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

net/publication/320168253

BIM Applications of Rule-Based Checking in Construction Site Layout Planning

Tasks

Conference Paper · July 2016

DOI: 10.22260/ISARC2016/0026

CITATIONS READS
14 451

3 authors:

Kevin Schwabe Markus König

Ruhr-Universität Bochum Ruhr-Universität Bochum
3 PUBLICATIONS   54 CITATIONS    235 PUBLICATIONS   2,457 CITATIONS   

SEE PROFILE SEE PROFILE

Jochen Teizer
Aarhus University
195 PUBLICATIONS   8,607 CITATIONS   

SEE PROFILE

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

System and Parameter Identification Methods for Ground Models in Mechanized Tunneling - SFB 837 View project

BIMKIT View project

All content following this page was uploaded by Markus König on 29 March 2021.

The user has requested enhancement of the downloaded file.

 33rd International Symposium on Automation and Robotics in Construction (ISARC 2016)

BIM Applications of Rule-based Checking in Construction

Site Layout Planning Tasks
K. Schwabea, M. Königa and J. Teizerb
a
Department of Civil and Environmental Engineering, Ruhr-Universität Bochum, Germany
b
RAPIDS Laboratory, Ettlingen, Germany
E-mail: kevin.schwabe@rub.de, koenig@inf.bi.rub.de, jochen@teizer.com

Abstract equipment and temporary facilities for the construction

Site layout and logistics planning generally plays process. This includes the allocation and dimensioning
an important role for the successful execution of of elements like tower cranes, containers or storage
construction activities. The allocation of the right areas. Decisions during this planning phase have direct
amount and size and the timely use of resources play impact on cost development and occupational safety on
critical roles. Compared to other industrial sectors site during construction. Meeting all requirements, such
the construction industry shows a lack of as safety standards and minimizing on-site overheads, is
technological progress in site logistics and a complex process that follows specific rules or best-
fabrication planning. The automotive or ship practices. Experience is needed for the time-consuming
building industries, for example, have stringent evaluation and checking of all relevant rules.
production planning methods in place to the point One of the most frequent problems during traditional
that almost every step in the planning and construction and planning phases of a building is loss of
manufacturing processes is supported by digital data. Therefore, the most ambitious goal of
simulation and optimization. On the other hand, implementing BIM (Building Information Modeling) -
construction planning appears to remain a manual processes is to provide secure and consistent data
process slowly taking advantage of building management. This is achieved by adding relevant
information modeling (BIM) processes and information into a parametric 3D model. Once stored in
techniques. Our work investigates automated rule- the 3D model, the information can be retrieved at any
based checking in construction site layout planning time in the future. This similarly applies to the phase of
tasks to simplify the existing manual processes. A site layout planning. The 3D environment allows an
gap analysis of the traditional site layout planning exact representation of the actual equipment geometry.
process is performed to identify key areas that are of This can be used to easily derive site layout plans as 2D
high concern to practitioners. A rule-based checking drawings from the 3D model, which are most
algorithm for site layout planning embedded in a commonly used in the field. The exact geometry and
commercially-available BIM-platform was created additional semantic data can also be used to retrieve
and tested on specific cases in the site layout required information for the dimensioning of site layout
planning process of a realistic building. Promising elements, where strict code compliance is mandatory.
results and a discussion to the existing limitations of With the help of rule-based algorithms the process of
rule-based checking approaches for site layout checking the site layout model for code compliance can
planning are presented. A short outlook towards be automated. This could minimize human error, which
future research gives a potential path forward in increases economic benefit and occupational safety.
advanced construction site layout planning. In this research we propose a concept for BIM-based
site layout planning tasks. The concept includes
Keywords – predefined parametric site equipment models and rule-
building information modeling; construction based evaluation of site layouts. Rule-checking
schedule; logistics; rule-based checking; site layout algorithms are prototypically implemented to evaluate
planning our findings in a case study.

1 Introduction 2 Related Work

Before construction starts, site layout planning must The steadily progressing digitalization of
be performed in order to provide the necessary construction processes covers several construction
 BIM Applications of Rule-based Checking in Construction Site Layout Planning Tasks

phases, but cannot yet be described as exhaustive in impossible.

every discipline. For example, for the phase of site The application of rule checking in practice does not
layout planning no specialized BIM-based software is cover site layout planning tasks, but the similar topic of
available. However, the idea of digitally assisted site occupational safety can be found in recent research [11-
layout planning is not new. Researchers around the 12]. Safety-related conflicts are detected in a 3D model
world try to optimize the allocation of site equipment, and proposals for possible solutions are provided.
e.g. tower cranes, in order to improve the workflow and However, the calculation of possible solutions is very
logistics on construction sites. First approaches use case-specific, but it gives an impression of what rule-
optimization algorithms to find the optimal position of checking can look like. Knowledge-based solution
site equipment in a 2D site layout plan [1-3]. The next proposals belong to the field of artificial intelligence.
level of optimization and simulation approaches to site
layout planning was to consider 3D and 4D
environments [4-7]. The optimization of equipment 2.1 Rule checking
positioning is highly attractive for BIM-based site
The process of rule checking in building models, as
layout planning, but basic requirements have to be met
summarized by Eastman et al. [9], can be structured into
first. For example, the optimization of the crane’s
four steps:
position requires characteristic parameters. But the basic
process of site layout planning is the choice and
dimensioning of equipment in order to define these 1. Rule interpretation
parameters. Therefore, BIM-assisted dimensioning of 2. Building model preparation
site equipment will be the next step towards digital site 3. Rule execution
layout planning. Choice and dimensioning of equipment 4. Reporting of checking results
is more than a geometric problem, as it is highly time- In the first step the naturally written rule texts need
dependent [6]. For example, the size of storage areas is to be transcribed into an interpretable computer
not calculated for all required material, but for a peak language. The form or grammar of an adequate rule
consumption. This peak consumption per day can be language is still a matter of research, so that applications
derived from the construction schedule linked with the of rule languages remain to be customized. Eastman et
3D-model by dividing the volume of a construction al. [9] suggest parametric tables for rule interpretation.
element by the scheduled completion time. Another The second step deals with objects and their
important parameter that can be derived at this point is parameters in a 3D model. The parameters in the
the required workforce. Repeating this procedure for parametric table must be equal to the parameters of the
every construction phase takes the dynamic nature of 3D objects, so that adequate filtering is guaranteed. The
workflows on construction sites into account. most common parameter type is string.
Furthermore, databases or object catalogs for parametric Step three checks the rules for compliance. Both
site equipment are proposed to simplify element positive (true) and negative (false) return values are
selection and perform dynamic site layout planning [8]. added into the data structure. A false return value
Both optimization and dimensioning tasks are based detects a rule conflict. The form of a conflict depends
on constraints. These constraints consist of specific on the associated rule. Conflict information needs to be
rules and standards. In previous research the constraints transferred to the next step.
are assumed, but their source is not defined any further. The fourth and last step receives the conflict
The process of defining and systematically evaluating information from the rule execution und displays the
rules is called rule checking or code compliance results in two different ways: the textual report, e.g. IDs
checking [9]. The automated process of rule-based of intersecting elements, and the visual report, e.g.
model checking is still in early stages of research. The element selection or isolation in the model viewer.
biggest obstacle is the transcription of existing mostly This approach to rule checking is a subsequent
textual rules, standards and best-practices into an procedure. This means, that the rule check takes place
algorithmically readable form. The definition of such a after design decisions are made. The afore mentioned
form is called rule language. A consistent rule language commercial software Solibri Model Checker uses this
that enables users to intuitively define and check rules principle to check 3D models for design quality.
in their 3D model does not yet exist. Simplified Predefined rules include element specific clash
approaches to rule checking tasks are available in detection or element parameter comparison. The editing
commercial software, e.g. Solibri Model Checker [10]. of these predefined rule sets is possible, but very limited.
The software provides certain alterable rule sets, but the New rules cannot be defined.
user is not able to define custom rules. This makes a
comprehensive adoption of site layout planning rules
 33rd International Symposium on Automation and Robotics in Construction (ISARC 2016)

3 Methods 3.1 Interactive rule checking

Our general concept for BIM-supported site layout An extended use of rule checking that is still in early
planning is depicted in Figure 1. Two of the four stages of research describes an interactive process,
requirements are the building model and the where design decisions are supported by preprocessed
construction schedule. The other two requirements have rules. This means, that the supporting software
yet to be developed. They contain a database or so- visualizes rule-based constraints in advance. Figure 2
called object catalog of parametric models of all shows a conceptual example of interactive preprocessed
existing site layout equipment and a comprehensive rule checking. The light green area around the building
database of rules concerning site layout planning. These indicates where the tower crane can be placed. Rules
rules must be in form of an algorithmically readable rule that would prohibit the placement are considered, for
language that enables the transcription of all existing instance the installation of a site fence and its safe
rules, standards and best-practices. Company-specific distances towards the surrounding terrain or inclined
expert knowledge or best-practices that have not yet surfaces.
been written down formally, can and have to be Although we consider the application of interactive
included as well. With this information the automated preprocessed rule checking to be very attractive for
rule checking can support the dimensioning and design supporting software, the first step towards a rule-
placement of site layout elements. The 2D site layout based evaluation of site layout plans is the
plans that are needed in the field can be derived, costs implementation of subsequent rule checking. The
including site equipment can be calculated and logistics problem with currently available rule checking software
can be planned. is the limited ability to edit the predefined rules. The
solution is the development of a rule language, which
can be operated intuitively by the user to create custom
rules. This avoids the exclusion of the user from the rule
algorithm and leads to a better understanding of the
consequences of the self-made rule. Furthermore, the
user is able to reedit the custom rule, if the result does
not satisfy previous expectations.

Figure 2. Interactive rule checking: possible

placement area of a tower crane

3.2 Parametric site equipment

A required step of rule checking is building model
Figure 1. Concept of rule checking application preparation. This includes both the building model and
the parametric 3D objects of site equipment. A database
of predefined site equipment elements needs to be
 BIM Applications of Rule-based Checking in Construction Site Layout Planning Tasks

developed, serving as an object catalog during site Add-Ins can be designed for additional and custom
layout planning to simplify model creation significantly. interaction with the objects in the 3D model to extend
Parameters added to the 3D objects are required for the original software functionality. An evaluation or
adequate filtering and rule execution, e.g. crane radius testing of similar modelling software for this particular
or maximum load capacity. The digital site equipment purpose was not performed. The case study includes a
catalog should include the following element categories: realistic construction project of a five story office
building with surrounding terrain and urban cityscape
 Large equipment (construction machinery, cranes, (see Figure 3). The implementation follows the four
etc.); steps of rule checking mentioned earlier.
 Social services and office equipment (containers,
sanitary facilities, etc.);
 Traffic areas and transportation routes 4.1 Rule interpretation
(construction road, storage areas, etc.); At first we defined a tabular structure for the
 Supply and disposal (electricity, garbage transcription of the rules, as recommended by Eastman
containers, etc.); et al. [9]. The rule table consists of two different input
 Construction site safety (site fence, scaffolding, types. The first type is the rule category. Every category
etc.); invokes a specified algorithm. We defined three
 Temporary pit system (excavation support, slope, prototype categories for the implementation:
etc.).
The commercial 3D modelling software Autodesk  Radius rule: Checks, whether an element cuts the
Revit [13] provides a small portion of these elements as pan radius of another element
prototype objects. So-called Revit-families of a tower  Reach rule: Checks, whether a set of elements can
crane, concrete pump vehicles, containers and a material be reached by other elements
storage area can be seen in Figure 3.  Lift rule: Checks, whether the heaviest element of
a selected type can be lifted by another element
The second input type are rule parameters. The
amount and value of rule parameters depend on the
associated rule category. For example, the category
Reach rule requires information about which elements it
affects. Both element types then serve as input
parameters. Figure 4 shows the resulting rule table as a
CSV-file. The first column always contains the category,
so that the program can identify which algorithm to
invoke. After identifying the category, the input
parameters found in the subsequent columns are
transferred to the invoked algorithm.

Figure 3. Examples of 3D site equipment in

Autodesk Revit
Figure 4. Prototypical rules transcribed into a
rule table
4 Case Study The rule in line one transcribes the example rule text
For the prototypical implementation of rule checking ‘Containers may not be placed within the load pan
we chose the software Autodesk Revit. The 3D radius of tower cranes’. The keywords in italics
modelling software provides design of parametric signalize the input types for transcription, e.g. radius for
objects and an API (application programming interface) Radius rule. The second rule represents another example
that enables custom programming of so-called Add-Ins. rule.
 33rd International Symposium on Automation and Robotics in Construction (ISARC 2016)

Figure 5. Rule execution: conflict appearance

‘All building elements must be within reach of tower out before the actual rule execution starts. However,
cranes’. Line three applies the example rule text ‘Tower pre-checking is not part of this research.
cranes must be able to lift the heaviest precast column’.
In the fourth line a second reach rule is invoked with the
example rule text ‘All material storage areas must be 4.3 Rule execution & reporting of checking
within reach of tower cranes’. These example rules are results
derived from a German handbook for site layout
planning [14]. Further rules will be implemented. This For the process of rule execution the three example
procedure can be repeated until every desired rule is rule categories mentioned above are implemented using
categorized. If the user wants to include another rule the API of Autodesk Revit. At first the rule table must
and it can be assigned to one of the existing rule be read and saved into a data model. The data model
categories, another line must be added into the CSV-file. mainly consists of two classes: Rule and Conflict, as
But if the user wants to define a new category new seen in Figure 6. Other classes consider the user
source code has to be written in order to include the new interface or event handling and play no role in the given
rule-based algorithm. data model.

4.2 Model preparation

As seen in Figure 4 the rule parameters are in string
format. For adequate filtering the exact string must be
found among the parameters of the desired elements. Figure 6. Classes in the data model
For example, the keyword Tower_crane can be assigned
to the name of the element. The keyword Building A rule can cause zero or an infinite amount of
requires a different approach. Either the algorithm is conflicts. But one conflict can only be associated with a
taught which elements belong to a building, or the user specific rule. During the reading of the rule table the
assigns the string ‘Building’ to a custom parameter of rule objects are created and the fields category and
the necessary elements manually. In order to check parameter are filled with values. During rule execution
whether all necessary parameters have reasonable conflict objects will be created, if elements do not apply
values a so-called pre-checking process must be carried to the rule and added to the list of conflicts within the
 BIM Applications of Rule-based Checking in Construction Site Layout Planning Tasks

rule object. In addition the field checkResult will be set important a comprehensive rule data base is, so that
to false, if at least one conflict occurs. If no conflict is even rules that can easily be missed will be considered.
found the algorithm will return true. Another aspect that becomes obvious at this point is that
The results of the first rule check can be seen in a conflict cannot always be solved by relocating an
Figure 5. The 3D building and site layout model are object. In the case of the crane its load pan radius must
depicted on the right, the Add-In window on the left. be edited in the model and pan restrictions must be
The site layout objects include two containers (grey given to the crane operator.
cuboids at the bottom), a storage area (blue rectangle on
the right) and a tower crane, whose radius is represented
by a semitransparent green cylinder. The rule table 4.4 Implementation
introduced in Figure 4 can be found within the Add-In
The Revit API provides the ability to automate
window.
features the user would otherwise carry out manually.
Therefore the automation of rule checking in Revit is
limited to the features provided by the API. One of
those features for the analysis of object geometry is
called ‘SolidSolidCutUtils’. It allows intersections
between two solid geometries and returns useful
information, especially volume differences before and
after the cut. This makes it easier to observe and
evaluate object interactions. Otherwise the algorithm
would have to use bounding boxes. A bounding box
determines a cubic space that includes every
geometrical point of an object. The problem with
bounding boxes is that they remain cubic, even if the
object does not. In Figure 6 this problem is visualized.
Figure 7. Rule execution: conflicts solved by Both spheres obviously do not intersect, but their
layout changes bounding boxes do. Thus, the calculation of
intersections via bounding boxes cannot guarantee
Below a text field reports potential conflicts. The complete accuracy. This problem can be avoided, by
first rule check marked three of the four rules as false working with CSG (constructive solid geometry) -
(red rows) and one rule as true (green row). Clicking a objects. CSG describes a method that uses boolean
red row adds the textual report of a conflict to the text operators to interact with predefined mathematical
field. The textual report of the selected Reach rule states, geometries, e.g. cylinder, sphere etc. With this method
“28.5% of the area is covered by Tower_Crane”. It the calculation of intersections is mathematically precise
returned false, because the rule dictates 100% storage and time efficient.
area coverage. A click on the conflict text selects all
objects involved in the conflict, in this case the tower
crane. The Radius rule returned false, because both
containers are within the load pan radius of the tower
crane. The upper Reach rule returns false, because the
tower crane does not cover 100% of the building. The
next step is eliminating the conflicts. Therefore, a
second tower crane is added to the layout model and the
objects are rearranged. The conflict solution can be seen
in Figure 7. Now the two tower cranes reach 100% of
the building and storage area. The Lift rule still is
marked as true and the containers are moved out of the
load pan radius of the tower cranes. Figure 8. Intersections of objects and their
The scenario described above illustrates our bounding boxes
prototype implementation of subsequent rule checking.
Although the displayed checking results do not report
An obstacle of using ‘SolidSolidCutUtils’ is that it
any conflict, not all imaginable rules have been
cannot be used for all family types. Standard family
considered in this case: for example, that both cranes
types, e.g. floors, are restricted by Revit. The solution to
pan at the same height or that the crane pan radius is not
bypass these restrictions is to duplicate the element
affected by the surrounding objects. This indicates how
geometry. Each solid in the geometry needs to be read
 33rd International Symposium on Automation and Robotics in Construction (ISARC 2016)

and rebuild. However, this procedure is not practicable leveled by the available resources, the calculated storage
in every case, so that each case must be handled area for masonry in the given project is 20 m². This
individually. information is used by the site layout planner to define a
The pseudo code of the algorithm implemented for rectangular storage area with dimensions 5 x 4 m in the
the radius rule can be seen in Figure 7. First all elements model. The use of peak consumptions for dimensioning
in the BIM-model are filtered by the two element types of storage areas prevents the waste of useful space on
received from the input parameters. Both element lists the construction site. Time-dependent rules like these
then get iterated over. If an intersection between the two have not yet been implemented for this research.
elements is found they are added to a new element list
that functions as the input parameter for the newly
created conflict object.

Figure 10. Schematic calculation of storage area

size using peak consumption

5 Discussion
The implemented rules presented in this research
demonstrate only a small number of prototype rules and
Figure 9. Schematic pseudo-code of the rule
raise no claim to be exhaustive. It is questionable, if
algorithm involving intersections
more diverse and complex rules can be managed by a
tabular approach to rule transcription. In addition, the
exclusion of the user from the implemented rule
4.5 Dimensioning of site layout elements
algorithm further questions the use of rule tables.
The process of dimensioning site layout elements Nevertheless, it shows a realistic first application of
mostly includes time-dependent and logistic rules that subsequent rule checking in site layout planning tasks.
can be derived from the construction schedule by The concept of interactive preprocessed rule checking
calculating peak consumptions [6]. Figure 10 shows the introduced in this paper has not yet been tested, but
schematic calculation of the size of a storage area for should be addressed in future work. In order to calculate
masonry in a simple example. The required information a proposed design solution the program requires a
is derived from different sources. The first information comprehensive database of rules. Another database
is derived from the schedule. The completion time of including site layout equipment and machinery needs to
the masonry works in this case is 2 weeks. The volume be developed, so that choice and placement of elements
of masonry works is derived from the BIM-model: 80 is massively simplified, due to the connection of 3D and
m³. This can be calculated to a peak consumption of 40 characteristic parameters.
m³/week. The best-practice for planning temporary The choice of the commercial software Autodesk
storage of masonry material, found in the Revit for implementation was arbitrary and served only
aforementioned German handbook for site layout for testing purposes. Although Revit provides some of
planning [14], recommends a size of 2 m3 per square the necessary requirements for the implementation,
meter of masonry work during a given week. restricted or limited functions of the API leave
Considering the masonry work tasks altogether and implementing to be counterintuitive. Another important
 BIM Applications of Rule-based Checking in Construction Site Layout Planning Tasks

issue is the dependency on a single commercial software, Construction, 44, 129-139, 2014.
which prevents the application from contributing to [5] Astour, H. and Franz, V., BIM-and Simulation-
open-BIM. This could be solved by developing an based Site Layout Planning, Computing in Civil
external software dealing solely with rule checking and and Building Engineering, 291-298, 2014.
design optimization in site layout planning tasks using [6] Cheng, J. and Kumar, S., A BIM Based
independent data exchange formats, e.g. IFC (Industry Construction Site Layout Planning Framework
Foundation Classes) [15]. Early examples that require Considering Actual Travel Paths, The 31st
some human input and expert knowledge are shown in International Symposium on Automation and
[16, 17]. Robotics in Construction and Mining, 2014.
[7] Olearczyk, J., Lei, Z., Ofrim, B., Han, S., and Al-
Hussein M., Intelligent Crane Management
6 Conclusion Algorithm for Construction Operation (iCrane),
The 32nd International Symposium on Automation
The concepts presented in this research demonstrate and Robotics in Construction and Mining, 2015.
that BIM- and rule based site layout planning is highly [8] Hollermann, S. and Bargstädt, H., 4D Site
attractive for the engineers involved in this process. The Installation Planning in Virtual Reality for Multi-
most significant advantages are the intuitive 3D user, Computing in Civil and Building Engineering,
environment of the building model, where 2D drawings 777-784, 2014.
can easily be derived from. The permanent availability [9] Eastman, C., Lee, J., Jeong, Y., and Lee, J.,
of additional relevant information enables integrated site Automatic rule-based checking of building designs,
layout scheduling (4D) and cost calculation (5D). Automation in Construction, 18, 1011-1033, 2009.
Joining the relevant information leads to a better data [10] Solibri Model Checker, On-line:
management and prevents loss of data before and during http://www.solibri.com/products/solibri-model-
construction. In addition, the development of automated checker/, Accessed: 11/08/2015.
rule checking could result in minimized human error [11] Kim, K. and Teizer, J., Automatic design and
during planning phase, which increases monetary planning of scaffolding systems using building
benefit and occupational safety. information modeling, Advanced Engineering
Future work should consider time-dependent rules, Informatics, 28, 66-80, 2014.
which are necessary for the dimensioning of site layout [12] Zhang, S., Sulankivi, K., Kiviniemi, M., Romo, I.,
elements. Furthermore, the development of a designated Eastman, C., and Teizer. J., BIM-based fall hazard
rule language, that is able to comprehensively cover the identification and prevention in construction safety
diversity of existing rules and best-practices, should be planning, Safety Science, 72, 31-45, 2015.
addressed. If this is achieved, a rich rule data base and [13] Autodesk Revit, On-line:
the corresponding object catalog of site equipment can http://www.autodesk.com/products/revit-
be generated. In the process of implementing a rule family/overview, Accessed: 12/06/2015.
checking application the idea of open-BIM using [14] Schach, R., Otto, J., Baustelleneinrichtung,
independent data exchange formats should be favored. Vieweg+Teubner Verlag, Wiesbaden, Germany,
2011
[15] buildingSMART, On-line: http://www.
References buildingsmart-tech.org/ifc/IFC4/final/html/,
Accessed: 15/07/2015.
[1] Ning, X., Lam, K., and Lam, M., A decision- [16] Schwabe, K., Liedtke, S., König, M., and Teizer, J.,
making system for construction site layout BIM-based Construction Site Layout Planning and
planning, Automation in Construction, 20, 459-473, Scheduling, International Conference on
2011. Computing in Civil and Building Engineering
[2] Pradhananga, N. and Teizer, J., Congestion (ICCCBE), Osaka, Japan, 2016.
Analysis for Construction Site Layout Planning [17] Golovina, O., Teizer, J., Pradhananga, N. Heat
using Real-time Data and Cell-based Simulation map generation for predictive safety planning:
Model, Computing in Civil and Building Preventing struck-by and near miss interactions
Engineering, 681-688, 2014. between workers-on-foot and construction
[3] Yahya, M. and Saka, M., Construction site layout equipment, Automation in Construction,
planning using multi-objective artificial bee colony http://dx.doi.org/10.1016/j.autcon.2016.03.008.
algorithm with Levy flights, Automation in
Construction, 38, 14-29, 2014.
[4] Andayesh, M. and Sadeghpour, F., The time
dimension in site layout planning, Automation in

View publication stats