Advanced Engineering Informatics 28 (2014) 66–80

Contents lists available at ScienceDirect

Advanced Engineering Informatics

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

Automatic design and planning of scaffolding systems using building

information modeling
Kyungki Kim, Jochen Teizer ⇑
School of Civil and Environmental Engineering, Georgia Institute of Technology, 790 Atlantic Dr. N.W., Atlanta, GA 30332-0355, USA

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

Article history: Considering their significant impact on construction projects, scaffolding as part of the temporary facil-
Received 7 December 2012 ities category in construction must be thoroughly designed, planned, procured, and managed. The current
Received in revised form 17 December 2013 practices in planning and managing scaffolding though is often manual and reactive, especially when a
Accepted 19 December 2013
construction project is already underway. Widespread results are code compliance problems, ineffi-
Available online 8 January 2014
ciency, and waste of procuring and managing material for scaffolding systems. We developed a rule-
based system that automatically plans scaffolding systems for pro-active management in Building Infor-
Keywords:
mation Modeling (BIM). The scope of the presented work is limited to traditional pipe and board scaffold-
Building information modeling
Design and construction planning
ing systems. A rule was prepared based on the current practice of planning and installing scaffolding
Geometric reasoning for automated systems. Our computational algorithms automatically recognize geometric and non-geometric conditions
modeling in building models and produce a scaffolding system design which a practitioner can use in the field. We
Rules and regulations implemented our automated scaffolding system for a commercially-available BIM software and tested it
Safety in a case study project. The system thoroughly identified the locations in need of scaffolding and gener-
Temporary facilities (scaffolding systems) ated the corresponding scaffolding design in BIM. Further results show, the proposed approach success-
fully generated a scaffolding system-loaded BIM model that can be utilized in communication, billing of
materials, scheduling simulation, and as a benchmark for accurate field installation and performance
measurement.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Scaffolding systems have significant impact on successful pro-

ject delivery. The same research report by CII interviewed 56 own-
Temporary facilities such as scaffolding systems, formwork, and er and contractor companies. Results show that these companies
shoring systems are structures that assist mainly in the placement consider scaffolding systems, and its construction and deconstruc-
of bulk materials. Most construction projects utilize temporary tion, as one of the most challenging and wasteful component
facilities frequently, and construction safety, quality, speed, and among the 16 subcategories of indirect construction cost [4]. In
profitability are impacted by using temporary facilities [1]. Some addition, other concerns were identified which matter for project
aspects of these are elaborated further. success. The Occupational Safety and Health Administration
Temporary facilities can cause spatial and temporal conflicts be- (OSHA) [5] and Workplace Health and Safety Queensland (WHSQ)
tween other temporary facilities or activities which can then lead [6,7], for example, reported that there exist repeated code-compli-
to unsafe conditions and loss of productivity [2]. According to Ra- ance problems related to scaffolding systems. OSHA reported also
tay [3], many construction disasters occur due to the failures of that 65% of the construction workers are frequently on scaffolding
temporary facilities during construction. In addition, the Construc- systems. It further states that preventing accidents associated with
tion Industry Institute (CII) highlighted in a recent research study scaffolding systems exclusively can protect workers from about
that temporary facilities form one of the top four primary cost cat- 4500 injuries and 50 deaths annually and save American employ-
egories belonging to indirect construction costs [4]. In the author’s ers spending $90 million on lost workdays [8]. Since work site pro-
opinion, temporary facilities, to which scaffolding systems belong, ductivity and safety go hand-in-hand, these statistics justify
deserve much more attention due to the impact they have on labor additional efforts towards optimizing design, planning, and utiliza-
cost and construction schedule. tion of scaffolding systems.
Despite the aforementioned issues and the importance of tem-
porary facilities to a construction project, there are currently sev-
⇑ Corresponding author. Tel.: +1 404 894 8269; fax: +1 404 894 2278. eral problems in planning and managing these in the
E-mail address: [email protected] (J. Teizer). construction industry. First, most temporary facilities (incl. scaf-
URL: http://www.teizer.com (J. Teizer). folding systems) currently lack effective front-end planning and

1474-0346/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.aei.2013.12.002
 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 67

management. Architectural and bid drawings typically do not mated solution. The section thereafter presents the proposed
incorporate temporary facilities except for exceptionally complex framework and methodology of the rule-based system for scaffold-
temporary facilities such as cofferdams [9,10]. Although some con- ing system design, planning, and management. The customization
struction plans include important temporary facilities late in the of the automated system for scaffolding systems is presented in the
construction planning process, they are often installed at construc- following section. The final section shows the implementation of
tion sites when needed but without sufficient planning effort. Fur- the scaffolding planning system for a realistic building model.
thermore, due to the lack of time and understanding, calculations The results from the case studies validate the feasibility of the pro-
and drawings of scaffolding systems submitted by temporary facil- posed system by creating a realistic construction plan containing
ity vendors are most often reviewed only to assess their impact on scaffolding system. The final section discusses limitations and con-
the permanent part of the building [9]. tributions of this research, and potential future research.
Along with the lack of front-end planning and management,
current industry practices suffer from heavy reliance on the knowl-
edge and experiences of individual engineers. Even though soft- 2. Background
ware programs exist that are specialized in designing temporary
facilities (e.g., formwork and scaffolding), the functions of these 2.1. Current industry practice in temporary facility planning and
commercially-available software programs are often limited to ra- management
pid generation of temporary facility designs. Only human cogni-
tion, based on visual analysis of the building designs or the The production plans of construction projects differ from the
construction sites, provides the basis for detecting locations where production plans in other industries such as manufacturing. In
temporary facilities are required. It is today’s human decision mak- addition to the complex and dynamic nature of construction pro-
ers that determine the proper types of temporary facilities and jects, the need for potentially extensive temporary facilities distin-
they typically generate the corresponding detailed designs. guishes construction plans from manufacturing plans.
Due to the complex nature of construction projects and poten- Manufacturing plans for mechanical components describe what
tially imperfect human judgment, there are opportunities to re- manufacturing tools, e.g. drills and mills, should be used in what
duce or eliminate potentially erroneous temporary facility order to produce the desired shapes of the components [11].
planning; especially in projects which are planned manually based On the other hand, construction plans present an execution
on visual analysis of the construction site drawings and schedules. strategy. Temporary facilities as part of plans often assist the con-
Some examples of errors are: necessary temporary facilities can be struction activities conducted by workers and equipment. Project
omitted in a construction plan; improper types of temporary facil- locations, times, and durations which require temporary facilities
ities can be selected; temporary facilities design may not reflect should be defined before they are needed in the field. Materials
the design requirements; and, accordingly, the amount of materials for temporary facilities therefore should be readily available since
and associated cost for temporary facilities cannot be estimated on time installation and avoidance of pre-mature dismantlement
accurately. are keys to a successful project. As mentioned by many practitio-
Taking into account the impact on the entire construction pro- ners, temporary facilities should be designed with the same level
ject and the deficiencies of the current practices in planning and of detail as the permanent structures to avoid schedule delay and
managing temporary facilities, the industry needs to overcome additional costs. They should also comply with regulatory design
these drawbacks by enabling thorough front-end planning of tem- requirements and best practices. The plan should include ap-
porary facilities – and in this case – scaffolding systems. proaches to optimize the uses of temporary facilities. The examples
This research addresses the articulated problems by integrating of optimization are maximization of sharing of scaffolding between
scaffolding systems into BIM-based construction planning. While a multiple tasks [12] and minimization of formwork system ele-
growing number of construction projects utilize BIM technology ments which require customized design [13]. Any error in planning
and processes already in order to incorporate temporary facilities temporary facilities can eventually lead to productivity losses and
into the construction plans, most of such planning is done manu- safety issues.
ally. Taking advantage of the rich information available in BIM In order to utilize these temporary facilities safely and effec-
models, we attempt to automate most of the processes of: assess- tively, the construction industry currently focuses on checking reg-
ing the construction site condition that changes according to the ulations and guidelines provided by OSHA. Construction
construction schedule, detecting required scaffolding systems companies already apply advanced technology, for example, many
and generating the design, visualizing them in the building model construction projects utilize model-based technology such as BIM
and construction simulation, and generating periodical utilization to incorporate major temporary facilities into their construction
schedules and reports. plans. For the construction project of the University of Baltimore’s
In order to automate these processes, our research developed an Law School, a contractor utilized three-dimensional (3D) modeling
automatic rule checking system for temporary facility planning of formwork systems and scaffolding systems. By doing so, scaf-
focusing on traditional pipe and board scaffolding systems. Decid- folding systems for floor structures were created and inserted into
ing where scaffolding systems are required and how they can be the main building model [14,15]. Other contractors created a set of
built satisfying various design and safety requirements are part parametric formwork system models [16]. As reported, when para-
of the research questions. Automated type or vendor selection of metric models are utilized in a project, savings in cost and time
the scaffolding systems was not part of scope of the work. The re- were achieved by reducing manual efforts and using the accurate
search first established an initial rule related to scaffolding system 3D visualization of the formwork designs. During the construction
installation. Then, the developed rule checking system was used to of the Tokyo Sky Tree, the contractor inserted maintenance scaf-
check the model at an early construction planning stage. The pre- folding and ladders into the tower model for construction simula-
sented results are in the form of building models and construction tion [17]. Eastman et al. [18] introduced many other industry case
simulations loaded with scaffolding systems and automatically studies for incorporating temporary facilities such as tower cranes
generated reports. and scaffolding in BIM-based construction planning. As already
This paper is organized as follows. The background section re- experienced, the development of realistic construction planning,
views the current industry practices and technology in support of accurate quantity take-offs, accurate evaluation of constructability
scaffolding system planning and discusses the need for an auto- and construction safety, and better communication facilitated by
68 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

realistic 3D visualization, yields significant benefits. Nevertheless, about the existing best practices are: estimation of material quan-
these existing approaches of integrating temporary facilities into tity and cost plans are based on sketches and other paper-based
BIM-based construction planning have several limitations. tools rather than modern information modeling and exchanges;
First, the current methods of construction companies utilize li- manual recording is performed in generating utilization reports,
braries of pre-modeled temporary facilities which are provided in inspection tags, installation, and dismantlement request forms.
BIM software packages. They use existing functions to manually These forms contain essential data such as scaffolding ID, location,
create parametric models for temporary facilities [18]. Taking into users, usages, approximate size, erection date, and dismantlement
account various possible project variations, it is impractical to pre- date. Further manual processing is required to generate reliable
pare designs manually for all types and sizes of temporary facili- information which a user may need in decision making applica-
ties. Even more difficult is making all of these use cases comply tions. Although the construction industry considers these pro-
with the existing standards, regulations, and best practices, includ- cesses and techniques as practical, certain advances in
ing the assurance for structural stability [19]. Creating parametric information technology can provide a more sophisticated method-
temporary facility models using the modeling capabilities of BIM ology that optimizes and pro-actively plans for scaffolding
software packages also involves a lot of manual efforts and pro- systems.
duces temporary facility models which today are not tested for
any code-compliance and structural stability. Furthermore, the 2.3. Existing computer-assisted approaches for designing and planning
existing approaches rely heavily on the abilities and inefficient of temporary facilities
manual efforts of individual engineers that may or may not have
safety, structural, or other critical knowledge in scaffolding design Many approaches exist that focus on expanding the benefits of
and use. In the following section, we review the leading industry information technology. BIM is one example that supports plan-
practice of scaffolding systems in order to differentiate or confirm ning and managing of temporary facilities. We reviewed existing
the already identified drawbacks. industry and academic approaches to identify the problems that
information technology can solve. Table 1 summarizes the capabil-
2.2. Leading industry practices for scaffolding system planning and ities of existing approaches.
management In order to derive needed temporary facilities from a digital
model, both geometric conditions as well as non-geometric condi-
In this section, we investigate the industry practice for planning tion (i.e., materials and construction schedule) should be analyzed
and managing scaffolding systems based on the results from an by computers automatically. Currently, no approach enabled such
existing research report [12]. Industry practices for different types a reasoning task that automatically assesses the project situations
of temporary facilities may vary because of the different character- and identifies required temporary facilities for specific construc-
istics how temporary facilities can be associated to construction tion tasks. Most of the approaches conduct this process manually
tasks. Thus, an industry practice specific to one type of temporary based on visual analysis.
facility should be investigated separately from others in order to Only the safety management tool developed by Kim and Ahn
identify the drawbacks and needs. In general, their wide use and [10] automatically recognized the perimeter of a BIM model and
the importance on construction projects justify the efforts in opti- incorporated a scaffolding system model around the building mod-
mizing the planning and management of scaffolding systems. el. This research, however, did not present any detailed methodol-
In 2012, CII conducted data collection and interviews with 56 ogy to analyze a building’s geometry.
experts and compiled the results to determine the current leading Kim and Fischer [20] established a theoretical foundation in
industry practice in estimating, controlling, and managing scaf- support of automated temporary facility type selection. Even
folding systems [12]. Fig. 1 illustrates three process flow charts though they focused on automated type selection, they suggested
for estimating, controlling, and managing scaffolding systems. In a useful method to analyze the model’s geometric condition based
the estimating stage, a periodical cost plan for scaffolding is estab- on the relationship between the work face and the base surface.
lished. Project participants then identify the scaffolding needs con- While their approach enables automated selection of temporary
sidering the preliminary construction schedule, predict monthly facility types, the work face and base surface of each situation
scaffolding usage, and conduct scaffolding cost estimating using should be specified manually by a user. Nagi et al. [13] developed
the estimating worksheet. The estimating worksheet derives an an optimization tool for the aluminum formwork layout system
estimate on quantity and cost of materials based on a vaguely de- with the objective of maximizing the area of standardized form-
fined (sketch of a) scaffolding system. In the controlling stage, the work elements compared to the area of customized formwork ele-
scaffolding utilization is tracked monthly using scaffolding utiliza- ments. The results show increased relative portion of standardized
tion reports. The actual scaffolding utilization is compared to the formwork elements compared to the customized elements. How-
initial plan that was established during the estimating stage. Vari- ever, a major drawback that remains is the system still requires
ances and forecasted uses are reported and discussed in the man- users to manually specify corners of the formwork elements to rec-
agement stage. The reports from the controlling stage are ognize the geometric information of the building objects.
reviewed. Meetings are conducted periodically to optimize the uti- Some of other approaches support the generation of temporary
lization and execute a pro-active scaffolding management. Specific facility designs automatically or manually. The tool developed by
tasks in pro-active scaffolding management include change order Kim and Ahn [10] automatically designs scaffolding systems
management, prevention of pre-mature dismantlement, and max- around a building model. CADS [21] also generates pre-defined
imization of sharing of scaffolding between different tasks [12]. types of scaffolding systems automatically around walls. While
Based on the review of the process flow charts, tools, and asso- these two approaches assist users to generate scaffolding systems
ciated considerations checklists [12], we identified that this cur- rapidly, they design scaffolding systems exclusively for walls and
rent leading industry practice related to scaffolding systems has fail to distinguish scaffolding systems for different construction
several drawbacks. These arise mainly from the heavy reliance of tasks. Scia scaffolding [22] provides functions in its user-interface
the process on manual efforts. Some of these manual efforts can that assist users to design scaffolding systems manually. It also
be associated to work tasks like: identification of scaffolding needs, provides automated code-compliance and structural stability
periodical review of utilization and cost plans, and preparation of checking for scaffolding systems. Sulankivi et al. [19] and Kim
bid documents and construction schedules. Additional concerns and Ahn [10] incorporated safety features such as guardrails into
 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 69

Fig. 1. Process flow for estimating, controlling, and managing scaffolding systems [12].

Table 1
Computer-assisted approaches for temporary facilities design and planning.

Addressed the area of Kim and Kim and Ahn Sulankivi et al. Akinci Jongeling Scia CADS Lee et al. Nagi et al.
concern Fischer [20] [10] (2009) et al. [2] et al. [24] scaffolding (2012) (2007) [13]
[22]
Needs identification  
Design generation     
Structural soundness 
Temporary facility type 
selection
4D simulation     
Safety features  
Space uses 

the temporary facility models. Lee et al. [23] developed a tool that construction projects incorporate temporary facilities into building
generates the formwork layouts based on the prioritized design models in an attempt to simulate the construction sequences real-
requirements. istically. However, the main benefit from applying model-based
Several approaches incorporate the temporary facility models technology is often limited to visualizing the temporary facilities
into the main models and 4D construction simulations. While within the model. The essential tasks for planning and managing
some of them used the temporary facility-loaded models and sim- temporary facilities have not been incorporated to date into BIM
ulations for enhanced visualization and analysis of construction and thus users cannot take full advantage of advanced information
sites [24], Akinci et al. [2] specified the space occupied by a scaf- modeling technology.
folding system in an attempt to analyze the spatial conflicts be- The current state of technology does not suggest a methodology
tween spaces occupied by other construction activities or that overcomes the explained drawbacks. Little research has been
temporary facilities. conducted to automatically analyze the geometric and non-geo-
metric conditions of temporary facilities on a construction project.
2.4. The need for an automated system for temporary facilities Since identification and designs of required temporary facilities
planning and management still requires experienced engineers to understand complex build-
ing geometry and schedule, potential exists to assist them in their
Model-based technology such as BIM is regarded as one of the decision making. Furthermore, an integrated tool that utilizes the
solutions to improve construction management. Many rich and precise information available from digital building mod-
70 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

els, assesses the project conditions, and develops a temporary facil- (2) Building model preparation: building models should provide
ity utilization plan automatically does not exist. the information necessary to execute the specific rule check-
Recognizing these needs, we attempt to develop and test a ing. The information should be automatically derived from
framework that automatically generates a temporary facility plan the design program or the rule checking system.
that supports pro-active management utilizing the rich informa- (3) Rule execution: after the pre-checking process that verifies
tion available in digital building models. Even though the frame- that the building model contains information required for
work proposed in this research can potentially provide the basis the rule checking, the rules are executed against the building
to plan many other types of temporary facilities, the current scope model.
was limited to traditional pipe and board scaffolding systems. (4) Checking result reporting: the result of executing rules such
The following section explains the algorithms we developed as ‘‘pass’’ or ‘‘fail’’ is reported in the form of graphical reports
and utilized to: (a) recognize the geometric and non-geometric or references to the source rules.
conditions in building models and (b) suggest automatically gener-
ated scaffolding system designs to a user. The presented algo-
rithms will first detect locations that require scaffolding systems 3.2. A rule-based approach for identification of required temporary
and then provide the geometric information to each location. After facilities
obtaining the geometric information that is essential for the scaf-
folding design generation, design guidelines and best practices The intention of implementing a rule-based system in this re-
were then incorporated into a construction plan to generate the search is to: (a) automatically evaluate the construction site condi-
necessary scaffolding design and utilization reports. Through the tion, (b) identify the project locations and schedules that require
implementation of the proposed framework and methodology, scaffolding systems, and (c) provide solutions automatically and
we expect to improve the current practice to become less labor- interactively. There are regulations and industry practices that
intensive and time-consuming, and instead more efficient and can be applied as rules for planning scaffolding systems. The rules
accurate. Potential contributions of this research will be discussed may provide descriptions about the conditions where a certain
later in the conclusion and discussion section. type of temporary facility is required. The rules may also vary
depending on the types of temporary facilities and the characteris-
tics of the related construction tasks. Such rules are generally inter-
3. BIM-based rule checking for scaffolding systems planning preted and translated into machine-readable codes and
implemented to check digital building models automatically. A
3.1. Existing rule-based approaches rule-based platform then visualizes the results from the rule-based
checking engine according to the progress of the construction se-
In order to automate the planning of temporary facilities, we quence. Installation time and removal of scaffolding is finally
developed an automated rule-based building model checking sys- shown in the construction schedule.
tem. The rule-based checking approach is one of the techniques
that recognize features from the available digital model geometry.
Feature recognition (FR) techniques are generally used to extract 4. Framework and methodology for automated temporary
important features that are not explicitly expressed by the model. facility design and planning
Features obtained from Computer-Aided Design (CAD) systems are
used as the input for computer-aided analysis or automated man- Fig. 2 illustrates the framework that implements the proposed
ufacturing process planning systems. They save time and reduce rule-based system for temporary facility design and planning.
potential human error [25]. Babic et al. [26] identified rule-based The initial step is to collect project data from a building model
approaches as a robust method among various FR techniques such and construction schedule. BIM and schedule information provide
as hint-based, graph-based, artificial neural network-based and geometric information such as building object shapes and their
volumetric decomposition. According to Zhang et al. [25] rule- locations within the model and non-geometric information such
based systems are flexible and rational because the rules and facts as materials, task information, and the stakeholder in charge of
are separated in the system structure. the task. This information is the essential source to automatically
Applying a rule-based approach for assessing building designs is analyze the construction site condition that changes dynamically.
regarded as a desired research direction in the architecture, engi- The step is equivalent to the building model preparation stage in
neering, and construction (AEC) industry. A rule-based checking the rule checking process explained in Section 3.1.
system is equal to software that assesses a building design based Then, the rules for detecting required temporary facilities from
on its attributes, relations, and configuration of building objects the model are established based on best practices and regulatory
without making modifications to the design. A rule-based checking rules. These are provided either by industry, for example vendors
process can also report results [27]. Building design reviews can be and suppliers, or other closely related organization, for example
conducted more quickly and reliably using a rule-based approach. OSHA. This step is equivalent to the rule interpretation and logical
There are wide uses of the rule checking approach in the AEC structuring of the rules stage in the rule checking process.
industry. An early concept design can be tested for spatial require- Using such placement rules and project information, the model
ments, circulation requirements, energy consumption, and cost geometry is interpreted to identify the required temporary facili-
[28]. In the construction stage, a building model can also be evalu- ties. This step is equivalent to the rule execution stage in the rule
ated to predict hazardous situations and eliminate them in ad- checking process.
vance [29,30]. The temporary facility design is then generated in compliance
A rule checking process can generally be structured into four with corresponding OSHA regulations and industry best practices
stages [27]: related to temporary facility design. The process of identifying
the locations in need of temporary facilities and generating the de-
(1) Rule interpretation and logical structuring of the rules: signs corresponds to the reporting stage in the rule checking pro-
building design rules written in the form of human language, cess. Project participants can finally utilize the temporary
text, table, etc. are interpreted in a form that can be pro- facility-loaded model and automatically generate reports in order
cessed by computers. to facilitate better communication and decision making.
 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 71

Fig. 2. Framework to implement a rule-based checking system for temporary facilities in BIM.

For both needs, identification of required temporary facilities who analyzed the geometric conditions of a project and outlined
and final decision making for temporary facility design selection, the automation of the temporary facility type selection.
a variety of user-defined rules can be established. For example, Shown in an example, Fig. 3 illustrates a situation in a construc-
the height of a work place, the weight of construction material to tion project where the front face of a brick wall is the work face. In
be lifted and placed on scaffolding systems, and/or a combination order to evaluate the scaffolding height in this situation, the dis-
of both can determine if a scaffolding system is required. A rule tance from the adjacent floor to the top of the work face must be
based system may further suggest multiple scaffolding options identified. This value is then compared against the pre-defined va-
for a specific work task situation, such as support or suspended lue (best practice) and it is decided if a scaffolding system is
scaffolding. User-defined criteria such as the maximum total needed or not. The example in Fig. 3 illustrates the definitions
weight and cost of the scaffolding materials may further drive we have established for such calculation. Assuming there is only
the selection of the preferred scaffolding system type. Eventually, one floor (indoors), the scaffolding height is equal to the work face
automated selection or user-preferences for selecting a scaffolding height. Should a scaffolding system be needed outdoors for a two-
system can be incorporated into the decision making framework story building and the base face is on the ground level, the sum of
which scaffolding systems to use. the work face height and the distance to work face becomes the scaf-
folding height. Ceilings can also be considered in the same way as
the base surfaces. This would become an issue should a suspended
5. Development of the automated design and planning system scaffolding system be needed, which is a type of scaffolding system
for scaffolding that remains outside of the scope of the presented work.

5.1. Scaffolding system placement rules 5.2. Prototype of a rule-based algorithm for scaffolding system
installation
Many scaffolding systems provide construction workers with an
elevated platform until the related work task is completed. The fo- The proposed rule-based system is composed of two major
cus of our rule-checking system for scaffolding systems was to de- parts. One is the algorithm that implements the rule-based model
tect the instances where construction workers need to work at geometry interpretation. The other is the design and planning algo-
elevations high enough to demand an elevated platform. A first rithm that utilizes the information acquired from the geometry
rule was generated based on deciding the primary type of use of interpretation and generates the scaffolding design, incorporates
the scaffolding system. Existing OSHA rules regulate what type of it into the model, and generates the utilization reports.
system must be used when working at greater than a specified
height limits. This is in the US typically six feet. In this work, we 5.2.1. Rule-based model geometry interpretation
name this feature the scaffolding height. It is the maximum distance For the implementation of the rule-based geometry interpreta-
from the base surface that a worker has to conduct a scheduled tion, the building model geometry was analyzed so that it contains
work task. a feature that can be evaluated by the placement rule. The rule
In future approaches more comprehensive rules can be de- evaluates each work face in the model by reasoning the scaffolding
signed to identify required scaffolding systems for complex geo- height. The geometry interpretation algorithm first detects all the
metric conditions. Eventually these might be based on industry work faces in the model and then identifies the scaffolding height
know-how and best practices. Since companies, their projects, for each work face. Since the work face and scaffolding heights are
and tasks demand different requirements, scaffolding type and not explicitly represented in currently available commercial BIM
height of the elevated platforms are generally specified by a user. software platforms, the developed geometry interpretation algo-
A practitioner might have further input in the decision making of rithm derives such information from the model before executing
which type to prefer over another and adjust values more easily the rule against the model.
depending on the situation at site. Deriving such information from a building model requires rea-
We attempt to detect such situations in need of scaffolding sys- soning about the qualitative relationship between the building ob-
tems based on the relationship between the work face and the base jects. Rather than considering the absolute coordinate information
surface. A work face is a face where the effect of construction activ- of the building objects, it evaluates how the building objects are
ity takes place. A base surface is a face where a worker or a tempo- placed relative to each other. This then determines the condition
rary facility stands on. We base our work on Kim and Fischer [20], that is meaningful for placing scaffolding systems. Depending on
72 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

Fig. 3. Work face and base surface relationship enables a geometric condition analysis.

the relative positions of the building objects, a work face can be methodology. This methodology has been adopted in the fields of
composed of multiple parts belonging to multiple building objects. qualitative geometric reasoning, qualitative spatial reasoning, im-
Shown in an example for easier comprehension, Fig. 4 illus- age recognition, and geographic mapping, and elsewhere [31–
trates a portion of a realistic building information model. This 33]. Like Chinowsky and Reinschmidt [31] did for CAD model inter-
model was used to test the proposed geometry interpretation algo- pretation, we decomposed the physical building objects into geo-
rithm. Considering the configuration of two wall and three slab metric entities such as faces, edges, and vertices. We then
segments, there are one large exterior work face and three smaller utilized the coordinate information available for each geometric
interior work faces in the model. In order to derive an object that entity to form the basis in determining the qualitative relationship
functions as a work face, the walls have to be decomposed into between building objects and geometric entities.
faces and the faces have to be finally divided and reassembled into The data structure diagram for scaffolding placement is illus-
a larger work face. To evaluate the work face, the assembly of par- trated in Fig. 5. Each building object is decomposed into faces,
tial objects may also contain the scaffolding height as one of its recomposed into work faces and further into sets of work faces
attributes. However, existing BIM software and associated rule- linked together. To each face entity, the system added relational
checking tools currently do not have a function to support such ob- information about the relationships of floors and ceilings. Using
ject granularity and cannot conduct the qualitative reasoning the geometric and relational information, faces are combined to
about the project geometric conditions. We come back to this form work faces. Each work face entity then contains geometric
example to see how the scaffolding system is incorporated auto- information, relational information as well as scaffolding height.
matically once we have explained the second part of the rule-based The height requirements determine if a work face requires a scaf-
system: the design and planning algorithm. folding system. A bounding box created in front of the work face
further represents the work space required for the scaffolding sys-
5.2.2. Geometry interpretation process tem. Since the face-edge representations from a building object
In order to enable the qualitative reasoning in the geometry maintains its links to a BIM model object, important non-geometric
interpretation system, we implemented a face-edge representation information such as materials and work tasks in the schedule can

Fig. 4. Model geometry formed by two walls and three slabs.

 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 73

Fig. 5. Data structure diagram for work face detection in scaffolding placement.

also be accessed. These help in the effort to distinguish geometric

entities related to work that intends the erection or disassembly of
scaffolding systems.
The geometry interpretation process is composed of four steps:
(a) object decomposition, (b) work face composition, (c) work face
linking, and (d) scaffolding space specification. Further details to
each step are provided.

5.2.2.1. Object decomposition. In the object decomposition step,

every building object is decomposed into its geometric entities.
As shown in Fig. 6, we focused on wall objects and considered only
the two largest faces belonging to the wall. The bottom, top, and
sides of a wall were not considered since they generally do not re-
quire scaffolding. Even though the faces belong to a single wall ob-
ject, face objects were created separately in the geometry
interpretation system because they can be combined into different
work faces. The two faces were distinguishable either by different
face vectors and different coordinate information or their relation-
ships to other objects.

5.2.2.2. Work face composition. The next step was the work face
composition. The face entities were combined into larger work
faces. In addition to the coordinate information, relational informa- Fig. 6. Wall object decomposition.
tion was used to determine if the faces can be combined. The rela-
tional information included the wall face’s relations with floors and face. This tolerance was defined empirically and can be adjusted
ceilings. by a user. If there was no slab that satisfied both conditions, the
In order to detect the slabs of the floor and ceiling that belong to nearest one among the slabs that satisfied the horizontal condition
the face, two sets of points were extracted from bottom and top of and had the z-coordinates smaller than the wall face’s bottom z-
each face, respectively. Those points were moved two feet towards coordinate was identified as the nearest floor. Ceilings were iden-
the direction of the face’s normal vector. These two sets of points tified using the points moved from the top of the face. In addition,
were used to identify the adjacency between the faces and slabs. the slabs that satisfied the horizontal condition but had their z-
For a slab to be considered as a floor or a ceiling of a wall face, both coordinates between the bottom and top of the face were identi-
horizontal and vertical conditions should be satisfied. If a slab con- fied as the abutting slabs. The wall face that had abutting slabs
tained a certain number of the points that were moved from the was finally vertically subdivided into pieces.
bottom of the face within its boundary contour and also had its A face of a wall having a floor was not combined with any faces
z-coordinate within the tolerance of two feet from the wall face’s underneath the floor. Likewise, if the face of the wall has a ceiling,
bottom z-coordinate, the slab was considered as the floor for the this face was not linked to faces above the ceiling (see Fig. 7). Face
74 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

normal vectors were used as relational information to determine if

two faces are oriented towards the same direction. In order for two
different faces to be included in the same work face, at least the
following three conditions need to be satisfied:

(1) Face vector: two faces should have the same or similar vec-
tor indicating that the faces are parallel and oriented
towards the same direction.
(2) Face-to-face distance: two parallel faces should share the
same work plane, this can be determined by having a zero
or near-zero face-to-face distance. While the value for the
near-zero distance is currently left for user-input, the toler-
ance should be defined. Eventually the tolerance can be
based on the industry practice to develop an application pro-
grams utilizing this algorithm. The tolerance for developing
this algorithm was set to two feet, but it can be adjusted by a
user. Fig. 8 illustrates the distance between two parallel
Fig. 8. Face-to-face distance.
faces.
(3) Face adjacency: satisfying the previous two conditions
means that two faces with the same orientation are on a sin-
gle work plane. In addition to these conditions, two faces
like it is illustrated in Fig. 7, can then be detected by combining
should be in close proximity. The proximity is identified
the faces above the floor and the other work face can be detected
using two-dimensional bounding boxes of the two faces
by combining all the faces below the floor.
and a tolerance of two feet that can be modified by a user.
After all work faces were detected as a result of the work face
In order for two faces to be adjacent, two bounding boxes
composition, the slabs adjacent to the bottom and top of the work
should overlap or be within two feet away from each other
faces were detected again using the same process of detecting slabs
both horizontally and vertically.
adjacent to the wall faces.

While this method is simple and easy to apply, it can potentially

produce erroneous results because it approximates the building 5.2.2.3. Work face linking. In the workface linking stage, the gener-
object shapes into two-dimensional boxes. Therefore, future re- ated work faces were combined into sets of linked work faces again
search may apply more advanced methods in defining the spatial as construction activities can take place unceasingly along multiple
relationships between building objects. [34,35] might be applied work faces. An example is explained later (see Fig. 13). The conti-
to produce more accurate results. Integration of faces was propa- nuity between multiple work faces generates realistic scaffolding
gated from a selected face toward adjacent faces. One work face, designs. Therefore, the connections between scaffolding systems

Fig. 7. Work face composition.

 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 75

within a set of linked work faces must be processed to avoid any

discontinuity or overlapping of scaffolding systems.

5.2.2.4. Scaffolding space specification. The final step of the geome-

try interpretation is the scaffolding space specification where the
use of space is automatically specified. An example is shown in
Fig. 9. In front of each work face, a scaffolding space is created to
indicate the space occupied by placing the scaffolding system. A
work space is also created to indicate the space occupied by con-
struction workers. Even though these spaces can be used for the
visualization of spaces and many analyses such as spatial conflict
analysis, our research only used scaffolding spaces to contain the
geometric and non-geometric information that are essential for
designing and planning scaffolding systems.

5.2.3. Geometry interpretation test

In order to assess the performance of the geometric reasoning
approach, we applied the algorithm to several test case studies.
Fig. 10 illustrates the process of detecting work faces from a part
of a realistic BIM. The model, created in TeklaÒ Structure, is a
three-story building model that is composed of three slabs and
two wall objects. In the work face detection step, the geometric
reasoning system extracts all the faces from the targeted objects
(walls) and subdivides them into multiple faces depending on
the qualitative relationship between the slabs. Each interior face
in the test model was subdivided into three faces because of the
two slabs which exist between the bottom and top of the face.
Two exterior faces and six interior faces were detected in this step.
Next, all work faces were detected by merging the faces. As a result
of the test implementation, one exterior work face and three inte-
rior work faces were detected from the configuration of two walls
and three slabs which coincides with the realistic perception about
the situation. Depending on an architect’s design, the same geo-
metric condition as the tested model can be created using different
numbers and sizes of walls and slabs. However, the developed geo-
metric reasoning algorithm detects the same work faces by auto-
matically deciding how object faces are decomposed and
recomposed.

5.2.4. Scaffolding system generation Fig. 10. Visualization of the work face detection process in a test model.
Fig. 11 illustrates the scaffolding system planning procedure for
a detected work face. As a result of the rule-based geometry inter-
pretation, each work face contains a set of new information. The
information includes geometric information, relational informa-
tion, and task information: the geometric information explains
the location, height, length, orientation of the work face; the rela-
tional information explains if the work face is attached to other ob-
jects such as floors or ceilings; and task information contains the
information about schedule task and the stakeholder in charge.
The geometric information can include more detailed informa-
tion such as openings and curved surfaces that should be consid-
ered for scaffolding design.
The floor slab adjacent to bottom of the work face is identified
as a result of the geometric reasoning. The scaffolding height of
each work face can then be calculated by subtracting the z-coordi-
nate of the floor slab from the z-coordinate of the top of the work
face. If the calculated scaffolding height was greater than the pre-
defined value, the planning system generated a scaffolding system
inside the scaffolding space. However, the developed system pro-
vides an alert – as required by the OSHA regulation – that only reg-
istered professional engineers can design scaffolding systems
higher than 125 feet [36].
The generation of the scaffolding system designs followed a
top-down modeling approach that transforms a scaffolding space
Fig. 9. Scaffolding space and work space specification. to a scaffolding system design (see Fig. 11). After the work faces
76 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

Fig. 11. Scaffolding system design generation for a work face.

were detected, they were represented as the scaffolding spaces in generated that are currently in use by the industry.
front of them. The scaffolding design then replaced the scaffolding Commercially-available BIM packages such as TeklaÒ Structure
spaces. Furthermore, a three dimensional point array was created have functions and templates to create various types reports. The
based on the height, width, and length of a scaffolding space. These developed system uses the Application Programming Interface
points were used as the input for the generation of scaffolding (API) of TeklaÒ Structure to automatically generate the report of
materials such as pipes, boards, guardrails, and toe-boards. the required materials to each scaffolding system design. As a re-
The scaffolding elements were visualized using the following sult, a material quantity report can be created for each scaffolding
steps. A pipe element was created using start point, end point, system.
and radius (see Fig. 12). Likewise, four contour points were used In addition, the detected work faces were linked into sets of
to describe the boundary a scaffolding board. The spacing of the linked work faces. Work faces that were geometrically close were
points was decided according to the OSHA regulations related to grouped together if they can be considered as a set of work faces
scaffolding material spacing. For example, a light duty scaffold where construction workers need to work on continuously.
should have its posts to be spaced equal to or less than six feet Finally, the scaffolding systems were automatically linked to
apart horizontally and have the runners to be spaced within 6.6 the construction schedule as shown in a test case in Fig. 13.
feet vertically [37]. The intended load for each scaffolding system The existing capabilities of TeklaÒ Structure were used to visual-
was left for the user as input. ize the construction sequence incorporating the scaffolding sys-
After creating the 3D point array, the pipes and boards were tems and to assists pro-active procurement of scaffolding
created complying with other scaffolding design regulations such system materials by automatically notifying the upcoming needs.
as guardrails for fall protection, cross bracing, longitudinal bracing TeklaÒ Structure visualizes the construction sequence by creating
(as specified in [36]). All the elements created for each scaffolding the virtual links between construction activities and set of build-
system were grouped as assemblies. Using the assemblies of scaf- ing objects to be constructed by the activities. The developed
folding systems as the input, scaffolding material reports can be system enables the visualization of the scaffolding systems by

Fig. 12. Scaffolding system planning process for a work face.

 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 77

Fig. 13. Detecting the linked work faces and creating a scaffolding system design.

creating the links between the building objects to be constructed 6. Case study implementation and results
and the assemblies of the scaffolding system elements to be
placed in front of the building objects. Fig. 14 also includes an After experimentation with various test models, the rule-based
automatically generated bill of materials for one of the generated scaffolding planning system was implemented and validated in a
scaffolding systems. BIM used on a life construction project. The model for a hospital

Fig. 14. Incorporating schedule information to scaffold construction and generating automated reports (e.g., bill of materials).
78 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

model was 84, only 72 (85%) were recognized correctly. The geo-
metric reasoning algorithm yielded eight false positives (work
faces were detected which were not required or detected in incor-
rect shapes). In addition, four false negatives were detected (the
algorithm did not detect the work faces that were required).
The eight false positives were due to the lack of the algorithm to
detect and solve for complex spatial conditions; for example:
where a work face has other scaffolding spaces or building objects
in front of or adjacent to it. The four false negatives were due to
portions of the slabs intersecting partially with the faces of the
walls. The success rate of the developed algorithm thus largely de-
pends on the complexity of the provided model. Complex geomet-
ric conditions such as irregular floor levels or shapes intersecting at
Fig. 15. A realistic building project model.
the same point in the model impact currently the performance of
automatically generated solutions like the presented one here.
extension is shown in Fig. 15. The three-story medical facility used Other issues were observed, for example: some of the detected
scaffolding systems extensively for exterior work, for example, work faces adjacent to each other resulted in overlapping scaffold-
bricklaying and curtain wall installation. Especially when working ing spaces or collisions with building objects. These imperfections
at larger heights (typically above six feet), scaffolding systems pro- of the developed algorithm lead to the generation of impractical
vide a safe elevation to the assigned work crews. Five brick laying scaffolding design solutions. Although these issues were experi-
activities at the north, east, south, and west faces and the roof of enced, applying existing collision detection checking algorithms
the facility were posted in the construction schedule. These tasks could solve the problem of at least identifying the location of the
also included some brick laying work along several walls on the imperfection. Future research must focus on computational algo-
roof surface. The activities towards constructing a brick wall were rithm development that can solve for complex shapes which are
coordinated by linking the activity to the building objects (done likely to occur often in building information models.
manually by the architect/engineer). As described in the previous The presented case study further identified potential construc-
section, the geometric reasoning algorithm was applied to analyze tability issues in scaffolding design and construction. Among the
the site conditions using the geometric properties of every wall ob- linked work faces identified on the roof of the medical building
ject and its relative position to the adjacent walls and slabs. model, some were smaller than two feet wide. Since the scaffolding
In the vast majority of the cases, the developed geometric rea- heights were also smaller than the pre-defined value, such scaf-
soning algorithm successfully detected work faces from the build- folding spaces in front of these work faces were highlighted and
ing model. It also correctly connected the detected work faces so no scaffolding systems were applied. Although in this case no scaf-
that the adjacent scaffolding joined in practical and economical fold was required, slightly taller work faces which then would re-
ways. The result shown in Fig. 16 was finally visualized in the quire a scaffold, would make it impractical to install a standard
BIM of the medical facility using scaffolding models of a standard scaffolding system. Since the width would exceed the available
pipe and board scaffolding system. For those work faces which two feet in this narrow space, more costly lifting methods (e.g.,
had height values smaller than a user-defined value, only the po- scissor lift) may need to be applied should alternative design solu-
tential scaffolding spaces were highlighted. The BIM before and tions be rejected. As demonstrated, architectural and engineering
after the geometric reasoning algorithm was applied is shown in design decision may in the future be guided by what construction
Figs. 15 and 16, respectively. Finally, the scaffolding systems and methods are effective and efficient and given the availability of
scaffolding spaces were incorporated into the construction sche- existing rule checking systems as developed in the presented
dule, and the building and scaffolding components were color- approach.
coded in an existing 4D schedule visualization tool (see Fig. 17). In summary, a rule for scaffolding system placement was devel-
The 4D construction simulation containing the scaffolding systems oped and translated into a machine-readable computer code. Algo-
has the capability to alert the users about forecasted needs for scaf- rithms automatically analyzed a model’s geometry and generated
folding through color-coded visualization and automatically gener- scaffolding system designs according to the geometry and sche-
ated bill of materials. dule. The developed rule-based system for temporary facilities
The developed algorithm detected 80 work faces from 61 wall was successfully implemented on a test and a realistic building
objects. More details to the results are summarized in Table 2. model. The results indicate success of the developed algorithms
While the manual count of work faces that actually exist in the if geometries in the BIM are simple, though difficulties of the
developed algorithm were detected when it had to handle complex
and irregular shapes and arrangements of walls and floors. Adapt-
ing approaches like this work presents assists various practical
tasks needed in pro-active engineering and management of tempo-
rary facilities, such as planning scaffolding systems thoroughly be-
fore they are installed on construction sites, forecasting the
quantity needs of such systems, the length in a construction sche-
dule, and benchmarking field installations to estimate the cost,
quantity, and schedule of required scaffolding materials on future
projects.

7. Discussion and conclusions

This work presented the framework, algorithms, and tests for

Fig. 16. Scaffolding systems incorporated in a building model. automatically designing and planning scaffolding systems as part
 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80 79

Fig. 17. 4D construction simulation incorporating scaffolding systems.

Table 2 to the identification of required scaffolding systems, generation of

Geometric reasoning performance for selected activities. the corresponding designs, and generation of various utilization re-
Building Related Actual Number of Correctly False False ports. The major contribution of the presented research is therefore
Side building number detected work identified positives negatives in the automation of the process in design and planning of scaffold-
objects of work faces by ing systems. The geometric and non-geometric conditions and con-
(faces) faces algorithm struction schedule were interpreted automatically based on pre-
North 10 (20) 5 5 5 0 0 defined rules or commonly known and applied best industry prac-
South 20 (40) 15 15 12 3 0 tices. Thus, potentially, accurate plans for scaffolding systems can
West 8 (16) 22 19 18 1 3
be generated in the future more consistently and assist or reduce
East 13 (26) 22 21 17 4 1
Roof 10 (20) 20 20 20 0 0 the error-prone manual efforts eventually more significantly.
Total 61 (122) 84 80 72 8 4 The proposed system also visualizes the temporary facilities in
existing BIM software – in particular tube-and-board scaffolding
systems – as well as the work spaces occupied by it. Such visuali-
zation can increase a project stakeholder’s awareness about work
of Building Information Models (BIM). Previous efforts led by the tasks and spaces that occur concurrently and are likely to cause
construction industry and academia were usually concentrated time or space conflicts among competing crews. By increasing
on providing methods to properly design temporary facilities. the project participants’ awareness, planning errors and hazardous
However, the developed approach extends the range of automation situations arising from temporary facility uses can be realized and
80 K. Kim, J. Teizer / Advanced Engineering Informatics 28 (2014) 66–80

corrected before such situations take place, particularly during [10] H. Kim, H. Ahn, Temporary facility planning of a construction project using
BIM (Building Information Modeling), in: Proceedings of the 2011 ASCE
field implementation. In addition, the developed system is able
International Workshop on Computing in Civil Engineering, 2011, pp. 627–
to estimate the required materials for scaffolding systems to aid 634.
decision makers by providing more predictable quantity-based [11] J. Vandenbrande, A. Requincha, Spatial reasoning for the automatic recognition
takeoffs in BIM. While construction companies and temporary of machinable features in solid models, IEEE Trans. Pattern Anal. Mach. Intell.
15 (12) (1993) 1269–1285.
facility vendors and suppliers have their own databases, they typ- [12] Construction Industry Institute, Playbook of Leading Industry Practices for
ically apply a ‘‘rule-of-thumb’’ approach for estimating quantities Estimating, Controlling, and Managing Indirect Construction Cost, The
and schedules. The presented approach has a potential to provide University of Texas at Austin, 2012.
[13] R. Nagi, T. George, M. Ashwin, Automation of scheme preparation and BOQ
them with more quantifiable means and methods. Such more accu- calculation for L&T-aluform, in: 24th International Symposium on Automation
rate quantity-takeoff can potentially lead to project cost savings & Robotics in, Construction, 2007, pp. 273–280.
and as explained, in higher efficiency of work task execution and [14] John & Frances Angelos Law Center Schuster Concrete. <http://
www.tekla.com/us/Documents/BIM-awards-2011/concreteproject7.html>
constructability. (accessed 11.14.12).
There are limitations to the developed approach that must be [15] Schuster Construction. <http://www.schusterconstruction.com/projects/
overcome in order to realize its full potential. First, complex and angelos_law_center.html> (accessed 11.14.12).
[16] In Practice: DPR Construction, Tekla. <http://www.tekla.com/us/solutions/
unexpected project conditions as they often exist on construction references/Pages/DPRConstruction.aspx> (accessed 11.14.12).
sites can – at least with the current version of the developed ap- [17] Tokyo Sky Tree, Japan, Tekla. <http://www.tekla.com/international/solutions/
proach – not be satisfactorily handled. As such, objects or their references/Pages/tokyo-sky-tree-japan.aspx> (accessed 05.01.12).
[18] C. Eastman, P. Teicholz, R. Sacks, K. Liston, BIM handbook: A Guide to Building
attributes that are often not represented accurately in the model
Information Modeling for Owners, Managers, Designers, Engineers and
create problems; for example, irregular ground conditions that Contractors, John Wiley & Sons Inc., New Jersey, 2008.
are not modeled today. Today’s scaffolding designs derived from [19] K. Sulankivi, K. Kähkönen, T. Mäkelä, M. Kiviniemi, 4D-BIM for construction
a BIM model may further require modifications to fit into the exist- safety planning, in: Proceedings of W099-Special Track 18th CIB World
Building Congress, 2010, pp. 117–128.
ing site conditions of the construction site. [20] J. Kim, M. Fischer, Formalization of the features of activities and classification
In addition, the proposed framework relies on the assumption of temporary structures to support an automated temporary structure
that the building model and construction schedule are available. planning, in: Proceedings of the 2007 ASCE International Workshop on
Computing in Civil Engineering, 2007, pp. 338–346.
As it is commonly known, building renovation projects often re- [21] Smart Scaffolder, CADS. <http://www.smartscaffolder.com> (accessed
quire scaffolding systems, but digital building models are rarely 11.10.12).
ever available unless drawn for this purpose. For such projects, a [22] Scia Scaffolding: Providing an Accurate Design and Time-saving Workflow,
Nemetschek, 2009. <http://www.scia-online.com/www/websiteUS.nsf/0/
building model has to be created first and may require accurate e87baaf3b09f3439c125758a004a411a/$FILE/Scia-Scaffolding.pdf> (accessed
dimensions. Laser scanning and other sensing technology might of- 11.18.12).
fer a solution by providing such realistic and accurate as-built con- [23] C. Lee, S. Ham, G. Lee, The development of automatic module for formwork
layout using the BIM, in: Proceedings of ICCEM/ICCPM, vol. 3, 2009, pp. 1266–
ditions to generate a building model. 1271.
Future research may focus as well on developing more compre- [24] R. Jongeling, J. Kim, M. Fischer, C. Mourgues, T. Olofsson, Quantitative analysis
hensive rules for detecting required scaffolding systems from com- of workflow, temporary structure usage, and productivity using 4D models,
Autom. Constr. 17 (6) (2008) 780–791.
plex building geometry. Accordingly, more sophisticated
[25] H. Zhang, C. Van der Velden, Y. Xinghuo, C. Bil, T. Jones, I. Fieldhouse,
algorithms should be developed to evaluate various geometric con- Developing a rule engine for automated feature recognition from CAD models,
ditions and to generate corresponding scaffolding designs. Other in: Proceedings of Industrial Electronics 2009 IECON 09 35th annual
potential exists on focusing on scaffolding systems for specific conference of IEEE, 2009, pp. 3925–3930.
[26] B. Babic, N. Nesic, Z. Miljkovic, A review of automated feature recognition with
work tasks that often require scaffolds, such as masonry construc- rule-based pattern recognition, Comput. Ind. 59 (4) (2008) 321–337.
tion and painting. The proposed approach of automating tempo- [27] C. Eastman, J. Lee, Y. Jeong, J. Lee, Automated rule-based checking of building
rary facility planning can further provide such tangible benefits if designs, Autom. Constr. 18 (8) (2009) 1011–1033.
[28] C. Eastman, J. Lee, H. Sheward, P. Sanguinetti, Y. Jeong, J. Lee, S. Abdelmohsen,
it is customized for the characteristics of different construction Automated assessment of early concept designs, Archit. Des. 79 (2) (2009) 52–
activities. 57.
[29] S. Zhang, J. Teizer, J. Lee, C. Eastman, M. Venugopal, Building Information
Modeling (BIM) and safety: automatic safety checking of construction models
References and schedules, Autom. Constr. 29 (2013) 183–195.
[30] J. Melzner, S. Zhang, J. Teizer, H.-J. Bargstädt, Quantification and Visualization
[1] R. Ratay, Temporary structures in construction operations, in: Proceedings of a of U.S. and German fall protection standards using automated safety-rule
Session Sponsored by the Construction Division of the American Society of checking in building information models (BIM), Construct. Manage. Econom.
Civil Engineers in conjunction with the ASCE Convention in Atlantic, City, 31 (6) (2013) 661–674. Taylor & Francis, Special Issue on Occupational Health
1987, pp. 1–8. and Safety in the Construction Industry.
[2] B. Akinci, M. Fischer, R. Levitt, B. Carlson, Formalization and automation of [31] P. Chinowsky, K. Reinschmidt, Qualitative geometric reasoner for integrated
time-space conflict analysis, J. Comput. Civil Eng. 16 (2) (2002) 124–134. design, J. Comput. Civil Eng. 9 (4) (1995) 250–258.
[3] R. Ratay, Temporary structures in construction – USA practices, Struc. Eng. Int. [32] P. Lienhardt, Topological models for boundary representation: a comparison
14 (4) (2004) 292–295. with n-dimensional generalized maps, Comput. – Aided Des. 23 (1) (1991) 59–
[4] Construction Industry Institute, Leading Industry Practices for Estimating, 81.
Controlling, and Managing Indirect Construction Cost, The University of Texas [33] R. Howarth, H. Buxton, Analogical representation of space and time, Image Vis.
at Austin, 2012. Comput. 10 (7) (1992) 467–478.
[5] Top 10 Most Frequently Cited Standards, OSHA, 2011. <http://www.osha.gov/ [34] A. Borrmann, E. Rank, Specification and implementation of directional
Top_Ten_Standards.html> (accessed 11.17.12). operators in a 3D spatial query language for building information models,
[6] Statewide Compliance Campaign-Formwork, WHSQ, 2008. <http:// Adv. Eng. Inform. 23 (1) (2009) 32–44.
www.deir.qld.gov.au/workplace/resources/pdfs/audit_reports/ [35] A. Borrmann, E. Rank, Topological analysis of 3D building models using a
audit_formwork_2007.pdf> (accessed 11.14.12). spatial query language, Adv. Eng. Inform. 23 (4) (2009) 370–385.
[7] Scaffolding compliance campaign 2009–2010, WHSQ, 2012. <http:// [36] A Guide to Scaffold Use in the Construction Industry, OSHA, 2002. <https://
www.deir.qld.gov.au/workplace/industry/construction/high-risk/scaffolding/ www.osha.gov/Publications/osha3150.pdf> (accessed 11.17.12).
index.htm> (accessed 09.21.13). [37] Construction Safety and Health Outreach Program, OSHA, 1996 <https://
[8] Safety and health topics: scaffolding, OSHA, 2003. <http://www.osha.gov/ www.osha.gov/doc/outreachtraining/htmlfiles/scaffreg.html> (accessed
SLTC/etools/scaffolding/index.html> (accessed 11.17.12). 9.4.13).
[9] R. Ratay, Handbook of Temporary Structures in Construction: Engineering
Standards, Designs, Practices and Procedures, McGraw-Hill, New York, 1996.