Embed Size (px)
Citation preview
1
ONTOLOGY-BASED WEB SERVICES FRAMEWORK FOR BUILDING INFORMATION MODELING
By
LE ZHANG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2012
2
© 2012 Le Zhang
3
To my wife and my parents
4
ACKNOWLEDGMENTS
I would like to thank my committee chair, Dr. Raymond Issa, for his continuous
and rigorous guidance along the unusual way of finishing a Ph.D. degree in Design,
Construction and Planning. He has been an incredible mentor and role model in either
academic achievement or personal integrity. I also owe my gratitude to the members of
my committee. Dr. Randy Chow, my external advisor from Computer and Information
Science and Engineering, gave me invaluable support on the implementation of my
ideas. I am grateful to Dr. Ian Flood and Dr. Svetlana Olbina for their scholarly advice
on the technical issues.
I am also obliged to Dr. Douglas Lucas, whose mastery of the art of building
construction and cheerful motivation both in English and in Chinese languages make
this journey much easier and enjoyable. Many of my colleagues in the Center for
Advanced Construction Information Modeling (CACIM) and in the Rinker School of
Building Construction helped me one way or another. My gratitude goes to them all.
5
TABLE OF CONTENTS page
ACKNOWLEDGMENTS .................................................................................................. 4
LIST OF TABLES ............................................................................................................ 8
LIST OF FIGURES .......................................................................................................... 9
LIST OF ABBREVIATIONS ........................................................................................... 10
ABSTRACT ................................................................................................................... 13
CHAPTER
1 INTRODUCTION .................................................................................................... 15
Problem Statement ................................................................................................. 15
Research Summary ................................................................................................ 17
2 LITERATURE REVIEW .......................................................................................... 19
Information Technology (IT) in Construction ........................................................... 19
Construction Industry Characteristics ............................................................... 19 Overview of IT Application in the Construction Industry ................................... 22
Scope and methodology ............................................................................ 22 Computing IT ............................................................................................. 25
Communicating IT ...................................................................................... 27 IT strategy and management ..................................................................... 32
Building Information Modeling (BIM) ................................................................ 33
BIM basics ................................................................................................. 33 BIM history ................................................................................................. 37
BIM benefits ............................................................................................... 38 IPD and other BIM related issues .............................................................. 43
Industry Foundation Classes (IFC) ................................................................... 45
IFC specification ........................................................................................ 45 IFC application ........................................................................................... 47
IFC implementation .................................................................................... 48 IFCXML implementation ............................................................................ 50
Ontology ................................................................................................................. 52 XML .................................................................................................................. 52 Metadata and Semantic Web ........................................................................... 53 Ontology and Ontology Languages .................................................................. 55 Ontology Research in Construction .................................................................. 56
Web Services .......................................................................................................... 57 Distributed Computing ...................................................................................... 57
Web Services Definition and Benefits .............................................................. 60
6
Web Services Basic Model ............................................................................... 62
Semantic Web Services ................................................................................... 64 Difference with Similar Concepts ...................................................................... 65
3 METHODOLOGY ................................................................................................... 68
Literature Review .................................................................................................... 68 From Construction Point of View ...................................................................... 68 From The Technology Point of View ................................................................ 69
Framework Development ........................................................................................ 69
IFC-based Ontology and Partial Model Extraction ............................................ 69 Web Service Integration ................................................................................... 70
Framework Validation ............................................................................................. 71
4 RESEARCH BACKGROUND ................................................................................. 73
IT Application on Construction Jobsites .................................................................. 73 Construction Jobsite IT Application Status ....................................................... 73
Technology and Research on Construction Jobsite IT ..................................... 75 Limitations of Current Construction Jobsite IT Applications.............................. 76
Two Alternative Ways of Delivering BIM Applications ............................................. 79 IFC-based Ontology and Partial Model Extraction for BIM Models ......................... 81
5 WEB SERVICES FRAMEWORK FOR BIM ............................................................ 86
Framework Architecture .......................................................................................... 86
Interface Level .................................................................................................. 86
Service Level .................................................................................................... 87 Data Level ........................................................................................................ 88
Data Level Implementation ..................................................................................... 89 IFC Model Data Manipulation ........................................................................... 89 Ontology Manipulation ...................................................................................... 90
Service Level Implementation ................................................................................. 90 SOAP Messages .............................................................................................. 93
WSDL Agreement ............................................................................................ 94 CXF Framework and Tomcat Server ................................................................ 94 Assistant Service and Additional Services Implementation .............................. 96
Interface Level Implementation ............................................................................... 98
User Portal Implementation .............................................................................. 98 Web-based IFC Model 3D Visualization ......................................................... 100
6 USE CASES AND TESTING ................................................................................ 102
Use Case 1: Formatted Model Information Retrieval ............................................ 102 Use Case 2: Technical Term Translation .............................................................. 104 Use Case 3: Partial Model Extraction ................................................................... 106
7
7 CONCLUSIONS AND RECOMMENDATIONS ..................................................... 109
Conclusions .......................................................................................................... 109 Limitations and Recommendations for Future Research ...................................... 112
LIST OF REFERENCES ............................................................................................. 114
BIOGRAPHICAL SKETCH .......................................................................................... 123
8
LIST OF TABLES
Table page 3-1 Inheritance relation for IfcWindow .......................................................................... 83
9
LIST OF FIGURES
Figure page 2-1 Construction process as information and material subprocesses ...................... 23
2-2 The BIM approach: from separate definition to central definition of building objects ................................................................................................................ 36
2-3 Effect versus cost curve of making changes during construction ...................... 41
2-4 Basic Web services model ................................................................................. 63
4-1 Partial model extraction sample .......................................................................... 85
5-1 Web services framework architecture ................................................................. 86
5-2 W3C Web services technology stack ................................................................ 92
5-3 WSDL Web service agreement .......................................................................... 95
5-4 List of all services generated by CXF ................................................................. 97
5-5 Screen shot of the portal website ....................................................................... 98
5-6 Screen shot of the Core Service ......................................................................... 99
5-7 Screen shot of the element detail page .............................................................. 99
5-8 3D IFC model shown directly in a FireFox Web browser .................................. 101
6-1 A list of all IFC elements available in current model ......................................... 103
6-2 A list of all IFC instances according to the selection of IFC element ................ 103
6-3 Element property inquiry result ......................................................................... 104
6-4 User can input any string as inquiry keyword ................................................... 105
6-5 A list of all elements satisfies the inquiry .......................................................... 105
6-6 Inquiry result with 3D geometry ........................................................................ 106
6-7 A partial model extraction link at the bottom of the element details page ......... 107
6-8 Partial model extracted shown in a Web browser ............................................. 108
10
LIST OF ABBREVIATIONS
AEC Architecture, Engineering and Construction
AGC Associated General Contractors of America
AIA American Institute of Architects
API Application Programming Interface
BIM Building Information Modeling
CAD Computer Aided Design / Drafting
CIC Computer-integrated Construction
DOM Document Object Model
EDMS Electronic Document Management System
GDL Graphical Description Language
GIS Geographic Information System
GML Geography Markup Language
GSA General Service Administration
HTML HyperText Markup Language
HTTP HyperText Transfer Protocol
ICT Information and Communication Technology
IDL Interface Definition Language
IFC Industry Foundation Classes
IGES Initial Graphics Exchange Specification
IPD Integrated Project Delivery
IS Information System
ISO International Standard Organization
IT Information Technology
JAS Java Application Server
11
JAXB Java API for XML Binding
JSP Java Server Pages
LCA Life Cycle Analysis
MAS Multi-Agent System
NBIMS National Building Information Modeling Standard
NIBS National Institute of Building Science
NIST National Institute of Standards and Technology
OGC Open Geospatial Consortium
OWL Web Ontology Language
RDF Resource Description Framework
REST REpresentation State Transfer
RFI Request For Information
RPC Remote Procedure Call
SAX Simple API for XML
SEI Service Endpoint Interface
SOA Service-Oriented Architecture
SOAP Simple Object Access Protocol
SPF STEP Physical File
STEP STandard for the Exchange of Product model data
URL Uniform / Universal Resource Locator
URI Universal Resource Identifier
W3C World Wide Web Consortium
WebGL Web Graphics Library
WPMS Web-based Project Management System
WSDL Web Service Description Language
12
XML eXtensible Markup Language
XSD XML Schema Definition
13
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
ONTOLOGY-BASED WEB SERVICES FRAMEWORK
FOR BUILDING INFORMATION MODELING
By
Le Zhang
August 2012
Chair: Raymond Issa Major: Design, Construction and Planning
The construction industry is known for being fragmented, yet a construction project
is a multi-disciplinary team effort which requires information from many different
stakeholders of various domains. Effective information exchange is of great importance
to a project’s success. Currently existing paper-based information exchange methods
between different parties in the construction industry are not efficient, especially in the
context of delivering construction information to the construction jobsite. The application
of Building Information Modeling (BIM) is seeing great acceptance in the construction
industry, and the problem of how to push the application of BIM to the jobsite is a field
that needs more research.
This dissertation proposes a framework build on the Web services model for
applying BIM to the jobsite. The Web services framework is built on an ontology
developed on Industry Foundation Classes (IFC) and a partial IFC model extraction
algorithm using that ontology. With the framework, only a partial model extracted from
the complete model needs to be transferred via the network. The framework also
supports textual information inquiry on the model. All the functions, including 3D model
14
presentation, can be carried in a Web browser. The architecture and implementation of
both a core service and an assistant service are discussed in detail. The proposed
framework can be easily expanded as long as the same Web services interface and the
common ontology are observed. Different use cases were used to demonstrate and
validated the framework. Once implemented, the framework can be utilized by other
IFC-supported BIM applications, as well as jobsite workers without extensive knowledge
of BIM and IFC, for precise and consistent project information retrieval on the
construction jobsite.
15
CHAPTER 1 INTRODUCTION
Problem Statement
The application of information technology (IT) in the Architecture, Engineering and
Construction (AEC) industry is an indispensable contributing factor for assisting the
improvement of productivity within each specific subfield. Different software applications
from different software vendors are available for almost all the individual professional
tasks, like architectural design, structural simulation, and construction scheduling.
However, a construction project is a multi-disciplinary team effort that requires
combining valuable and unique inputs by stakeholders from various domains, including
owners, architects, engineers, contractors and facility managers. This fragmentation in
the AEC industry has been identified by many researchers (Shen and Chua 2011).The
successful application of IT in individual tasks does not guarantee a successful
application on a project as a whole. When the digital products from different parties are
put together to work on a project, the information exchange problem between software
applications adversely impacts the overall project productivity (Chaaya and Jaafari 2001;
Pietroforte 1997). The information exchange between applications tends to be inefficient
or even impossible, resulting in increased costs and schedule delays in the design and
construction of a project. Even after a project is built, sometimes extra money is spend
during facility management on re-inventing and re-inputting data that should have been
stored during the design and construction phases of the project.
The two types of fragmentation identified by researchers are vertical interface
fragmentation among different activities through the lifecycle of a construction project,
and horizontal collaboration fragmentation among different stakeholders (London and
16
Kenley 2010). Efficient information exchange is of vital importance for the success of a
construction project, as well as the continuous development of the industry. In a report
by National Institute of Standards and Technology (NIST), it is estimated that over 15.8
billion dollars are wasted due to the lack of interoperability in construction industry
(Gallaher et al. 2004).
Building Information Modeling (BIM) has been proposed as a tool to address the
fragmentation and information exchange problem throughout the construction industry
through the introduction of a single parametric 3D model, which could be reused in all
analysis and services during the lifecycle of a construction project. Industry Foundation
Classes (IFC) is the only international openBIM standard that regulates the data
exchange format between different BIM software applications. Tizani and Mawdesley
(2011) deemed information modeling as “the foundation for future directions.” One of
the trends in BIM research is the need to add detailed digital representation on a wider
range of aspects, which is corresponding to the expectation of BIM as a complete
information depository for the whole life-cycle of a building project.
However, they also recognized that construction information modeling need to
address the “operational practice” and “discipline-dedicated views of information.”
Different use case scenarios require different information to be extracted from the model
(Venugopal et al. 2012). A completed BIM model often requires a minimum threshold
effort from computing resources and trained personnel, which are not always readily
available on construction jobsites. The IFC specification is not designed for end users
and is too complicated to use without special training. To push the application of BIM
17
further down the worker hierarchy on construction jobsites, a standard yet simple
methodology beyond traditional IT deployment will be required.
Research Summary
The goal of this research is to improve and facilitate the correct and timely
information exchange between different parties in the AEC industry, with a focus on the
delivery of model information from the design team which is located at a remote office to
the construction team which is working at the construction jobsite.
The approach is through the use of an ontology-based Web services framework
under the referential BIM model. In this approach a Web services framework is built on
the construction industry ontology developed according to the widely used openBIM
standard -- IFC specification. The framework can be used by any IFC-supported BIM
software application, as well as by persons without extensive knowledge of IFC
specifications, for more precise and consistent project information retrieval, which is
expected to further the effort of open standardization of BIM and IFC, and enhance
interoperability in the AEC industry.
The research objectives include: understanding the special requirements of the
jobsite for construction IT and BIM application with a focus on how to make technology
accessible to construction jobsite workers; exploring the possibilities of utilizing Web
technologies to deliver the information stored in a BIM model, namely through ontology-
based Web services frameworks; and providing insight to the AEC community on which
direction to follow in the future regarding the use of BIM on construction jobsites.
This dissertation is organized into seven chapters. Chapter 1 is the introduction of
the research, with problem statement and summary of the dissertation. Chapter 2 is the
literature review. In the first part of Chapter 2 the IT application and research in the
18
construction industry are reviewed, with a focus on BIM and IFC technologies. The Web
and related technologies including ontology engineering and Web services are briefly
reviewed in the second part of Chapter 2. Research methodologies are summarized in
Chapter 3. Next, as the direct background of this research, how IT is applied on
construction jobsite is reviewed in Chapter 4. Also reviewed in this chapter are the two
alternative BIM application system frameworks as well as the process of IFC-based
construction industry ontology and the partial model extraction algorithm previously
developed in this research. The high level architecture and implementation details of the
Web services framework proposed in this research are discussed in Chapter 5, followed
by use cases testing of the system in Chapter 6. Finally the dissertation is concluded
with a discussion of research conclusions and recommendations for future research
directions.
19
CHAPTER 2 LITERATURE REVIEW
Information Technology (IT) in Construction
The construction industry is regarded as information intensive by many
researchers (Brandon and Betts 1995; Wang et al. 2010). Two kinds of information are
identified: design information that directly specifies the end building product, such as
drawings and specifications, and other management information such as schedules,
contracts and purchase orders. New developments in Information Technology (IT) have
changed many industries in recent years. The effect is so profound that this change is
often referred to as the Information Revolution (Fang 1997). The construction industry is
no exception. The research and application of IT in the construction industry is also
known as Information System (IS) in construction, Information and Communication
Technology (ICT) in construction, Computer-integrated Construction (CIC), etc. In this
section the research and application of IT in the construction industry and related
research efforts are reviewed, with focus on Building Information Modeling (BIM) and
Industry Foundation Classes (IFC).
Construction Industry Characteristics
Encyclopedia Britannica defines construction as (the process of) assembly or
erection of large structures (eb.com 2012). A broader definition of construction is the
human activity of producing artifacts such as buildings, plants, roads and bridges (Bjork
1999). The “artifacts producing” part of construction definition is similar to the
manufacturing industry, but a construction artifact is a unique product specifically
designed for a particular location. The duration of the production process is usually long,
especially when the whole life cycle of the building is being considered, including
20
periods of the manufacturing of the building materials before the construction and the
operation and maintenance after occupancy. The construction industry has very unique
characteristics that are different from traditional engineering and manufacturing domains,
which are summarized by Elvin (2003) and Beetz (2006) as follows:
• Uniqueness of production process: Learning curve specifies that with repetitive
production the workers tend to be much more efficient and accurate. A
construction project is often mentioned as “one of a kind”. The uniqueness of a
construction process makes it very difficult for the team to take advantage of the
learning through repetition. Another effect of the unique product is that the high
cost embedded into detailed drawings and/or models is not distributable among
the mass production of thousands of products.
• Outdoor and less controlled production environment: Modern manufacturing
industry has a tightly controlled production environment in a factory. But most of
the activities for a building construction need to be carried out on the particular
jobsite. The jobsite is usually an outdoor circumstance, which is easily affected by
the surrounding environment and the weather conditions.
• Heterogeneity and scale of participants: It is estimated that over 90% of
participants in construction industry are small and medium enterprises (SME) with
very limited IT budget, in contrast to the monopolized car manufacturing industry
where the leading companies have the ability and resources needed to implement
large scale research and development efforts, as well as the ability to influence
both the IT vendors and subcontractors.
21
Due to the above mentioned characteristics, the construction industry is
categorized as the most inefficient industry, with about 30% cost waste, which is about
US $10 billion (Brown and Beaton 1990; Elvin 2003). Many factors may contribute to
this waste, but one of the major factors is the separation of design and construction (de
la Garza 1994; Anumba et al. 1997; Pena-Mora and Li 2001). This is also referred to as
traditional Design-Bid-Build (DBB) delivery system, in which organizationally designers
are separated from the contractors and sequentially the design process is separated
from the construction process by a bid process. The designer is responsible for the
design of a project and the contractor is responsible for completion of the project
according to the design handed off from the designer. There is little collaboration
between the two parties, and sometimes there is even a controversial attitude between
them on who should be responsible for problems occurred during construction.
The return of the Design-Build (DB) is one of the efforts the industry is using to
address this challenge, through the integration of design and construction under one
entity which holds a single responsibility to the owner. Under this system, the
information exchange, as well as potential disputes that may occur along the
information exchange process, between two legal entities is now changed into issues
resolvable within one organization, which bears much less cost and risk. The aim of the
integration is to eliminate waste and duplication. Fergusson (1993) confirmed the
correlation between increased integration and improved quality.
The transformation from DBB to DB brings many revolutions in the construction
industry, including construction information management. In DBB, because of contract
responsibilities and potential risk of future dispute, the information being transmitted
22
from the designer to the contractor is intentionally limited. But under DB, the contractual
responsibility is in one organization and information exchange becomes internal. More
information means more collaboration and faster implementation as well as lower cost
to reinvent the wheel when it is needed, so both the designer and the contractor retain
as much information as possible during the information exchange process from one
party to the other. Building Information Modeling (BIM), which is discussed in detail in
next section, is fulfilling this new requirement as part of this transformation of the
industry.
Overview of IT Application in the Construction Industry
Scope and methodology
The concept of IT in construction can be very broad and subjective (Jung 2011).
The definition of IT is adopted from the research of Bjork (1999), in which it is defined as
all kinds of electronic machines and programs and other technologies used for the
processing, storage, transfer, manipulation and presentation of information. As shown in
Figure 2-1, he also proposed a construction process model based on the IDEF0 activity
modeling methodology, in which the material and information flow in construction
process are distinguished as two interacting sub-processes. Accordingly, an IT
application in construction is defined as the use of information technology to facilitate
and re-engineer the information process component of construction, ranging from
coding in computer languages to technology management strategies, so as to improve
the effectiveness of construction process by better utilizing construction information
(Bjork 1999; Jung 2011). IT in construction research is concerned with how the tools
and techniques that are used in the information sub-process of a construction project as
well as the interface between the information and material sub-processes. Generally
23
two IT application and research approaches are identified, namely “technology push”
(new technology that appeared in IT finds its application in construction) and “problem
driven” (a problem that appeared in construction finds its solution in IT) (Bjork 1999).
1
Information
process
2
Material
process
The client’s needsInformation
Raw materials
Products
Energy
Detailed drawings,
schedules,
procurement orders
etc.
Buildings
Products
Observations,
Measurements
Computing and
telecommunications
equipments (IT)Machines
and tools
Design know-how,
Building regulations,etc.
Client’s brief,
schematic
drawings,
etc.
Designers,
construction
managers etc.
Construction
workers
Construction
know-how
Figure 2-1. Construction process as information and material subprocesses (Bjork 1999)
It is argued that one important standard of a technical field becoming mature is
widely accepted methodologies (Fernandez-Lopez 1999). On that account, IT in
Construction is regarded as a relatively new research field, which still lacks a clearly
defined boundary of research scope and widely accepted methodology (Bjork 1999). IT
in construction is different from other engineering disciplines that are based on basic
science whose research could be tested in laboratories. The result of IT application is
24
very difficult to quantitatively measure. The research results in construction IT are also
very difficult to be proved by statistics, as the research concept demos and prototypes
from academia are generally far from the scale requirement of industry application, as
was noticed as early as the 1970s during the software crisis (Sommerville 2004).
It would be relatively straightforward to use standard social science sampling and
interviewing techniques in empirical studies of IT applications in construction where the
effects of IT applications are observed and measured in individual company or project
case studies or bench-marking studies comparing performance of different companies.
The potential benefits of many IT application depend on economies of scale which are
only achievable through industry level standardization and collaboration among different
stakeholders, which is not available during the individual research periods.
The only one technique that could be used to prove the validity of construction IT
research is the ability to replicate the research or experiment result according to the
literature, which is not rigorously applied either because of the complexity of modern IT
systems and the limited resource an individual researcher may access (Bjork 1999).
For example, in conducting jobsite tablet computer efficiency research, Elvin (2003)
measured the impact of tablet computers on project communication at the task level,
which is one of the few research projects that take into consideration measurable
quantitative parameters. Specific parallel demo projects were designed and carried out
in a controlled laboratory environment. Accuracy, timeliness, completeness and
efficiency of information exchange were measured. Again, there is a huge difference in
both the scale and the knowledge level of participants in the research projects and real
25
construction projects. The measurement parameters chosen by the researcher were not
substantiated.
Computing IT
According to the definition of IT discussed previously, researchers often neglect
that there are two completely different aspects of IT applications in the construction
industry. The first one is to utilize the computing power of modern computers to process
a task that is tedious or even impossible by hand, such as structural calculations or
photo rendering. This kind of IT technology will be designated for the purpose of this
study as Computing IT. The other one is to use IT technologies for information
exchange between different parties, such as transmitting drawings across continents in
seconds via fax or email. This kind of IT technology will be designated for the purposes
of this study as Communicating IT. These two aspects of IT require completely different
technologies and their effects may not always be consistent. Although this distinction
was mentioned by other researchers, they are seldom discussed as separate topics.
For example, Bjork (1999) regarded the information processes which generate new
information as “primary activities,” and all the others as “secondary activities.”
The characteristic of Computing IT is that during the information processing new
information is generated. The use of Computing IT has a long history in AEC industry. In
fact, the early days of IT application in construction were almost exclusively Computing
IT (Bjork 1999). Currently Computing IT is recognized as a powerful means to substitute
tedious manual calculation and preparation and to improve productivity (Peansupap and
Walker 2005; Law 2011). A sample of fields where Computing IT finds its application in
construction industry include but is not limited to: computer graphics, engineering
analysis, and virtual simulations.
26
In 1963 the first Computer-Aided Drafting (CAD) software “Sketchpad” was
developed by Ivan Sutherland during his Ph.D. study at MIT (Myers 1998). The
application of CAD in the construction industry proliferated during 1980s. CAD uses the
same methodology as traditional drafting, only with the computer as a tool instead of
paper and pen. CAD is good and accurate at geometric expression of design intent
solutions with unintelligent points and lines, but lacks the semantic information needed
to make automatic design decisions when it comes to the qualitative information
required especially in the early design stages when specific design details are not yet
defined (Chinowsky and Reinschmidt 1995). CAD drawings may be finished and stored
in a computer system, but the end products are still often plotted, copied and distributed
to the authority, owner and contractor in paper format.
The other kind of Computing IT technologies are those that do not need very
powerful computing resources but rely on extensive databases. Expert systems are one
of the examples. The expert systems are expected to formalize human knowledge. With
the help of expert system shells, individual researchers can build meaningful systems
with simple logic base. But such systems are generally limited to a highly specialized
field or a certain sub-task. To build a system suitable for real projects, which may
consist of hundreds of such sub-tasks, a huge database as the reasoning foundation is
required.
The advancements in IT hardware / software are also reflected in the
advancement of requirements in construction industry. As computer programs have
become more complex and legacy programs accumulate, one of the trends in software
development is identified as the shift from coding of specific functional modules to
27
integration and reuse of different existing functional modules, which leads to the
popularization of the Web services model discussed later on in this chapter (Law 2011).
It is relatively late in the process that researchers and industry realized that
Computing IT alone may not generate the significant productivity increase expected.
Bjork (1999) estimated that the general productivity gain through CAD is about a factor
of 2 to 3. Because in real work only a fraction of working hours per day is dedicated to
drawing production, while the rest of the time is for different kinds of information retrieval
and communication, which is discussed in the next section. Another reason is that such
kind of Computing IT applications are usually on a task level, which is relatively lower
when we consider the information flow process from a company-wide point of view.
Without re-engineering of the business process, the productivity increase of single IT
technology might be very limited.
Communicating IT
As discussed earlier, the construction industry is information-intensive and multi-
disciplinary. Different parties need to transfer documents, drawings, and engineering
models all the time. The importance of communication to the success of a construction
project has been confirmed by many researchers (Pocock et al. 1997; Kumaraswamy
and Chan 1998). But to a large extent Communication IT applications in the
construction industry are not through conscious re-engineering and lack relevant
extensive research (Bjork 1999).
There are two different aspects regarding communicating IT. The first one is data
delivery, which means getting the required data to the point where it is needed. The
second one is data interoperability, which means the data receiver can understand the
information stored in the data being transmitted from the data sender.
28
Data delivery. Data delivery may appear in many different forms, including
personal communication either on the phone or email, information search and retrieval,
or information distribution. The early days of IT assisted data communication is limited
to several persons viewing the same image from the same screen during design, or
information retrieval from different terminals connected to the same computer (Bjork
1999). The advent of the Web and Internet, including mobile network and cloud
computing, brought new aspects for IT application in information communications in the
construction industry. Networks enable easy access to distributed software applications
and computing resources, and make it possible for the transition from stand-alone
software applications to interconnected software services to occur. Linking different
resources and information from the same website is proven to be useful (Ibrahim 2004).
With development of dedicated technology, the resources linked on a website are no
longer limited to files and drawings; both the Graphical Description Language (GDL) by
Graphisoft and i-drop by Autodesk enable representation and exchange of graphical
information directly on the webpage (Graphisoft 2012; Autodesk 2012).
Being flexible yet powerful, Web-based applications are regarded as especially
suitable for the work environment of construction projects (Nitithamyong and
Skibniewski, 2006). The scattered geographical location of construction projects, which
is an important factor preventing efficient data transmission, is efficiently handled in
such systems utilizing web technology. The construction industry has seen various
applications of web-based systems, ranging from more general-purpose project
management systems to more specifically designed systems like sub-contractor
evaluation systems or construction document processing systems (Skibniewski and
29
Abduh 2000; Arslan et al. 2008). Examples of commercial web-based electronic
document management systems (EDMS) include Buzzsaw and e-Builder. Construction
information portal sites are one of the most popular and direct application of the Web in
the construction industry. Also called Web-based project management systems
(WPMS), this kind of application utilizes private or public Web for information exchange
within organizational boundaries or between project partners. Many studies have
agreed on the benefits of WPMS in improved communication, information exchange and
document transfer, as well as resulting cost and time savings (Ibrahim 2004; Berning
2003).
The shift of software development from coding to integration is also reflected in the
AEC industry. With the Web and Internet, code integration goes beyond the limitations
of the programming libraries installed on the local machine, and covers anywhere in the
world. As a new technology facilitating code integration over the Web, Web services
applications are able to extend software use beyond individual applications to support
collaboration on a project, in a corporation or even at the industry level. The details of
Web services are further discussed in Section 2.3.
Data interoperability. Interoperability is the ability to communicate electronic
product data between different software platforms. The breadth and complexity of a
building project crosses many disciplines, and the life cycle of a building facility lasts for
many years. Many applications from various sources are expected to be used in a
building project. Almost every discipline or task in the construction industry has its
specific software applications. Normally those applications use the vendor’s own
proprietary data file formats. Collaboration and interoperability between the software
30
applications are expected, but only a small portion of data files can typically be directly
opened by different software applications.
The media for the information exchange in a construction projects is evolving from
traditional paper-based documents into electronic models. The transition brings great
benefit to the industry, including visual advantages and direct use of the model in all
kinds of computer-based analyses. However, associated with this trend there are also
some problems. Interoperability is one of the most significant problems. Unlike paper-
based documents that typically do not need any interpretation accessories, an
electronic model must be rendered in a specific software application to make the
information understood by humans. For example, a Revit model stored in an .rvt file
must be opened in Autodesk Revit software, or a compatible software application.
Otherwise, the information is locked in the model. A construction project is a multi-
disciplinary team effort combining inputs from various domains. The information for
specific tasks is typically generated using different software packages. When everything
is put together, information interoperability will be of critical importance. Currently data
exchange between different software applications is adversely impacting the overall
productivity of the construction industry. BuildingSMART estimates that information
about a project is re-entered an average of seven times during the project life cycle
(ASHRAE 2009). NIST research indicates that inefficient interoperability results in a
total added cost of $15.8 billion per year in US alone (Gallaher 2004).
One simple interoperability solution is to develop pair-wise translators that can
change the data from one format into the other. This approach is difficult to develop and
manage with the continuously increase of potential target data formats. Another
31
approach is standardization, i.e. to use files exporting and importing via a common
neutral industry standard file format specification into which software vendors could map
their internal data structure.
The effort to address data exchange has existed since the early days of 2D CAD.
The Initial Graphics Exchange Specification (IGES) is one of the first vendor neutral
data exchange formats for CAD and CAM, published by US National Bureau of
Standards (known as NIST currently) in 1980 (NIST 2008). Initially IGES only covered
the graphic and geometric characteristics of a product, which is relevant to the content
covered by the CAD applications it supports. However, IGES is under constant progress
and topological and non-geometric product data support is added in the newer versions.
The newest available version is version 5.3 published in 1996, with version 6 in ballot
for an ASME/ANSI standard (US PRO 1996; NIST 2011).
The STandard for the Exchange of Product Model Data (STEP), which is an
international standard numbered 10303 in the International Standard Organization (ISO),
was the first interoperability standard promoted as a product model information
exchange format between different computer programs. STEP tried to define product
model structures for all branches of industry, including mechanical, furniture, electrical,
and ship manufacturing. However, this approach has also been seriously questioned by
several researchers (Bjork 1999).
EXtensible Markup Language (XML) is another technology addressing data
interoperability in the construction IT industry. Almost all data can be described in XML
format, and it is relatively easy to transform data from one XML format into another XML
format. Many domains in the AEC industry are taking advantage of XML to develop
32
domain specific XML data formats for tagging the model data which is then used in
different analyses. Such initiatives in the construction industry include agcXML, gbXML,
CityGML, as well as ifcXML.
The early work of data exchange and interoperability focused on certain fields.
With the progress of BIM, which promotes a single model reused throughout the
lifecycle of a building, interoperability is pushed to go beyond domain boundaries. The
interoperability tool for BIM, IFC, includes not only the description of a building and its
product components, but also the process and project information. IFC was initiated by
several commercial CAD software vendors in the early 1990s (Liebich 2010), and is
discussed separately in the following section.
IT strategy and management
A successful IT framework is more than just the technical advancement discussed
above, but also includes management policy. This is referred to as the tactical vs
strategic IT research in the construction industry (Betts 1995). Basically this kind of
research aims at solving what the company should do to get the best result out of their
existing IT tools or their future IT investments. It has been noticed that different
countries are using different IT strategies tailored to their specific legal and technical
context (Ting and Wang 2000).
Three levels of strategic IT normally cited are industry level, organization
(enterprise) level and project level (Jung 2011). Very little research also address the
even higher national (government) level (Ting and Wang 2000). A special field is at an
even lower task level and is focused on how IT is applied on the construction jobsite,
and is discussed in detail in Chapter 3.
33
Building Information Modeling (BIM)
According to the National Building Information Modeling Standard (NBIMS) by
National Institute of Building Science (NIBS), BIM is defined as “a digital representation
of physical and functional characteristics of a facility” (NIBS 2007). Normally a BIM
model has 3D objects as a higher level user interface and in the form of a database at
lower level data storage. It is expected to store all the information about the building and
its components, and to work as a basis for decision making during the building’s life
cycle. As a process, BIM is the human activity of using BIM software, hardware and
methodology to create, use, and manage BIM models. In this section the basic, history,
benefits and some key technologies of BIM are discussed.
BIM basics
A building is 3D by its nature. Since there is no easy way to document a 3D object
directly on 2D paper, the traditional way of documenting a building design is to work on
multiple views to describe a building indirectly. The views include basic 2D drawings as
a combination of geometric lines and arcs and schedules. For example, a wall is often
defined as two parallel lines in the plan view, where the two lines do not have a
substantial relationship. Each view is essentially another copy of the building, with only
part of its features (or information) being defined and reflected. Each additional view
introduces additional information on top of other existing views, making the definition of
the building more accurate. But on the other hand this also adds redundancy to the
features of the building that have already been documented, hence increasing the
chances of error. For example, in the plan view only the length and width of a wall could
be defined. In order to define the height of a wall, a section or elevation view must be
introduced. However, in the new view (suppose it is an elevation) the length of the wall
34
will be defined again and introduces a redundancy, which leads to potential errors. The
length definition in the plan and elevation views should be the same logically, but do not
have substantial relationship in the lower level data representation, so it is perfectly fine
in a CAD application if a wall is defined as 10 feet long in plan view while 12 feet long in
elevation. There are other documents describing different but related information. For
example, besides architectural drawings, there are also structural drawings, mechanical
drawings, electrical drawings, plumbing drawings as well as relevant tables, schedules,
and specifications. The more complex a building is, the more views are needed to
define it precisely. The number of documents in a project may reach thousands (Ibrahim
2004). In the meantime, the more information overlap and redundancy will occur.
Besides the potential omissions and conflicts and document management problem
created from the multi-view representation, more coordination problems are probable
when there are design changes. If any building feature is to be changed, all the views
need to be updated and revised manually accordingly. The errors and inconsistencies
are especially prone to occur during this change process. In short, traditionally one
object is defined by several irrelevant geometry-based or other views and is error-prone.
The inconsistency is embedded among all the different documents besides drawings
and specifications generated among the life cycle of a building project. The same
information is entered several times by different people and is kept in different places in
the information systems. An integrated information system in which such errors and
inconsistencies are eliminated automatically is highly desirable and expected.
The 3D model is the center of BIM. Different 2D drawings and documents are
simply a view of the 3D shape information in a BIM model, and hence are consistently
35
generated. The length of a wall is defined only once and stored in one place in the lower
level data structure. Different views of the wall are all generated from this central
representation, eliminating data redundancy and inconsistency. If a value change is
needed, it is changed only once, and all different views are updated accordingly.
3D CAD models are not a new thing. However, the 3D objects under the CAD
context only provide graphical object representation, or highly symbolic information of
the building. The reader of the model still needs to interpret the meaning and
representation of the object, which is basically the same way 2D drawings are used,
although with a little bit more ease visually. The lack of rich information or data besides
geometries is the key difference between 3D CAD and BIM.
The building blocks of a BIM model as well as its drawings are not 2D lines and
arcs, but intelligent object-oriented 3D objects and assemblies, which are digital
representations of the actual building blocks of a building, such as walls, columns, doors,
windows and slabs. This follows the object-oriented (OO) trend in programming
languages. From the point of computer science, an object is the smallest independent
procedure on the code level that contains the instructions and data to perform tasks
based on messages or instructions received (Ibrahim 2004b).
Besides graphical information that describes its shape and dimension, the objects
also store other information about the object itself (material, specification, building code,
assembly procedure, price, manufacturer, warranty, etc.) as well as its relationship with
others (behavior). All different pieces of element information are either stored within the
element definition, or linked to external information resources. They are defined by
parameters (or properties) that specify certain characteristics of an element. A
36
parameter is related to a set of conditions (rule set). It is a value that is constant under
certain conditions but will change if the conditions change. For example, if the value of
airflow is changed in a diffuser, then the linked duct will change its size automatically
(ASHRAE 2009). A BIM model is such a digital parametric model of the building that
aggregates all the object-oriented building elements stored in a database. Figure 2-2
illustrates this approach from separate definition to central definition.
Figure 2-2. The BIM approach: from separate definition to central definition of building objects
By moving from 2D to 3D as well as information-supported automatic model
coordination, BIM helps to increase productivity, to lower design cost and to improve
design quality. The 3D model is only a starting point. With cost, schedule and other
building information (product data, supplier information, warranty, etc.) added into the
element definition, it is easily augmented to 4D (scheduling), 5D (estimating) or even
higher. Some commonly mentioned higher dimensions include 6D facility management
and 7D procurement solutions (Silva 2012).
An information-rich BIM model can generate all the traditional 2D design view
documentation components of a building, including but not limited to plans, sections,
elevations, perspectives, details, and schedules. Since the views are generated from
37
the element’s object definition, if an element is to be changed, the change occurs in a
central location and is reflected by all the views automatically.
BIM history
There are different opinions on the origin of BIM, but one thing is certain that the
concept of BIM is not new. Some argue the concept has existed since as early as
1940s, when it was presented and discussed academically without traceable
documentation (Silva 2012). The research article by Eastman describing a building
description system is widely cited as the first prototype research on BIM, in which the
basic concept of modern BIM were elaborated, including single database, visual
advantage, cost estimating, scheduling and material ordering (Eastman 1975). Another
burst of research under the name “building product modeling” followed in around 1985-
1990. They defined the basic expectations of BIM that one coherent model being used
along the life cycle of a building, and started several research initiatives whose products
last until today, including IFC (Bjork 1999). A general rule of IT technology research and
transfer in the construction industry, which the application of CAD has roughly followed,
is that academically proven technologies will take about 10 years to become the best
practice expectation of leading companies and about 20 years to become standard
common practice of the industry (Bjork 1999).
Commercially, Autodesk, a US company, released its first version of AutoCAD in
1982. Two years later in 1984, a Hungarian company founded in 1982 named
Graphisoft launched ArchiCAD. This is recognized as the first BIM software in the world,
although at that time it was known under a different name of Virtual Building (VB).
Although these two software applications appeared almost at the same time, the
methodology behind the two products is vastly different. AutoCAD is based on 2D CAD,
38
while ArchiCAD is based on 3D modeling. Pioneers may not always be readily
acceptable to their peer, which is again proved by the popularity of 2D CAD in following
20 years. One of the reasons is the similarity of CAD to the existing design technology
environment using paper and pen at that time, another reason is that only recent
development in computer software and hardware have made it possible to have a
practical deployment of 3D models (Law 2011).
Currently BIM is moving from a selling point into a standard practice in the
construction industry. The application of BIM is also crossing the boundary of traditional
construction industry into facility management, government code review, emergency
management, geographical information systems, etc.
BIM benefits
The most direct benefit of BIM is better visualization and easier information
access. As we live in a 3D world, it is easier for human brain to interpret 3D scenes
(Mao 2011). But under the traditional way of documenting much training is needed to
understand the 2D drawings and build a virtual 3D in the brain. It is also difficult to
understand the document collection of all the project information as it is scattered in
different views and needs to be viewed in the context with reference to detail drawings
or specifications. It is the viewer’s responsibility to isolate specific documents and find
the relevant information quickly. In BIM, both the view of the building as a whole and the
elements within the building model are in 3D, so the viewer with minimal training could
understand its geometry and element relationship between each other. All the
information required for an element, either geographical or textual, is associated with
the element in a single place and same interface, so it is much easier to find.
39
Ideally, a BIM model could be built as early as possible and used for exploring a
project’s key physical and functional characteristics digitally in the early stages of the
design before it is built. Its parametric design methodology can reduce the design cost
through more consistent models with less errors and conflicts as well as better
visualization. The contractor will benefit from better coordination among different trade
models, constructability analysis and smoother jobsite work. During the project design
and construction progress, new data is added and existing data is continuously refined
and updated. The data can also be used in more precise quantity take-off, building parts
production by the manufactures, off-site building component prefabrication, or even
automation of construction. 4D scheduling and 5D cost information is also a key benefit
where the progress and cost of a building could be visualized according to the time
specified in a schedule. All leads to safer jobsite and shorter construction. After the
project is finished, the data could be used by the owner in commissioning and facility
management to understand the building operation, to locate the specific location of any
maintenance work order and to link to a database of building parts with manufacturer
and warranty information. And finally the model could be used to support the adaptation,
renovation as well as demolition of the building.
Although BIM is advantageous in many aspects as mentioned above, including
early visualization, and parametric design, the true value of BIM is in its integration of
building information between different stakeholders along the whole life cycle of a
building project, from inception through design and construction, then facility
management and finally demolition and disposal. A BIM model is much more than a
collection of drawings but a service hub for all project information. The data stored in a
40
BIM model captures and maintains the vast knowledge required for building design and
construction, and could be used to support various calculations, analyses and decision
making activities of other trades such as cost estimating, construction planning,
engineering simulation, building performance analysis, building element production and
construction document generation. Each trade does not need to build own models for
different analyses.
Traditionally it is very difficult to do design optimization and interference checking
among different disciplines. Many conflicts across discipline lines, such as conflicts
between HVAC ducts and structural members, will not be discovered until they are
actually carried out on the jobsite, resulting in Requests for Information (RFIs) and
change orders. Generally they will be solved by the “first trade there” rule, and the later
discipline has to use offsets, work-arounds, or even unfit components to accommodate
the situation. The result is waste of time and money as well as performance
compromise of the building system. With BIM, the whole building could be optimized
under the coordination of a single model, with the potential of higher design quality
control. With better 3D visualization and automated clash detection, most of the
problems could be discovered and solved early among different trades during the
design. Responses and modifications could also be synced back into the model through
use of consistent, parametric model data. As a result, the more consistent and complete
data from a BIM model enabled multidiscipline collaboration starting early, way before
the traditional way and the actual construction on the construction jobsite. Figure 2-3
shows the relationship between the time of making changes and the cost/difficulty.
41
Basically the earlier in the building’s life cycle, the easier and less expensive to make
changes.
Figure 2-3. Effect versus cost curve of making changes during construction (ASHRAE 2009)
Geographic Information Systems (GIS) are used extensively in urban planning on
the level of the whole city. It is noted that GIS application is also being changed from 2D
to 3D (Mao 2011). Focusing on a more detailed level of describing the specific buildings
in a city, BIM and GIS are complementing each other in description and analysis of the
built environment, and are becoming the important information archives of the modern
city.
42
In summary, the BIM methodology itself reduces data duplication and redundancy
via parametrically related model components, and significantly saves both time and
money. BIM enabled collaboration furthers this effort by making early collaboration
between different trades and the owner possible, resulting in optimized early decision
making when it is still easy and less expensive to make changes.
The owner is regarded as the final beneficiary of the BIM application. After all, it is
the owner who has to pay the architects, the engineers and the contractors. The owner
is usually the one who is not familiar with the actual building construction and the one
who is going to occupy and operate the building for the rest of its existence. With the
early design visualization and optimization as well as 4D and 5D support, the owner can
be much more confident on what end product to expect, how much it is going to cost
and how long it will take. It is also a fact that the owners are impacting and pushing the
AEC industry in the application of BIM. The General Service Administration (GSA), the
biggest public owner of the United States, has been requiring the delivery of IFC-
compatible BIM models in all major federal building projects since fiscal year 2007.
Building construction is a social activity. The building is occupied by people and is
an important part of the community and the society. The use of BIM models can also be
outside the AEC industry. BIM enables quick and easy considerations of alternatives in
early stage Green Building and Life-Cycle Analysis (LCA), which results in a design
choice with higher performance and lower environmental impact. Real-time monitoring
and adjustment of building performance like temperature and humidity with the aid of
the BIM model could improve the occupants’ comfort and work productivity. A BIM
model could be used by the safety and emergency management for emergency
43
responses. A better rescue and evacuation route designed with the aid of a BIM model
could save lives.
IPD and other BIM related issues
BIM promises to improve the overall quality of building delivery and is expected to
become the dominant building delivery technology. The proliferation of BIM use brings
challenges beyond pure technology and affects all aspects of the construction industry.
As discussed above, BIM is a tool for integrated design and construction, and it
will further this integration with Integrated Project Delivery (IPD). IPD is a summary of
the key factors that should be addressed in the new project delivery process applying
BIM. IPD identifies and proactively manages risk within the delivery process. It requires
a new way of working across stakeholders, companies and locations, with collaboration
in the center and with the value of the project (instead of the profit of single organization)
in the mind of the stakeholders. One does not need to use BIM to follow IPD, but BIM is
a tool that facilitates communication and enables the team to cooperate and to
collaborate more efficiently in IPD. For example, it enables the owner and contractor to
explore a project’s key physical and functional characteristics digitally before it is built.
New legal and contractual concerns are also a side effect of applying BIM in the
design and construction process, which will affect not only the contractors but also all
the other parties. First, a single BIM model blurs the traditional allocation of contractual
responsibilities that the industry is already familiar with. Some main concerns include
the ownership of the model, the cost of model management, new insurance policies,
risk of software defects and failure, and others (Popovsky et al. 2012). Contract
documents developed specifically with BIM and IPD in consideration include the
Integrated Form of Agreement (IFoA) in UK and ConsensusDOCS300 supported by
44
Associated General Contractors (AGC) in US (Lichtig 2005). The American Institute of
Architects (AIA) has also developed a new contract family dedicated to IPD (AIA 2007).
However, there are also researchers who regard BIM as a tool that will not alter the
legal framework fundamentally (Popovsky et al. 2012).
BIM is no more than just digital models sitting in the computer system; it is the
team and the people who can exert the power of the BIM process. Love et al. (2011)
claimed that the error-reduction advocacy of BIM is misleading and many design errors
fall into the category that cannot be simply captured and corrected by software. The
application of new technology brings changes in roles and responsibilities as well as
work processes among the stakeholders. The successful implementation of BIM
requires a cultural change besides pure technology, especially considering the initial
information producer may not really enjoy the down-stream benefits promised by BIM.
For example, some authors regard BIM as a “living historical database of every material,
component, assembly, and system used in the building,” which will ultimately benefit its
renovation, restoration and demolition. However, technically paper based building
delivery will achieve the same effect, as long as all the information has been recorded
and indexed properly. BIM as a technology is just making things simpler and more cost-
effective but a BIM approach will not achieve this expectation automatically, since these
data still need to be inputted into the model by somebody. A poorly maintained BIM
model without any information update will soon become obsolete and useless.
Jung (2011) noted that at this relatively early stage the advocated benefit of BIM is
still difficult to justify. As more applications of BIM are appearing in the industry, the
knowledge about BIM expands and the data about real BIM projects are captured. More
45
empirical research is needed to verify the claimed benefits and to compare the
performance of expectation and reality. Only solid data will generate more focused
research and further the application of BIM. Research on the cultural, educational, even
the psychological aspects of BIM are all needed.
Industry Foundation Classes (IFC)
IFC specification
The data interoperability discussed above also exists in the BIM era. Since BIM
expects a single model to be reused by different parties along the lifecycle of a building,
the data interoperability issue is even more critical. Technically, if all the parties are
using software applications from the same vendor, the data will be interoperable even
with the vendor’s proprietary data format and the interoperability issue will not be a
problem. But this is rarely the case in real projects.
IFC is the set of internationally standardized object-oriented definitions describing
the consistent data representation of building components, developed by
buildingSMART, formerly known as International Alliance for Interoperability (IAI).
Addressing the interoperability problem in the BIM era, IFC is designed to be a vendor-
neutral, open source common language for the storage and exchange of intelligent
building objects between disciplines across the building lifecycle, and is currently the
only widely used openBIM standard.
The STEP Physical File is the native data file format for current IFC documents,
usually with an extension .ifc, sometimes with .stp. This format is based on plain text,
and will become quite large if used to store all the building information in one file
(Campbell 2007). This is a factor that contributes to IFC not being used in any software
for internal data storage, computation and real-time rendering. However, as an instance
46
of the ISO 10303 international standard, one advantage of IFC is that it is an open
standard and everyone has full access to the information within. Therefore it is ideal for
transferring data between different software platforms. Another advantage of IFC is its
built-in support for XML, which allows any IFC model to be described in ifcXML format in
a standard XML file. As a result, an IFC model is very versatile for use outside
traditional construction software applications. For example, the information stored in a
model could be easily transferred to human readable web pages (AEC3 2009).
The object specification in IFC, which includes not only the geometric information
but also properties and relationships, describes a set of well defined ways of identifying
object information and endows the IFC objects with intelligence (Vanlande et al. 2008;
Eastman 2008). Generally the data in IFC could be classified into three types. The first
one is geometric data, which is the 3D shape representation of the physical parts of the
building. The second one is property data: The property data is description of the object,
and could be further divided into two kinds: attributes and properties. The attributes are
often metadata about the object, like GUID and OWNERHISTORY. They are part of the
IFC specification, but not directly related to the object as a building element. The
properties define the object as a building element, like material and color of the wall.
They are often predefined in standard sets called property sets (Pset). Users and
applications could also create specific custom property sets for the information that is
not defined in IFC. Although in most BIM applications the classification of attributes and
properties is not explicit. For example, both the attributes and properties are called
parameters in ArchiCAD. The third one is relationship data: Relationships in IFC link
different building components and give them meanings. For example, the containment
47
structure is defined by the IfcRelContainedInSpatialStructure relationship. The
relationship data is the key to the semantic ability of the IFC specifications, through
which additional meaning exploration is made possible.
The development of the IFC specification started in 1995. IFC is deemed as a
particular implementation of ISO 10303 standards. This ISO standard has several
different representations, including Part 1 (EXPRESS schema), Part 21 (STEP Physical
File, SPF), and Part 28 (XML). The IFC 2x3 Technical Corrigendum 1 is the most recent
stable version. With the newest version of IFC 2x4 Release Candidate 3, the IFC
specification is seeking a new ISO standard (ISO 16739) with updated documentation
format, but the new version and the standardization process has not yet been finished.
By defining an innovative, globally available open standard for the description of
AEC objects, IFC specification transcends both the technology and market limitations
(Graphisoft 2004). Proven benefits of the IFC-based approach include improved quality,
reduction of errors, better coordination, and new services to owners and other partners
in the development process (CIFE 2002).
IFC application
As mentioned earlier, IFC is not currently used as the internal data format in
current BIM authoring software applications. The basic application is to view a BIM
model in IFC format using one of the many available IFC viewers. The viewers generally
will also have the basic ability to read and explore parametric data besides the 2D/3D
geometric information.
The IFC format is also used to translate the BIM model from one software
application to another. It is also used to combine different disciplinary BIM models into
one single model for model coordination and clash detection.
48
The IFC format is widely used in Web-based BIM applications. Normally a Web-
based BIM application is setup for the purpose of interdisciplinary collaboration with
partners from different domains and maybe outside the organization. IFC as the
interoperability enabling technology is particularly suitable for such environments.
However, showing the IFC model in a Webpage, especially when the model is large, is
still a problem that needs further research.
IFC implementation
Most of the IFC models are generated from BIM models already built in BIM
authoring software packages. Because of its complexity, it is almost impossible to build
an IFC model from scratch directly. In addition, the BIM authoring software provides
different kinds of tools and functions that make the BIM model building process more
efficient.
However, after an IFC model is generated, it is possible to manipulate it directly.
Currently there are two ways to do so. According to Nisbet and Liebich (2007), there are
four methods in working with IFC data:
The first one, which is used most, is using toolkits (or toolboxes) to operate directly
on the SPF (.ifc) file, which essentially is a pure text file, the native format for IFC
specifications and the only recognizable format for most of the IFC viewers. Some
toolkits available for this purpose include: 1) ECCO Express Toolkit by PDTec, which
works directly under the ISO 10303 standard; 2) Express Data Manager (EDM) by EPM
which works under the ISO 10303 standard also; 3) IFCEngine DLL by TNO, which only
works on the IFC file format. Several software implementations, including almost all the
viewers, are built on these toolkits.
49
The second one is using Express STEP SDAI bindings to a product model server.
A model server stores and manipulates building data in IFC format over a LAN/WAN
network environment, using either TCP or HTTP protocols. Some high level functions
are often provided, such as coordination and project management, life cycle and
operational data management. According to Graphisoft (2004), besides simple data
storage and speed / performance requirement, other key aspects of a building model
server include: discipline (partial model) and merge; concurrency and transaction
processing; team collaboration and audit (user roles and tracking); version control; data
protection; and queries. One of the most prominent model servers is the bimServer.org
developed as an open source project (bimServer 2011).
The third one is using database query and modification languages including
Express-X. The fourth one is using XML technologies on the ifcXML file, which requires
transforming an .ifc file to an .ifcxml file first and then using the XML libraries available
in high-level computer languages, e.g. Simple API for XML (SAX) and Document Object
Model (DOM) libraries in Java. If the modification of the .ifcxml file is valid, it can always
be transformed back into an .ifc file.
It is extremely difficult to implementing .ifc directly because of its complexity. For
example, in IFC, there are two ways to represent shapes: extrusion or brep. Extrusions
are typically modeled with IfcExtrudedAreaSolid, while brep is modeled with
IfcFacetedBrep. Most extrusions are straight. Breps can model anything, but breps are
considered to be a surface only, instead of a solid model. For the purposes of the IFC
Coordination View, extrusions (solids) are required. Boolean operations could be used
to modify the IFC shape representations. For some specific modifications IFC has
50
dedicated entities. For example, there is a specific IfcOpening element to model
openings for doors and windows, as long as the opening and what they are applied to
are extrusions (Lipman 2011).
IFCXML implementation
Recommended by World Wide Web Consortium (W3C), XML is a simple and
flexible hardware- and software-independent, general purpose tagging specification for
text, and could be extended to create custom markup languages. It is used for storing,
sharing, exchanging and transmitting data. XML schema, or XML Schema Definition
(XSD), is used to describe the structure of an XML document and to validate an XML
document. The purpose of XML is to facilitate the exchange of structured data.
XML is open source standard, but there are many proprietary extension and
adaptation for specific purposes. For example, gbXML is defined to facilitate the building
information exchange regarding green building analyses, and has been widely
supported by many CAD, BIM and engineering software applications. IfcXML is a similar
application of XML in IFC, proposed to address the gap between the increasing demand
for IFC and the difficulty of IFC implementation in EXPRESS modeling languages. As
an extension of the standard EXPRESS-based IFC specification, ifcXML is an XML
implementation of the IFC specification, targeting the XML development community.
The first ifcXML for IFC 2x2 was announced in 2004. The current version is ifcXML 2x3,
corresponding to IFC 2x3. The newest version of IFC 2x4 does not yet have a
corresponding ifcXML schema.
Similar to the difference between XML schema and XML files, ifcXML could also
be separated into two parts: first, the ifcXML schema (XSD), or part 28 edition 2
implementation of ISO 10303, which is an automatic conversion from the ISO 10303
51
part 1 EXPRESS schema representation of IFC specification; and second, the ifcXML
data file, which is equivalent to a corresponding SPF or .ifc file, and is the instantiation
of the schema. IfcXML data files are normally with the file extension name .xml,
sometimes with .ifcxml or .ifx. An ifcXML data file must conform to, and can be validated
by, the ifcXML schema. Each data element in a SPF file can also be expressed in an
ifcXML document (Nisbet and Liebich 2007).
One of the disadvantages is the much larger file size of ifcXML. As we have
mentioned above, a model file in IFC is already larger than the internal data structures
of BIM applications, the repeating tags in ifcXML makes the situation worse, sometimes
even 10 times larger (Nisbet and Liebich 2007). The other disadvantage is that the
derivation of ifcXML schema (.xsd) from Express schema (.exp) will result in information
loss. Nisbet and Liebich (2007) identified three limitations, including inverse
relationships, derived attributes, and constraints like the WHERE rules. One of the
initiatives of the IFC-based ontology development is to recover those lost constraints in
the form of ontology so that the combination of ifcXML and the ontology will remain
equivalent to the original Express schema.
On the other hand, XML is the W3C recommended basis of eCommerce and Web
services. Compared with EXPRESS format, XML enjoys a much wider range of support,
from tools to manipulate XML documents to developers who understand the details of
the technology. For example, with XSLT support, the same intelligent information could
be easily transformed into other formats for different use scenarios, e.g. same model on
different platforms like smart phones; or documents or messages generated based on
52
the model. With the extension of ifcXML, IFC is expecting much wider international
acceptance.
In general, IFC will remain as the primary and standard file format for data
exchange, especially when the volume of information involved is large. It will mostly be
used in the background and when the information being exchanged does not need any
manipulation. When the information accessibility is emphasized (e.g. partial model
extraction or building report generation), or information need to be accessed in a
generic method (e.g. in a database scenario or by a GIS application), ifcXML will be the
primary format developers will target.
Ontology
XML
XML, which has been mentioned earlier in the Data Operability and ifcXML
sections, is also the backbone of both ontology and Web services. XML is a platform
independent data representation. The basic idea of XML is adding tags that include
context information into a text document. On the Web, these tags are hidden to human
readers but can be used by machines and programs to make intelligent reasoning and
automatic decisions. The resulting XML file supports structured document interchange
and processing.
XML is a standard yet extensible data format, aiming to solve the interoperability
issue between different computing platforms. More details about XML are beyond the
scope of this dissertation. One of the major benefits of the Web services as discussed
later is platform independence. A Web service could be consumed by clients written in
almost any modern programing language. XML works as the intermediary bridge and
the fundamental enabling technology of this language transparency.
53
Metadata and Semantic Web
Currently the Internet and World Wide Web has become an indispensable part of
the daily life. A tremendous variety of information and services are available on the Web,
sometimes at a much lower cost or even free, especially when compared with the cost
of traveling to different locations in order to finish the tasks in person. However, the
intelligence of and integration between different websites and / or services is far less
than ideal. The World Wide Web was initially designed as a document system whose
content is meant to be displayed by Web browsers and is only meaningful to human
readers rather than to computers (Cardoso 2007). The data and information expressed
on a web page is difficult for a computer to extract and understand, thus preventing
further automated information processing (Berners-Lee 2001). As a result, from the
point of view of a computer, the current Web is similar to a huge library without any
indexing system. The information it needs is stored somewhere, but there is no way to
find it. People still have to manually navigate to different websites to finish different
tasks. Giving some random information on a webpage, people have to make their own
judgment about the authenticity of the information. As a result, the current Web is
sometimes referred to as Syntactic Web, or Web 2.0.
The World Wide Web is undergoing an upgrade from Web 2.0 to Web 3.0, or
Semantic Web. The Semantic Web project was initiated in the hope of giving order and
meaning to the unstructured information available on the Web by adding contextual
information (i.e. metadata) to existing information. As a result, computers will have the
intellectual ability to discover, understand and process the data from diverse sources
automatically without human interference. The goal of this upgrade is to transform the
Internet into a distributed computing platform that could utilize the scattered computing
54
resources over the web to finish given tasks intelligently and automatically. This
transformation will also affect how traditional desk-top software works, including BIM
and other construction industry applications.
The first step towards the Semantic Web is to add metadata – data about data –
into the webpages by inserting “tags”. Metadata is the building brick of the Semantic
Web. The tagging ability of XML allows this task to be accomplished without much
difficulty. It allows users to insert tags into the current content on the web to label the
content. The context (or semantic) information stored in the tags will be available to
software applications (the so-called agents) reading the content, and make the agents
able to realize the difference between similar data. For example, by adding different
tags like <equipment>, <bird> and <constellation> around the word “crane”, the
computer will be able to differentiate a piece of construction equipment from a kind of
bird or the constellation, although they may appear with the same name “crane”.
The second step of the task is to make the computers really understand the
meaning of the metadata by classifying the metadata in accordance with formal
ontologies. The concept of ontologies was actually put forward earlier than Semantic
Web. They originated in research about knowledge management (Neches et al. 1991).
The Semantic Web brought even greater need for ontologies. Because the same
concept may appear as different terms under different contexts, and the same concept
may be defined by different parties in different ways simultaneously, the stand-alone
metadata from different parties needs to be unified and linked together to be useful for
all Internet users.
55
Ontology and Ontology Languages
Ontologies play a central role in the Semantic Web (Knublauch 2004). Ontologies
are “formal, explicit specifications of shared conceptualizations,” i.e. the special
documents that define metadata terms (Cardoso 2007). Being used as a formal model
to capture the knowledge about some domain of interest, an ontology is represented as
a set of concepts within a domain and the description of the relationships between the
concepts (Akinci 2008). Ontologies work like an encyclopedia for computers, joining
heterogeneous metadata information, explaining all the definitions and listing all the
synonyms, therefore giving global structure to the data on the Web and allowing the
data to be understood and shared across multiple applications and communities.
The mathematical foundation of ontology is logics. Different ontology languages
exist, each with their own characteristics. The formal syntax of ontology language is
essential for the computer to comprehend the content of an ontology, and do reasoning
by itself. Through ontology language, complex concepts can be built from simple
concepts via boolean operators like intersection and union. Semantic reasoners based
on Description Logic can be used to maintain the hierarchical consistency of an
ontology.
In 1999 W3C published the Resource Description Framework (RDF), a
recommended specification for a system on how to locate and describe information.
Using RDF, each individual tag is linked to a Universal Resource Identifier (URI), which
contains the definition of a concept mentioned in the tag. Moreover, RDF uses triplets
(subject + verb + object) to link the individual XML metadata tags to form rules that
could be used in the reasoning of computer programs.
56
Besides the computer languages specifically designed for ontology engineering,
other software engineering tools like the Unified Modeling Language (UML) or Entity-
Relationship (ER) diagrams can also be used for modeling ontologies (Corcho 2002).
Endorsed by W3C, OWL is one of the most updated and widely supported ontology
languages (Horridge 2011). The most current industry effort for formalization of ontology
modeling is the Ontology Definition Model (ODM) which is expected to be a common
formal notation of ontology modeling (OMG 2009).
The Semantic Web makes Web information understandable and processable to
machines by adding annotation tags and linking the tags with ontologies. However, the
tags and ontologies are still static domain knowledge. To discover and actually make
use of this knowledge, Web services and Semantic Web services need to be used.
Ontology Research in Construction
The construction industry is considered to be information intensive. How to utilize
the information produced by the building process “intelligently” has been the focus of
several research efforts (Fidan 2010; Beetz 2009; Fernandez-Lopez 1999). Ontologies
are being used in various information-related areas, such as knowledge management
and data fusion, and are becoming an increasingly important research area in the field
of construction. Domain ontologies define concepts, activities, objects and the
relationships among elements within a certain domain. Several sources have been
explored to build an industry ontology for the construction industry. A detailed review of
construction industry ontology was conducted by Zhang (2012).
Construction industry knowledge management is among the first disciplines
focusing on the building and application of industry-wide ontologies. Several projects in
Europe have addressed this problem. The e-COGNOS project emphasized ontology as
57
“a basis for knowledge indexing and retrieval (Wetherill et al. 2002)”. Semantic
information is identified to be helpful in reducing ambiguities and improve accuracy in
the integration of heterogynous spatial data (Stadler and Kolbe 2007). In 2006, Open
Geospatial Consortium (OGC) examined the feasibility of representing Geography
Markup Language (GML) in OWL as part of the preliminary effort to extend existing
services, encodings and architectures with Semantic Web technologies (Akinci et al.
2008). Currently studies are being undertaken to investigate the opportunities to
leverage the current IFC model to derive ontologies and develop standard models of the
knowledge within the domain. The ISTforCE project explored the development of an
ontology to decode IFC models (Katranuschkov 2002). Besides the OGC and IFC
initiatives, building codes seem to be a promising alternative to build ontologies (Cheng
et al. 2008).
When several ontologies are available, it is necessary to choose one of them to
use for a specific Web Services call. Semantic and ontology matching and mapping is
becoming an interesting topic since it plays an important role in joining heterogeneous
ontologies to work together (Paolucci et al. 2002; Cheng et al. 2008).
Web Services
Distributed Computing
It is common that a software application is divided into hierarchical subtasks (or
subroutines, functions, procedures), which are then compiled and linked into a complete
software package. Some of the subtasks are in the form of standard libraries provided
by the programming language, like Core Java or Microsoft .Net; some are provided by
third party professionals as a variety of APIs, like the Jena API for ontology and the
OpenIFCTools API for IFC model manipulation used in this dissertation; and others may
58
be coded by the program author. The purpose of dividing a software application into
subtasks is for code management and to facilitate reuse of existing codes.
As time goes by, as more software subtasks are completed and become readily
available to solve most of the routine tasks, there is a trend in software development
focus that is shifting from coding from scratch to integration of existing subtasks
(Beringer et al. 1998). Software integration refers to the process of linking different
subtasks into a software package. Two forms of integration are identified: tightly
coupled or loosely coupled (Law 2011). The integration through code reuse is often
categorized as tightly coupled software integration, in which the subtasks are often
under single administrative entity.
On the other hand, the development of network and communication technology
makes it possible to access different, geographically distributed services instead of
keeping everything local in a central location. Distributed computing systems or
distributed computing environments refer to a system with multiple diverse, autonomous
computers or programs, each with its own memory or address space, to communicate
through message passing over hardware/software protocol stacks via a network and
achieve a common goal. The computing devices participating in such a system can
range from large servers to handhelds. Since the message passing in a distributed
computing system is much slower and less stable, many unique questions need to be
addressed in a distributed computing system, including more complex system
configuration and trickier debugging as a result of unpredictable network latency,
concurrency handling and partial failure (W3C 2004).
59
In a distributed system, the program subtasks mentioned above could be deployed
to different computers and the programmer does not need to be aware of the physical
location of the subtask being invoked. This technique is often named Remote Procedure
Call (RPC), a software development framework that emerged in the early 1990s (Kalin
2009). RPC has the client/server architecture, where a client invokes a procedure that is
executed on the server. Arguments are passed to the server and results are returned
back to the client. An Interface Definition Language (IDL) works as the service contract
between the message exchanges. The RPC is facilitated by lower-level network
protocols such as TCP/IP. On top of the lower-level protocols and standards, many
higher-level object-oriented protocols are developed to facilitate the direct use of
programmers. The early generation of distributed computing frameworks includes
CORBA (Common Object Request Broker Architecture), DCOM (Distributed Common
Object Model) from Microsoft, as well as RMI (Remote Method Invocation) from Java.
RPC is a pioneering technology in distributed computing. However, there are
some drawbacks. One of the problems of the RPC is that a lot of data needs to be
downloaded from the server when dealing with programmer defined datatypes. Another
problem is that the binary stream communicated is proprietary to a specific programing
language. But the fact is that modern software systems are written in many different
languages, and legacy systems are going to be hosted on many different platforms. In
the traditional RPC approach it is very difficult to reuse the libraries or subroutines
written in a different programming language or running on a different platform. The
introduction of XML is expected to change this situation with open standard structured
XML files working as intermediaries in document interchange and processing. Proposed
60
in late 1990s, XML-RPC is the evolution of RPC adopting XML as the communication
media. Although it was a simple and lightweight framework that only supports limited
elementary datatypes, XML-RPG became the foundation of Web services standards
later, with two key features implemented, namely language neutrality with XML
document and separate transport via HTTP (Kalin 2009).
The development of the distributed systems described above makes it easier to
integrate existing software components across different platforms via network.
Accordingly, current software application is transforming from stand-alone desktop
application utilizing subtasks compiled locally to distributed application utilizing different
computers located remotely. This type of integration is often referred to as loosely
coupled integration.
Combining the two trends mentioned above (the trend of coding to integration
software development and the trend of stand-alone to distributed computing), a new
software development model more suitable for distributed computing and software
integration was developed, which is the Web services model.
Web Services Definition and Benefits
Web services are a modern, lightweight approach of distributed computing (Kalin
2009). According to W3C (2004), Web services are defined as “a software system
designed to support interoperable machine-to-machine interaction over a network.” It is
a distributed computing system whose components can be across distinct devices
(Kalin 2009). Web services encapsulate certain functions, which can be accessed by
other applications over the Internet via an appropriate interface. Through predefined
query syntax Web services can retrieve specific information for a user from the Web
and/or finish specific tasks using the resources available on the Web. In plain words, a
61
Web service is a function over the Internet to perform some task or to provide some
information (Law 2011).
Web service is an implementation of RPC concept that uses Internet or Intranet as
a communication medium. Web service has become popular in not only integrating
newly developed software but also extending existing software applications as a
wrapper application outside legacy applications and making them accessible via
Internet.
Distributed system build on Web services are loosely coupled, which means the
client and service may not know the details of the other. This type of interaction
between components is defined formally by the Service-Oriented Architecture (SOA),
which is further discussed in Chapter 4.
W3C (2004) summarized the situations in which Web service is most suitable,
including Internet operations where reliability and speed is not guaranteed; deployment
management where centralized upgrade is not possible; and system components that
are from diverse platforms and vendors. The biggest advantage of Web services is that
it integrates different applications via a platform-independent standard-based framework
(Cardoso 2007). The Web services technology is based on W3C standards like HTTP
and XML. The XML documents as the communication media are in plain text and can
be inspected, validated and processed. Almost all the modern programming languages
have an existing library for processing such documents. SOAP also provides data
binding mechanisms to handle complex user-defined datatypes so both the service side
and the client side only needs locally available libraries for message processing. It hides
the implementation details from the clients and provides a standard means to integrate
62
different software solutions from different programming languages and/or on different
platforms. On the hardware side, Web services can be deployed on distinct devices,
ranging from business grade servers to PCs and handhelds.
Web Services Basic Model
According to the W3C standard, the basic working model of Web services consists
of two roles: a service consumer and a service provider. A service consumer (or
requester) is an entity that needs the function provided by some Web service provider.
A service provider is an entity that could finish certain computing tasks and return the
result to the consumer. The provided service may be implemented in arbitrary
languages or platforms, but the service must have a Web service description interface
described in the Web Service Description Language (WSDL), which is a machine-
processable XML file working as the service agreement that both the consumer and
provider agree on. The interaction between the consumer and the provider uses Simple
Object Access Protocol (SOAP) messages, which again is an XML file. Generally the
transport of the message is via HyperText Transfer Protocol (HTTP), which is the
standard Web protocol, but other lower level protocols are also used for carrying the
SOAP messages.
The key is that the service provider is remotely located and communicates with the
consumer through a predefined interface via Internet protocols. The service provider
may provide the service based on its local resources, but more importantly, it could
search the web and provide the result on the fly by revoking other Web services.
In the reality, mostly the consumer is the one who initiates the message exchange
between the provider in a request/response message exchange pattern. Before a Web
service is invoked, the consumer and provider first need to be known to each other, or
63
at least the initiator must have knowledge of the address of the other party. As long as
the consumer knows the URL address of the provider service, as well as the
parameters for invoking the service, it can communicate with a specific service provider
directly. If not, a third party, the service registry, may be needed to match the consumer
to the provider according to the service request it specifies. Each service is going to be
registered in a public Universal Discovery Description and Integration (UDDI) registry.
The UDDI registry is also a Web services standard for storing WSDL files describing the
Web services so that they can be discovered by clients. This basic model is shown in
Figure 2-4.
Figure 2-4. Basic Web services model
Web services have developed into a very complicated and comprehensive system,
with over 70 standard initiatives in 10 categories, including WS-Interoperability, WS-
Security, and WS-Messaging, to name a few. Besides the standard SOAP-based Web
services, REpresentation State Transfer (REST-style or RESTful) Web service is
another important branch. A review of these details is beyond the scope of this research.
64
Semantic Web Services
As previously mentioned, as a static knowledge depository, ontology alone is
hardly useful in the real world. Many researchers have resorted to Web services to
access the power of domain knowledge stored in an ontology. On the other hand, Web
services is promising in providing a platform for cooperation, but the WSDL contract
only governs the mechanics of the service interaction and lacks semantic representation
capabilities on the meaning and purpose of the service being described, therefore a
Web service alone is not capable of automatic integration (W3C 2004; Cardoso 2007).
Semantic Web services have thus been proposed to extend the Web services concept
by adding machine-interpretable semantics, making the Web services tasks, including
discovery, composition, execution and integration processes automatic. By combining
the power of Web services and the semantic Web, Semantic Web services are
becoming an important application area for ontologies and the Semantic Web.
In semantic Web services applications, Web services are described by ontologies
or similar semantically rich languages, enabling automatic discovery and execution of
the Web Services. All the functional and non-functional aspects of describing Web
services are defined in a single framework. Currently such functional Web services
description frameworks include Web Service Modeling Ontology (WSMO), WSDL-S,
OWL-S and SWSF. The benefit of using ontologies for the description of Web services
is that the reasoning ability of a formal ontology is an important prerequisite for
automating Web services tasks. The common understanding of an ontology in a specific
domain is expected to increase the usage of Web services descriptions (de Bruijn 2009).
Wetherill et al. (2002) suggested a knowledge management platform based on the
Web services model. Although they applied the ontology concept and used the
65
construction domain ontology as “a basis for knowledge indexing and retrieval”, they
failed to specify the details of building a working ontology for the system.
The Construction Information and Knowledge Protocol/Portal (CIKP) framework
proposed by Zhang (2010) is a standard Web-based application build to address the
information distribution inefficiency problem in the AEC industry. A domain-level
ontology is used to encapsulate knowledge about actors and roles in the AEC industry,
which is extended to an application-level ontology and used to support the semantics in
the framework. The system integrates publish/subscribe features in the Semantic Web
and claims to breakdown linear communication through social involvement.
The effort of Vacharasintopchai et al. (2007) to build a working semantic Web
services framework for computational mechanics is a good example of combining the
Semantic Web and Web services together to work in the real world rather than
academic laboratories. Their framework is built on a smart phone rather than normal
desktop operating systems, which is very promising for mobile computing requirements
such as on a construction jobsite.
The core standards of the Web services model were finished in 2004. The
Semantic Web services concept is still under development. Besides a complicated
infrastructure, a working Semantic Web needs much more effort to fill content into this
frame. Web services, especially with semantic capabilities, are still very rare.
Difference with Similar Concepts
A clarification of the meaning of Web services is necessary to prevent confusion.
Since the concept of Semantic Web and Web services is relatively new, these terms
may have been used by different authors in different contexts. While W3C formally
defined the Web services as described above, the phrase is often misused as simple
66
combination of “Web” and “service” to denote the ability of the application to utilize the
Internet or the World Wide Web via an arbitrary communication protocol. In a working
group note, W3C (2004) developed a strict definition of Web services. Its components,
including interface description and message passing, must conform to specific Web
standards. According to this definition, some so-claimed Web services applications are
in fact web-based or web-enabled applications (Chen et al. 2006; Vacharasintopchai
2007).
Another research field that is closely related to Web services is the multi-agent
systems (MAS). MAS are identified by Beets (2006) to be particularly suitable for
distributed collaboration in the AEC industry in the context of the Semantic Web. The
review of Ren and Anumba (2004) is a good summary of the basic agent and multi-
agent system concepts and their applications in construction research. Agent, or smart
agents, is defined as an autonomous program that is capable of cooperating with and
learning from other agents and the environment. Being “active” to “perceive, reason, act
and communicate” is the key property of agents (Huhns and Singh 1998). The concept
of agents was put forward before the concept of Web services. As both the concepts
evolved with the developments in the Semantic Web, it is evident that the line between
these two are being blended (Beetz 2006), especially when the non-semantic Web
service protocols could be enhanced by semantic annotations. Currently there is debate
on the difference between an agent system and a Web service. W3C states that a Web
service is an “abstract notion” of the functionality being provided, which needs to be
implemented by the “concrete” software and hardware of an agent (W3C 2004). One
may implement two different agents using different computer languages to serve as the
67
same Web service. Some scholars even claim they are essentially the same (Pathak
2006). The concept adopted in this study is that Web services is a special kind of
software agent, which conforms to W3C Web services standards (W3C 2004).
68
CHAPTER 3 METHODOLOGY
Literature Review
From Construction Point of View
The advent of BIM is built on a long history of IT applications in the AEC industry
and is the continuance of integration of CAD and other IT solutions available in the time
that the industry is rethinking the fragmented relationship between different
stakeholders. The research starts with a literature review on the history and status of
different IT applications and research in the construction industry. The unique
characteristics identified for the construction industry brings forward special IT
requirement. Two different categories of IT, namely Computing IT and Communicating
IT, are proposed to classify the different IT technologies; and their difference and
relationships are discussed.
After the general literature review on IT construction, the application and research
on BIM is studied, including its basic concepts, history, benefits, as well as other related
topics including IPD. Different from other engineering fields, the discussion on BIM is
very difficult to evade the specific vendor software solutions. Instead of endorsing any
specific vendors, IFC, the only openBIM standard currently available, is reviewed in
detail as a general reference implementation of BIM concepts.
The application of IT on construction jobsites, especially for construction
production workers, is a special field that is often neglected by researchers. With the
introduction of BIM, if production worker can be more involved in the work flow and
information exchange process, it will greatly improve the construction jobsite efficiency
and model data accuracy with little cost overhead. On the other hand, the popularization
69
of Internet and World Wide Web among general users provides us a good example of
how technology can be accepted in a much wider user base.
From The Technology Point of View
The World Wide Web and the Internet are good models of connecting
geographically scattered computing resources. With the help of wireless networks, the
Internet can literally reach every corner of the world and hence is a good fit of IT
deployment on construction jobsites, where the environment is harsh to electronic
devices and yet high portability and real-time response is required for building
information delivery.
Web service is an application of Internet with strict structures for distributed
computing and interoperability between different platforms and programming languages.
With the help of Web services, loosely coupled components could work under one
single framework. The research on ontology continues the research on artificial
intelligence in the construction industry and tries to link scattered information into a web
of reusable and machine-processable knowledge, where deductions can be carried out
by a computer program automatically.
Framework Development
IFC-based Ontology and Partial Model Extraction
With strict syntax and broad coverage, IFC is an international openBIM standard.
But the EXPRESS modeling language used in IFC is difficult to manipulate and rarely
used by general programmers. Raising the existing IFC specification onto a formal
ontology in OWL enables its access from widely used ontology tools. The IFC-based
ontology also can work as the basis for other model manipulation, such as partial model
extraction.
70
Model data manipulation is a critical part. Normally an IFC file is generated by BIM
authoring software such as Autodesk Revit. The IFC files are mostly used for
transferring model data between different BIM software applications, and are rarely
used for direct model manipulation. Several IFC and EXPRESS toolkits are tried and
the Java-based OpenIFCTools API is chosen as the tool to manipulate the model files.
The ontology and partial model extraction are developed completely in a traditional
desktop computing environment initially.
Web Service Integration
Service-Oriented Architecture (SOA) is the proposed solution in the IT world for
interoperability. SOA is a form of distributed computing systems, or a system of loosely
coupled Web services, with a relatively stable and fixed service interface separated
from the maybe frequently changing and updating service implementation. W3C (2004)
summarized the characteristics of SOA, including: logical view; message orientation;
description orientation; granularity; network orientation; and platform neutral. Basically
the service is formally defined in terms of the message exchange but the properties of
the service implementation detail are “wrapped” and not disclosed. The service
description is machine-processable metadata, often in platform-neutral XML-based files,
which is the key for the interoperability between different platforms and programing
languages.
Although the end result is the same, there are two styles as to how a new Web
services framework is developed, namely contract first and code first. In the contract
first style, the development starts with a WSDL service description and associated XML
schema, followed by detailed coding for actually implementing the service. Contract first
style guarantees the separation of interface and implementation, and hence is
71
recommended by experts in the SOA community, but this approach has been criticized
as being too complex to start and lacking proper tool support, as well as yielding
unpredictable results. The code first style is more suitable for exposing existing function
implementations to the Web via Web services. Service-related special code is inserted
into existing function-related code and the WSDL and XML schema are generated as a
result. Different platforms have different service-related code. In Java’s JAX-WS
adopted in this research, special code called annotations is required to be inserted into
original code to have the WSDL generated.
Based on previous research, the IFC model can be accessed through a desktop
computer in a Java-based prototype application. So the focus of this dissertation
research is to see how such applications can be brought to the Web under the code-first
Web service development methodology. The Web services framework proposed in this
chapter is an object-oriented implementation of a building information retrieval system
via SOA and the Web services model. Open industry standards are used wherever
possible. Both a core service and an assistant service are built to demonstrate and
validate the use of the framework. The framework can be easily expanded as long as
the same Web services model is observed.
Framework Validation
Besides the difference between an academia prototype and the requirement of an
industry application generally discussed in software engineering, it is also noted the
“catch-22” dilemma existing in the construction IT research (Lu 2002). It is very difficult
to validate the claimed benefits of an IT research without large scale industry application,
while on the other hand it is very difficult to implement large scale industry application
72
without validated benefits. As a result, most construction IT research resort to “prove-of-
concept” validations using test cases.
The framework prototype is tested using three different use cases, ranging from
simple text-based model information query to complex partial model extraction and 3D
geometry representation in a Web browser. All the use cases are extracted from real
world construction jobsite requirement for the model data delivery. Each use case
includes a background scenario, an input requirement and description of how the
proposed framework prototype can be used to solve the problem.
73
CHAPTER 4 RESEARCH BACKGROUND
IT Application on Construction Jobsites
Construction Jobsite IT Application Status
Although more components are being pre-fabricated off-site, the construction
jobsite remains the primary point-of-production in the construction industry. The
application of IT on construction jobsites is a special subfield. It is even lower than the
lowest “project level” of the levels of IT management discussed above, and is beyond
the general IT refactoring policies available in construction companies. Some
researchers categorize jobsite IT as “task level” in particular (Elvin 2003). In the IDEF0
model proposed by Bjork (1999), the author noticed that under a certain level, oral
communication will become the primary information exchange format and take the place
of the formalized paper-based document exchange, and below this level detailed formal
documents are no longer produced. Bjork also noticed a trend that over the time more
information is being explicitly formulated in project documents, i.e. the level from which
oral communication starts is becoming lower over time. In ancient times even grand
construction projects left very few documents, but heavily relied on the master builder’s
craftsmanship. Bjork argues the reason for this trend is more complicated building
systems and ever increasing division of labors in the construction industry. Our study
argues that the development of IT technology is also a very important driving factor in
this trend. It is the development of IT that makes it commercially possible to document
the information that was too difficult or too costly to record and transmit at a much
detailed lower level. Ultimately the detail of the documented information will reach a
level that a robot could finish all the construction routines. Currently the level where oral
74
communication prevails is on the construction jobsite, which is part of the reason that
normal IT technology is failing on the construction jobsite.
On the other hand, unique conditions exist on the construction jobsite besides the
construction industry characteristics discussed above. Portability is one of the special
requirements for construction jobsites, which is very similar to the requirement for
medical personnel and couriers. The requirement for high mobility makes even laptop
computers look cumbersome. Elvin (2003) uses the term “rugged” to summarize
another special condition on a construction jobsite, meaning its tendency to damage
both paper document and electronic equipment. Real-time response is also a special
requirement for jobsite work. Since this is where the concrete is being placed and the
bricks are being laid, if the required information needs to be waited on for delivery, the
project progress is being delayed; if the required information is wrong or out of date,
once the work is done it would be very difficult and costly to make changes.
Those special conditions make it very difficult to access required information in a
timely manner. Paper-based documents, either manually or electronically drafted, are
still the primary media for information delivery to the construction jobsite. First, paper-
based documents tend to be damaged. Second, large volumes of paper documents are
very difficult to manage and to distribute to different locations working concurrently.
Finally, when they are actually used, chances are that the information on the paper
documents are out-of-date, especially when concurrent engineering exists in DB project
delivery systems. As a result, the construction jobsite is regarded as the place that
information exchange breakdown is most apparent (Hameri et al. 1999), and waiting for
75
updated construction information is a leading factor that causes only one third of the
jobsite worker’s time to be engaged in productive work (Sweet 1994).
Technology and Research on Construction Jobsite IT
The data delivery of the “Communicating IT” aspect became the most demanding
IT application problem on the construction jobsite. Normal electronic devices are fragile
and cannot satisfy mobility requirements. Special technologies are required for work on
construction jobsites. Different IT technologies are being researched for various jobsite
functions, including wireless networks, companion computing, and RFID, in conjunction
with the Web technologies and project management site portal discussed in Chapter 2.
Mobile and wireless network is a technology especially suitable for the jobsite. By
transferring data directly to the point of work, the drawbacks of traditional paper
transmission, including media damage and content latency can be eliminated (Elvin
2003). Managers on construction jobsites can constantly connect to the corporate IT
system. With the higher bandwidth of 3G/4G wireless networks, such devices can
accomplish many tasks not possible on traditional wireless devices and networks. The
wireless and mobile networks working together with barcode and RFID tags are widely
used in the retail industry for inventory management and in the transportation industry
for packet tracking. The research on those identification technologies started as early as
the 1980s (Bell and McCullouch 1988).
Mobile computing, or companion computing which is newly marketed by Intel,
includes netbooks, tablet computers, and smart phones. Similar technologies include
digital pen and paper, digital hardhats, wearable computers, PDA and palmtop
computers (Bowden et al. 2004). They are being researched as their high mobility suits
the requirement of the construction jobsite well. Elvin (2003) is the first scholar to
76
promote the use of wireless-enabled tablet computers on the construction jobsite to
enhance the information exchange between jobsite workers and off-site collaborators.
Tablet computers are smaller in size and with large touch-screen, whose usage is very
similar to how paperwork is signed off on jobsite. With full HTML browsers preinstalled,
the fast development of smart phones and tablets makes the computing power of
mobile devices even greater. Many researchers have confirmed the advantage in
facilitating communication of using these technologies and the resulting improvement in
quality improvement as well as time and money savings due to reduced rework. A
number of implementation issues have also been found (Siegel 1995; Liu 2000; Elvin
2003; Thorpe et al. 2005).
Limitations of Current Construction Jobsite IT Applications
Thorpe et al. (2005) noticed that although a lot of research is done on jobsite IT
applications, most of it is focused on various site records keeping (Scott 1990; Cox and
Issa 1996; Cox et al. 2002). Except for the material delivery research which is largely
part of the supply chain management, the research targets in most of them are
“knowledge workers” instead of the production workers who actually lay the bricks
(McCullouch and Gunn 1993). The point-of-production workers are not involved in the
information exchange, neither information retrieval nor information contribution.
Reasons for this limitation include equipment cost, equipment life-span in the tough
jobsite environment and worker’s literacy.
During Elvin’s research in 2003, commercial-off-the-shelf tablet computers were
“available on the market for about the same cost as a typical high-end desktop
computer,” a price he thought was already acceptable. Now their price is even lower
than high-end desktop computers. There is no problem with construction firms affording
77
to buy the hardware. Compared to the magnificent hardware advance, the development
of software applications that could justify the requirement of information delivery to the
construction jobsite is much slower (Thorpe et al. 2005). One of the issues identified is
the limited integration of jobsite IT devices with the enterprise IT systems. Besides
some of the mobile versions for smart phones of the software vendor’s proprietary
products like AutoCAD (Autodesk 2012), currently the only integration available is
simple general purpose enterprise email and file synchronization. The disadvantage is
that emails and document files are limited in the information they can carry, which are
often not rich enough for complex construction jobs, and resulting in data redundancy.
Considering the large file sizes that come with BIM models, file synchronization will fail
on the limited wireless bandwidth available on construction jobsites.
As discussed previously, BIM is bringing a new paradigm into the AEC industry
with whole life-cycle support. However, it is noticed that this point of view is still largely
from architects, whose job is drawings and bid documents preparation (Ibrahim 2004).
From the point of view of contractors working on construction jobsites, there are still
limitations on BIM application, including technical, financial and managerial limitations.
Technical limitation refers to the requirement of computing powers by BIM applications.
Modern BIM software generally requires the support of advanced computer hardware,
which is not always readily available on construction jobsites. Financial limitations refer
to the cost of hiring technicians experienced on operating BIM software. Managerial
limitations refer to the fact that BIM applications are not always supported by company
executives because of the lack of concrete research and proof on the benefits of the
generally high-cost BIM applications.
78
Users may have different requirements as to the details required in a model. This
problem is addressed by different scales and Level of Details (LODs) defined by OGC
and AIA in the GIS world (OGC 2012; AIA 2008). When a larger area is requested, for
example, on the scale of a city, the data density required is much lower than that on the
scale of a block. Accordingly, lower LOD with less information and fewer details are
generated from higher LODs to keep the amount of data in the manageable range using
different generalization algorithms (Mao 2011). Information access on different levels is
rarely addressed in the current BIM applications. The same detailed information is
provided to different information users, either an engineer or a construction worker.
If research and development of BIM software applications can catch up with the
advancement in hardware development, construction jobsite productivity is expected to
be greatly improved by the timely and accurate delivery of construction information from
BIM models to jobsite workers.
As a summary of this section, a construction scenario example is presented of the
BIM application on a construction jobsite. Suppose a construction worker installing a
door needs to access the information about the door. Technically, this worker may not
have access to a powerful computer that can be carried around the jobsite to the point
of work. The worker may have a smart phone, but only can handle basic text
information. Financially, the subcontractor may not have the ability to invest in IT and
buy a dedicated BIM software license. Managerially, the worker may not have the
training to operate BIM software and understand the parts outside his specialty.
Accordingly, since only the information about the door and the connected wall might be
of interest, the other part of the model can be filtered out.
79
Two Alternative Ways of Delivering BIM Applications
Most of the current BIM applications are based on the stand-alone system
structure, with computing resources consumed and the model file stored locally. It is
difficult for partners to access the model from remote sites. The BIM application vendors
have realized this problem, and proposed several different solutions. The concept of
BIM should not be confined by the pattern defined by the software platform one may be
using (Eastman et al. 2007). Ibrahim et al. (2004) proposed a classification of BIM
implementation: the integrated (or all-purpose) application and the distributed (or
referential) approach. Most popular BIM software tools currently available fall into the
category of “all-purpose”, featuring a big model under a single platform storing all the
relevant building information. The advantage of this approach is efficient data storage
and transfer inside the platform, and maybe some powerful tools available exclusively
on the platform. The disadvantage is that the data are hidden in the vendor’s proprietary
data structure black box and difficult to enable cooperation between different platforms.
The users have to use the designated platform to fulfill all related tasks. Another
disadvantage is that the model file tends to become huge as the building proceeds and
new information is being added.
The other approach, the referential one, utilizes a central hub model that only
stores the link to different component models. Each component model is handled by
specialized software tools designed for specific building elements or building processes.
Although much less powerful than an all-purpose tool, those tools could finish the
specific tasks in a more efficient way. The BIM system under this alternative approach is
designed to share its data with others instead of hiding the data. The growth in size of
the central referential model is limited. For example, instead of storing the manufacturer
80
and model of a diffuser in the BIM model locally, only a link to an entry in the vendor’s
online catalog is stored. If the vendor calls back this product or decides to stop the
production of this product, all those linked to this entry will be notified automatically. The
problem, in this scenario, is that the link between the central model and the component
models must be valid all the time. For example, if they are distributed over the net, then
the models must be online all the time.
It is expected that even in an integrated all-in-one BIM solution, specialized fields
such as mechanical and structural are highly modularized. The key difference between
a modularized integrated BIM approach and a referential BIM approach lies in whether
the data structure is open and whether any other software platform is welcome to
interpret the data and make changes. In a referential BIM solution, the data is stored
under publicly available specifications, and the users are free to choose any capable
platform or software tool to finish the task.
Tardif (2008) classified BIM applications into “authoring tools”, “audit and analysis
tools” and “concept design tools”, but did not make detailed distinctions between
different categories. Actually, there are some overlaps for this classification. For
example, Autodesk Revit is known as an authoring tool, but its “massing” function is
targeted at the conceptual design stage. By extending Ibrahim’s alternative referential
BIM approach and considering current BIM applications, a categorization of all available
BIM applications is hereby proposed as: BIM authoring tools, BIM updating tools, and
BIM viewing tools.
BIM authoring tools have long been the focus of the industry and are expected to
be the core of BIM implementation. They are powerful, complex and costly software
81
applications that allow one to build a BIM model from scratch. Examples of BIM
authoring tools include Autodesk Revit, Bentley and ArchiCAD. BIM viewing tools are
those software applications that can only view the content of an existing model without
making any changes. A lot of IFC model viewers are available on market and most of
them are free to the user. Besides the simple viewing and exploration function, the
viewing tools should also be equipped with basic database search and query functions.
While BIM authoring tools and BIM viewing tools are widely available, BIM
updating tools are unexpectedly scarce. We define those tools as those software
applications that can make specific updates to an existing model. While less powerful
than BIM authoring tools, BIM updating tools should be specifically designed for some
tasks, simple to learn and use and highly responsive and they should also be much less
expensive. The highly limited application field of those tools is correct the niche for them
to survive and prosper. The system proposed in this research fits in this category.
IFC-based Ontology and Partial Model Extraction for BIM Models
One of the reasons that file-based information exchange is not suitable for real-
world data exchange needs is the huge file size generated for BIM models, which is
usually much larger than a few tens of megabytes (MB) (Adachi 2002). In a previous
research, an ontology-based partial model extraction method is explored on how to
reduce the size of the file that actually needs to be transmitted, which is briefly reviewed
in this section (Zhang 2012). As a result, instead of a whole model, only a partial model
is being transmitted. This section is a brief review of the research as the “partial model
extraction” function background of the Web service developed in this research.
In the 3D model world, there is always a gap between semantic information and
geometric information. For example, in the CityGML data model specification widely
82
used in geospatial applications, two sets of independent model hierarchies are defined:
the semantic model (buildings, rooms, doors, etc.) and the geometric model (solids,
surfaces, polygons, etc.). Different approaches which can be generally separated into
two categories have been proposed to bridge this gap. The first is to tag the existing
geometric information with semantics tools (e.g. tagging a box as a room); the second is
the other way around, namely store geometry information as spatial property of the
primary semantic objects (e.g. linking the shape property of a building element to a
dedicated shape representation, as used in IFC). Each of these approaches has their
own advantages and disadvantages (Stadler 2007).
Ontology could be used by a semantic reasoner to explore the hidden knowledge
and relationship between classes and individuals that are not explicitly declared. This is
one of the main reasons for building an ontology-based application (Dickinson 2009).
The domain-specific ontology for the construction industry has been explored in various
ways, including through the IFC specification as the primary ontology source in this
research. As noted earlier, the IFC specification itself uses the second approach of
storing geometric information as element property. Our approach of using ontology
could be deemed as adding the information in the first approach to bridge the gap
between geometric and semantic information, namely to use ontology to tag existing
geometric information in an IFC model. The extra link is stored as external index
information that makes efficient partial model extraction possible.
The first step is basic ontology development. This covers the part of the ontology
components that can be derived from the IFC specifications directly. The contents of the
IFC specifications can fulfill most of the ontology components requirements, and forms
83
the basics of the whole ontology. The classes or concepts requirement of an ontology is
about the nature or definition of certain terminology. They are also known as entities
(this is the term used in IFC) or sets. The “summary/definition” section of the IFC
specification gives formal definitions of a window, including the definition of the building
element from ISO as well as an explanation of other IFC entities or types used by or
related to the entity. The Uniform Resource Locator (URL) of the webpage can be used
as the URI to identify the term in the ontology. Classes are usually organized in
taxonomies with inheritance information. This information is available in the IFC
“inheritance graph” section, which traces the inheritance relationship back to the
abstract entity IfcRoot, the ancestor of all independent IFC entities. The inheritance
could be expressed by a subclass in the ontology. Table 3-1 shows the inheritance
relationship from IfcRoot to IfcWindow. The (abs) after the entity name indicates the
entity is an abstract entity.
Table 3-1. Inheritance relation for IfcWindow
IFC Entity IFC Schema
IfcRoot (abs) IfcObjectDefinition (abs)
IfcObject (abs) IfcProduct (abs)
Core – Kernel
IfcElement (abs) IfcBuildingElement (abs)
Core – Product Extension
IfcWindow IfcWindowStandardCase
Shared – Shared Building Elements
In IFC, the attributes of a class are included in the following sections: property set
use definition, geometry use definitions and attribute definitions. Property sets are most
typical attributes information. Each property is described in a word (string) or a number,
84
which are referred to as IfcPropertySingleValue in IFC. Other sections are also sources
of entity attributes, for example, the “geometry use definition” section includes details
about the height and width of the window, defined in the OverallHeight and OverallWidth
attributes, with each value represented as a positive number.
The relations in an ontology are also called roles. They denote how the classes or
entities are associated with other classes. Most of the relations are binary, meaning two
classes are involved. In IFC, most of the relations are defined as subclasses of
IfcRelationship, with prefix IfcRel. IfcRelationship is an abstract entity inherited from
IfcRoot, on the same level of IfcObjectDefinition shown in Table 3-1. The relations
between the IFC class and other classes are described in the “summary/definition” as
well as “containment use definition” (new in IFC 2x4) sections.
The next step is extended ontology development, which involves the ontology
components that are not originally included in the IFC specifications but are added
according to the requirement of the specific system or requirement. While the basic
ontology remains stable with each release of the IFC specifications, the extended
ontology can be more versatile and be updated more frequently according to the
specific requirements of the different systems that the ontology is being used for.
In formal DL, there is a separation between an ontology of the axioms defining the
classes and relations (TBox), and an ontology of the axioms of the individuals (ABox).
The ontology mentioned above is an ontology TBox without any information of specific
IFC elements in a concrete model. The first step in the extraction is when an IFC file is
read into the system, it is processed against the IFC ontology to generate an ontology
augmented IFC index file, which is essentially an ontology ABox with all the IFC
85
elements represented as ontology individuals. Meanwhile, a tree structure is also
generated in the internal storage format.
Based on the ontology-enhanced tree structure, a two-pass partial model
extraction algorithm is developed. The partial model extraction starts with some kind of
location information as an input parameter. The first pass is going up the tree from the
element to locate a proper container that could hold all the elements required for the
partial model. After the first pass, the second pass is going down the tree from this
container element to traverse all the potential elements. The elements that are
connected with the starting element and other elements under the same container are
checked. The location of each element is compared with the starting element. If the
distance between the two elements is in a certain range specified as the second
parameter of the algorithm, it will be selected for inclusion into the partial mode. Finally,
all the selected elements are reassembled into a new partial model. Figure 4-1 shows
the result of the partial model extraction algorithm, with a window and connecting wall
extracted from a complete model.
Figure 4-1. Partial model extraction sample
86
CHAPTER 5 WEB SERVICES FRAMEWORK FOR BIM
Framework Architecture
The conceptual architecture of the framework is shown in Figure 5-1. The system
is divided into three levels: the data level, the service level and the interface level.
Figure 5-1. Web services framework architecture
Interface Level
On the interface level there are two modules: a Web portal user interface and an
application interface. The Web user interface is a portal Web page provided for normal
end users to interact with the model data directly via the inputting of queries. It provides
operation options based on the user’s role in the system. It also accepts user input and
87
calls the relevant Web services, as well as shows the result of the operations. A special
function of the Web user interface is the ability to show an IFC model in a Web page
directly.
The application interface is reserved for external applications and application
developers. An external client program initiates the service by sending a service request
using the SOAP protocol to the Web service interface, and receives the returned value
from the Web service. Since Web service is an open standard, although the Web
service is implemented in Java Enterprise Edition, the client application does not need
to be using the same programming languages. Almost all modern programming
languages have SOAP Web service APIs available, including but not limited to
JavaScript, Ruby, Peal, and C#. To some extent the Web user interface can be treated
as a special external client implemented in Web page format integrated with the Web
service interface so that it can trigger the Web services directly.
Service Level
On the service level is the collection of all the Web services modules. The primary
part in the service level is the core service, or model service, deployed in a Java
application server. The core service is the only one in the system that can access the
model data. After receiving the requests from the interface, triggered either by Web
portal or by other applications, the core service analyzes the client’s enquiry and
invokes its internal logic to determine how to accomplish the request, which assistant
service to be activated if any, and send the query result or the detailed information of a
certain element specified by the user back to the client after the internal operation is
finished.
88
Other assistant services are also available for use under the same interface. The
“Dictionary Service” assistant service is shown in Figure 5-1 as an example of assistant
services. The function of this service is to translate between plain English words and
IFC terms.
The actual implementation of each service module may be different, including the
internal architecture, programming logic or even programing language. Some of the
services could also be implemented by third party applications. As long as they all
conform to the uniform Web services interface and other standard they should work
together. For example, instead of the ontology-based model access implementation
specified in this study, the core service may also be implemented through a relational
database that queries model data stored in a SQL database. The assistant service
could be written in another programming language that runs remotely.
When necessary, a service module can directly use another service module’s
functions. An example would be when a user searches for “window” in the core service,
the core service may consult the translation service to translate the “window” into
“IfcWindow” in order to run a query on the building model. But the translation service
could also provide its service independently. For example, if a user just wants to know
what the corresponding IFC term for a window is, the user could just run the translation
service and get the result. So although they could work together, the different services
are independent and modules may not necessarily know the physical existence of each
other.
Data Level
On the data level are those static data waiting to be exploited. First there is the
BIM model file in the IFC format (an .ifc file) which is ready to be queried by the core
89
service, along with the ontology augmented IFC index file (an .owl file), which is an
index file resulting from the preprocessing the IFC model file against the IFC ontology.
The IFC ontology discussed in Chapter 3 is also on the data level. Unlike some
ontology applications that display an ontology tree directly on the user interface,
(although the ontology is widely used in the internal programming logic of almost all the
service modules) it is not explicitly visible to the user. The consideration is that the
concept and operations of the ontology are beyond the targeted user population of the
system. Indeed, one of the goals of the system design is to shield normal users like
construction jobsite workers from the technical details of the building information model.
Hence, it does not make sense to expose them to the burden of yet another technology
called ontology.
Data Level Implementation
IFC Model Data Manipulation
The BIM model data used in the system are stored in .ifc files. As discussed in
Chapter 2, IFC is the open and neutral standard for BIM data exchange. Currently about
150 applications support IFC format BIM models (buildingSMART 2012). The IFC file
could be exported from different BIM authoring software applications. As the IFC
specification is still in development, not all model data in a BIM authoring software can
be represented in an IFC file, and different software vendors may have different
implementations on the IFC export functionality.
After an IFC file is generated, different toolkits are available to access the
information in it, as discussed in Chapter 2. OpenIFCTools (2011) is selected in this
research because of its complete Object-Oriented implementation in Java languages,
which is also the computing language that most of the framework is implemented in.
90
It is also possible to transform an IFC file into an IfcXML file as the lower level
model data storage format. The IfcXML file is a standard XML file which can be
manipulated with XML access technologies available in most programing languages;
however, currently there is no efficient BIM tool available to work on IfcXML files directly,
e.g. for building information visualization.
Ontology Manipulation
The ontology used in the system is in OWL ontology language, and is stored in
an .owl file. As discussed in Chapter 3, the basic ontology TBox, which contains the
ontology terms and relations, is developed by hand. Hence only a small fraction of the
complete IFC specifications are currently implemented in formal ontology format.
The ontology building is finished in Protégé (2011). Protégé is a Java based, free
and open source ontology editor developed at Stanford University. Protégé provides a
suite of tools to construct knowledge-based application with ontology. The ontology
which is built in Protégé can be exported into different formats.
Jena (2011) is an Apache project that deals with OWL ontology directly through
the Java programing language. It provides both lower-level RDF API and higher-level
OWL API. So it can both interpret an OWL ontology expressed in RDF/XML syntax and
access its RDF statement directly. It can work on both the ontology TBox and ABox,
namely both ontology classes and ontology instances.
Service Level Implementation
The core service, named “Model Service”, will receive and analyze the client’s
inquiry and return the result. Other assistant services are also available for use under
the same interface. The “Dictionary Service” is shown as an example of assistant
91
services in Figure 5-1. The function of this service is to translate between plain English
words and IFC terms.
The core service is implemented as one single service endpoint, which wraps its
multiple functions or operations into one service endpoint interface (SEI). An SEI (as
well as the WSDL generated from the SEI) is the part that is shared with both the
service provider that actually implement the service business logic and the service
consumer that make requests on the service, and works as the agreement between
them. In the code-first service development methodology adopted in this study, since
the basic functionalities have already been implemented as described in Section 4.3,
the function of the SEI is to expose the business logics via Web services so they could
be invoked by the service consumers remotely.
The SEI is an abstract Java interface without any actual implementation code.
Besides the SEI, a Service Implementation Bean (SIB) is needed to extend the interface
and to include the code for implementation of the declared operations. This
implementation class can further call other utility or helper classes in its implementation,
which further includes data accesses like reading the model file or the database.
The operations in the SEI are implemented as independent, self-contained
methods, in other words, stateless. This means the timing of each method being called
is not important. All the method calls do not have impact on the state of the system. The
end result of each method call is not depending on the specific sequence of previous
method calls. The only exception is that at the initialization the user needs to specify
which model to work on. If not, a default current model will be populated.
92
SOAP provides the Java API for XML Binding (JAXB) as well as other data binding
systems to exchange user defined complex datatypes, but in order to ensure neutrality
and interoperability between different programming languages, standard simple
datatypes are used in the system wherever possible. The XML Schema Definition (XSD)
type system is the default datatype system in SOAP, and is used to act as a translator
between the Web service’s datatypes (Java datatypes) and the client’s datatypes, which
may be in any programming languages.
Figure 5-2 is an illustration of the complete technology stack that could be used to
implement the core service published by W3C (2004). The technologies covered in this
dissertation research are mainly the ones included in the middle box with Processes,
Descriptions and SOAP Messages parts.
Figure 5-2. W3C Web services technology stack (W3C 2004)
93
SOAP Messages
The communication between the service and the client, as well as between
different service modules when necessary, is under the request/response message
exchange pattern. The client will initiate the message exchange by making a language-
neutral remote procedure call, invoking methods or operations in the service.
The messages to and from the Web services are implemented through the Simple
Object Access Protocol (SOAP). SOAP is an XML-based standard protocol for
packaging and exchanging XML messages used for communication between a service
and its client (W3C 2004). The SOAP specification defines the envelope structure,
encoding rules, and conventions for representing remote procedure calls and responses.
SOAP messages are essentially XML files over a variety of network protocols, normally
in HTTP, to describe service request/response messages. It provides a platform and
programming language independent way for Web services to exchange information.
Although the SOAP message itself has a complex structure, different APIs are
provided to cover the details and generate the messages automatically. So SOAP
messages are mostly hidden infrastructure working in the background. On the server
side, the remote procedures are specified by defining methods in an interface written in
the Java programming language. The developer also codes one or more classes that
implement those methods. Client programs are also easy to code. A client creates a
proxy (a local object representing the service) and then simply invokes methods on the
proxy. With JAX-WS, the developer does not generate or parse SOAP messages. It is
the JAX-WS runtime system that converts the API calls and responses to and from
SOAP messages.
94
Besides being language neutral, SOAP is also transport neutral, so clients and
web services have a big advantage of being able to use any transport technologies,
either standards defined by the W3C, or other proprietary transport protocols. However,
HTTP, which is the protocol for the Web, is still the dominant transport protocol.
WSDL Agreement
The clients need information about the services so that they can initiate the
request. The Web Service Description Language (WSDL) is used to describe each
service’s function syntax. WSDL is an XML-based language for describing a service and
its messages used to exchange information between the service consumer and the
provider. It includes both the abstract description and its binding information to a
concrete network protocols like HTTP. The information includes the service’s qualified
name in the form of namespace + local name and URL address, message format,
operation parameters and datatypes, transport protocols and other parameters needed
to advertise and invoke a Web service and the format that the result will be in. The
WSDL file is an agreement governing the mechanics of the service, and is referred to as
the service “contract.” XML-based WSDL is vendor neutral. As long as the WSDL
agreement stays the same, the implementation detail behind the WSDL on the server
side can be in any language or any platform. The client can also be implemented in
different languages or platforms. Figure 5-3 shows part of the Core Service WSDL that
is a standard XML file.
CXF Framework and Tomcat Server
The above mentioned Web service technologies are implemented in the CXF
framework and are published in Apache Tomcat Web server. CXF is an open-source
framework for developing Web services. It is a merger of previously separate Java Web
95
Figure 5-3. WSDL Web service agreement
service frameworks Celtix and XFire, and is now included as an Apache incubator
project. The frontend APIs that CXF supports include JAX-WS and JAX-RS, the
protocols it supports include SOAP, XML/HTTP, or CORBA. Spring is the lower level
platform that CXF builds on. Spring is an enterprise Java application development
framework. It is built on basic Java EE technologies and forms an infrastructure on
which business logics could be build directly.
96
The Web services as the end result of this research can be published in multiple
ways. The easiest involves the simple use of the publish method embedded in the
Endpoint in Core Java, without involving any Web servers. But this publishing method
lacks the ability to use the features integrated with Web servers like security
authentication, and it is very difficult to expand with dynamic Web content. Another
method of Web service deployment is to use a full-fledged Java Application Server
(JAS). The application server is a complex system with multiple components. The Web
container required for Web service publishing is a component in the Application Server.
Different Java application servers currently available include GlassFish from Oracle,
JBoss from RedHat and WebSphere from IBM. Besides Web content publication, JAS
also supports messaging, naming and directory, security functions needed in an
enterprise environment through other components. The publication in JAS requires
additional system configurations and the complexity may lead to many errors.
Balancing the functionality and the development difficulty, Apache Tomcat server,
which is a light-weight Java servlet container and the reference implementation of Java
Web container used in Java’s GlassFish application server, is adopted for the
deployment of the Web services in this research.
Assistant Service and Additional Services Implementation
The assistant service is implemented as a separate and independent SEI. When
necessary, a service module can directly use another service module’s functions. An
example would be when a user searches for “window” in the core service, the core
service may consult the translation service to translate the “window” into “IfcWindow” in
order to run a query on the building model. But the translation service can also provide
its service independently. For example, if a user just wants to know what the
97
corresponding IFC term for a window is, the user can just run the translation service and
get the result. So although they can work together, the different service modules may
not necessarily know the physical existence of each other.
Similarly, other services can be developed as independent components of the
system. Figure 5-4 is a screenshot generated from CXF that lists all the active services
available at that instant in the system. The link leads to a standard WSDL document
that can be consumed by different programming languages. The actual implementation
of each service module may be different, as long as they all conform to the uniform Web
Services interface. For example, the core service may be implemented through a
client/server database query via Structured Query Language (SQL). Some of the
services can be implemented by third party applications.
Figure 5-4. List of all services generated by CXF
98
Interface Level Implementation
User Portal Implementation
The client portal is the user interface available to the end users. Mainly three
levels of Web technologies are used. The static content of the webpage is built in plain
HyperText Markup Language (HTML); the dynamic part of the portal webpage is
implemented in JavaScript; finally the Java Server Pages (JSP) is used to connect to
the backend Web services. These technologies are all widely used Web standards or
semi-standards, and can easily be transported to mobile devices.
Figure 5-5 is the portal page the user will encounter upon entering the system.
Listed on the portal page are the currently loaded model and available options, as well
as a link to the service list and WSDL files.
Figure 5-5. Screen shot of the portal website
99
The core service page is shown in Figure 5-6. All available IFC elements in the
current model are listed in a tree structure. Each node is a link that leads to the detailed
information of the element, as shown in Figure 5-7. The portal also talks to the Web
services through the SOAP messages and the tree structure are generated on the fly.
Figure 5-6. Screen shot of the Core Service
Figure 5-7. Screen shot of the element detail page
100
Web-based IFC Model 3D Visualization
As discussed above, the Internet and World Wide Web have become a very
important information media. With the assistance of wireless network technology, the
Internet is changing IT applications in both the entertainment and business world. Being
able to show a 3D model in a web browser is a critical step toward wider BIM
applications adoption.
The traditional method to display a 3D object in a Web browser is to use vendor-
specific plug-ins, including Adobe Flash, Microsoft Silverlight, or Java. Different
methods are not compatible with each other. Besides, this requires the user to install
the plug-in, which may be limited by different company policies.
The Web Graphics Library (WebGL) is the preferred 3D graphics implementation
in HTML5. WebGL is a JavaScript API that can render 3D graphics in the HTML5
canvas element in supported Web browsers. Different high level libraries and APIs have
been developed to abstract the lower level graphics manipulation commands. The one
used in this research is MyBimShare (myBimShare 2012).
The first release of WebGL was in 2011. Currently only Google Chrome and
Mozilla Firefox support WebGL. However, Microsoft Internet Explorer can support
WebGL indirectly by installing a Chrome Frame plug-in. WebGL is also supported by
smart-phone based browsers including BlackBerry, Nokia and Firefox for mobile
(iclkevin 2011). Figure 5-8 is a screenshot of the sample IFC model as displayed in
Firefox Web browser.
The advantage of using WebGL is its native support by Web browsers without the
need to install any third party plug-ins. It is argued that the disadvantage of WebGL is
the requirement of running lower-level code in the computer’s Graphics Processing Unit
101
(GPU), which would be difficult for computers with less powerful GPU, and may lead to
application crash when the model is complicated. Another concern is that WebGL
actually needs to download the 3D object information into the client browsers, which
may be against security protocols of many design companies. In that case, other
technologies, such as Flash, O3D, VRML and X3D, must be used to manipulate 3D
graphics directly in a Web browser (Iglesias 2012).
Figure 5-8. 3D IFC model shown directly in a FireFox Web browser
102
CHAPTER 6 USE CASES AND TESTING
Three case studies were used one for the core service and the other one for the
term-translation assistant service, as well as a third one combining the two.
Use Case 1: Formatted Model Information Retrieval
This use case is a simple case showing the function of the core service, i.e. the
model service. BIM is regarded as the next paradigm of information technology
application in the construction industry after Computer Aided Design (CAD). However,
the majority of current BIM applications are still limited to the building design phase. BIM
applications in the construction field or on small projects are rare. One of the reasons is
that BIM models are usually computationally intensive. Handling such a model requires
powerful computing capability, which is not always available on construction sites. On
the other hand, most of the daily work on a construction site is being accomplished by
specialty subcontractors, who only deal with small portions of the project and do not
need to access the whole model file. In addition, due to the cost of BIM implementation,
subcontractors may not be equipped with the required BIM software tools or the
relevant training.
The proposed Web services framework suits the construction jobsite condition. As
most of the computing task would be accomplished on a remotely located server, or in
the cloud, locating computing power on the jobsite is no longer critical. Through the IFC
manipulation of the model, Web service is able to deliver the information of the part of
the whole model that is actually needed.
This case study shows the ability of the system to use standard IFC specification
terms to invoke simple inquiries of the model. Figure 6-1 shows the starting point of the
103
inquiry, with the first dropdown list giving all the IFC elements currently available in the
model for the user to choose from. Figure 6-2 shows according to the first selection
(IfcWindow), all the window instances within the model are listed in the second
dropdown list. Accordingly, the third dropdown list shows all the available property
information for the window elements. Figure 6-3 shows the result of the inquiry, with the
width of window #577 shown as the result.
Figure 6-1. A list of all IFC elements available in current model
Figure 6-2. A list of all IFC instances according to the selection of IFC element
104
Figure 6-3. Element property inquiry result
Use Case 2: Technical Term Translation
This use case is a simple case showing the function of term translation via the
Dictionary Assistant Service, which can be used, for example, in code-checking of a
building model against the building code.
As most building codes have already been released online, they are available to
be retrieved by software applications automatically. On the other hand, IFC-compatible
BIM is becoming a standard submittal requirement from owners. Technically, it would be
possible to check the model against the building codes, but in reality this is difficult to
achieve. Among other issues, term-matching is a problem. The information stored in the
model is under the IFC naming convention, e.g. a window is under the element name of
“IfcWindow”. The correlated term used in the building code would be simply “window” in
plain English. In order to enable the model and the building code to talk to each other, a
translator is needed. The technique currently adopted by the model-checking software
applications is to use a hard-coded dictionary, in which all the terms in building code are
linked to the corresponding ones in IFC. This approach, while feasible, has major
drawbacks when maintenance of the dictionary is required. The IFC specification is
constantly evolving. On the other hand, sometimes new terms need to be added by the
user. As the users of the software may lack the necessary knowledge of IFC
105
specifications and/or code checking terminologies, it would be difficult for the user to
add new specific terms into the dictionary when needed. It would be even worse if some
special computer programming language is involved, e.g. EXPRESS, the native
language used in IFC. Also, since each copy of the dictionary is locally stored with the
software on one computer, the update is difficult to spread to other computers or to be
shared by the other users.
By implementing the term translation service as a Web Service, all the update and
maintenance of the dictionary can be implemented in one central place. The users can
input any term freely as the inquiry keyword, as shown in Figure 6-4. The system will
match the inputted keyword with the ontology dictionary and list all the relevant
elements available in the current model, as shown in Figure 6-5. The user can then
continue to click on any element to access the detailed information of that element
including its geometric information in 3D, as shown in Figure 6-6.
Figure 6-4. User can input any string as inquiry keyword
Figure 6-5. A list of all elements satisfies the inquiry
106
Figure 6-6. Inquiry result with 3D geometry
Use Case 3: Partial Model Extraction
The research and application of building information modeling (BIM) has been
focused on the entire project and the complete life cycle. However, the daily routine on
a construction jobsite has specific requirements and bears certain limitations regarding
the usage of information stored in a BIM model. The limitations include scarcity of
computing power and trained personnel. One of the special requirements for daily
routine on jobsite is to be able to view the partial model that the team is actually working
on instead of the original complete model. The partial model may be defined by certain
location parameters such as floor numbers and/or building grid lines.
An ontology-based method to extract a partial model from a complete BIM model
is the approach taken in this study. The partial model, as well as the complete model,
should be defined in IFC format, which is the widely supported open standard data
exchange format for BIM. The extraction is based on an IFC-based ontology which
107
defines the necessary building blocks of a valid IFC model and the rules of extraction.
The whole process is to be implemented as a Web service allowing remote accessibility
from various computing platforms. The Web service system can also be linked to other
construction software applications for automating construction management functions.
The partial model extraction starts with a building element like a door or window. In
the system, the user must first use normal query methods described in earlier sections
to get to a specific building element, like a window, as shown in Figure 6-6. If a partial
model could be extracted based on the element, at the bottom of the element detail
page a link is provided for a partial model extraction, as shown in Figure 6-7. Figure 6-8
shows the extracted partial model as it appears in the Web browser.
Figure 6-7. A partial model extraction link at the bottom of the element details page
108
Figure 6-8. Partial model extracted shown in a Web browser
109
CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS
Conclusions
The application of BIM is part of the general IT infrastructure refactoring in the
construction industry against the background of integrated design and construction. The
benefits of BIM in information integration in the AEC industry have been determined by
many researchers and the construction industry has shown great interest in the
application of BIM. It is very difficult to deploy general IT technologies on a construction
jobsite because of its special requirements for real-time data, high mobility and its harsh
work environment. This research explores methods of enabling BIM and IFC on the
construction jobsite via Web technology.
After the basic concepts and implementations of BIM have been largely
established and standardized by now, the next step in BIM application in the AEC
industry is to extend its application from the industry leaders to the smaller companies
and from offices to the construction jobsite. Contrasting the slow adoption of new
technologies in the AEC industry, the fast spread of Internet and World Wide Web
provides us a model for new technology adoption of a much wider user base. By
promoting an ontology-based Web services framework for BIM, this dissertation
explores the possibilities of applying Web technology to promote the application of BIM
on the jobsite at a much lower entry threshold.
The Web services model is one of the many possibilities made available by the
Internet. Software components, including legacy systems and geographically distributed
procedures, can be dynamically integrated into one loosely coupled system and work
together under the Web services model. The framework proposed in this dissertation
110
builds on the Web services model and its versatility and simplicity. Language
transparency is one of the major benefits of the Web service framework. With proper
library support, the Web services framework described in this study can be queried in
any programming language.
The benefits of the framework include:
Standard. The framework is developed and deployed using both construction and
information industry standards that are vendor-independent and open-source,
including the XML and SOAP used in the Web services building and the IFC and
ontology used in lower-level data handling. The adoption of those standards
ensures the interoperability between different platforms and vendor products.
Language transparency. Not only the developed Web service can be used by
clients written in different programming languages, the existing Web service
framework itself can be extended by adding new service modules written in
different languages. XML as well as XML-based Web service standards including
SOAP and WSDL enable the interoperability between different programing
languages.
Modular design. Distributed BIM model is used to define the framework structure.
The Web services are developed in small units, and new service modules can be
developed and easily integrated with existing services. Furthermore, different
modules do not need to be centrally located, but can be running on different
machines.
111
Easier object data access. Data representation and access completely utilized in
Web page. The display of 3D model data does not need any plug-in or terminal
client software installation. Accessing model information does not require
extensive technical knowledge of IFC specification.
Authoritative information. Centralized model storage structure on the server makes
the model file management easy and guarantees the authority of the model
content to be up to date.
The construction jobsite is regarded as the “missing link” in realizing the full
potential of IT applications in the AEC industry (Elvin 2003). The framework explored in
this research is different from similar research on web-based project management
solutions. It consists of Web services working as the backend of the portal website,
which conforms to W3C standards and can be accessed from any programing
languages or operating systems. The integration of IFC-based ontology provides
intelligence to interpret and extract information from a BIM model without involving the
technical details of the IFC specification or requiring powerful computers. As a result,
the framework is able to reach the lowest level in a construction project hierarchy and to
take jobsite production workers directly into the information workflow of a project using
BIM and IFC.
Once implemented, the framework can be utilized in combination with other IFC-
supported BIM applications as well as by personnel like construction jobsite workers for
more accurate, consistent and up-to-date construction information access and to
improve the productivity of construction workers on construction jobsites. By simple
additions to the current framework, bidirectional information exchange can be realized
112
incorporating the real time information collected on the jobsite, which is expected to
improve the data reliability and robustness, and ultimately be helpful for construction
automation in the future (Thorpe et al. 2005).
Limitations and Recommendations for Future Research
The information exchange issue in the construction industry especially how the
model information can be delivered to the jobsite in the BIM era, is a fairly new and
complicated research field. Internet and Web technology has shown their suitability and
benefits from previous research studies. The frameworks proposed in this research only
address a small part of the issue, namely the timely information retrieval from BIM
models via Web technologies. There are other technologies that can achieve similar
functionality, with various levels of difficulties and costs. Other considerations that
should be taken into account for such a system to work on real construction projects
include, but are not limited to, model security, multiple model management, and user
authorization.
The primitive frameworks explored in this research only have limited functions. A
complete industry-wide framework requires much more work to build the ontology, the
functional services as well as constant maintenance. How to integrate the new services
into the currently existing enterprise IT frameworks is also a topic that needs more
research. Although Web services is expected by the IT community to be adopted as the
main application distribution platform in the future due to its advantage as a pay-per-use
service model, its commercial application in the construction industry needs more
research.
As discussed in Chapter 2, a successful IT application deployment should be a
combined development on two levels: the strategic level and the tactical level. This
113
research is mainly focused on the technical aspect on a tactical level. How to evaluate
and justify the benefits of adopting new IT applications is also an interesting and
complicated topic. At last, even a technically proven IT application need to be supported
by regulations and enterprise management policies on the strategic level in order to be
successfully deployed.
114
LIST OF REFERENCES
Adachi, Y. (2003). "Overview of partial model query language." 10th International Conference on Concurrent Engineering, Madeira, Portugal.
AEC3 (2010). "BIM services." <http://www.aec3.com/6/6_04.htm>. (Sep. 11, 2010).
AIA (2007). "Integrated Project Delivery (IPD) Family ", <http://www.aia.org/contractdocs/AIAS076706>. (Feb. 29, 2012).
AIA (2008). "Building Information Modeling Protocol Exhibit." <http://www.aia.org/aiaucmp/groups/aia/documents/pdf/aiab083007.pdf>. (Mar. 15, 2012).
Akinci, B., Karimi, H., Pradhan, A., Wu, C.-C., and Fichtl, G. (2008). "CAD and GIS interoperability through semantic web services." Journal of Information Technology in Construction, 13, 39-55.
Anumba, C. J., Baron, G., and Evbuomwan, N. F. O. (1997). "Communications issues in concurrent life-cycle design and construction." BT Technology Journal, 15(1), 209-216.
ASHRAE (2009). "An introduction to Building Information Modeling." ASHRAE, Atlanta, GA.
Autodesk (2012). "i-drop Indicator." <http://usa.autodesk.com/adsk/servlet/index?siteID=123112&id=2753219&linkID=9240618>. (Mar. 15, 2012).
Autodesk (2012). "AutoCAD WS: Extend AutoCAD to Web & Mobile Devices." <http://usa.autodesk.com/autocad-ws/>. (Mar. 15, 2012).
Beetz, J., van Leeuwen, J., and de Vries, B. (2006). "Towards a topological reasoning service for IFC-based building information models in a semantic Web context." Joint International Conference on Computing and Decision Making in Civil and Building Engineering, Montreal, Canada, 3426-3435.
Beetz, J., van Leeuwen, J., and de Vries, B. (2006). "Distributed collaboration in the context of the semantic Web." 8th International DDSS Conference, Eindhoven University of Technology.
Bell, L., and McCullouch, B. (1988). "Bar code applications in construction." Journal of Construction Engineering and Management, 114(2), 263-278.
Beringer, D., Tornabene, C., Jain, P., and Wiederhold, G. (1998). "A language and system for composing autonomous, heterogeneous and distributed megamodules." DEXA International Workshop on Large-Scale Software Composition, IEEE Computer Society, Vienna, Austria, 826.
115
Berners-Lee, T., Hendler, J., and Lassila, O. (2001). "The Semantic Web." Scientific American, 284(5), 34.
Berning, P. W., and Flanagan, P. (2003). "E-Commerce and the Construction Industry: User Viewpoint, New Concerns, Legal Updates on Project Web Sites, Online Bidding and Web-Based Purchasing." www.constructionweblinks.com.
Betts, M. (1995). "Technology planning frameworks to guide national IT policy in construction." Automation in Construction, 6, 251 - 266.
bimServer (2011). "Factsheet." <http://bimserver.org/faq-2/en/>. (Mar. 15, 2011).
Bjork, B. C. (1999). "Information technology in construction: domain definition and research issues." International Journal of Computer Integrated Design and Construction, 1(1), 1-16.
Bowden, S., Dorr, A., Thorpe, A., and Anumba, C. J. (2004). "Mapping site processes for the introduction of mobile IT." ECPPM 2004 eWork and eBusiness in Architecture, Engineering and Construction, A. A. Balkema, The Netherlands, 491-498.
Brandon, P. S., and Betts, M. (1995). Integrated construction information, Taylor & Francis.
Brown, C., and Beaton, H. (1990). "Looking back at design, looking forward to construction." Journal of Management in Engineering, 6(3), 342-349.
buildingSmart (2010). "IfcWindow." <http://www.iai-tech.org/ifc/IFC2x4/beta3/html/ifcsharedbldgelements/lexical/ifcwindow.htm>. (May 20, 2010).
buildingSmart (2010). "Industry Foundation Classes release 2x4 (IFC2x4) release candidate 2." <http://www.iai-tech.org/ifc/IFC2x4/rc2/html/index.htm>. (Mar. 15, 2010).
Campbell, D. A. (2007). "Building Information Modeling: the Web3D application for AEC." 12th International Conference on 3D Web Technology, ACM, New York, NY, 173.
Cardoso, J. (2007). Semantic Web services: theory, tools and applications, IGI Global, Portugal.
Chaaya, M., and Jaafari, A. (2001). "Cognizance of visual design management in life-cycle project management." Journal of Management in Engineering, 17(1), 49-57.
Chen, Y., and Kamara, J. M. (2011). "A framework for using mobile computing for information management on construction sites." Automation in Construction, 20(7), 776-788.
116
Cheng, C. P., Lau, G. T., Pan, J., Law, K. H., and Jones, A. (2008). "Domain-specific ontology mapping by corpus-based semantic similarity." 2008 NSF CMMI Engineering Research and Innovation Conference, Knoxville, Tennessee.
Chinowsky, P. S., and Reinschmidt, K. F. (1995). "Qualitative geometric reasoner for integrated design." Journal of Computing in Civil Engineering, 9(4), 250-258.
Cholakis, P. (2012). "Cloud computing and BIM." The Military Engineer, 104(675), 59-60.
Chuang, T.-H., Lee, B.-C., and Wu, I.-C. (2011). "Applying cloud computing technology to BIM visualization and manipulation." 28th International Symposium on Automation and Robotics in Construction, Seoul, Korea.
Corcho, O., and Fernandez-Lopez, M. (2002). "Ontological engineering: what are ontologies and how can we build them?" Semantic Web services: theory, tools and applications, J. Cardoso, ed., IGI Global, Portugal.
Cox, R. F., and Issa, R. R. A. (1996). "Mobile field data aquisition for construction quality control and ISO 9000 certification." International Workshop on Computing in Civil Engineering.
Cox, S., Persomo, J., and Thabet, W. (2002). "Construction field data inspection using pocket PC technology." International Conference on Application of IT in the AEC Industry.
Day, A. (1996). "The maquette, the model and the computer: organizational futures for design and construction." Engineering, Construction and Architecture Management, 3(1).
de Bruijn, J., Kerrigan, M., Zaremba, M., and Fensel, D. (2009). "Semantic Web services." Handbook on Ontologies, S. Staab, and R. Studer, eds., Springer, Berlin, Germany, 617 - 636.
de la Garza, J. (1994). "Value of concurrent engineering for AEC industry." Journal of Management in Engineering, 46-55.
Dickinson, I. (2009). "Jena ontology API." <http://jena.sourceforge.net/ontology/index.html>. (Jul. 4, 2011).
Eastman, C. (1975). "The use of computers instead of drawings in building design." AIA Journal, 63(3).
Eastman, C., Teicholz, P., Sacks, R., and Liston, K. (2008). BIM handbook: a guide to Building Information Modeling for owners, managers, designers, engineers, and contractors, Wiley, Hoboken, New Jersey.
117
eb.com (2012). "Construction." <http://www.britannica.com/EBchecked/topic/134411/construction>. (Feb. 29, 2012).
Elvin, G. (2003). "Tablet and wearable computers for integrated design and construction." ASCE Construction Research Congress, Honolulu, Hawaii.
Fang, I. (1997). A history of mass communication: six information revolutions, Focal Press, Boston.
Fergusson, K. J. (1993). "Impact of integration on industrial facility quality." Center for Integrated Facility Engineering, Stanford University, Palo Alto, CA.
Fernandez-Lopez, M. (1999). "Overview of methodologies for building ontologies." 16th International Joint Conference on Artifical Intelligence, Stockholm, Sweden.
Fidan, G., Dikmen, I., and Birgonul, M. T. (2010). "Using multi agent systems in construction claim negotiation." International Conference on Computing in Civil and Building Engineering and XVII Workshop on Intelligent Computing in Engineering, Nottingham University Press, Nottingham, UK.
Gallaher, M. P., O'Connor, A. C., Dettbarn, J. L., and Gilday, L. T. (2004). "Cost Analysis of Inadequate Interoperability in the U.S. Capital Facilities Industry." NIST.
Graphisoft (2004). "IFC 2x Edition 2 Reference Guide."
Graphisoft (2012). "Object Technology." <http://www.graphisoft.com/products/gdl-object-technology/>. (Mar. 15, 2012).
Hjelseth, E. (2010). "Overview of concepts for model checking." 27th International Conference on Applications of IT in the AEC Industry and Accelerating BIM Research Workshop, Cairo, Egypt.
Horridge, M. (2011). "A practical guide to building OWL ontologies using Protege 4 and CO-ODE tools, Edition 1.3."
Ibrahim, M., Krawczyk, R., and Schipporeit, G. (2004). "Two approaches to BIM: a comparative study." 22nd Conference of Education in Computer Aided Architectural Design in Europe, Citeseer, Copenhagen, Denmark.
Ibrahim, M. M., Krawczyk, R. J., and Schipporeit, G. (2004). "A Web-based approach to transferring architectural information to the construction site based on the BIM object concept." 9th Conference of the Association for Computer-Aided Architectural Design Research in Asia, Seoul, Korea.
iclkevin (2011). "WebGL on mobile devices." <http://www.ichemlabs.com/1375>. (Mar. 15, 2012).
118
Iglesias, D. G. (2012). "Design and implementation of 3D building integration for a WebGL-based virtual globe."
Issa, R. R. A., and Mutis, I. (2006). "Ontology based framework using a semantic Web for addressing semantic reconciliation in construction." International Workshop on Intelligent Computing in Engineering, 348-367.
Jena (2011). "Jena - a Semantic Web framework for Java." <http://jena.sourceforge.net/>. (Mar. 15, 2011).
Jung, Y., and Joo, M. (2011). "Building Information Modeling (BIM) framework for practical implementation." Automation in Construction, 20, 126-133.
Kalin, M. (2009). Java Web services: up and running, O'Reilly, Sebastopol, CA.
Katranuschkov, P., Gehre, A., and Scherer, R. J. (2003). "An ontology framework to access IFC model data." Journal of Information Technology in Construction, 8, 413.
Knublauch, H., Fergerson, R. W., Noy, N. F., and Musen, M. A. (2004). "The Protege OWL plugin: an open development environment for semantic Web applications." 3rd International Semantic Web Conference.
Kumar, B., and Cheng, J. C. P. (2010). "Cloud computing and its implications for construction IT." 13th International Conference on Computing in Civil and Building Engineering, Nottingham, UK.
Kumaraswamy, M. M., and Chan, D. W. (1998). "Contributors to construction delays." Construction Management and Economics, 16, 17-29.
Law, K. H. (2011). "Web services in civil and structural engineering simulations." International Symposium on Innovation & Sustainability of Structures in Civil Engineering, Xiamen University, China.
Law, K. H., Cheng, J. C. P., and Han, C. S. (2010). "Web services in architectural and civil engineering." 10th International Conference on Construction Applications of Virtual Reality.
Lee, J.-S., Min, K.-M., Lee, Y.-S., Kim, J.-H., and Kim, J.-J. (2008). "Building ontology to implement the BIM focused on pre-design stage." 25th International Symposium on Automation and Robotics in Construction, Vilnius, Lithuania, 350-354.
Lichtig, W. (2005). "Integrated Form of Agreement (IFoA)." <http://www.thechangebusiness.co.uk/TCB/ifoa.html>. (Feb. 29, 2012).
Liebich, T. (2010). "Unveiling IFC2x4 - the next generation of openBIM." 27th International Conference on Applications of IT in the AEC Industry and Accelerating BIM Research Workshop, Cairo, Egypt.
119
Lipman, R. (2011). "Solid model." <http://groups.google.com/group/cis2-group/browse_thread/thread/e952b914b422d1ab>. (Jul. 11, 2011).
Love, P. E. D., Edwards, D. J., and Han, S. (2011). "Bad apple theory of human error and Building Information Modelling: a systemic model for BIM implementation." 28th International Symposium on Automation and Robotics in Construction, Seoul, Korea, 349-354.
London, K. A., and Kenley R. (2010). "An industrial organization economic supply chain approach for the construction industry: a review." Construction Management and Economics, 19(8), 777-788.
Mao, B. (2011). "Visualisation and generalisation of 3D city models." Doctoral Thesis, Royal Institute of Technology, Stockholm.
McCullouch, B. G., and Gunn, P. (1993). "Construction field data aquisition with pen based computers." Journal of Construction Engineering and Management, 119(2), 374-384.
McGraw-Hill (2009). "The business value of BIM: Getting Building Information Modeling to the bottom line." McGraw-Hill Construction, New York, NY.
myBimShare (2012). "BimShare." <http://www.mybimshare.com/>. (Mar. 15, 2012).
Myers, B. A. (1998). "A brief history of human-computer interaction technology." Interaction, 5(2), 44-54.
Neches, R., Fikes, R. E., Finin, T., Gruber, T., and Patil, R. (1991). "Enabling technology for knowledge sharing." AI magazine, 12(3), 36.
NIBS (2007). "National Building Information Modeling Standard, Versioni 1 - Part 1: Overview, Principles, and Methodologies."
NIST (2008). "The Initial Graphics Exchange Specification (IGES)." <http://ts.nist.gov/standards/iges/>. (Sep. 18, 2011).
NIST (2011). "About the Current Version of IGES." <http://ts.nist.gov/Standards/IGES/ecoList.cfm>. (Sep. 18, 2011).
OGC (2012). "OpenGIS City Geography Markup Language (CityGML) Encoding Standard." <http://www.opengeospatial.org/standards/citygml>. (Mar. 15, 2012).
OMG (2009). "Ontology definition metamodel." <http://www.omg.org/spec/ODM/>. (May 10, 2010).
OpenIFCTools (2011). "Open IFC Tools." <http://openifctools.com/Open_IFC_Tools/Home.html>. (Jul. 1, 2011).
120
Paolucci, M., Kawamura, T., Payne, T., and Sycara, K. (2002). "Semantic matching of web services capabilities." 6th IEEE International Symposium on Wearable Computers, Springer, Seattle, WA, 333.
Pathak, J. (2006). "Difference between Agents and (Semantic) Web Services." <http://lists.w3.org/Archives/Public/public-sws-ig/2006Oct/0074.html>. (Mar. 15, 2011).
Pena-Mora, F., and Li, M. (2001). "Dynamic planning and control methodology for design/build fast-track construction projects." Journal of Construction Engineering and Management, 127(1), 1-17.
Pietroforte, R. (1997). "Communication and governance in the building process." Construction Management and Economics, 15, 71-82.
Pocock, J. B., Liu, L. Y., and Kim, M. K. (1997). "Impact of management approach on project interaction and performance." Journal of Construction Engineering and Management, 123(4), 411-418.
Popovsky, C. L., Roberts, C. G., and Kipnis, A. M. (2012). "Whose idea was this? - Design control and delegation - The road to a project utopia." American Bar Association Forum on the Construction Industry, Houston, Texas.
Protege (2011). "The Protege ontology editor and knowledge acquisition system." <http://protege.stanford.edu/>. (Mar. 15, 2011).
Ren, J., and Li, C. (2003). "Comparison of the application of e-commerce in construction industry in and out of China." Architecture and Construction, 61.
Scott, S. (1990). "Keeping better site records." Project Management, 8(4).
Shen, L., and Chua, D. K. H. (2011). "Application of Building Information Modeling (BIM) and Information Technology (IT) for Project Collaboration." EPPM, Singapore.
Shen, W., Hao, Q., and Xue, Y. (2010). "An agent-based service-oriented approach for facility lifecycle information integration and decision supports." IEEE International Conference on Systems Management and Cybernetics, 1945-1952.
Silva, V. (2012). "Evolution, methodologies & concepts of CAD & BIM software tools in the AEC industry." ArchiMAG.
Sommerville, I. (2004). Software engineering, Pearson Education.
Sosnoski, D. (2007). ""Code First" Web Services Reconsidered." <http://www.infoq.com/articles/sosnoski-code-first>. (Feb. 01, 2012).
121
Stadler, A., and Kolbe, T. H. (2007). "Spatio-semantic coherence in the integration of 3D city models." 5th International ISPRS Symposium on Spatial Data Quality, Enschede, The Netherlands.
Studer, R., Benjamins, V. R., and Fensel, D. (1998). "Knowledge engineering: principles and methods." Data & Knowledge Engineering, 25(1-2), 161-197.
Sweet, J. (1994). Legal aspects of architecture, engineering and the construction process, West Publishing, St. Paul, MN.
Tardif, M. (2008). "BIM: reaching forward, reaching back." <http://info.aia.org/aiarchitect/thisweek08/0321/0321rc_face.cfm>. (Nov. 11, 2009).
thbs.com (2012). "Cloud computing services - a comparison." <www.thbs.com/pdfs/Comparison%20of%20Cloud%20computing%20services.pdf>. (Mar. 15, 2012).
Thorpe, A., Ward, M. J., Bowden, S., and Price, A. D. F. (2005). "Facilitating the link between point-of-production workers and corporate ICT systems in construction." 22nd International Conference on Application of IT in the AEC Industry.
Ting, S. K., and Wang, F. (2000). "Alternative strategies in the use of information technology in the construction industry of different countries." 17th International Symposium on Automation and Robotics in Construction.
Tizani, W., and Mawdesley, M. J. (2011). "Advances and challenges in computing in civil and building engineering." Advanced Engineering Informatics, 25(4), 569-572.
TNO (2010). "IFC Engine DLL." <http://www.ifcbrowser.com/ifcenginedll.html>. (Mar. 15, 2010).
USPRO (1996). "Initial Graphics Exchange Specification IGES 5.3."
Vacharasintopchai, T., Barry, W., Wuwongse, V., and Kanok-Nukulchai, W. (2007). "Semantic Web services framework for computational mechanics." Journal of Computing in Civil Engineering, 21(2), 65.
Vanlande, R., Nicolle, C., and Cruz, C. (2008). "IFC and building lifecycle management." Automation in Construction, 18, 70.
W3C (2004). "Web services architecture." <http://www.w3.org/TR/ws-arch/>. (Feb. 11, 2010).
W3C (2009). "OWL 2 Web Ontology Language Primer." <http://www.w3.org/TR/2009/REC-owl2-primer-20091027/>. (Mar. 15, 2011).
122
Wang, H.-H., Boukamp, F., and Elghamrawy, T. (2010). "An ontology-based approach to context representation and reasoning for managing context-sensitive construction information." Journal of Computing in Civil Engineering.
Wetherill, M., Rezgui, Y., Lima, C., and Zarli, A. (2002). "Knowledge management for the construction industry: the e-COGNOS project." Journal of Information Technology in Construction, 7, 183.
Zhang, J. (2010). "A social semantic Web system for coordinating communication in the architecture, engineering and construction industry." Ph.D. dissertation, University of Toronto, Toronto, Canada.
Zhang, L. (2012). "IFC-based construction industry ontology for information retrieval from building information models." Master thesis, University of Florida, Gainesville, FL.
123
BIOGRAPHICAL SKETCH
Le Zhang is a scholar with broad interest from China. He was admitted into the
construction project management program of Tianjin University, Tianjin, China in 2000,
where he earned his Bachelor of Management degree. During his undergrad study, he
also finished a second degree of Bachelor of Arts in English. In 2005 he continued his
study in international construction contract documents in Tianjin University and earned
his Master of Management degree. He was admitted into the Ph.D. program in M.E.
Rinker, Sr. School of Building Construction, University of Florida, Florida, USA in 2007
and started research on Building Information Modeling (BIM). He also pursued a non-
traditional concurrent degree program with a Master of Science in computer engineering
while at the University of Florida.
