplers were
different, which eventually induced the bar pull-out failure, an exact cause of the signifi-
cant degradation of the seismic performance of LZ2 under positive loading. This unfavor-
Buildings 2023, 13, 62 15 of 19

Remarkably, even at a drift ratio of only 1.5%, the B-J interface rotation contribution
recorded 69% for LZ2, which was almost twice that of LZ1 and LZ3. This further confirmed
a significant B-J interface slip had occurred in LZ2.

4. Post-Test Check and Discussion

4.1. A Close-Up View of LZ-2
The failure of LZ-2 was unexpected. As an afterthought, it should be attributed to the
use of the mechanical couplers in LZ-2. In an attempt to elucidate the genesis of the failure
of LZ-2, the specimen was broken apart after testing to check the embedded reinforcing
bars and the threaded couplers. Figure 13 shows a close-up view of the disassembled
LZ-2. Apparently, the lengths of the bottom headed bars anchored into the couplers were
different, which eventually induced the bar pull-out failure, an exact cause of the significant
degradation of the seismic performance of LZ2 under positive loading. This unfavorable
Buildings 2023, 12, x FOR PEER REVIEW 17 of 21
pull-out failure was probably due to the steel congestion in the joint core, which prevented
the tightening of the headed bars, resulting in a “slop” between the couplers and the
headed bars.

Mechanical couplers
used in the beam

Headed bars in joint

Figure 13. Headed bars and mechanical couplers in LZ-2.

Figure 13. Headed bars and mechanical couplers in LZ-2.
4.2. Tensile Behavior of Different Bar Connectors
To further
4.2. Tensile examine
Behavior the tensile
of Different performance of different bar connectors, a test rig was
Bar Connectors
designed to carry out tensile experiments
To further examine the tensile performance on the
of threaded couplers
different bar and grouted
connectors, sleeves
a test rig was
after the cyclic tests, as shown in Figure 14. Two D22 steel deformed bars
designed to carry out tensile experiments on the threaded couplers and grouted sleeves (the same as the
beam
after thebottom bars) as
cyclic tests, were screwed
shown into the
in Figure coupler
14. Two D22or sleeve,
steel with their
deformed bars upper andas
(the same lower
the
ends bottom
beam clamped. Two
bars) LVDTs
were were into
screwed arranged on the or
the coupler leftsleeve,
and right
withsides
theirofupper
the connector
and lowerto
measure
ends the vertical
clamped. elongation
Two LVDTs were of the whole
arranged on assembly.
the left and The elongation
right sides of was then divided
the connector to
by the gauge length to obtain the tensile strain of each specimen.
measure the vertical elongation of the whole assembly. The elongation was then divided
by the gauge length to obtain the tensile strain of each specimen.
Buildings 2023, 12, x FOR PEER REVIEW 18 of 21

Buildings2023,
Buildings 2023,13,
12,62x FOR PEER REVIEW 18 of
16 of1921

(a) Mechanical coupler (b) Grouted sleeve

Figure 14. Tensile test setup for different bar connectors.

Figure(a)15 shows thecoupler

Mechanical tensile stress–strain curves obtained for different
(b) Grouted sleeve bar splicing
methods, along with that of the bare steel bars. In this figure, “MC1” represents that two
Figure
steel
Figure 14.Tensile
14.
bars joined test
Tensilein setup
the
test for
fordifferent
coupler
setup bar
were torqued
different barconnectors.
such that the bars were bearing against each
connectors.
other (i.e., a snug-tight condition), “MC2” means that one bar was fully tightened whereas
Figure 15 shows the tensile stress–strain curves obtained for different bar splicing
another was only
Figure screwed
15 shows theintensile
2/3 of stress–strain
the length it should
curves be, “MC3”for
obtained represents
differentthat
barone bar
splicing
methods, along with that of the bare steel bars. In this figure, “MC1” represents that two
was fully fastened
methods, whereas
along with that ofanother
the bare wassteel
secured
bars.inIn1/2 offigure,
this the length
“MC1”it should be. The
represents three
that two
steel bars joined in the coupler were torqued such that the bars were bearing against each
cases
steelwere set up in
bars joined to the
investigate the influence
coupler were torqued of suchthethat
tightness
the bars of were
the steel barsagainst
bearing in the me-
each
other (i.e., a snug-tight condition), “MC2” means that one bar was fully tightened whereas
chanical coupler
other (i.e., on the overall
a snug-tight tensile
condition), “MC2” stress–strain
means that behavior.
one bar was fully tightened whereas
another was only screwed in 2/3 of the length it should be, “MC3” represents that one
Fromwas
another the only
results of Figure
screwed in 15,ofit the
2/3 canlength
be seen it that
shouldthebe,tensile
“MC3”stress–strain
represents curve
thatbe.of the
one bar
bar was fully fastened whereas another was secured in 1/2 of the length it should The
grouted
was fullysleeve was almost coincident with that of the bare steel bars. This explained why
three casesfastened
were setwhereas another was
up to investigate the secured
influence inof
1/2the
of tightness
the lengthofit the
should
steelbe. The
bars inthree
the
the seismic
cases were performance
set up to of LZ-3 was
investigate the close to that
influence of of
theLZ-1.
tightness of the steel bars in the me-
mechanical coupler on the overall tensile stress–strain behavior.
chanical coupler on the overall tensile stress–strain behavior.
600From the results of Figure 15, it can be seen that the tensile stress–strain curve of the
grouted sleeve was almost coincident with that of the bare steel bars. This explained why
the500
seismic performance of LZ-3 was close to that of LZ-1.

600
400
σ (MPa)

500
300 Steel bar
Grouted sleeve
MC1
400
200
MC2
σ (MPa)

MC3
300
100 Steel bar
Grouted sleeve
MC1
200
0
0 10000 20000 30000 MC2
40000 50000
MC3
με
100
Figure 15. Tensile
Tensile behavior of different bar splicing methods.
0
From 0 the results
10000 of 20000
Figure 15, 30000 40000
it can be 50000
seen that the tensile stress–strain curve of the
με
grouted sleeve was almost coincident with that of the bare steel bars. This explained why
the seismic performance of LZ-3 was close to that
Figure 15. Tensile behavior of different bar splicing of LZ-1.
methods.
On the other hand, with the reduction of the length of steel bars in the coupler, the
resulting tensile performance appeared drastically degraded. In the cases of “MC2” and
Buildings 2023, 13, 62 17 of 19

“MC3”, the maximum tensile loads were not even half of the yield strength of the steel
bar. This fully demonstrated the high sensitivity of the couplers to the degree of bar
tightness. Therefore, the mechanical couplers should be used cautiously in areas with
dense reinforcement. In comparison, with regard to the new semi-precast construction
method discussed here, using half-grouted sleeves is more recommended for connecting
beam bottom bars and joint headed reinforcement.
It is noteworthy that in French et al.’s study [14], it was also found that, in a precast
connection specimen using ordinary threaded couplers, large slippages were captured
between the reinforcement and the couplers. This phenomenon was confirmed in the
current study. To eliminate such kinds of slip, taper-threaded splices were recommended
by French et al. [14]. Specifically, threads were mechanically cut on the rebar to eliminate
“slop” between the bar and the splice. As a result, the bar slippage was eliminated, and
the threaded rebar connections with taper-threaded splices exhibited favorable seismic
performance. This solution appeared to be more reliable than using ordinary rebar couplers,
holding promise for use in regions of moderate to high seismicity [14].

5. Conclusions
In a new construction method for semi-precast industrial facilities, the steel bars in
precast beams are required not to protrude from the beam ends to facilitate the temporary
fixation of robotic cranes. This leaves engineers with the problem of choosing the form of
connection between beam bottom bars and headed reinforcement in joints. To address this
question, three large-scale exterior beam–column connections were fabricated and tested.
A detailed account of the cyclic behaviors of the specimens is presented in this paper. From
the obtained results the following conclusions can be drawn:
(1) The precast connection using half-grouted sleeves showed ductile seismic perfor-
mance, being almost as good as its cast-in-place counterpart. A drift ratio of 3.5% could
be attained, and the specimen exhibited beam hinging failure without significant beam
bar slip. Therefore, the use of the combination of headed bars and half-grouted sleeves
is recommended by the writers to facilitate the mechanical anchorage of the bottom steel
reinforcement in the precast beam;
(2) The failure of the precast connection using mechanical couplers was not expected.
A wide beam–joint interface crack was observed. Furthermore, the specimen showed
sudden load shedding in the positive loading direction (i.e., when the beam upward
loading was applied), resulting in substantial reductions in the post-peak strength and
energy dissipation capacities. Despite being easy to install, the use of the ordinary threaded
couplers cannot emulate the cyclic behavior of the cast-in-place connections. Consequently,
the use of these couplers should be restricted for the new construction method, and this is
especially true in high-intensity seismic regions;
(3) As revealed by this study, as loads were cycled, the reinforcement was worked
back and forth in the ordinary couplers such that the bars were no longer in bearing, thus
inducing tensile slippage of the reinforcement relative to the coupler. When the steel bars
are not fully tightened, the tensile strength of the coupler connection may be reduced by
half. According to previous studies, the threaded rebar connection with taper-threaded
splices is easy to manufacture and may be a better alternative to ordinary couplers. In
the future, it is necessary to conduct more in-depth research on characterizing the tensile
performance of ordinary threaded couplers under large inelastic cyclic deformations.

Author Contributions: Conceptualization, M.J., W.C. and X.Z.; data curation, S.Z. and X.Z.; formal
analysis, M.J., S.Z. and X.Z.; funding acquisition, W.C. and X.Z.; investigation, M.J., W.C., S.Z. and
X.Z.; methodology, M.J., S.Z. and X.Z.; project administration, W.C. and X.Z.; resources, M.J. and X.Z.;
software, S.Z. and X.Z.; supervision, M.J. and X.Z.; validation, S.Z. and X.Z.; visualization, S.Z. and
X.Z.; writing—original draft, X.Z.; writing—review & editing, M.J. All authors have read and agreed
to the published version of the manuscript.
Buildings 2023, 13, 62 18 of 19

Funding: This research was funded by Guangdong Provincial Key Laboratory of Modern Civil
Engineering Technology (grant number 2021B1212040003) and Science and Technology Program of
Guangzhou Municipal Construction Group Co., Ltd. (grant number [2019]-KJ006 and [2021]-KJ035).
Data Availability Statement: Data sharing is not applicable to this article.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Buyle, M.; Braet, J.; Audenaert, A. Life cycle assessment in the construction sector: A review. Renew. Sustain. Energy Rev. 2013, 26,
379–388. [CrossRef]
2. Huang, L.; Krigsvoll, G.; Johansen, F.; Liu, Y.; Zhang, X. Carbon emission of global construction sector. Renew. Sustain. Energy Rev.
2018, 81, 1906–1916. [CrossRef]
3. Norouzi, M.; Chàfer, M.; Cabeza, L.F.; Jiménez, L.; Boer, D. Circular economy in the building and construction sector: A scientific
evolution analysis. J. Build. Eng. 2021, 44, 102704. [CrossRef]
4. Lima, L.; Trindade, E.; Alencar, L.; Alencar, M.; Silva, L. Sustainability in the construction industry: A systematic review of the
literature. J. Clean. Prod. 2021, 289, 125730. [CrossRef]
5. Park, R. A perspective on the seismic design of precast concrete structures in New Zealand. PCI J. 1995, 40, 40–60. [CrossRef]
6. Elliott, K.S. Precast Concrete Structures; CRC Press: Boca Raton, FL, USA, 2016.
7. Wu, P.; Low, S.P. Lean management and low carbon emissions in precast concrete factories in Singapore. J. Archit. Eng.-ASCE
2012, 18, 176–186. [CrossRef]
8. Belleri, A.; Brunesi, E.; Nascimbene, R.; Pagani, M.; Riva, P. Seismic performance of precast industrial facilities following major
earthquakes in the Italian territory. J. Perform. Constr. Fac.-ASCE 2015, 29, 04014135. [CrossRef]
9. Belleri, A.; Torquati, M.; Marini, A.; Riva, P. Horizontal cladding panels: In-plane seismic performance in precast concrete
buildings. Bull. Earthq. Eng. 2016, 14, 1103–1129. [CrossRef]
10. Cao, X.Y.; Shen, D.J.; Feng, D.C.; Wang, C.L.; Qu, Z.; Wu, G. Seismic retrofitting of existing frame buildings through externally
attached sub-structures: State of the art review and future perspectives. J. Build. Eng. 2022, 57, 104904. [CrossRef]
11. Breccolotti, M.; Gentile, S.; Tommasini, M.; Materazzi, A.L.; Bonfigli, M.F.; Pasqualini, B.; Colone, V.; Gianesini, M. Beam-column
joints in continuous RC frames: Comparison between cast-in-situ and precast solutions. Eng. Struct. 2016, 127, 129–144. [CrossRef]
12. Englekirk, R.E. Seismic Design of Reinforced and Precast Concrete Buildings; Wiley: New York, NY, USA, 2003.
13. Englekirk, R.E. Development and testing of a ductile connector for assembling precast concrete beams and columns. PCI J. 1995,
40, 36–51. [CrossRef]
14. French, C.W.; Hafner, M.; Jayashankar, V. Connections between precast elements—Failure within connection region. J. Struct.
Eng.-ASCE 1989, 115, 3171–3192. [CrossRef]
15. Soubra, K.S.; Wight, J.K.; Naaman, A.E. Cyclic response of fibrous cast-in-place connections in precast beam-column subassem-
blages. ACI Struct. J. 1993, 90, 316–323.
16. Choi, H.K.; Choi, Y.C.; Choi, C.S. Development and testing of precast concrete beam-to-column connections. Eng. Struct. 2013, 56,
1820–1835. [CrossRef]
17. Dal Lago, B.; Negro, P.; Dal Lago, A. Seismic design and performance of dry-assembled precast structures with adaptable joints.
Soil. Dyn. Earthq. Eng. 2018, 106, 182–195. [CrossRef]
18. Senturk, M.; Pul, S.; Ilki, A.; Hajirasouliha, I. Development of a monolithic-like precast beam-column moment connection:
Experimental and analytical investigation. Eng. Struct. 2020, 205, 110057. [CrossRef]
19. Correal, J.F.; Bernal, S.A.; Hassan, W.M. Seismic performance of full-scale moment connections of one-story precast industrial
facilities. ACI Struct. J. 2021, 118, 179–192.
20. Nzabonimpa, J.D.; Hong, W.K.; Kim, J. Nonlinear finite element model for the novel mechanical beam-column joints of precast
concrete-based frames. Comput. Struct. 2017, 189, 31–48. [CrossRef]
21. Ghayeb, H.H.; Razak, H.A.; Sulong, N.R. Seismic performance of innovative hybrid precast reinforced concrete beam-to-column
connections. Eng. Struct. 2020, 202, 109886. [CrossRef]
22. Li, Z.H.; Qi, Y.H.; Teng, J. Experimental investigation of prefabricated beam-to-column steel joints for precast concrete structures
under cyclic loading. Eng. Struct. 2020, 209, 110217. [CrossRef]
23. Liu, J.; Xue, Y.T.; Wang, C.K.; Nie, J.G.; Wu, Z.H. Experimental investigation on seismic performance of mechanical joints with
bolted flange plate for precast concrete column. Eng. Struct. 2020, 216, 110729. [CrossRef]
24. Ye, M.; Jiang, J.; Chen, H.M.; Zhou, H.Y.; Song, D.D. Seismic behavior of an innovative hybrid beam-column connection for
precast concrete structures. Eng. Struct. 2021, 227, 111436. [CrossRef]
25. Master Computers 4.0: Production Control Systems in the Technological Transition. Available online: https://www.bft-
international.com/en/artikel/bft_Master_computers_4.0_production_control_systems_in_the_technological_2650315.html (ac-
cessed on 5 November 2022).
26. Reichenbach, S.; Kromoser, B. State of practice of automation in precast concrete production. J. Build. Eng. 2021, 43, 102527.
[CrossRef]
27. GB50010-2010; Code for Design of Concrete Structures. China Architecture and Building Press: Beijing, China, 2010.
Buildings 2023, 13, 62 19 of 19

28. Wallace, J.W.; McConnell, S.W.; Gupta, P.; Cote, P.A. Use of headed reinforcement in beam-column joints subjected to earthquake
loads. ACI Struct. J. 1998, 95, 590–606.
29. Lee, H.J.; Yu, S.Y. Cyclic response of exterior beam-column joints with different anchorage methods. ACI Struct. J. 2009, 106,
329–339.
30. GB50011-2016; Code for Seismic Design of Buildings. China Architecture and Building Press: Beijing, China, 2016.
31. ACI Committee 318. Building Code Requirements for Structural Concrete (ACI 318-19); American Concrete Institute: Farmington
Hills, MI, USA, 2019.
32. ACI Committee 374. Acceptance Criteria for Moment Frames Based on Structural Testing and Commentary (ACI 374.1-05) (Reapproved
2019); American Concrete Institute: Farmington Hills, MI, USA, 2005.
33. Naito, C.J.; Moehle, J.P.; Mosalam, K.M. Evaluation of bridge beam-column joints under simulated seismic loading. ACI Struct. J.
2002, 99, 62–71.
34. Ehsani, M.R.; Wight, J.K. Exterior reinforced concrete beam-to-column connections subjected to earthquake-type loading. ACI
Struct. J. 1985, 82, 492–499.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.