Journal of Cleaner Production 205 (2018) 552e564

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

A meta-network-based risk evaluation and control method for

industrialized building construction projects
Tao Wang a, Shangde Gao a, Xiaodong Li b, *, Xin Ning c
a
Department of Engineering Management, School of Management Science and Engineering, Central University of Finance and Economics, Tupei Building
403, South College Road 39, Haidian District, Beijing 100081, China
b
Department of Construction Management, School of Civil Engineering, Tsinghua University, Beijing 100084, China
c
School of Investment & Construction Management, Dongbei University of Finance and Economics, Dalian, Liaoning Province 116025, China

a r t i c l e i n f o a b s t r a c t

Article history: Construction methods in developing countries such as China are gradually industrializing due to rising
Received 1 January 2018 labor costs, new techniques, tools, procedures and management methods. This makes it essential to
Received in revised form develop new risk evaluation and control methods. Previous research seems to ignore the risk manage-
3 August 2018
ment of the application of new construction technologies and procedures. The identification of risk
Accepted 15 September 2018
Available online 17 September 2018
mechanisms is unclear, leading to a lack of clear guidance for risk avoidance and control. This paper
describes a network model based on meta-network analysis of project objectives, risk events, risk factors
and stakeholders in the construction process of building industrialization. According to ISO 31000, the
Keywords:
Building industrialization
main processes of risk management in the model has three parts: risk identification, risk analysis and
Risk identification evaluation, and risk treatment and control. Risk factors are identified from previous literature and site
Risk evaluation investigation, and the indirect impact on project objectives is analyzed and calculated with the networks
Risk control in the meta-network. The order of importance is evaluated and used as the foundation of risk treatment
Meta-network analysis and control. The analysis of the crucial risk factors makes it possible to identify the stakeholders who
Project stakeholders influence them, leading to suggestions for relevant control strategies. A residential building construction
project in South China that uses the building industrialization construction system serves as a case study
to verify the feasibility and applicability of the risk management system. Results show that critical risk
factors are construction-related factors and design-related factors. The water and electricity engineers,
the project manager, the project secretary, the field engineer, the project supervisor and the main
contractor engineer have a significant influence on the risk factors. A stakeholder supervision system
with targeted risk control strategies is proposed.
© 2018 Published by Elsevier Ltd.

1. Introduction advantage of prefabrication is that better supervision on improving

Building industrialization has had various definitions and con-
cepts in each period along with increasing demands and economic
necessities. It is the integration of a wide range of new concepts and
techniques, including e but not limited to e automation, robotics,
reproduction, preassembly, standardization, mechanization, pre-
fabrication and off-site construction (Zabihi, 2013). Industrialized
construction has had a significant influence on improving effi-
ciency, quality and the environmental impact of projects (Aye et al.,
2012; Gibb, 1999; Wong and Yeh, 1985). The most significant
* Corresponding author.
E-mail address: eastdawn@tsinghua.edu.cn (X. Li).
the quality of prefabricated products can be conducted (Tam et al.,
2007). The movement of the construction industry toward indus-
trialization in the fabrication system includes the trend of helping
the building industry achieve benefits and share them with the
end-user (Martinez et al., 2008). This trend gains considerable
attention and support from the government in China (Ji et al., 2017).
In the transformation process of building industrialization, new
concepts and techniques create greater requirements for project
management for several reasons. First, compared to the traditional
construction process, construction workers typically need to learn
new techniques and become more familiar with the process of
prefabricated building components assembly. Second, pre-
fabricated components and new construction equipment (e.g., self-
climb form, which is an intelligent wall formwork) greatly increase

https://doi.org/10.1016/j.jclepro.2018.09.127
0959-6526/© 2018 Published by Elsevier Ltd.
 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564
the construction speed and enhance the flexibility of project
planning (Jaillon and Poon, 2009). New management modes, such
as interspersed construction e a new construction planning tech-
nology (Feng and Hu, 2016) e, have been applied to project man-
agement and have had significant influences on project duration
and cost. Building industrialization brings positive changes in the
construction process (Lovell and Smith, 2010) as well as complexity
and uncertainty to project implementation. Industrialized con-
struction projects are generally subjected to more risks compared
to traditional projects (Karimiazari et al., 2011; Luo et al., 2015).
Risk factors (e.g., the seam material is not securely fastened)
increase with the new construction system of building industrial-
ization (Luo et al., 2015), and they are more complex as more re-
lationships exist among project objectives, risk events and risk
factors. Some risk events can even lead to the failure of realization
of project objectives, and risk factors can lead to risk events, so
more strict requirements for risk management are necessary. Thus
the relationships among project objectives, risk events and risk
factors need to be considered (Bu-Qammaz et al., 2009; Fang and
Marle, 2015; Fang et al., 2012; Guan and Guo, 2014; Luo et al.,
2015; Tavakolan et al., 2017). However, there are still some limita-
tions that need to be addressed. First, effective risk control methods
are necessary for the successful risk management of industrialized
building construction projects, while the risk management and
control skills of managers of different levels (project managers,
field engineers, supervision engineers) may fail to meet the re-
quirements of a new construction management system (Liu, 2017).
But very limited studies focus on proposing targeted risk control
methods based on how the risk factor lead to the deviation of
project objectives and affect each other (Kuo and Lu, 2013; Wang
et al., 2016). Second, previous project risk management has
analyzed and assessed risks from the perspective of the whole
project (e.g. economics risks, market risks, risks from the external
environment) (Luo et al., 2015; Xiahou et al., 2018), and there has
been little attention focused on the potential risks associated with
the implementation of building industrialization, which are listed
in the case study of this paper. Third, a substantial proportion of
researchers in this field have analyzed only the direct influences of
risks (Luo et al., 2015), while overlooking the interaction between
one risk event and other risk events (Han et al., 2008) (e.g. the
interaction between “design changing” and “unreasonable con-
struction planning”). Assumptions for some research methods of
risk management limit certain kinds of associations. For example,
the analytic hierarchy process (AHP) assumes that risk factors are
isolated, which is the prerequisite of the comparison for the factors'
weight judgment (Saaty, 2004). In the Bayesian network analysis
process, closed loops among risk factors and events are not
considered (Borsuk et al., 2004).
In order to bridge the knowledge gaps of risk management and
control of industrialized building construction projects, our
research was based on meta-network analysis (MNA); thus, it is
included a network analysis model composed of stakeholders, risk
events, risk factors and project objectives. The meta-network was a
complex network composed of various entities and connections
among them (Li et al., 2015). MNA can conceptualize a project as
multi-node with multi-link meta-networks composed of different
node entities and their interdependencies (Zhu and Mostafavi,
2015). MNA has been applied in many fields, such as the social
networks and supply chains in the e-commerce market
(Wakolbinger and Nagurney, 2004), the integration of social net-
works with knowledge networks (Nagurney and Dong, 2005), a
social media analysis (Carley et al., 2013a) and an assessment of
public health systems (Lenz, 2012). The goal of our research was to
systematically study the risk assessment and responding mecha-
nisms for the deviation from project objectives in the application of
553
building industrialization. In this model, the association of risk
events and factors that led to the deviation of project objectives was
revealed, and targeted risk control strategies e based on the anal-
ysis of the relationship between stakeholders and risk factors e are
proposed.
The structure of this paper is as follows: In the next section,
previous studies of risk management in building industrialization
projects are reviewed, and the research gap is analyzed. In section
3, the establishment of the meta-network and our calculation
methods are described in detail. To show how the risk evaluation
and control model can be applied, a case study of a residential
building construction project in South China, which applied a
building industrialization construction system, is discussed in
section 4. Related results and future work are described in section 5.
2. Literature review
2.1. Building industrialization and its influence on construction
projects
Previous research has shown that building industrialization has
several notable characteristics. The first is standardization e
including in the design phase, component production, actual con-
struction and management. Standardization has a significant in-
fluence on the improvement of the quality of construction projects
in the existing building system (Roy et al., 2005). Standardization
and generalization of building components leads to the second
characteristic: prefabrication in factories (Jaillon and Poon, 2009).
Prefabrication refers to the mass production of building compo-
nents in factories; these components are then delivered to a project
site to be used in construction. The third characteristic is scientific
management (Yan et al., 2004). Overall planning and new tech-
nology are the bases of scientific management in industrial con-
struction. Building industrialization has been widely adopted
around the world (Pan and Sidwell, 2011). According to Mao et al.
(2013), in 1996, the levels of prefabrication in the construction in-
dustry in Germany, the Netherlands and Denmark already were
31%, 40% and 43%, and the size of the industrialized construction
industry in the U.K. was £6 billion in 2006. Compared to these
developed countries, building industrialization in China is still in its
infancy (Luo et al., 2015), and developers in China lack the expe-
rience of industrialized construction approaches (Zhang and
Skitmore, 2012).
Building industrialization brings changes to the construction
process and affects the achievement of project objectives. Tech-
nology associated with building industrialization e especially the
use of prefabricated building components, such as prefabricated
slabs, beams, precast concrete (PC) roofs, PC balconies and inte-
grated bathrooms e has been gradually introduced in the resi-
dential building construction industry in China (Gan et al., 2017).
This has had several effects on the construction process, including
compressing construction duration (Goodier and Gibb, 2007),
increasing building costs (Chiang et al., 2006; Lihong et al., 2013;
Mao et al., 2013) and improving building quality (Gan et al.,
2017). Assembled building provides an example to illustrate the
changes and effects brought by building industrialization. In
traditional construction processes, waste arises from design
changes, design error, materials remaining, packaging and non-
recyclable consumables, and inclement weather (Faniran and
Caban, 1998). Xu and Zhao (2010) proposed that assembled build-
ing could significantly reduce the disadvantages of traditional
building processes while decreasing construction time, costs and
environmental impact yet increasing quality. However, at this time,
building industrialization is more costly than the conventional
construction approach. By comparing assembled building projects
554 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564
and cast-in-place projects in a case study, Lihong et al. (2013)
discovered that the cost of the assembly construction process was
12.26% higher than it was for traditional projects. The main reason
was related to the prefabricated components and installation, for
there it was costlier to purchase PC components, and the salaries of
workers who had experience with installation were higher than
those without this specialized experience. In terms of construction
safety, Rubio-Romero et al. (2014) concluded that depending on the
project type, the safety of a prefabricated building might not be as
good for a traditional building. In regard to environmental impact,
by employing a life-cycle assessment (LCA), Hong et al. (2016)
determined that building industrialization (assembled buildings)
could reduce 4%e14% of the total life-cycle energy consumption,
which covers the whole life cycle of the prefabricated components,
including prefabrication manufacturing, transportation, on-site
assembling and recycling in the demolition phase. With a similar
method, Pons and Wadel (2011) analyzed a school project that
adopted building industrialization in Catalonia, and the project
made outstanding achievements in energy conservation and
emission reduction. Overall, compared to traditional projects,
building industrialization has improved efficiency and effective-
ness, but it is still much more difficult for some projects to achieve
their objectives (e.g. objectives about project cost, duration, quality,
safety and environmental impact). A life-cycle assessment suggests
that it is possible for the total cost to be lower, but in actual con-
struction processes, the costs are much higher because of the use of
precast concrete components and the lack of experienced pro-
fessionals (Lihong et al., 2013).
2.2. Risk management methods of construction projects
Changes in the achievement process of the project objectives
leads to changes in the risk system. Therefore, successful risk
management is critical. Previous research has described many risk
management methods. Elsawah et al. (2016) adopted a risk matrix
combining probability and influence from expert judgment. Zhang
and Zou (2007) integrated fuzzy logic and the analytic hierarchy
process in risk evaluation of critical infrastructure construction.
Abdelgawad and Fayek (2011) combined fuzzy logic and fault tree
analysis to reveal the sources of risks. Lin et al. (2015) built a multi-
hierarchy risk analysis framework with risk matrices and applied it
to risk examination and evaluation. Zhou and Zhang (2011) applied
Fuzzy Bayesian Analysis (FBA) with experts' experience and a
related construction database to deep basement projects and found
that FBA had a significant influence on risk avoidance.
Stergiopoulos et al. (2016) found that fuzzy logic and risk depen-
dence charts could be used to effectively analyze the interdepen-
dence between different projects, and critical factors could be
found by simulations of specific scenarios.
To study building industrialization, Bari et al. (2012) analyzed
and evaluated the types of critical factors in the construction pro-
cess of building industrialization projects, including project factors
(e.g. repeated use of design, molds or construction techniques from
previous projects), economic factors and contractor factors (e.g. a
contractor's staff and workers). Based on the related literature, Gan
et al. (2017) proposed that in China, there were three kinds of risk
factors related to project quality: design-related factors,
production-related factors, and construction-related factors. Based
on a Strengths, Weaknesses, Opportunities and Threats (SWOT)
analysis, Jiang et al. (2018) analyzed 107 scientific papers and 85
government documents and determined the opportunities and
challenges of building industrialization in the new urbanization in
China. Jiang mentioned that the main challenges at this stage
included high upfront costs, low collaboration between experts and
project participants, incomplete regulations and standards, and low
market acceptance. Li et al. (2017) applied a system dynamics
model to analyze the investment risk management of assembled
buildings in China and classified the risk factors into five categories:
economic risks, internal risks, technical risks, policy and legal risks,
and market risks. They proposed related risk control strategies.
From this brief literature review, we can observe that previous
research has focused on total market risks, and there has been a
lack of identification of the risk factors inherent in adopting the
industrialized construction approach in the building construction
process. Quantitative analysis of the interactions between risks has
been insufficient, and risk response strategies after the risk
assessment have not been given enough attention.
We identified the following problems from our consideration of
previous studies on the industrialization of buildings and the risk
management studies of construction projects: (1) the level of
achievement of project objectives in building industrialization
projects was different from traditional projects, and the risk system
should be changed; (2) most previous project risk research
considered each risk as an independent event, and the correlation
between risks has received limited attention; and (3) risk man-
agement research has usually stopped at the risk evaluation phrase
with little attention given to risk response strategies.
2.3. Meta-network analysis
Notably, meta-network analysis (MNA) e which can conceptu-
alize construction projects as multi-node and multi-link meta-
networks composed of different node entities and their in-
terdependencies e can be applied to analyze the risk association
and to propose risk control strategies (Zhu and Mostafavi, 2015).
The structure of MNA includes the identification of entities and
associations between these entities, and it proposes an analysis
based on these associations and entities. The advantages of this
method include the following: (1) risks are regarded as the re-
lationships between different entities, and the MNA, which used
the network in the analysis, is more intuitive than previous studies;
(2) the mechanism of risk control by stakeholders' direct inter-
vention in regard to the risk factors is revealed; (3) MNA provides
an effective way to express the complex interactions of various
factors involved in the project and extends the scope of analysis to
multiple dimensions and forms an integrated project network that
covers the relationships between the various influencing factors;
and (4) MNA provides multi-dimensional perspectives and the
ability to construct complex models that can accurately measure
and diagnose the performance of project goals and tasks, helping to
optimize task scheduling and distribution (Zhu and Mostafavi,
2016).
MNA has been applied in some research. Li et al. (2015)
compared construction projects of 11 auto dealers and built a
meta-network with entities that included stakeholders, informa-
tion and tasks. Zhu and Mostafavi (2015) put forward a compre-
hensive framework of organizational vulnerability assessment for
complex construction projects to analyze the problem of cost
management. In these studies, each project was broken down into
different entities, such as stakeholders, information, resources and
tasks, and the meta-network was formed with links between these
entities. By analyzing the relationship between various types of
project factors, MNA can conduct risk analysis and assessment
under the condition of uncertain factors, so it is suitable for the risk
management of these projects.
In the MNA model proposed in this paper, the elements e the
nodes of the meta-network e were divided into four categories:
project objectives, risk events, risk factors and stakeholders. The
standards for project success included the objectives of duration,
cost, quality, influence on the environment and safety (health)
 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564
(Kerzner, 1989). These standards formed the category of “project
objective.” The failure to achieve an objective and the deviation
from the project objectives were defined as project risks (Purdy,
2010), which are described below. Risk events were the direct
reflection of a loss in a project and had a direct impact on project
objectives. Risk factors referred to factors that lead to a risk event.
Risk factors may be related to individual actions or come from
various broader conditions (including societal conditions, legisla-
tion, politics, economics and nature). Stakeholders were the project
participants who had a direct or indirect influence on the project.
We introduced these elements to the meta-network because (1) the
purpose of risk management is to reduce the deviation from the
objectives in the implementation process of the project, while the
deviation is reflected as the consequence of the occurrence of risk
events; (2) to conduct risk control, the risk factors which lead to the
occurrence of risk events need to be discovered by analyzing how
risk events happen; and (3) operative risk control strategies should
be proposed from the responsibility assigned to the stakeholders.
The process of MNA is described in detail in the following sections.
3. Risk system identification and MNA in building
industrialization
Fig. 1 shows the process of MNA for risk management. The first
step is risk identification and collection of risk levels. The second
step is the establishment of a meta-network, including trans-
forming entities into nodes and weighting the links between the
nodes. Stakeholders, risk factors, risk events and project objectives
are transformed into nodes and combined with weighted links to
form several networks, which are integrated into a meta-network.
The final step is risk analysis and calculation, including risk eval-
uation and proposing risk control strategies. Risk levels of the risk
factors are transferred to the hierarchy of objectives through the
meta-network, and relationships between stakeholders and
Fig. 1. The process of a meta-network analysis (MNA).
555
different risk factors are identified through the networks. Major
factors are evaluated based on the model's networks between risk
factors and project objectives. The strategy for risk response is
based on the analysis of the networks that relate stakeholders and
risk factors, and specific measures are designed for specific stake-
holders. The results can provide an objective and comprehensive
reference for the risk management of building industrialization
projects.
3.1. Risk identification
Because the new risk system for building industrialization is
different from the system for traditional projects, the risk events
and risk factors should be reidentified, and a new risk library for
building industrialization construction needs to be established. The
unique risks for industrialization construction processes include
the risks arising from the introduction of new construction tech-
nology, building materials, construction equipment and manage-
ment methods (Li-zi et al., 2015). Risk identification for the process
of introducing building industrialization construction methods in-
volves identifying the potential risks and creating a risk database
that can be used to construct the meta-network. There are two
sources for the risk library: the summary of previous research on
building industrialization and interviews with professionals and
project managers who have related experience. Analysis of the risk
management methods and conclusions in previous studies helps to
identify potential risks, and interviews of the professionals and
project managers help to revise the risk list to yield the risk library.
The risk library was the basis of the following analysis.
3.2. Establishment of a meta-network based on risk identification
Based on the risk identification, connections can be made be-
tween the identified project objectives, risk events and risk factors
556 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564

to form a risk evaluation system with three categories. There are
interactions among these entities. The occurrence of one risk event
may result in a chain reaction leading to a series of other risk
events, and there also may be interactions among risk factors. Risk
factors may not directly affect project objectives, but they may have
an indirect influence due to the occurrence of risk events. The
dividing line between risk factors and risk events is not strict. The
main basis for the division is related to how directly they affect the
project objectives. Based on the above relationships among project
objectives, risk events and risk factors, the risk level of risk factors
will ultimately influence the achievement of objectives through
risk events.
To achieve targeted risk control, nodes and related networks of
stakeholders are added to the meta-network. Stakeholders are the
main participants in the project, and they include project man-
agers, developers, designers, construction workers and suppliers. In
this research, the responsibilities of stakeholders for specific risk
factors are analyzed, and stakeholders are linked to the risk factors
with logic connections. If one risk factor has to be controlled,
related stakeholders need to be found for targeting responses. The
nodes of stakeholders are the ends of the risk control system. Fig. 2
shows the relationships between the four kinds of nodes.
As displayed in Fig. 2, the four types of nodes can connect the
following seven networks: (1) the project objective network formed
by the association between the project objective nodes, (2) the risk
impact network formed by the association between the project
objective nodes and the risk event nodes, (3) the risk event network
formed by the association between risk event nodes, (4) the risk
decomposition network formed by the association between risk
event nodes and risk factor nodes, (5) the risk factor network formed
by the association between risk factor nodes, (6) the social network
formed by the association between stakeholder nodes and (7) the
risk distribution network formed by the association between risk
factor nodes and stakeholder nodes.
3.3. Calculation processes in a meta-network analysis
To quantify the meta-network, risk levels of risk factors and
weights of links in the networks need to be assigned values. The
network conveys the level of risk through the connections between
the nodes, where the sources of risk level are the risk factors
because they are the smallest units that contain the risk content.
This is because the risk factors lead to the occurrence of risk events
and the deviation of project objectives. The connections between
nodes for the level of risk transfer efficiency may be different; their
practical significance is that the influence from risk factors to risk
events or from risk events to objectives is different, which needs to
be judged. The following algorithm is proposed for the calculations:
First, from the experts' judgment, we assigned values of risk
level to different risk factors, leading to the vector risk factor:
risk factor ¼ ðf1 ; f2 ; …; fn Þ (1)
The vectors risk events and project objective were established
without values:
risk event ¼ ðe1 ; e2 ; …; en Þ (2)
project objective ¼ ðo1 ; o2 ; …; on Þ (3)
Next, the links between the six networks that formed the meta-
network were assigned values that represented the different
transfer efficiency of risk level from the risk factors. From a math-
ematical perspective, networks can be regarded as matrices, and
the elements in the matrices represent the weight of the links. The
relationships among vectors in the networks are shown below:
risk event ¼ risk factor  risk factor network
 risk decomposition network  risk event network
(4)
project objective ¼ risk event  risk impact network
 project objective network (5)
By assigning values to risk levels for risk factors and transferring
the risk levels through the networks, the risk level of the project

Fig. 2. Relationships among different nodes in the meta-network.

 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564
objectives can be obtained. For risk control, the risk distribution
network and social network were established after introducing
stakeholders to the meta-network. Unlike the risk evaluation sys-
tem, links in these networks were not weighted. Because they only
represented the interactions between stakeholders and the re-
lationships of stakeholders to risk factors, their influence was not
relevant. The risk factor network also was considered to express the
association among risk factors. The actual influence between
stakeholders and risk factors was obtained from the nodes of the
stakeholders:
0
risk distribution network ¼ social network
 risk distribution network
 risk factor network (6)
These calculations were used as a comparison of the correlations
between stakeholders and risk factors, and critical stakeholders for
one specific risk factor were identified. Meanwhile, in the meta-
network, the changes of nodes and links led to perturbations of
the project objectives, and changes in the nodes or links and related
vectors and networks changed accordingly. Therefore, with an
effective risk control measurement, controlling some risk factors or
the links related to these factors will ultimately reduce the risk level
of the project objectives.
4. Case analysis
To verify the feasibility and applicability of the meta-network
analysis of risk evaluation and to provide direction for risk con-
trol, a residential building construction project in South China that
applied building industrialization was analyzed. The developer of
this project is a famous Chinese residential real estate developer
who aims to improve construction efficiency and quality by intro-
ducing industrialized construction approaches. Therefore, our
research focused on the risk evaluation and control in the appli-
cation of the new construction system of building industrialization.
The basic information for this developer and the project is shown in
Table 1.
The process of applying the proposed risk evaluation and con-
trol method included the following steps: (1) formulate the specific
risk factor list based on the literature related to building industri-
alization and interviews with project managers and determine the
relationship between these risk factors and the new construction
system; (2) establish hierarchies of project objectives, risk events,
risk factors and stakeholders and form the framework of the meta-
network based on the results of interviews and the scientific
literature; (3) conduct a survey of project managers and collect the
judgment of risk level and weight of links from the experts and
then assign values to the meta-network; and (4) carry out risk
evaluation based on the calculation methods and propose a risk
control strategy.
4.1. Risk identification and establishment of the meta-network
In our research, the risk identification and development of the
Table 1
Basic information of the project and the developer in this case.
Type of company One of the top five real estate developers in China
Structure type Reinforcement concrete shear wall
Floor number 28 with two underground floors
Greening ratio 60%
Total project investment
557
risk factors list were mainly based on interviews with project
managers and related information of the new construction system
of building industrialization in that project. Interviews were con-
ducted with the project manager, the project assistant manager, the
field engineers, the water and electricity engineers and the project
supervisor. Because the project was a normal residential building
project, our research focused on the risk evaluation and control in
the application of the new construction system of building indus-
trialization. In this case, the project objectives were similar to
traditional projects, and the risk events were related to the appli-
cation of construction technologies of the new system. The stake-
holders were the major participants in this project. The foundation
of the establishment of the meta-network included two parts: the
actual experience of project managers from the construction pro-
cess, which was determined by interviews and the relevant infor-
mation obtained from technicians involved in this project. We
established a preliminary list of risk factors by analyzing and
summarizing technique information. We then interview project
managers to get to revise the preliminary list into the final risk
factor list (see Table 2). Fig. 3 shows the process of the establish-
ment of the meta-network risk database:
The risk list shown in Table 2 includes all the nodes in the meta-
network, which can be divided into four categories: project ob-
jectives, risk events, risk factors and stakeholders. According to the
demonstration of networks in section 3.2, there were seven net-
works in the meta-network: (1) the project objective network to
disclose the interactions among the project objectives; (2) the risk
impact network to present how risk events lead to the bias of the
project objectives; (3) the risk event network to show the in-
teractions among the risk events; (4) the risk decomposition network
to disclose how risk factors lead to risk events; (5) the risk factor
network to show the interactions among the risk factors; (6) the
social network to show the relationships among stakeholders; and
(7) the risk distribution network to show how the stakeholders can
prevent the risk factors. To determine the existence and weight of
the links in the meta-network, surveys were conducted with the
project managers. The questionnaire directly collected their judg-
ments of the existence and weights of links. Fig. 4 summarizes the
results of the questionnaires and the construction of the network
by ORA-NetScenes (Carley et al., 2013b). The meta-network was the
foundation of the MNA.
In Fig. 4, the meta-network has round nodes which represent
the project stakeholders; square nodes represent the project ob-
jectives; diamond nodes represent the risk events and triangle
nodes represent the risk factors. The links in the meta-network
appeared to show that risk factors had an indirect influence on
project objectives by causing risk events, and stakeholders could
bring down the risk levels of the project objectives by controlling
risk factors. Therefore, critical risk factors which had the most
significant impact on project objectives could be determined by
comparing the influence levels on objectives of risk factors.
Meanwhile, by calculating the influence of the stakeholders on the
risk factors, the targeted risk control strategies could be proposed
and assigned to the project stakeholders who had a great impact on
the critical risk factors.
Location Guangdong Province
Land area 20,234 m2
Plot ratio 2.78
1.5 billion RMB (about $237 million U.S.)
558 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564

Table 2
Nodes and their contents for the meta-network in this case.

Project objectives

1 Time 2 Cost 3 Quality

4 Safety
Risk events
1 Prefabricated wall panel technology
4 Structural seam technology
7 Ceramic insulation board technology
10 Self-healing waterproofing technology
13 PVC wallpaper technology
Risk factors
1 The supply of wall panels is unstable.
4 The mounting of the flexible connector is
not horizontal or vertical.
7 The openings in the dry section are not
blocked off on time.
10 Drainage in the wet section is not installed
on time.
13 Poor construction quality of basement.
16 Cannot implement correction after the
installation finish.
19 Low quality due to lack of adequate
supervision.
22 Conservation work is not timely.
25 Drawings of water, electricity and the
garden are not completed before
construction.
28 The flatness of structural wall panels and
stability of supports do not meet
requirements.
31 The pipes crossing the beams deviate from
the design direction.
34 Concrete leakage into the water stop.
Stakeholders
1 Project manager
4 Project secretary
7 Subcontractor's engineers (decoration)
10 Supplier
5 Environment
2 Full cast-in-place exterior craft
5 Aluminum formwork
8 High-pressure water roughen technology
11 Integrally curing floor technology
14 Interspersed construction technology
2 The quality of wall panels is poor.
5 The seam material is not securely fastened.
8 Drainage in the dry section is not installed on time.
11 Inefficient costs because of a failure to intercept water
from other floors.
14 Large differences between design and site conditions.
17 Filling wall and structural shear wall displacement and
deformation of vertical flexible joints during concrete
pouring.
20 Inaccurate positioning of the starting point causes the
assembly to fail or causes the stairs to be distorted.
23 The flatness of the board and the base are not up to
standards.
26 Worker- and goods-related equipment installation failed
to be completed on time.
29 Fastening of the model is loose, and the shape of support 30 Mortar leakage in the embedded lamp box.
changes during structural board construction.
32 Oblique cracking exists while installing the wire box.
2 Field engineers
5 Project supervision
8 Subcontractor's engineers (garden)
11 Designers
3 Prefabricated components
6 Intelligent climbing formwork technology
9 Precise positioning of water and electricity equipment
12 Toilet cubicle technology
15 Water interception between floors technology
3 Cracks exist at wall joints.
6 Horizontal seam materials are not installed after pouring
floor concrete and before the concrete's first
solidification.
9 The openings in the wet section are not blocked off on
time.
12 Workers get hurt by high pressure water.
15 Faults at the end of the wall panel.
18 The supply of aluminum formwork delayed by design
changes.
21 The concrete is damaged due to uncompleted concealed
construction.
24 Conflicts between design drawings of the wall panel
parts and the construction site.
27 Decoration work is impacted by low-quality
interception of water from other floors.
33 Deviations exist during distribution box installation.
3 Water and electricity engineers
6 Main contractor's engineers
9 Workers

4.2. Risk evaluation with the meta-network
The next step in the risk evaluation system was to analyze the
risk factors to determine their relative impact on the project ob-
jectives. After establishing the meta-network that connected the
risk factors, risk events and project objectives, the risk levels for the
risk factors were determined. Risk levels referred to the probability
of risk factors. Questionnaires were used to collect the judgments of
experts who also were the participants in the risk identification
phase. The survey results are shown in Fig. 5.
The influence of risk factors on project objectives that were
derived from the results of the questionnaires were put into the
equations listed in section 3. This involved transferring experts'
judgments on risk factors to the hierarchy of project objectives
through weighted links in the meta-networks. As demonstrated in
section 3, the essence of the meta-network is the matrix, and the
transfer of risk levels from risk factors to risk events and project
objectives is essentially a multiplication of vectors of risk levels and
matrices in different networks to get the influence on the risk
events and project objectives. The resulting matrix of these calcu-
lations is shown in Fig. 6.
In this matrix, red cells indicate high risk level, while green
signify a low risk. On the whole, for the five objectives, safety
received the least influence from the risk factors, which showed
that the application of the new building industrialization con-
struction system had less impact on safety. To select the critical risk
factors, the Pareto principle was introduced to the analysis. We
considered the first 20% of the risk factors as the major source of
risk and formulated a risk control strategy based on them. Table 3
shows the risk factors at the first 20% risk level for the 34 risk
factors.
In the calculation of risk evaluation, these risk factors had the
largest impact on the project objectives when combining the risk
level and the transfer effect of the network in the matrix of Fig. 6.
There are several reasons for this. First, the most important influ-
ence on the project from building industrialization construction
methods is on the construction of the main buildings. Assembly
construction, prefabricated components and module building all
affect the process of the main building construction. Previous
studies (Bari et al., 2012; Gan et al., 2017) have suggested that the
critical factors in the application of building industrialization in
China that influence project quality can be divided into three cat-
egories: design-related factors, production-related factors, and
construction-related factors. For the critical risk factors listed in
Table 3, Risk Factor_18 represented a design-related and
production-related factors, including overall design issues and the
template production problems caused by the use of aluminum
templates. Risk Factor_3, 4 and 15 were construction-related fac-
tors, including the issues of wall panel installation, the problems in
the construction process of the structure and the matter of quality
control. In essence, during the process of change from the tradi-
tional construction methods to a new type of industrial construc-
tion, the equipment, technology and materials were new to the
works of design, management and construction. If the design phase
 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564 559

Fig. 3. The process for the establishment of the meta-network risk database.
560 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564

Fig. 4. Meta-network of the risk management for the new construction system of building industrialization in this case.

Fig. 5. The probability of the risk factors in the meta-network.

RF_1 RF_2 RF_3 RF_4 RF_5 RF_6 RF_7 RF_8 RF_9 RF_10 RF_11 RF_12 was not perfect, substantial design changes might be necessary
Time 23.1 51.975 329.175 16.8 16.8 58.8 105.3 89.7 89.7 89.7 91.8 1.95 (Risk Factor_18), which could lead to faulty construction processes,
Cost 23.4 52.65 333.45 16 16 56 99.9 85.1 85.1 85.1 89.7 1.85
such as casting issues in the panel and connection installation and
Quality 21.9 49.275 312.075 15.6 15.6 54.6 97.2 82.8 82.8 82.8 85.5 1.8
Safety 10.4 23.4 148.2 8.6 8.6 30.1 59.4 50.6 50.6 50.6 51 1.1
shear walls (Risk Factor_3, 4 and 15). Second, in the industrialized
Environment 23.1 51.975 329.175 16.8 16.8 58.8 105.3 89.7 89.7 89.7 91.8 1.95 construction process, the new project planning method e inter-
RF_13 RF_14 RF_15 RF_16 RF_17 RF_18 RF_19 RF_20 RF_21 RF_22 RF_23 RF_24 spersed construction technology (Feng and Hu, 2016) e was used to
74.3 32.1 107 21.4 102.4 214 192.6 10.7 16.4 4.1 31.2 39.9
70.2 31.35 104.5 20.9 99.6 209 188.1 10.45 14.4 3.6 29.6 38.75
improve project efficiency. Interspersed construction technology
68.5 30.6 102 20.4 97.2 204 183.6 10.2 14.8 3.7 28.8 37.8 means that a sub-project was interspersed during the construction
42 16.8 56 11.2 53.4 112 100.8 5.6 9.6 2.4 17.6 21.2 of another sub-project; therefore, it was necessary to provide a
74.3 32.1 107 21.4 102.4 214 192.6 10.7 16.4 4.1 31.2 39.9
clean and dry workplace for the next sub-project after the previous
RF_25 RF_26 RF_27 RF_28 RF_29 RF_30 RF_31 RF_32 RF_33 RF_34
25.9 14.8 33.3 4.1 4.1 31.2 7.8 7.8 31.2 7.8 sub-project. Thus, floor water interception could meet the
26.6 15.2 34.2 3.6 3.6 29.6 7.4 7.4 29.6 7.4 requirement of interspersed construction. Floor water interception
24.5 14 31.5 3.7 3.7 28.8 7.2 7.2 28.8 7.2 means intercepting the wasted water from other floors e which
14 8 18 2.4 2.4 17.6 4.4 4.4 17.6 4.4
25.9 14.8 33.3 4.1 4.1 31.2 7.8 7.8 31.2 7.8
were doing wet construction works e to provide a clean and dry
workplace for the following works (e.g., decoration work). Low
Fig. 6. Influence matrix between risk factors (RF) and project objectives. quality of floor water interception will pollute the workplace of
 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564 561

Table 3
Critical risk factors and related analysis of the risk evaluation.

Risk Factor ID Risk Factor Analysis

Risk Factor_3 Cracks exist at wall joints. Wall joints are critically important components in the
technology of prefabricated interior walls. If cracks exist, they
will have a significant influence on the quality of the
prefabricated walls and perhaps the entire building.
Risk Factor_18 The supply of aluminum formwork delayed by design Unlike the timber formwork, the aluminum formwork cannot
changes. be changed at the last minute when there are design changes.
This will lead to a delay of the formwork supply and have a great
impact on the project duration.
Risk Factor_11 Inefficient costs because of a failure to intercept water If there is a failure to intercept the wasted water from other
from other floors. floors, the decoration work area will be polluted, and the quality
of decoration work will be affected. This will lead to additional
costs (e.g., rework and cleaning).
Risk Factor_15 Faults of the ends of wall panels. The structure walls are the components that bear the load, so if
the faults exist, the load-bearing capacity of the entire building
will be weakened.
Risk Factor_4 The mounting of the flexible connector is not horizontal It is necessary for the mounting of the flexible connector to meet
or vertical. the design requirements, which normally requires the
connector to be mounted horizontally or vertically. If not
meeting these requirements, the related construction will be
negatively affected.
Risk Factor_27 Decoration work is impacted by low- quality Similar to Risk Factor_11, the decoration work area will be
interception of water from other floors. polluted, and the efficiency and quality of decoration work will
be negatively impacted.

decoration works, negatively impacting quality and efficiency. For influence on the risk factors.
the critical risk factors listed in Table 3, Risk Factor_11 and 27 were These results were consistent with the field study. One possible
the risk factors related to floor water interception. In summary, reason is that Stakeholder_1 and 4 were the highest-level man-
because of the critical risk factors listed in Table 3, the project agers of the project; they were responsible for the organization and
quality may not reach the expected outcome. Serious mistakes may coordination of the entire project. Stakeholders_2, 3, 5 and 6 were
lead to the need for much rework, which significantly increases the the project managers lowest in the hierarchy; they were directly
total project costs and duration. responsible for the construction activities. These workers were
These risk factors were not completely avoided in this test case, from the project team, construction units and supervision unit. It
as illustrated by Risk Factor_3. According to the experience of field can be concluded that for projects applying building industriali-
engineers and experts in the case study, there were a large number zation, the major risk stakeholders were the top-level managers
of issues in the installation of wall panels, and cracks appeared in and basic managers from these three categories. The water and
the connecting parts. However, with interspersed construction electricity engineers, project manager and secretary had more
technology, the effectiveness of the project was so high that there
was not enough time for workers to totally solve these issues. The
result was that even after the decoration phase, many cracks were
still obvious, which had a significant influence on the project
RF_1 RF_2 RF_3 RF_4 RF_5 RF_6 RF_7 RF_8 RF_9 RF_10 RF_11 RF_12
quality. A comparison of the results of MNA and the actual condi- Project manager 4 3 4 3 3 3 4 4 4 4 3 2
Field engineers 3 2 4 3 3 3 4 4 4 4 3 1
tions of this case showed that the risk factors identified from MNA Water and electricity engineers 4 2 4 4 4 4 4 4 4 4 3 2
were nearly the same as the potential risks. This means the process Project secretary 4 2 4 3 3 3 4 4 4 4 3 2
Project supervision 4 2 4 3 3 3 4 4 4 4 4 1
of risk identification and evaluation was effective. Risk control Main contractor's engineers 4 2 3 3 3 3 4 4 4 4 3 2
Subcontractor's engineers (decoration) 4 2 3 3 3 3 3 3 3 3 2 2
strategies are analyzed in the next phase. Subcontractor's engineers (garden) 4 2 3 3 3 3 3 3 3 3 2 2
Workers 2 1 4 3 3 3 4 4 4 4 3 2
Supplier 3 2 2 1 1 1 1 1 1 1 1 0
4.3. Risk control based on MNA Designers 2 1 4 3 3 3 4 4 4 4 3 2
RF_13 RF_14 RF_15 RF_16 RF_17 RF_18 RF_19 RF_20 RF_21 RF_22 RF_23 RF_24
3 3 3 2 2 2 4 2 3 2 2 3
Based on the above process, stakeholders with major re- 3 2 3 1 1 2 4 2 3 1 2 2
4 3 4 1 2 3 4 3 4 2 0 2
sponsibilities for the critical risk factors were identified by the 3 3 3 1 2 2 4 2 3 2 2 3
3 2 3 1 1 2 4 3 3 1 2 3
calculation of social network, risk factor network and risk distribution 3 2 3 2 2 3 3 2 3 2 1 4
network, and the calculation results are shown in Fig. 7. 3 2 3 2 2 3 3 2 3 2 0 3
3 2 3 2 2 3 3 2 3 2 0 3
In this matrix, red cells indicate a high level of influence level, 3 3 3 0 2 2 4 2 3 2 2 2
1 1 1 1 0 0 2 1 1 0 1 0
while green means there was a low level of influence. Fig. 7 shows 3 3 3 0 2 2 4 2 3 2 2 2
that Stakeholder_3 (the water and electricity engineers), Stake- RF_25 RF_26 RF_27 RF_28 RF_29 RF_30 RF_31 RF_32 RF_33 RF_34
5 3 3 3 3 3 3 3 1 1
holder_1 (the project manager) and Stakeholder_4 (the project 5 3 3 4 4 4 4 4 1 1
4 4 4 3 3 3 3 3 0 0
secretary) were the critical stakeholders with the highest levels of 5 3 3 3 3 3 3 3 1 1
influence, and the influence levels of Stakeholder_2 (the field en- 5 3 3 3 3 3 3 3 1 1
4 3 3 3 3 3 3 3 1 1
gineer), Stakeholder_5 (the project supervisor) and Stakeholder_6 3 3 3 2 2 2 2 2 0 0
3 3 3 2 2 2 2 2 0 0
(the main contractor engineer) were near the influence levels of the 5 3 3 3 3 3 3 3 1 1
critical stakeholders. Therefore, all six stakeholders were consid- 2 1 1 1 1 1 1 1 0 0
5 3 3 3 3 3 3 3 1 1
ered to propose risk control strategies. The basic responsibilities of
these stakeholders are listed in Table 4 in the order of the level of Fig. 7. Matrix of influence between stakeholders and risk factors (RF).
562 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564

Table 4
Critical stakeholders and related responsibilities.

Stakeholder ID Stakeholder Responsibility

Stakeholder_3 Water and electricity engineers Construction and installation of water and electricity equipment (e.g. water pipe, electricity wire).
Stakeholder_1 Project manager The top manager of this project; responsible for planning, organizing, leading and controlling the entire project.
Stakeholder_4 Project secretary The support staff for the project manager; responsible for coordination and communication.
Stakeholder_2 Field engineer The management of detailed construction activities.
Stakeholder_5 Project supervisor Supervision of the project activities to ensure that the project is implemented according to plan.
Stakeholder_6 Main contractor engineer The management of detailed construction activities and the arrangement of workers.

impact on the risk factors, while the field engineers from the
development units, main contractor engineers from the construc-
tion units and project supervisors from the supervision units
played supporting roles.
After agreeing on the critical stakeholders, risk control strategies
were proposed based on the results of the risk evaluation. For
construction-related factors and water interception related factors
e such as Risk Factor_3 and 11ecritical stakeholders should have
adopted several measures. As the highest-level managers of the
project, the project manager and project secretary should have
strengthened the communication of the engineers and proposed a
flexible construction plan based on the real-time conditions of the
project. As members of the construction units, the main contractor
engineers bear the brunt of risk events and should implement strict
control methods for the workers and provide correct instruction
about construction technology. This can reduce the probability of
the risk factors. Water and electricity engineers and field engineers
from the project team should be more familiar with the new con-
struction methods, related technologies and design, so they should
guide people higher in the hierarchy. They need real-time monitors
at all project venues to ensure that design ideas are implemented
and that the building industrialization construction methods are
implemented correctly. As support for the project team, supervi-
sion engineers should monitor the project from the perspective of
project quality and preventing accidents by reducing risks, such as
Risk Factor_3, 4 and 15. They should provide for careful examina-
tion during and after the construction process. For design-related
risk factors, such as Risk Factor_18, the major risk stakeholders
are the members of the project team. For example, in the imple-
mentation of the “full cast-in-place exterior craft,” changing water
and electricity lines directly leads to changes in the aluminum
formwork; thus, the water and electricity engineers and field en-
gineers should revise the requirements for the aluminum form-
work as soon as possible to reduce the impact on project duration
and total cost. Also, the project manager and secretary should give
assistance and coordinate the communication between the project
and the design teams. Based on these strategies, a strict re-
sponsibility system for the industrialized construction process
should be established among these six critical stakeholders e
namely the water and electricity engineers, the project manager,
the project secretary, the field engineer, the project supervisor and
the main contractor engineer. First, as the highest-level managers
of the project, the project manager and project secretary are
responsible for the communication and coordination of the stake-
holders inside and outside the project, such as the design team and
the basic managers of the project. Second, to achieve the greatest
degree of risk aversion, a supervision mechanism is needed among
the basic managers. Engineers from the main contractor, who are
directly responsible for implementation of the industrialized con-
struction approach, should receive supervision from the water and
electricity engineers, field engineers and project supervisors, who
are more familiar with these approaches. The quality control of the
project supervisors should be examined by the engineers from the
project team.
5. Conclusions
Our research evaluated the risk levels in an industrialized
building construction project and suggested targeted risk control
measures based on the proposed integrated risk evaluation and
control approach. We used meta-network analysis to incorporate
the experts' knowledge and data from our on-the-spot investiga-
tion for the final model of risk management, supplementing the
body of knowledge in the area of industrialized construction pro-
jects. First, we identified risks by using the scientific literature and
previous construction experience. Next, the results were converted
into a meta-network that consisted of many associated networks.
Finally, risk evaluation and control were carried out with reference
to the meta-network. From a theoretical perspective, this frame-
work can thoroughly analyze risk factors in a building industriali-
zation construction project, study the influence of the associations
among risks and offer effective risk control strategies based on risk
evaluation. Compared to previous risk management methods, the
advantages of MNA are that it can identify the critical risk factors
more visibly (by graphing the meta-network) and provide effective
guidance for risk control by analyzing the participation of stake-
holders for controlling risk factors. The traditional methods usually
stopped at the risk evaluation phase. Meanwhile, the use of this
method is simple. For building industrialization projects, project
managers can somewhat easily handle the risk management
related to the risk factors identified by MNA.
We applied our proposed MNA method to a residential building
construction project in South China that introduced building
industrialization to the construction process. During the construc-
tion process in this project, because of new construction and
management approaches of industrialized construction, risk factors
in the construction process were quite different than those in a
traditional project. Therefore, risk identification for the industrial-
ized construction process was conducted by interviewing related
experts and experienced engineers, and the results of these in-
terviews were used to create a meta-network of risk. As shown in
the results of the risk evaluation process, six risk factors in three
categories were considered critical risk factors: (1) construction-
related factors, such as Risk Factor_3 (cracks exist on wall panel
joints and other unqualified low-quality factors); (2) design and
production-related factors, such as Risk Factor_18 (the supply of
aluminum model was delayed due to design changes); and (3)
water interception-related risk factors, such as Risk Factor_27
(decoration work was impacted by low-quality interception of
water from other floors). The results were consistent with the field
study results. The risk factors were not completely avoided during
the project. A risk control analysis was based on the results of the
risk evaluation to find the critical stakeholders who had a signifi-
cant influence on the risk factors, and these stakeholders were
Stakeholder_3 (the water and electricity engineers), Stakeholder_1
(the project manager), Stakeholder_4 (the project secretary),
Stakeholder_2 (the field engineer), Stakeholder_5 (the project su-
pervisor) and Stakeholder_6 (the main contractor engineer). To
avoid risk factors and events, risk control strategies were proposed.
 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564
In addition, a responsibility system among these stakeholders
should be established. First, the project manager and project sec-
retary should be responsible for the communication and coordi-
nation of the stakeholders inside and outside the project, such as
the design team and the basic managers of the project. Second, a
supervision mechanism is needed among the basic managers. En-
gineers from the main contractor should receive supervision from
the water and electricity engineers, field engineers and project
supervisors. The quality control of project supervisors should be
examined by the engineers from the project team. In this risk
control strategy, all stakeholders should minimize the possibility of
risk factors occurring within their area of responsibility.
In actual practice, risk managers should build a hierarchy of
project objectives, risk events, risk factors, stakeholders and related
networks based on their actual situation. The meta-network pro-
posed for this project is just an example; the establishment of a
network should be based on each project's unique nature. These
results render novel and practical ideas for risk management,
particularly to evaluate and control the risks of industrialized
building construction projects. Future work should use MNA in
other scenarios and add more entities and relationships to the
meta-network to make the method more effective and feasible.
Acknowledgement
The study is financially supported by the National Science
Foundation of China (No. 71871235). The authors would like to
thank the managers and workers participated in this study.
References
Abdelgawad, M., Fayek, A.R., 2011. Fuzzy reliability analyzer: quantitative assess-
ment of risk events in the construction industry using fuzzy fault-tree analysis.
J. Construct. Eng. Manag. 137, 294e302.
Aye, L., Ngo, T., Crawford, R., Gammampila, R., Mendis, P., 2012. Life cycle green-
house gas emissions and energy analysis of prefabricated reusable building
modules. Energy Build. 47, 159e168.
Bari, N.A.A., Yusuff, R., Ismail, N., Jaapar, A., Ahmad, R., 2012. Factors influencing the
construction cost of industrialised building system (IBS) projects. Procedia Soc.
Behav. Sci. 35, 689e696.
Borsuk, M.E., Stow, C.A., Reckhow, K.H., 2004. A Bayesian network of eutrophication
models for synthesis, prediction, and uncertainty analysis. Ecol. Model. 173,
219e239.
Bu-Qammaz, A.S., Dikmen, I., Birgonul, M.T., 2009. Risk assessment of international
construction projects using the analytic network process. Can. J. Civ. Eng. 36,
1170e1181.
Carley, K.M., Pfeffer, J., rgen, Liu, H., Morstatter, F., Goolsby, R., 2013a. Near real time
assessment of social media using geo-temporal network analytics. In: Ieee/acm
International Conference on Advances in Social Networks Analysis and Mining,
pp. 517e524.
Carley, K.M., Reminga, J., Storrick, J., Pfeffer, J., Columbus, D., 2013b. ORA User's
Guide 2013. Ora Users Guide.
Chiang, Y.-H., Chan, E.H.-W., Lok, L.K.-L., 2006. Prefabrication and barriers to
entryda case study of public housing and institutional buildings in Hong Kong.
Habitat Int. 30, 482e499.
Elsawah, H., Bakry, I., Moselhi, O., 2016. Decision support model for integrated risk
assessment and prioritization of intervention plans of municipal infrastructure.
J. Pipeline Syst. Eng. Pract. 7, 04016010.
Fang, C., Marle, F., 2015. A Framework for the Modeling and Management of Project
Risks and Risk Interactions.
Fang, C., Marle, F., Zio, E., Bocquet, J.C., 2012. Network theory-based analysis of risk
interactions in large engineering projects. Reliab. Eng. Syst. Saf. 106, 1e10.
Faniran, O., Caban, G., 1998. Minimizing waste on construction project sites. Eng.
Construct. Architect. Manag. 5, 182e188.
Feng, W., Hu, J., 2016. Application of BIM Technology in Interspersed Construction of
High-rise Residences. Construction Technology.
Gan, Y., Shen, L., Chen, J., Tam, V.W., Tan, Y., Illankoon, I., 2017. Critical factors
affecting the quality of industrialized building system projects in China. Sus-
tainability 9, 216.
Gibb, A.G., 1999. Off-site Fabrication: Prefabrication, Pre-assembly and Modular-
isation. John Wiley & Sons.
Goodier, C., Gibb, A., 2007. Future opportunities for offsite in the UK. Construct.
Manag. Econ. 25, 585e595.
Guan, D.J., Guo, P., 2014. Constructing Interdependent Risks Network of Project
Portfolio Based on Bayesian Network. International Conference on Management
563
Science & Engineering, pp. 1587e1592.
Han, S.H., Du, Y.K., Kim, H., Jang, W.S., 2008. A web-based integrated system for
international project risk management. Autom. ConStruct. 17, 342e3568.
Hong, J., Shen, G.Q., Mao, C., Li, Z., Li, K., 2016. Life-cycle energy analysis of pre-
fabricated building components: an inputeoutput-based hybrid model. J. Clean.
Prod. 112, 2198e2207.
Jaillon, L., Poon, C.S., 2009. The evolution of prefabricated residential building
systems in Hong Kong: a review of the public and the private sector. Autom.
ConStruct. 18, 239e248.
Ji, Y., Zhu, F., Li, H.X., Al-Hussein, M., 2017. Construction industrialization in China:
current profile and the prediction. Appl. Sci. 6.
Jiang, R., Mao, C., Hou, L., Wu, C., Tan, J., 1 February 2018. A SWOT analysis for
promoting off-site construction under the backdrop of China's new urbanisa-
tion. J. Clean. Prod. 173, 225e234.
Karimiazari, A., Mousavi, N., Mousavi, S.F., Hosseini, S., 2011. Risk assessment model
selection in construction industry. Expet. Syst. Appl. Int. J. 38, 9105e9111.
Kerzner, H., 1989. Project Management: a Systems Approach to Planning, Sched-
uling, and Controlling. Wiley.
Kuo, Y.C., Lu, S.T., 2013. Using fuzzy multiple criteria decision making approach to
enhance risk assessment for metropolitan construction projects. Int. J. Proj.
Manag. 31, 602e614.
Lenz, R., 2012. Topology of local health officials' advice networks: mind the gaps.
J. Publ. Health Manag. Pract. 18, 602e608.
Li-Zi, L., Chao, M., Li-Yin, S., Zheng-Dao, L., 2015. Risk factors affecting practitioners'
attitudes toward the implementation of an industrialized building system: a
case study from China. Eng. Construct. Architect. Manag. 22, 622e643.
Li, M., Li, G., Huang, Y., Deng, L., 2017. Research on investment risk management of
Chinese prefabricated construction projects based on a system dynamics
model. Buildings 7, 83.
Li, Y., Lu, Y., Li, D., Ma, L., 2015. Metanetwork analysis for project task assignment.
J. Construct. Eng. Manag. 141, 04015044.
Lihong, L.I., Geng, B., Baoku, Q.I., Lei, Y., Luan, L., 2013. Empirical Analysis on the Cost
of Prefabricated Construction Contrast with Cast-in-place Building. Construc-
tion Economy.
Lin, Y.C., Paul, A., Corotis, R.B., Liel, A.B., 2015. Framework methodology for risk-
based decision making for transportation agencies. ASCE-ASME J. Risk Uncer-
tain. Eng. Syst. Part A Civ. Eng., 04015006
Liu, H., 2017. A new perspective of building industrialization to promote rapid
development of China's construction. World Construct. 6.
Lovell, H., Smith, S.J., 2010. Agencement in housing markets: the case of the UK
construction industry. Geoforum 41, 457e468.
Luo, L.Z., Mao, C., Shen, L.Y., Li, Z.D., 2015. Risk factors affecting practitioners' atti-
tudes toward the implementation of an industrialized building system: a case
study from China. Eng. Construct. Architect. Manag. 22, 622e643.
Mao, C., Shen, Q., Pan, W., Ye, K., 2013. Major barriers to off-site construction: the
developer's perspective in China. J. Manag. Eng. 31, 04014043.
Martinez, S., Jardon, A., Navarro, J.M., Gonzalez, P., 2008. Building industrialization:
robotized assembly of modular products. Assemb. Autom. 28, 134e142.
Nagurney, A., Dong, J., 2005. Management of knowledge intensive systems as
supernetworks: modeling, analysis, computations, and applications. Math.
Comput. Model. 42, 397e417.
Pan, W., Sidwell, R., 2011. Demystifying the cost barriers to offsite construction in
the UK. Construct. Manag. Econ. 29, 1081e1099.
Pons, O., Wadel, G., 2011. Environmental impacts of prefabricated school buildings
in Catalonia. Habitat Int. 35, 553e563.
Purdy, G., 2010. ISO 31000: 2009dsetting a new standard for risk management.
Risk Anal. 30, 881e886.
Roy, R., Low, M., Waller, J., 2005. Documentation, standardization and improvement
of the construction process in house building. Construct. Manag. Econ. 23,
57e67.
Rubio-Romero, J.C., Su~ a, r.-C.M., Abad, J., 2014. Modeling injury rates as a function of
industrialized versus on-site construction techniques. Accid. Anal. Prev. 66,
8e14.
Saaty, T.L., 2004. Fundamentals of the analytic network process d dependence and
feedback in decision-making with a single network. J. Syst. Sci. Syst. Eng. 13,
129e157.
Stergiopoulos, G., Kotzanikolaou, P., Theocharidou, M., Lykou, G., Gritzalis, D., 2016.
Time-based critical infrastructure dependency analysis for large-scale and
cross-sectoral failures. Int. J. Crit. Infrastruct. Protect. 12, 46e60.
Tam, V.W.Y., Tam, C.M., Zeng, S.X., Ng, W.C.Y., 2007. Towards adoption of prefabri-
cation in construction. Build. Environ. 42, 3642e3654.
Tavakolan, M., Etemadinia, H., Tavakolan, M., Etemadinia, H., 2017. Fuzzy weighted
interpretive structural modeling: improved method for identification of risk
interactions in construction projects. J. Construct. Eng. Manag. 143.
Wakolbinger, T., Nagurney, A., 2004. Dynamic supernetworks for the integration of
social networks and supply chains with electronic commerce: modeling and
analysis of buyer-seller relationships with computations. Netnomics 6,
153e185.
Wang, T., Wang, S., Zhang, L., Huang, Z., Li, Y., 2016. A major infrastructure risk-
assessment framework: application to a cross-sea route project in China. Int.
J. Proj. Manag. 34, 1403e1415.
Wong, A.K., Yeh, S.H.K., 1985. Housing a Nation : 25 Years of Public Housing in
Singapore.
Xiahou, X., Yuan, J., Liu, Y., Tang, Y., Li, Q., 2018. Exploring the driving factors of
construction industrialization development in China. Int. J. Environ. Res. Publ.
564 T. Wang et al. / Journal of Cleaner Production 205 (2018) 552e564
Health 15.
Xu, X., Zhao, Y., 2010. Some Economic Facts of the Prefabricated Housing. Industry
Report, vol. 3. Rutgers Business School, Newark, NJ.
Yan, W., CAO, Y.-h., LI, G.-r., 2004. Development of assembly-type RC structure and
building industrialization. J. Chongqing Jianzhu Univ. 26, 131e136.
Zabihi, H., 2013. Definitions, concepts and new directions in industrialized building
systems (IBS). Ksce J. Civ. Eng. 17, 1199e1205.
Zhang, G., Zou, P.X.W., 2007. Fuzzy analytical hierarchy process risk assessment
approach for joint venture construction projects in China. J. Construct. Eng.
Manag. 133, 771e779.
Zhang, X., Skitmore, M., 2012. Industrialized housing in China: a coin with two
sides. Int. J. Strat. Property Manag. 16, 143e157.
Zhou, H.B., Zhang, H., 2011. Risk assessment methodology for a deep foundation pit
construction project in shanghai, China. J. Construct. Eng. Manag. 137,
1185e1194.
Zhu, J., Mostafavi, A., 2015. Ex-ante assessment of vulnerability to uncertainty in
complex construction project organizations.
Zhu, J., Mostafavi, A., 2016. Metanetwork framework for integrated performance
assessment under uncertainty in construction projects. J. Comput. Civ. Eng. 31,
04016042.