Automation in Construction 124 (2021) 103553

Contents lists available at ScienceDirect

Automation in Construction
journal homepage: www.elsevier.com/locate/autcon

BIM-integrated construction safety risk assessment at the design stage of

building projects
Ying Lu *, Peizhen Gong, Yuchun Tang, Shuqi Sun, Qiming Li
Department of Construction Management and Real Estate, School of Civil Engineering, Southeast University, Nanjing, 211189, China

A R T I C L E I N F O A B S T R A C T

Keywords: The construction industry has a higher occupational casualty rate than other industries. As a proactive approach
Construction safety to safety management, Construction Hazard Prevention through Design (CHPtD) can significantly eliminate or
Safety risk assessment reduce the construction safety risk. However, this concept is not implemented effectively in practice because the
Building information modeling (BIM)
technical issues that underlie CHPtD have not been addressed. This paper proposes a novel method of quanti­
Design stage
tative construction safety risk assessment for building projects at the design stage. This method consists of three
indexes: Likelihood, consequence and exposure. These indexes are calculated using occupational injury, fatality
and specific construction planning data which are accurate and objective. A plug-in that links building infor­
mation modeling (BIM) with safety risk data is developed in Autodesk Revit, which can automatically calculate
construction safety risk to help architects and structural designers quickly select design alternatives. A case study
is presented to demonstrate the feasibility and effectiveness of the proposed method.

1. Introduction for Occupational Safety and Health (NIOSH), is considered to be an

Currently, the occupational fatality and injury rate in the construc­
tion industry is relatively high due to the dynamic characteristics of the
construction work environment. The Ministry of Housing and Urban-
Rural Development of the People’s Republic of China (MOHURD) indi­
cated that 734 construction safety accidents and 840 deaths occurred in
2018, which is an increase of 6.1% and 4.1% respectively [1]. The Bu­
reau of Labor Statistics (BLS) of the US Department of Labor showed that
the average fatality rate in the construction industry in 2017 was 9.5
fatalities per 100,000 workers. However, the all-worker fatality rate was
3.5 fatalities per 100,000 workers during the same year [2]. A report of
construction statistics in Great Britain revealed 30 fatalities among
workers in 2018/19 [3]. The worker fatality rate was 1.31 fatalities per
100,000 workers while the average fatality rate was 0.45 fatalities per
100,000 workers across all industries. Construction safety performance
has become a recognized issue in the construction industry.
Since design schemes often decide construction methods and
schedules, many construction safety accidents are caused by improper
design, which could have been avoided if properly designed. Risk
assessment in the design stage, as an important content of safety man­
agement, can fundamentally reduce the risk at its source [4]. Prevention
through Design (PtD), as originally proposed by the National Institute
effective construction risk reduction strategy [5–8]. The construction
occupational injuries and fatalities during construction, maintenance,
repair, and demolition can be prevented by PtD; the cost and effort that
are required in adjusting design schemes can also be reduced [9]. As an
extension of the PtD theory, Construction Hazard Prevention through
Design (CHPtD) specifically emphasizes the identification and removal
of construction safety hazards of a building project at the design stage
[10]. Behm [11] reviewed 224 death investigation reports and demon­
strated that 42% of deaths would been reduced or eliminated when
CHPtD is utilized at the design stage. Zhang et al. [12] identified po­
tential fall hazards and discovered that placing the guardrail posts at the
edges of the slab in the design phase would reduce fall hazards in a case
project. Golabchi et al. [13] indicated that many interventions exist to
reduce construction risks in the design stage, including the design of
workplaces, design of sequence of tasks, design of shape and size, design
of human actions, and other interventions. Integrating these designs can
increase safety in the construction process. The concept of CHPtD is
beneficial to the safety performance of the construction industry.
Although many studies have recommended the use of CHPtD as a
proactive safety intervention for reducing the construction safety risk,
this concept has not been implemented effectively in practice. Most of
the architects and structural designers lack a sufficient knowledge of

* Corresponding author.
E-mail address: luying_happy@126.com (Y. Lu).

https://doi.org/10.1016/j.autcon.2021.103553
Received 7 July 2020; Received in revised form 23 October 2020; Accepted 5 January 2021
Available online 23 January 2021
0926-5805/© 2021 Elsevier B.V. All rights reserved.
Y. Lu et al.
construction processes and safety risk which is necessary to perform
CHPtD. Only 3.2% of architects and 9.8% of structural designers stated
that their firm has guidelines for reviewing CHPtD [14]. The main
reason for this is that the technical issues underlying CHPtD in order to
help architects and structure designers better perform CHPtD have not
yet been addressed [15]. Safety risk assessment in the design stage is
necessary to perform CHPtD. Traditional approaches to assessing con­
struction safety risk are implemented after design, such as risk checklists
and risk management systems, which rely on manual operations and are
time-consuming and inefficient in practical construction projects [9].
Therefore, as noted by Hardison and Hallowell [16], the key points of
technical issues include how to use risk-based methods to select design
alternatives that optimize construction safety, and how to use novel
tools to improve the selection of an optimal design solution. The avail­
ability of CHPtD computer tools can be a favorable stimulus for CHPtD
implementation [4]. Building information modeling (BIM) provides a
platform to enable other BIM-based development to integrate external
data with the parametric BIM environment and perform further auto­
matic risk analysis in the design of buildings, which can help architects
and designers quickly select or improve design alternatives [17]. The use
of BIM-related technologies in building design and safety management
are growing rapidly. However, the majority of previous research
remained in the hazard identification stage [18,19]. Few studies have
focused on the use of the BIM platform to quantify and calculate
comprehensive safety risk. Therefore, the BIM-integrated construction
safety risk assessment at the design stage was proposed in this study to
further implement the concept of CHPtD.
This paper proposes a quantitative construction safety risk assess­
ment model that considers three indexes: likelihood, consequence and
exposure. These indexes are calculated using occupational injury, fa­
tality and specific construction planning data which are accurate and
objective. A plug-in that links BIM with safety risk data is developed in
Autodesk Revit. The plug-in can automatically calculate construction
safety risk to help architects and structural designers quickly select
design alternatives that optimize construction safety. In addition, a case
study is presented to demonstrate the feasibility and effectiveness of the
proposed method for construction safety risk assessment at the design
stage.
2. Literature review
2.1. Safety risk assessment in construction
Scholars have proposed several methods from different perspectives
to quantify construction safety risk, which can be summarized into three
aspects. The first aspect is a project-based approach that concentrates on
analyzing the causes of construction hazards and comprehensively
estimating the construction safety risk level. Some scholars express that
organizing for safety supervision and equipment management, compli­
ance with applicable safety regulations, attitudes about risk taking and
management behavior, and high quality and frequent safety communi­
cation are critical factors that affect the safety behavior of construction
workers [20,21]. Gunduz and Laitinen [22] proposed a dynamic risk
assessment model that consists of a 14-item checklist and a two-scale
risk matrix. Risk factors include order and tidiness, chemical hazards,
psychology factors, etc. It is necessary to establish various types of safety
communications between all parts [23]. The safety risk assessment
methods include fuzzy sets, the analytic hierarchy process and the
weighted aggregated sum product assessment method [24–26].
The second approach is activity-oriented, which highlights creating a
work breakdown structure, analyzing related hazards, and hierar­
chically evaluating construction safety risk. Gurcanli et al. [27] estab­
lished the work breakdown structures of 25 building projects and listed
all construction activities of each construction project to determine
different safety risks. Hallowell and Gambatese [28] identified 13 con­
struction activities of building concrete formwork, and quantified each
2
Automation in Construction 124 (2021) 103553
safety risk. Esmaeili et al. [29] proposed an attribute-based risk assess­
ment approach to quantify the risk of construction activities.
The third approach is occupation-oriented, which concentrates on
analyzing injury and fatality data of different occupations from the
construction industry viewpoint. Baradan and Usmen [30] analyzed the
nonfatal injury and fatality risks of sixteen construction occupations.
The results indicated that ironworkers and roofers have the highest
safety risk. Choe and Leite [31] demonstrated that each occupation has a
unique risk data structure with regard to injury sources. Occupation-
oriented approaches are unable to identify further hazards generated
from the subsequent construction process. However, occupation-
oriented approaches have several advantages over project-based ap­
proaches and activity-based approaches in the design stage. First, the
data collection of occupation-oriented approaches is more accurate and
objective. The previous two approaches are mainly based on historical
data or survey-based data [32–34], while occupation-oriented ap­
proaches can extract real statistical data from existing occupational
databases. Second, different occupations have different levels of casu­
alty frequency and severity, even in the same construction project or
construction activity. Therefore, safety risks should be quantified ac­
cording to the unique nature of different occupations.
2.2. Quantitative model for construction safety risk assessment
Different scholars have different interpretations of the concept of
safety risk. Esmaeili and Hallowell [35] proposed that safety risk in­
cludes likelihood and consequence. Likelihood represents the number of
injuries or fatalities per hour. Consequence describes the severity of
injuries or fatalities. Some scholars use likelihood and consequence to
establish a quantitative model and evaluate construction safety risk.
Hallowell et al. [36] established a safety risk scale that considers all
possible likelihoods and consequences. Azadeh-Fard et al. [37] intro­
duced a new consequence scoring system that considers multiple injury
consequence factors, and employed it as part of a novel risk assessment
matrix. Mohandes et al. [38] developed a novel risk quantification
model with a probability parameter and severity parameter where the
Fuzzy Best Worst Method is exploited to obtain the weights of each
parameter to calculate the safety risks of construction workers.
Exposure is also an important index of safety risk evaluation. The US
Occupational Information Network (O*NET) program [39,40] empha­
sized that construction work conditions involve many hazardous expo­
sures such as contaminants, noise, fall risks and other hazards. The
program quantified the hazard exposure by giving an exposure score to
each occupation according to the number of workers and exposure
duration. Some scholars proposed that exposure can be described as the
time of contact with the dangerous situation [41]. Risk can be identified
as the product of likelihood, consequence and exposure [42–45]. Frijters
and Swuste [46] utilized a risk quantification table to estimate the
likelihood and consequence, and determined the exposure by calcu­
lating the total duration of the construction processes. Hallowell and
Gambatese [28] proposed a risk scale that includes likelihood and
consequence, and calculated exposure in terms of worker-hours. Gang­
olells et al. [47] rated the likelihood and consequence of a construction
activity according to the opinions of experts, and assessed the exposure
according to construction project documents. Dharmapalan et al. [48]
distributed questionnaires to general contracting superintendents to
determine the risk scale of likelihood and consequence, and quantified
the exposure based on worker-hours per unit.
However, the values of these three indexes are typically conducted
through expert interviews and questionnaires, which are easily influ­
enced by subjective judgment [49]. Therefore, solving superior data
source issues, which should be accurate and objective, is urgent.
2.3. Application of BIM in safety risk
As an object-oriented parametric digital representation, BIM is
Y. Lu et al. Automation in Construction 124 (2021) 103553

Data Collection
Fatality Days-Away Injury Unit Labor
Hours
CFOI Number Number SOII
Required Labor
Resource RSMeans

Construction
Occupation Material

Occupation Type SOC

Employment Size OES

Injury, Fatality Construction

Database Planning Database

Data Mapping
BIM Model

Identity
IDs of Building
Elements

Establish
Construction
Process

Acquire Acquire
Construction
Quantity
Occupation

Select Select
Injury, Fatality Unit Labor
Data Hours Data

Data Analysis

Likelihood Consequence Exposure

Safety risk assessment

API

Automatic
Calculation of Safety
Risk

Fig. 1. Theoretical framework.

becoming a systematic tool that is changing the project planning, design,
construction, operation and maintenance phases. In the design stage,
BIM is an effective way to link risk data with design elements in BIM
[50]. Many scholars combine BIM with relevant databases to realize the
concept of PtD, which aims to help architects and structure designers
identify hazards and provide solutions [51]. Two main trends exist in
achieving PtD: using BIM software for clash detection and integrating
BIM and knowledge-based systems to detect hazards [9]. For instance,
Hu et al. [52] created a model of a continuous and dynamic simulation
of the entire construction process and then applied BIM technology for
collision inspection. Zhang et al. [12] integrated safety rules into BIM to
identity and prevent the hazards associated with falls. Guo et al. [53]
combined BIM with design safety rules to automatically identify possible
safety problems in the design phase. Hussein et al. [54] developed a risk
review system that was integrated with BIM to help architects and
structural designers check design elements. Cortés-Pérez et al. [55] in­
tegrated an occupational health and safety risk assessment into BIM
projects in Spain and parameterized BIM objects according to Spanish
regulations and hazard preventive measures in the design phase. The
assessment results were shown with color codes based on the risk rating
level of each object and then the corresponding measures were
implemented.

3
Y. Lu et al.
Architects and structural designers could benefit from BIM-
integrated safety management system [56]. However, the majority of
previous research has only identified specific types of hazard based on
rules. Few studies focus on using the BIM platform to comprehensively
calculate safety risk. A novel approach for quantitative construction
safety risk assessment and a BIM-based automatically calculation plug-
in is urgently needed to help architects and structural designers iden­
tify safer design alternatives.
3. Methodology
The methodology in this paper includes a theoretical framework and
a developed plug-in. The proposed theoretical framework includes data
collection, data mapping and data analysis, which are employed to
quantitatively calculate construction safety risks by occupation, as
shown in Fig. 1. The developed plug-in implements the proposed theo­
retical framework and serves as a platform to automatically calculate
construction safety risk.
3.1. Proposed theoretical framework
3.1.1. Data collection
As a visual modeling tool, BIM can provide the information needed in
the planning, design, construction and operation stages. However, BIM
lacks data related to safety risk assessment. Safety risk data should be
systematically captured in the data collection phase. Therefore, two
databases that integrate all the data for risk assessment were newly built
in the data collection phase: The Injury and Fatality Database and the
Construction Planning Database. The former database recorded the ca­
sualty information and the employment size of each construction
occupation. The latter collected data on the specific construction pro­
cess, unit labor hours, and required labor resources. All the information
aggregated in these databases were based on the US construction
industry.
In the Injury and Fatality Database, the casualty types were divided
into days-away injury and fatality, according to the BLS Injuries, Ill­
nesses, and Fatalities (IIF) program. The BLS IIF program collects annual
data from diverse state, federal, and independent data sources in the US
to identify and verify the number of occupational days-away injuries and
fatalities of all trades [2]. The data of the construction industry extracted
from the 2018 BLS IIF Program were applied in this study. Specifically,
the data of days-away injury were collected from the Survey of Occu­
pational Injury and Illness (SOII) of the IIF program. The fatality data
were collected from the Census of Fatal Occupational Injuries (CFOI) of
the IIF program. The BLS SOII only collects data from the private sectors,
while the BLS CFOI collects data from the private sectors, government
workers, and self-employed workers by industry. Therefore, data from
the private sectors were used in this study for consistency. Nineteen
construction occupations were selected for analysis in this paper based
on the 2010 Standard Occupational Classification (SOC) system. The
total number of workers employed in the private sector were collected
from the BLS Occupational Employment Statistics (OES). All of these
data are integrated in the Injury and Fatality Database.
In the Construction Planning Database, the unit labor hours, required
labor resources and construction materials were acquired from
RSMeans. RSMeans is an existing mature construction cost database
owned by Gordian Company; the database contains the most compre­
hensive data of unit material costs, unit labor costs, unit equipment
costs, labor sources, city cost indexes, equipment rentals, location fac­
tors and other information of construction projects in North America
[57]. RSMeans can reflect the average level of the US construction in­
dustry. In this study, only three types of data were applied to calculate
the exposure index: the unit labor hours, required labor resources, and
construction materials. Therefore, only these three types of data were
separately acquired from RSMeans to integrate into the Construction
Planning Database.
4
Automation in Construction 124 (2021) 103553
Table 1
Indexes of construction safety risk assessment.
Data source Index name
Injury and Fatality Database Number of fatalities (NFi)
Number of injuries (NIi)
Employment size (ESi)
Median days-away for occupation i (MDAi)
Construction Planning Database Construction process j
Unit labor hours
Required labor resources
Based on the information stored in these databases, the indexes of
construction safety risk assessment can be calculated automatically ac­
cording to the corresponding formulas. Detailed calculation methods
will be introduced in Section 3.1.3. All the indexes and their data sources
are preliminarily listed in Table 1 to illustrate the relationship between
the risk evaluation process and the databases.
3.1.2. Data mapping
BIM is a modeling technique that is oriented to family objects. By
establishing a custom family library of basic building components, more
dimensional information can be integrated beyond dimensional infor­
mation. The families were divided according to the construction pro­
cesses. The two previously mentioned databases were linked to BIM, and
DB Link was applied to extract and update data. As shown in Fig. 2, the
process of data mapping includes two steps. First, the required labor
resources, required occupation ID, and unit labor hour of each con­
struction process stored in the Construction Planning Database were
mapped into the extension parameters of the Revit family objects. Sec­
ond, the required occupation IDs obtained from the first mapping were
mapped to the Injury and Fatality Database. Then, the likelihood and
consequence indexes of the corresponding occupations can be obtained.
The quantities of the construction materials are automatically calculated
in Revit. The exposure index can be calculated according to the quan­
tities and unit labor hour. Overall, data mapping is responsible for
linking the collected data with the BIM platform to quantify the safety
risk.
3.1.3. Data analysis
The logic diagram of data analysis is shown in Fig. 3. The likelihood
index is divided into the fatality frequency index and the days-away
injury frequency index. The fatality frequency index for occupation i
(FFIi) is calculated by the number of fatalities (NFi) and employment size
(ESi). The days-away injury frequency index for occupation i (DAIFIi) is
calculated by the number of injuries (NIi) and employment size (ESi), as
shown in Eqs. (1) and (2) [31].
FFI i = NF i /ESi (1)
DAIFI i = NI i /ESi (2)
To compare with the average occupational injury rate, the relative
fatality frequency index (RFFIi) and relative days-away injury frequency
index (RDAIFIi) for occupation i are created, as shown in Eq. (3) and Eq.
(4), respectively.
/
RFFI i = FFI i FFI average (3)
/
RDAIFI i = DAIFI i DAIFI average (4)
Considering the same consequence of fatality, the relative fatality
severity for occupation i (RFSi) is set to 1. For the relative days-away
injury severity for occupation i (RDAISi), the median days-away for
occupation i (MDAi) is combined to consider the different levels of
severity of construction occupational injuries, as shown in Eqs. (5) and
(6).
RFSi = 1 (5)
Y. Lu et al. Automation in Construction 124 (2021) 103553

Revit data table Construction Planning Database Injury and Fatality Database

- Building element ID - Revit family ID - Occupation ID

- Building element - Required labor resource - Likelihood
name - Occupation ID - Consequence
- Revit family ID - Unit labor hour
- Size information

Fig. 2. Data mapping.

Building element Construction Process

Building element X Process1 Process2 ĂĂ Process j ĂĂ Process m

Occupation Occupation Occupation Occupation

…… …… …… ……
1 1 i code

Likelihood Consequence Exposure Likelihood Consequence Exposure ……

FFI1 DAIFI1 EI1=∑ Unit labor hours *Quantity EIi ……

RFFI1 RDAIFI1 RFS1 RDAIS1 REI1

RFR1 RDAIR1

TFR TDAIR

Fig. 3. Logic diagram of data analysis.

/ /
RDAISi = MDAi MDAaverage (6) REI i = EI i EI average (9)

The exposure index for occupation i (EIi) is determined by the Safety risk is identified as the product of likelihood, consequence and
quantity (Qj) related to the material of construction process j, and the exposure. The relative fatality risk (RFRi) and relative days-away injury
unit labor hours involved in construction process j. Moreover, the risk (RDAIRi) for occupation i are proposed, as shown in Eqs. (10) and
average exposure index (EIaverage) is determined by the total exposure (11). A construction process usually requires that multiple occupations
index (EItotal) and the total number of occupations (n) in the construction to collaborate, so the total risk of the construction process is the sum of
process. To compare with the average exposure index of the occupa­ the occupational risks. The total fatality risk (TFR) and total days-away
tions, the relative exposure index for occupation i (REIi) is proposed, as injury risk (TDAIR) are proposed, as shown in Eq. (12) and Eq. (13),
shown in Eqs. (7), (8) and (9). respectively.
∑
m
RFRi = RFFI i × REI i × RFSi (10)
EI i = Qj × Unit labor hoursj (7)
(11)
j=1
RDAIRi = RDAIFI i × REI i × RDAISi
/
EI average = EI total n (8)

Revit modeling platform 3D model

RFFI
Automatic calculation of
RFS
Building component BIM for occupational construction safety risk:
RDAIFI
family library safety risk assessment TFR
RDAIS
TDAIR
REI

Parameters of geometric Extract and update

dimension information
Custom parameters of
family objects

Construction Planning Injury and Fatality

Database Database

Fig. 4. Workflow of automatic calculation with construction safety risk.

5
Y. Lu et al.
Fig. 5. User interface.
∑
n calculation process of construction safety risk is shown in Fig. 4.
TFR = RFRi (12)
i=1
∑
n To realize the proposed theoretical framework and embed the risk
TDAIR = RDAIRi (13)
i=1
3.2. Developed plug-in
Autodesk Revit is an application platform that integrates the func­
tions of architectural design, structural design and equipment and
pipeline design. All series of Revit provide an application programming
interface (API). Users can integrate the application into Revit via Revit
API for further development. This paper combines BIM with specific
casualty data and construction planning data and designs a BIM-based
safety risk automated calculation plug-in, so that the visual and auto­
mated results can improve the efficiency of selecting design alternatives
that optimize construction safety.
3.2.1. Plug-in description
The objective of this plug-in is to realize automatic calculation of
safety risk in construction processes. The plug-in needs occupational
injury, fatality and specific construction planning data. Quantities can
be extracted from the BIM model. Occupational injury and fatality data
can be retrieved from the historical Injury and Fatality Database. The
specific construction processes, occupations and unit labor hours can be
obtained from the Construction Planning Database. The automatic
6
Automation in Construction 124 (2021) 103553
3.2.2. Main development
data analysis into the BIM model, a functional plug-in is designed and
developed. As previously mentioned, relevant risk data can be collected
using historical information. The real challenge is to obtain the quan­
tities of the construction materials in BIM and match the unit labor
hours. This study uses Revit 2018 to build the model and C# language
for further development based on Microsoft Visual Studio 2017. Func­
tion extension can be achieved in two ways. The first way is the external
command mode, the second way is the external application mode. This
paper uses the external command mode to extend the specific func­
tionality of Revit. Querying or updating the historical data in the injury,
fatality and construction planning database is realized by Sequential
Query Language (SQL) commands.
The main development process consists of the following three steps:
Step1: Constructing a BIM model that integrates the Construction Plan­
ning Database based on custom family modeling technology.
BIM technology is a modeling technology that is oriented toward
various family objects. Based on the custom family modeling technology
of Revit software, this study encodes all components in the Construction
Planning Database, and conducts the modeling of standardized families
according to the dimensions and materials provided by RSMeans to form
a complete BIM family library for building projects. By creating shared
parameters for each family, the two parameters in the Construction
Planning Database—required labor resources and unit labor hours—are
Y. Lu et al. Automation in Construction 124 (2021) 103553

Table 2
Available data.
BIM model Construction process Quantity Required labor Unit labor
resource hours

Plan A
Shop-Fabricated Wood Trusses 5 2 Carpenters 0.800 h/
Roof Trusses 2 Construction Ea.
Common wood, 2′′ × 4′′ metal plate laborers
connected, 24” O.C., 4/12 slope, 1 Construction
1’overhang, 32′ span equipment operator
Prefinished Plywood Paneling 3202 2 Carpenters 0.032 h/
Paneling, Plywood S.F.
Plywood, prefinished, 1/4”thick, 4′ × 8′
sheets with vertical grooves. Birch- faced,
minimum
Damp proofing 3202 1 Roofer 0.012 h/
Bituminous Asphalt Coating S.F.
Brushed on, below grade, 1 coat
Metal Shingles 3202 1 Carpenter 0.016 h/
Aluminum shingles S.F.
Mill finish, 0.019 thick
Plan B
Shop-Fabricated Wood Trusses 5 2 Carpenters 0.889 h/
Roof Trusses 2 Construction Ea.
Common wood, 2′′ × 4′′ metal plate laborers
connected, 24” O.C., 8/12 slope, 1 Construction
1’overhang, 32′ span equipment operator
Prefinished Plywood Paneling 3651 2 Carpenters 0.038 h/
Paneling, Plywood S.F.
Plywood, prefinished, 3/4′′ thick, stock
grades, minimum
Damp proofing 3651 1 Roofer 0.016 h/
Bituminous Asphalt Coating S.F.
Brushed on, below grade, 2 coats
Metal Shingles 3651 1 Carpenter 0.016 h/
Aluminum shingles S.F.
Mill finish, 0.020 thick

extended to the attributes of the corresponding coded family. The
element filter (FilteredElementCollector) in Revit API is used to traverse
all the objects in the BIM model and return the objects that satisfy the
conditions to extract the number of standardized components and
required completion hours.
Step2: Realizing information interaction between databases and BIM
model based on Revit DB Link.
To calculate the safety risks of all occupations at a construction site,
the Construction Planning Database and Injury and Fatality Database
need to be connected by the occupations ID, which extend the RFFI,
RDAIFI, RFS, and RDAIS data of each occupation in the Injury and Fa­
tality Database to the occupations’ information. Considering that all
casualty data are updated annually, the Revit DB Link, which is a plug-in
that is often used to maintain the connection between the BIM model
and the SQL Server database, can be used to import the updated data to
the BIM model to render the calculation of occupational safety risks
more accurate.
Step3: Inputting the calculation results on the plug-in interface based on
Visual Studio.
Windows Presentation Foundation is employed to realize this func­
tion. Windows Presentation Foundation provides a unified program­
ming model, language and framework, which can be used to design a
new multimedia interactive user graphical interface. When using
Microsoft Visual Studio and Revit API to develop Windows Presentation
Foundation, users can create interfaces that meet their needs and access
all information obtained from the BIM model, the Injury and Fatality
Database and the Construction Planning Database. In this study, all the
formulas in Section 3.1.3 are compiled into callable functions in Visual
Studio. After reading the RFFI, RFS, RDAIFI, RDAIS and EI data that are
synchronously associated with the occupations ID in the plug-in inter­
face, the calculation results of REI, RFR, RDAIR, TFR and TDAIR are
automatically fed to the plug-in interface through the Revit API.
3.2.3. User interface
Once a building design model is loaded into Revit, the user interface
of the plug-in can be launched through the add-in menu. As shown in
Fig. 5, the interface consists of four modules. In the basic information
module, the interface automatically obtains the ID and the name of the
building element selected by the user. In the second module, users can
enter and select the intended construction process. The number of
workers, occupation types and unit labor hours in the Construction
Planning Database can be accessed according to the chosen construction
materials. The quantities of the construction materials are automatically
calculated. Different quantity calculation methods exist for different
construction components. Three types of construction materials exist;
the quantities of these materials are presented by the number, length
and surface area. 1) The number of certain materials can be obtained
directly in the self-contained list of building categories. 2) The lengths of
certain materials can be identified by the automatic dimensioning
function in Revit. 3) The surface areas of certain materials can be
calculated by programming in Microsoft Visual Studio and using the
external command mode to load, so as to achieve Revit secondary
development. In the third module, the interface can automatically
retrieve and calculate the corresponding relative fatality frequency
index (RFFI), relative fatality severity (RFS), relative days-away injury
frequency index (RDAIFI), and relative days-away injury severity
(RDAIS) within the Injury and Fatality Database. The exposure index (EI)
and relative exposure index (REI) can also be calculated automatically
with the data in the second module. The relative fatality risk (RFR) and
relative days-away injury risk (RDAIR) can be calculated with the rela­
tive fatality frequency index (RFFI), relative fatality severity (RFS),
relative days-away injury frequency index (RDAIFI), relative days-away
injury severity (RDAIS), and relative exposure index (REI). In the last
module, the interface can automatically calculate the total fatality risk
(TFR) and total days-away injury risk (TDAIR) of these occupations.

7
Y. Lu et al. Automation in Construction 124 (2021) 103553

Table 3
Relative frequency and severity for construction occupations.
Construction occupations (SOC code) Employment size Fatality Days-away injury

Number RFFI RFS Number MDA RDAIFI RDAIS

Construction trades workers (47–2000) 4,497,490 731 1 1 61,150 11 1.0 1.0

Boilermakers (47–2010) 13,870 1 0.4 1 130 23 0.7 2.1
Brickmasons, blockmasons, and stonemasons (47–2020) 76,240 21 1.7 1 940 12 0.9 1.1
Carpenters (47–2030) 718,730 86 0.7 1 12,640 10 1.3 0.9
Carpet, floor, and tile installers and finishers (47–2040) 83,740 7 0.5 1 2380 4 2.1 0.4
Cement masons, concrete finishers, and terrazzo workers (47–2050) 189,130 11 0.4 1 840 25 0.3 2.3
Construction laborers (47–2060) 1,001,470 259 1.6 1 20,430 10 1.5 0.9
Construction equipment operators (47–2070) 433,690 51 0.7 1 2350 25 0.4 2.3
Drywall installers, ceiling tile installers, and tapers (47–2080) 120,220 13 0.7 1 1300 20 0.8 1.8
Electricians (47–2110) 655,840 80 0.8 1 6350 7 0.7 0.6
Glaziers (47–2120) 50,940 0 0 1 560 6 0.8 0.5
Insulation workers (47–2130) 56,440 10 1.1 1 710 7 0.9 0.6
Painters and paperhangers (47–2140) 231,200 31 0.8 1 2130 9 0.7 0.8
Pipelayers, plumbers, pipefitters, and steamfitters (47–2150) 476,140 37 0.5 1 5780 15 0.9 1.4
Plasterers and stucco masons (47–2160) 24,870 0 0 1 260 5 0.8 0.5
Reinforcing iron and rebar workers (47–2170) 18,360 1 0.3 1 120 30 0.5 2.7
Roofers (47–2180) 128,680 96 4.6 1 2060 8 1.2 0.7
Sheet metal workers (47–2210) 131,570 5 0.2 1 1280 30 0.7 2.7
Structural iron and steel workers (47–2220) 77,410 15 1.2 1 800 28 0.8 2.5
Solar photovoltaic installers (47–2230) 8950 4 2.7 1 80 15 0.7 1.4

Building element Construction Process

Roof
(plan A) Shop-Fabricated Wood Trusses Prefinished Plywood Paneling Dampproofing Metal Shingles

1
2
2 Construction 2 1 1
Construction
Carpenters equipment Carpenters Roofer Carpenter
laborers
operator

Likelihood Consequence Exposure Likelihood Consequence Exposure Likelihood Consequence Exposure Likelihood Consequence Exposure

157.7 4.0 4.0 38.4

0.7 1.3 1 0.9 3.2 1.6 1.5 1 0.9 0.1 0.7 0.4 1 2.3 0.1 4.6 1.2 1 0.7 0.8

2.4 3.8 0.1 0.1 0.1 0.1 3.6 0.7

6.2 4.6

Roof
(plan B) Shop-Fabricated Wood Trusses Prefinished Plywood Paneling Dampproofing Metal Shingles

1
2
2 Construction 2 1 1
Construction
Carpenters equipment Carpenters Roofer Carpenter
laborers
operator

Likelihood Consequence Exposure Likelihood Consequence Exposure Likelihood Consequence Exposure Likelihood Consequence Exposure

201.6 4.4 4.4 58.4

0.7 1.3 1 0.9 3.1 1.6 1.5 1 0.9 0.1 0.7 0.4 1 2.3 0.1 4.6 1.2 1 0.7 0.9

2.3 3.6 0.1 0.1 0.1 0.1 4.1 0.8

6.6 4.6

Fig. 6. Construction safety risk value for plan A and B.

4. Case study
To demonstrate the theoretical framework and operability of the
developed plug-in, a case study is presented. This case is a three-story
private building that is mainly employed to analyze the construction
safety risk involved in roofing work. Roofing work includes tasks such as
cutting, aligning and attaching plywood sheets, asphalt felt, shingles,
flashing to the roof structure. In this case, two construction plans have
been established, namely, plan A and plan B, as shown in Table 2. The
Construction Planning Database provides required labor resource and
unit labor hours for construction processes. According to quantity
calculation rules of the US Construction Specification Institute, the
quantity of construction material is calculated based on their actual
data. In the BIM environment, the quantities of the construction mate­
rials can be automatically calculated. In this case study, the quantity of
wood trusses is presented by the number of this kind of material. The
quantities of plywood, damp proofing and metal shingles of the roof
element are calculated by the surface area of the single-story roof. The

8
Y. Lu et al.
Fig. 7. Automated calculation results of plan A.
corresponding secondary development code mainly realizes the
following functions: (1) identify all these layers for calculation, (2)
determine the angles between the normal vector of each surface and the
positive Z-axis, (3) identify all faces with angles greater than 0, and (4)
sum the areas of all these selected surfaces.
According to Eqs. (1), (2), (3), (4), (5) and (6), Table 3 shows the
likelihood and consequence index of 19 construction occupations using
the injury and fatality data of 2018. Only 4 kinds of occupations are
shown in this case. The findings conclude that the relative fatality fre­
quency index (RFFI) of roofers (47–2180) and the relative days-away
injury frequency index (RDAIFI) of construction laborers (47–2060)
are relatively high. Additionally, the relative days-away injury severity
(RDAIS) of construction equipment operators (47–2070) is 3.3 times
that of roofers (47–2180).
The relative exposure index (REI), relative fatality risk (RFR) and
relative days-away injury risk (RDAIR) for construction occupations in
plan A and plan B are calculated by Eqs. (7), (8), (9), (10) and (11). The
total fatality risk (TFR) and total days-away injury risk (TDAIR) for
roofing work in plan A and plan B can be calculated by Eqs. (12) and
(13), respectively, as shown in Fig. 6.
In this process, Revit API is used to convert the data into computer
language. The specific operations are listed as follows: (1) selecting roof
element; (2) loading the developed plug-in; (3) selecting the corre­
sponding construction process in the user interface; (4) clicking the
“Calculation” button in sequence to realize automatic calculation of the
construction safety risk. The automatic calculation results with plan A
and plan B are shown in Fig. 7 and Fig. 8, respectively.
9
Automation in Construction 124 (2021) 103553
The manual calculation result matches the automatic calculation
result obtained from the developed plug-in. This result validates that
using this plug-in in Revit can quickly calculate the construction safety
risk of different design plans in a very short time and shows that the
plug-in developed in this paper has excellent engineering application
value.
The final results obtained from the developed plug-in of this case
show that plans A and B have the same total days-away injury risk
(TDAIR), but plan A has a lower total fatality risk (TFR). Architects and
structural designers should quickly select design plan A. Fig. 6 shows
that carpenters (47–2030) and roofers (47–2180) have a higher relative
fatality risk (RFR) and relative days-away injury risk (RDAIR) in plan A.
Thus, the architects and structural designers should focus on the con­
struction work of carpenters and roofers. Effective protective design
should be considered at the design stage.
5. Discussion
5.1. Generality of the methodology
All the data of this research were collected from the US construction
industry. The numbers of occupational days-away injuries and fatalities
collected from the 2018 BLS IIF Program and the unit labor hours,
required labor resources and construction materials extracted from the
RSMeans Database were based on the US construction industry.
Different countries have different casualty statistics, which reflects
various safety management situations. Abueisheh et al. [4] investigated
Y. Lu et al.
Fig. 8. Automated calculation results of plan B.
the implementation situation of design for safety in the Palestine con­
struction industry which differed from the developed countries and
generated effective strategies that have been implemented in Palestine.
Therefore, the calculation results of the case study cannot reflect the
actual safety risk situation in other countries. However, the proposed
quantitative risk assessment method in Section 3 can be directly applied
to other countries if they have detailed statistics on injuries and fatalities
and the quota system data of different construction occupations.
5.2. Integration of TFR and TDAIR
When selecting design alternatives in practice, in some situations,
the total fatality risk (TFR) index of one design plan is higher than
another design plan, while the total days-away injury risk (TDAIR) index
is lower. More than two design alternatives may also exist. In these
possible scenarios, to the separate calculation of the values of the two
indexes is not enough. The designers should comprehensively consider
the plan according to the policy, expert opinions and other circum­
stances. By the certain weight determination method, the comprehen­
sive objective function of risk can be determined by combining both the
TFR index and the TDAIR index. Similarly, decisions can be made based
on the values of this objective function when more than two design al­
ternatives are available.
To determine the specific weights of these two indexes in the
objective function, many factors have to be considered. Differences may
exist in different countries, different regions, and different companies.
From the perspective of countries and regions, different social, political
10
Automation in Construction 124 (2021) 103553
and economic situations may also be considered. From the perspective of
construction companies, policy requirements, the extent to which death
and injury indexes affect the reputation of the company, and the legal
penalties for death and injury need to be considered while determining
the weights.
6. Conclusion
As a proactive safety intervention for reducing the construction
safety risk, the use of CHPtD is recommended in practice. The technical
issue that underlies CHPtD is how to use risk-based methods to select
design alternatives that optimize construction safety. This paper pro­
posed a quantitative construction safety risk assessment method that
consists of three indexes: likelihood, consequence and exposure. These
indexes can be calculated using occupational injury, fatality and specific
construction planning data which are accurate and objective. These data
are closely related to the design features of a building project. The
technical issue that underlies CHPtD is how to use novel tools to improve
the selection of an optimal design solution. A plug-in that links BIM with
safety risk data has been developed in Autodesk Revit. This plug-in can
automatically calculate construction safety risk to quickly help archi­
tects and structural designers select design alternatives that optimize
construction safety. This approach resolves the problem that the archi­
tects and structural designers cannot execute CHPtD in practice due to a
lack of safety knowledge. In addition, a case study is presented to
demonstrate the feasibility and effectiveness of the proposed method for
construction safety risk assessment at the design stage. The final results
Y. Lu et al.
obtained from the developed plug-in of the case show that plan A has a
lower total fatality risk (TFR). In plan A, carpenters (47–2030) and
roofers (47–2180) have a higher relative fatality risk (RFR) and relative
days-away injury risk (RDAIR). The architects and structural designers
can consider an effective protective design for the construction work of
carpenters and roofers at the design stage.
The data of fatality and days-away injury were collected by 19
construction occupations in this paper. For the same occupation, no
further analysis of safety risks arose from different construction activ­
ities. Neither BLS SOII nor BLS CFOI provides risk data according to the
causes of death or injury of construction workers with the same occu­
pation. Using the occupational casualty data of different construction
activities for further analysis after refining the statistical caliber is more
reasonable. Furthermore, in the traditional method of procuring-
construction, design-bid-build, the separation of design and construc­
tion precludes a designer from considering construction site safety
during the design process. In practice, the effect of the CHPtD concept
may gain momentum especially in Design-Build (DB), Engineering
Procurement Construction (EPC) or Integrated Project Delivery (IPD)
project delivery methods where the integration of design and con­
struction can be realized [11]. In the design phase, the team consists of
not only designers, but also architects and engineers [58], which enable
a collaborative team relationship. The engineers can provide detailed
data of the construction process in advance for better implementation of
CHPtD.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
The authors wish to express their sincere gratitude to the National
Social Science Foundation of China (Grant No. 19BGL238), the Natural
Science Foundation of Jiangsu Province, China (Grant No. BK20181276)
and Social Science Foundation of Jiangsu Province, China (Grant No.
17GLB002).
References
[1] Ministry of Housing and Urban-Rural Development (MOHURD), Available online:
http://www.mohurd.gov.cn/wjfb/201903/t20190326_239913.html, 2019
(accessed on 2 July 2020).
[2] BLS, Injuries, Illnesses, and Fatalities, Available online: https://www.bls.gov/iif/,
2017 (accessed on 2 July 2020).
[3] Health and Safety Executive, Workplace Fatal Injuries in Great Britain, Available
online: http://www.hse.gov.uk/statistics/pdf/fatalinjuries.pdf, 2019 (accessed on
2 July 2020).
[4] Q. Abueisheh, P. Manu, A.M. Mahamadu, C. Cheung, Design for safety
implementation among design professionals in construction: The context of
Palestine, Safety, Science 128 (2020), https://doi.org/10.1016/J.
SSCI.2020.104742.
[5] J.A. Gambatese, Research issues in prevention through design, J. Saf. Res. 39 (2)
(2008) 153–156, https://doi.org/10.1016/j.jsr.2008.02.012.
[6] W. Creaser, Prevention through design (PtD) safe design from an Australian
perspective, J. Saf. Res. 39 (2) (2008) 131–134, https://doi.org/10.1016/j.
jsr.2008.02.018.
[7] F.A. Manuele, Prevention through design (PtD): history and future, J. Saf. Res. 39
(2) (2008) 127–130, https://doi.org/10.1016/j.jsr.2008.02.019.
[8] M.D. Martinez Aires, M.C. Rubio Gamez, A. Gibb, Prevention through design: the
effect of European directives on construction workplace accidents, Saf. Sci. 48 (2)
(2010) 248–258, https://doi.org/10.1016/j.ssci.2009.09.004.
[9] J.F. Yuan, X.W. Li, X. Xiahou, N. Tymvios, Z. Zhou, Q.M. Li, Accident prevention
through design (PtD): integration of building information modeling and PtD
knowledge base, Autom. Constr. 102 (2019) 86–104, https://doi.org/10.1016/j.
autcon.2019.02.015.
[10] T.M. Toole, J. Gambatese, The trajectories of prevention through design in
construction, J. Saf. Res. 39 (2) (2008) 225–230, https://doi.org/10.1016/j.
jsr.2008.02.026.
11
Automation in Construction 124 (2021) 103553
[11] M. Behm, Linking construction fatalities to the design for construction safety
concept, Saf. Sci. 43 (8) (2005) 589–611, https://doi.org/10.1016/j.
ssci.2005.04.002.
[12] S. Zhang, K. Sulankivi, M. Kiviniemi, I. Romo, C.M. Eastman, J. Teizer, BIM-based
fall hazard identification and prevention in construction safety planning, Saf. Sci.
72 (2015) 31–45, https://doi.org/10.1016/j.ssci.2014.08.001.
[13] A. Golabchi, S.U. Han, S. Abourizk, A simulation and visualization-based
framework of labor efficiency and safety analysis for prevention through design
and planning, Autom. Constr. 96 (DEC) (2018) 310–323, https://doi.org/10.1016/
j.autcon.2018.10.001.
[14] N. Tymvios, J. Gambatese, D. Sillars, Designer, Contractor, and Owner Views on
the Topic of Design for Construction Worker Safety, Construction Research
Congress, American Society of Civil Engineers, 2012, pp. 341–355, https://doi.
org/10.1061/9780784412329.035.
[15] J.A. Gambatese, M. Behm, S. Rajendran, Design’s role in construction accident
causality and prevention: perspectives from an expert panel, Saf. Sci. 6 (4) (2008)
675–691, https://doi.org/10.1016/j.ssci.2007.06.010.
[16] D. Hardison, M. Hallowell, Construction hazard prevention through design: review
of perspectives, evidence, and future objective research agenda, Saf. Sci. 120
(2019) 517–526, https://doi.org/10.1016/j.ssci.2019.08.001.
[17] R.S. Nizam, C. Zhang, L. Tian, A BIM based tool for assessing embodied energy for
buildings, Build. Energy Efficiency 170 (2018) 1–14, https://doi.org/10.1016/J.
ENBUILD.2018.03.067.
[18] K.F. Chien, Z.H. Wu, S.C. Huang, Identifying and assessing critical risk factors for
BIM projects: empirical study, Autom. Constr. 45 (2014) 1–15, https://doi.org/
10.1016/J.AUTCON.2014.04.012.
[19] W. Solihin, J. Dimyadi, Y.C. Lee, C. Eastman, R. Amor, Simplified schema queries
for supporting BIM-based rule-checking applications, Autom. Constr. 117 (2020),
https://doi.org/10.1016/J.AUTCON.2020.103248.
[20] D. Langford, S. Rowlinson, E. Sawacha, Safety behaviour and safety management:
its influence on the attitudes of workers in the UK construction industry, Eng.
Constr. Archit. Manag. 7 (2) (2000) 133–140, https://doi.org/10.1108/eb021138.
[21] B. Zhang, Z. Chu, L. Cheng, N. Zou, A quantitative safety regulation compliance
level evaluation method, Saf. Sci. 112 (2019) 81–89, https://doi.org/10.1016/j.
ssci.2018.10.016.
[22] M. Gunduz, H. Laitinen, Construction safety risk assessment with introduced
control levels, J. Civ. Eng. Manag. 24 (1) (2018) 11–18, https://doi.org/10.3846/
JCEM.2018.284.
[23] V. Zamani, S.Y. Banihashemi, A. Abbasi, How can communication networks among
excavator crew members in construction projects affect the relationship between
safety climate and safety outcomes? Saf. Sci. 128 (2020) https://doi.org/10.1016/
j.ssci.2020.104737.
[24] A. Pinto, QRAM, a qualitative occupational safety risk assessment model for the
construction industry that incorporate uncertainties by the use of fuzzy sets, Saf.
Sci. 63 (2014) 57–76, https://doi.org/10.1016/j.ssci.2013.10.019.
[25] T. Dejus, J. Antucheviciene, Assessment of health and safety solutions at a
construction site, J. Civ. Eng. Manag. 19 (5) (2013) 728–737, https://doi.org/
10.3846/13923730.2013.812578.
[26] S. Aminbakhsh, M. Gunduz, R. Sonmez, Safety risk assessment using analytic
hierarchy process (AHP) during planning and budgeting of construction projects,
J. Saf. Res. 46 (2013) 99–105, https://doi.org/10.1016/j.jsr.2013.05.003.
[27] G.E. Gurcanli, S. Bilir, M. Sevim, Activity based risk assessment and safety cost
estimation for residential building construction projects, Saf. Sci. 80 (2015) 1–12,
https://doi.org/10.1016/j.ssci.2015.07.002.
[28] M.R. Hallowell, J.A. Gambatese, Activity-based safety risk quantification for
concrete formwork construction, J. Constr. Eng. Manag. 10 (2009) 990, https://
doi.org/10.1061/(ASCE)CO.1943-7862.0000071.
[29] B. Esmaeili, M.R. Hallowell, B. Rajagopalan, Attribute-based safety risk assessment.
I: analysis at the fundamental level, J. Constr. Eng. Manag. 141 (8) (2015), https://
doi.org/10.1061/(ASCE)CO.1943-7862.0000980.
[30] S. Baradan, M.A. Usmen, Comparative injury and fatality risk analysis of building
trades, J. Constr. Eng. Manag. 132 (5) (2006) 533–539, https://doi.org/10.1061/
(ASCE)0733-9364(2006)132:5(533).
[31] S. Choe, F. Leite, Assessing safety risk among different construction trades:
quantitative approach, J. Constr. Eng. Manag. 143 (2017), https://doi.org/
10.1061/(ASCE)CO.1943-7862.0001237.
[32] Y. Sun, D. Fang, S. Wang, M. Dai, X. Lv, Safety risk identification and assessment
for Beijing Olympic venues construction, J. Manag. Eng. 24 (1) (2008) 40–47,
https://doi.org/10.1061/(ASCE)0742-597X(2008)24:1(40).
[33] B. Esmaeili, M.R. Hallowell, B. Rajagopalan, Attribute-based safety risk assessment.
II: predicting safety outcomes using generalized linear models, J. Construct. Eng.
Manag. 141 (8) (2015), https://doi.org/10.1061/(ASCE)CO.1943-7862.0000981.
[34] V. Sousa, N.M. Almeida, L.A. Dias, Risk-based management of occupational safety
and health in the construction industry - part 1: background knowledge, Saf. Sci.
66 (2014) 75–86, https://doi.org/10.1016/J.SSCI.2014.02.008.
[35] B. Esmaeili, M. Hallowell, Integration of safety risk data with highway construction
schedules, Constr. Manag. Econ. 31 (6) (2013) 528, https://doi.org/10.1080/
01446193.2012.739288.
[36] M.R. Hallowell, W. Blaney, J. Teizer, Application of sensing technology to safety
management, in: Construction Research Congress 2010: Innovation for Reshaping
Construction Practice - Proceedings of the 2010 Construction Research Congress,
2010, pp. 31–40, https://doi.org/10.1061/41109(373)4.
[37] N. Azadeh-Fard, A. Schuh, E. Rashedi, J.A. Camelio, Risk assessment of
occupational injuries using accident severity grade, Saf. Sci. 76 (2015) 160–167,
https://doi.org/10.1016/j.ssci.2015.03.002.
Y. Lu et al.
[38] S.R. Mohandes, H. Sadeghi, A. Mahdiyar, S. Durdyev, Assessing construction
labors’ safety level: a fuzzy MCDM approach, J. Civ. Eng. Manag. 26 (2) (2020)
175–188, https://doi.org/10.3846/JCEM.2020.11926.
[39] The Center for Construction Research and Training, Construction Chart Book,
Available online: https://www.cpwr.com/sites/default/files/publications/The_6th
_Edition_Construction_eChart_Book.pdf, 2018 (accessed on February 2018).
[40] U.S. Department of Labor, Employment and Training Administration, O*NET,
2017. Available online: http://www.onetonline.org/ (Accessed on April 2017).
[41] M.R. Hallowell, A Formal Model of Construction Safety and Health Risk
Management, Ph.D, Thesis, Oregon State University, Oregon State, USA, 2008, htt
p://hdl.handle.net/1957/8541.
[42] O.A. Jannadi, S. Almishari, Risk assessment in construction, J. Constr. Eng. Manag.
129 (5) (2003) 492–500, https://doi.org/10.1061/(ASCE)0733-9364(2003)129:5
(492). http://hdl.handle.net/1957/8541.
[43] G. Reniers, W. Dullaert, B. Ale, K. Soudan, Developing an external domino accident
prevention framework: Hazwim, J. Loss Prev. Process Ind. 18 (3) (2005) 127–138,
https://doi.org/10.1016/j.jlp.2005.03.002.
[44] F. Henselwood, G. Phillips, A matrix-based risk assessment approach for addressing
linear hazards such as pipelines, J. Loss Prev. Process Ind. 19 (5) (2006) 433–441,
https://doi.org/10.1016/j.jlp.2005.10.005.
[45] P.K. Marhavilas, D.E. Koulouriotis, A risk-estimation methodological framework
using quantitative assessment techniques and real accidents’ data: application in
an aluminum extrusion industry, J. Loss Prev. Process Ind. 21 (6) (2008) 596–603,
https://doi.org/10.1016/j.jlp.2008.04.009.
[46] A.C.P. Frijters, P.H.J.J. Swuste, Safety assessment in design and preparation phase,
Saf. Sci. 46 (2) (2008) 272–281, https://doi.org/10.1016/j.ssci.2007.06.032.
[47] M. Gangolells, M. Casals, N. Forcada, X. Roca, A. Fuertes, Mitigating construction
safety risks using prevention through design, J. Saf. Res. 41 (2) (2010) 107–122,
https://doi.org/10.1016/j.jsr.2009.10.007.
[48] V. Dharmapalan, J.A. Gambatese, J. Fradella, A.M. Vahed, Quantification and
assessment of safety risk in the design of multistory buildings, J. Constr. Eng.
Manag. 141 (2015), https://doi.org/10.1061/(ASCE)CO.1943-7862.0000952.
12
Automation in Construction 124 (2021) 103553
[49] P.K. Marhavilas, D. Koulouriotis, V. Gemeni, Risk analysis and assessment
methodologies in the work sites: on a review, classification and comparative study
of the scientific literature of the period 2000-2009, J. Loss Prev. Process Ind. 24 (5)
(2011) 477–523, https://doi.org/10.1016/j.jlp.2011.03.004.
[50] S. Azhar, M. Khalfan, T. Maqsood, Building information modelling (BIM): now and
beyond, Construct. Econom. Build. 12 (4) (2015), https://doi.org/10.5130/AJCEB.
v12i4.3032.
[51] J.A. Gambatese, J.W. Hinze, C.T. Haas, Tool to design for construction worker
safety, J. Archit. Eng. 3 (1) (1997) 32–41, https://doi.org/10.1061/(ASCE)1076-
0431(1997)3:1(32).
[52] Z. Hu, J. Zhang, Z. Deng, Construction process simulation and safety analysis based
on building information model and 4D technology, Tsinghua Sci. Technol. 13 (S1)
(2008) 266–272, https://doi.org/10.1016/S1007-0214(08)70160-3.
[53] H. Guo, Y. Yu, W. Zhang, Y. Li, BIM and safety rules based automated identification
of unsafe design factors in construction, Procedia Eng. 164 (2016) 467–472,
https://doi.org/10.1016/j.proeng.2016.11.646.
[54] M.A. Hossain, E.L.S. Abbott, D.K.H. Chua, T.Q. Nguyen, Y.M. Goh, Design-for-
safety knowledge library for BIM-integrated safety risk reviews, Autom. Constr. 94
(2018) 290–302, https://doi.org/10.1016/j.autcon.2018.07.010.
[55] J.P. Cortés-Pérez, A. Cortés-Pérez, P. Prieto-Muriel, BIM-integrated management of
occupational hazards in building construction and maintenance, Autom. Constr.
113 (2020), https://doi.org/10.1016/J.AUTCON.2020.103115.
[56] Z. Jin, J. Gambatese, D. Liu, V. Dharmapalan, Using 4D BIM to assess construction
risks during the design phase, Eng. Constr. Archit. Manag. 26 (11SI) (2019)
2637–2654, https://doi.org/10.1108/ECAM-09-2018-0379.
[57] RSMeans From Gordian, The 2020 RSMeans Building Construction Costs Book,
Available Online, https://www.rsmeans.com/products/books/2020-building-cons
truction-costs-book, 2020.
[58] T.M. Toole, J.A. Gambatese, D.A. Abowitz, Owners’ role in facilitating prevention
through design, J. Prof. Issues Eng. Educ. Pract. 143 (2017), https://doi.org/
10.1061/(ASCE)EI.1943-5541.0000295.