<%BANNER%>

Integration of Construction Worker Fall Safety in Design through the Use of Building Information Modeling

Permanent Link: http://ufdc.ufl.edu/UFE0043660/00001

Material Information

Title: Integration of Construction Worker Fall Safety in Design through the Use of Building Information Modeling
Physical Description: 1 online resource (165 p.)
Language: english
Creator: Qi, Jia
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: bim -- construction -- design -- safety
Design, Construction and Planning -- Dissertations, Academic -- UF
Genre: Design, Construction, and Planning Doctorate thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Abstract: The construction industry has incurred the most fatalities of any U.S. industry in the private sector in recent years. While many factors may contribute to this statistic, one likely cause is due to designers who often lack design for construction safety knowledge, which results in many safety hazards being built into project models. To improve the current situation, this research was undertaken to identify the possible influence of Building Information Modeling technology on construction safety. After identifying the extent of the positive impact of BIM technology on safety, the research entailed the development of a design for construction worker safety tool which efficiently makes designing for safety suggestions available to designers and constructors. Particular emphasis was placed on fall accidents since falls are the most frequently occurring cause of fatalities on construction sites. The Design-Build delivery method is helpful for implementing this tool. The traditional Design-Bid-Build approach limits the abilities of contractors to contribute their knowledge to the project during the design phase, when they could add significant value to the project. With this designing for safety tool, it is possible for project participants to work together to optimize the building models/drawings, which provide valuable downstream benefits. Meanwhile, using the checking results from this tool, the constructors have the opportunity to take preliminary safety measures to address construction site hazards from the beginning of the project.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jia Qi.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Hinze, Jimmie W.
Local: Co-adviser: Issa, R. Raymond.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043660:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043660/00001

Material Information

Title: Integration of Construction Worker Fall Safety in Design through the Use of Building Information Modeling
Physical Description: 1 online resource (165 p.)
Language: english
Creator: Qi, Jia
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: bim -- construction -- design -- safety
Design, Construction and Planning -- Dissertations, Academic -- UF
Genre: Design, Construction, and Planning Doctorate thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Abstract: The construction industry has incurred the most fatalities of any U.S. industry in the private sector in recent years. While many factors may contribute to this statistic, one likely cause is due to designers who often lack design for construction safety knowledge, which results in many safety hazards being built into project models. To improve the current situation, this research was undertaken to identify the possible influence of Building Information Modeling technology on construction safety. After identifying the extent of the positive impact of BIM technology on safety, the research entailed the development of a design for construction worker safety tool which efficiently makes designing for safety suggestions available to designers and constructors. Particular emphasis was placed on fall accidents since falls are the most frequently occurring cause of fatalities on construction sites. The Design-Build delivery method is helpful for implementing this tool. The traditional Design-Bid-Build approach limits the abilities of contractors to contribute their knowledge to the project during the design phase, when they could add significant value to the project. With this designing for safety tool, it is possible for project participants to work together to optimize the building models/drawings, which provide valuable downstream benefits. Meanwhile, using the checking results from this tool, the constructors have the opportunity to take preliminary safety measures to address construction site hazards from the beginning of the project.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jia Qi.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Hinze, Jimmie W.
Local: Co-adviser: Issa, R. Raymond.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043660:00001


This item has the following downloads:


Full Text

PAGE 1

1 INTEGRATION OF CONSTRUCTION WORKER FALL SAFETY IN DESIGN THROUGH THE USE OF BUILDING INFORMATION MODELING By JIA QI 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 2011

PAGE 2

2 2011 J ia Q i

PAGE 3

3 To my parents, Yinghong Song and Fengzhou Qi

PAGE 4

4 ACKNOWLEDGMENTS The first group of people I would like to acknowledge is my committee members. I a m grateful to my advisor, Jimmie Hinze, for his tireless support and selfless mentoring. He provided me with a role model as a researcher. I would like to thank Dr. Raymond Issa for his encouragement and generosity. Dr. Issa helped me throughout my endeavo r and offered me the possibility to discuss my research study with members of his vast network of industry and academic contacts I am extremely grateful to Dr. Olbina for all of her hard work and dedication to her students and the teaching profession. I w ould also like to thank Dr. Chow. I greatly appreciate all of his help and support in completing this study. I am indebted to BIMserver, AEC3, and Solibri for their generosity in authorizing me to use their software that has been invaluable to my research study. I am also very appreciative of all those who helped me in the course of my research: Dr. William Eas t Lon van Berno, Nichlas Nisbet, Ruben de Laat, Wei Wu, Yogesh Veeraraghavan Le Zhang, Zhe Wang Rui Liu and many, many more.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................8 LIST OF FIG URES .........................................................................................................................9 LIST OF ABBREVIATIONS ........................................................................................................12 ABSTRACT ...................................................................................................................................14 CHAPTER 1 INTRODUCTION ..................................................................................................................16 Project History ........................................................................................................................17 Research Questions and Research Objectives ........................................................................18 2 LITERATUR E REVIEW .......................................................................................................20 Design for Construction Worker Safety .................................................................................20 The Concept of Design for Construction Safety .............................................................20 Implementing the Design for Safety Concept .................................................................23 Barriers to Implementation ..............................................................................................24 The Legal Framework of Constr uction Health and Safety ..............................................27 Design for Safety Resources, Tools, and Processes ........................................................32 Building Information Modeling ..............................................................................................39 Adoption of Advanced Information Technologies in Construction Industry ..................39 Concepts of Building Information Modeling ..................................................................41 Definition of Building Information Modeling .........................................................41 Characteristics of Building Information Modeling ..................................................42 Legal iss ues ..............................................................................................................43 Building Information Modeling Software .......................................................................44 IFC interface .............................................................................................................44 Classif ication of BIM software ................................................................................45 Code checking software ...........................................................................................47 Delivery Method Influence on Safety Performance ........................................................48 Integrated Project Delivery method .........................................................................51 Integrated Project Delivery with Building Information Modeling ...........................54 Computer Aided Critiquing System .......................................................................................54 Computer Aided Critiquing System in Design and Construction ...................................54 Summary of Related Work ..............................................................................................55 3 METHODOLOGY .................................................................................................................63

PAGE 6

6 Finding the influence of BIM technology on Construction Safety .........................................64 Design for Construction Safety Process .................................................................................65 Defining the Rule Source ........................................................................................................66 Existing Design for Construction Worker Safety Suggestions .......................................67 New Design for Construction Worker Safety Suggestions .............................................68 Classification of Design Suggestions ..............................................................................69 Architecture and Functionality of the desired System ............................................................72 Platform Chosen .....................................................................................................................77 Data Saved in a Relational Databas e ...............................................................................80 Data in IFC/Ifcxml Schema .............................................................................................82 Data in Other XML Schema ............................................................................................85 Structuring the Rule Source ....................................................................................................87 Creation of Rule Sets ..............................................................................................................89 Rule Sets Based on Relational Database .........................................................................89 Rule Sets Based on BIMserver ........................................................................................91 Direct approach ........................................................................................................92 Indirect approach ......................................................................................................94 Rule Sets Based on Solibri Model Checker ....................................................................98 Rule Sets Based on Bimservices ...................................................................................101 4 RESULTS .............................................................................................................................108 BIM technologys impact on construction safety .................................................................108 Safety Planning ..............................................................................................................108 3D/Virtual Reality .........................................................................................................109 Schedule/4D ..................................................................................................................110 Conducting Clash Detection ..........................................................................................112 Increasing Prefabrication ...............................................................................................113 Design for Construction Worker Safety Software Tool .......................................................115 Checking Results with a Relational Database ...............................................................118 Checking Results with BIMserver .................................................................................121 Checking Results with Solibri Model Checker .............................................................125 5 CONCLUSIONS AND RECOMMENDATIONS ...............................................................135 Conclusions ...........................................................................................................................135 Drawbacks ............................................................................................................................141 Recommendations for Future Research ................................................................................143 APPENDIX A FALL PROTECTION BEST PRACTICS ............................................................................146 B SAMPLE CODES: NET AREA VALUE OF IFCSLAB ....................................................149 C SAMPLE CODES: IFCOPENINGELEMENT AREA VALUE RETRIVEAL ..................152 D SAMPLE CODES: QUERYING RELATIONAL DATABASE WITH SQL .....................155

PAGE 7

7 LIST OF REFERENCES .............................................................................................................156 BIOGRAPHICAL SKETCH .......................................................................................................165

PAGE 8

8 LIST OF TABLES Table page 21 C omparison of the hierarchy of safety requirements between U.S. and U.K. ...................28 22 Computer aided critiquing systems ...................................................................................62 31 Four different mark up colors ..........................................................................................103 51 Comparison computable rule compiling process based on three software platforms ......137 52 Comparison of functiona lities between three software platforms ...................................138

PAGE 9

9 LIST OF FIGURES Figure page 11 Number and rate of fatal occupational injuries by industry sector in 2009 .......................16 21 Time/safety influence curve ...............................................................................................21 22 The hierarchy of control measures with practical examples ..............................................23 23 Distribution of designer activity in addressing safety. .......................................................24 24 Hierarchy of the designing for safety resources. ...............................................................33 25 Design for construction worker safety process ..................................................................35 26 IFC Interfaces .....................................................................................................................45 27 BIM roadmap .....................................................................................................................46 28 Participant relationships in the design bidbuild method ..................................................49 29 Teams ability to affect project variables ...........................................................................50 210 Traditional and redefined project phases ...........................................................................52 211 Participant relationships in a project under the IPD method .............................................53 212 Block diagram of a system for checking a building information model against SMARTcodesTM ...............................................................................................................60 213 Predominant checking systems ..........................................................................................61 31 Classification of design for construction worker safety best practices ..............................71 32 Architecture of Design for Construction Safety tool .........................................................74 33 Functionality of the D4S software tool ..............................................................................77 34 Querying BIM model/database with different programming languages and derived platforms/servers ................................................................................................................80 35 Building features in a relational database ..........................................................................81 36 Worksheet created by the IFA ...........................................................................................82 37 Linking the IfcWindow to IfcW all in HelloWall example ................................................84 38 Information lost during data transfer process ....................................................................86

PAGE 10

10 39 Interoperability through ISO standards ..............................................................................88 310 Table listing window types ................................................................................................90 311 Table listing levels .............................................................................................................90 312 Sequencing floor levels based on Level value in Floors spreadsheet ...........................90 313 All windows above the first floor in Windows Spreadsheet ...........................................91 314 Filtering windows by SillHeight ........................................................................................91 315 The returned result of noncompliance ..............................................................................91 316 IFC hierarchy for opening area value retrieval ..................................................................93 317 Dimension parameters of an IfcOpeningElement ..............................................................94 318 IFC 23 object diagram IfcOpeningElement ..................................................................96 319 Parameters of rectangle profile definition ..........................................................................97 320 The Rule Set Manager interface for browsing and editing rules .......................................99 321 The Libraries View ..........................................................................................................100 322 Workspace View ..............................................................................................................100 323 Parameter View ................................................................................................................101 324 Encoding D4S suggestions into safety Constraint Model ................................................102 325 Sample rule of bimServices .............................................................................................104 326 Review ing IFC opening element properties in IfcStoreyView ........................................105 327 Three windows in building model 4351.ifc .....................................................................106 328 Opening elements in bui lding model 4351.ifc .................................................................107 41 Photograph of BIM computer lab at Center for Advanced Construction Information Modeling at the University of Florida .............................................................................110 42 Construction project life cycle safety with a D4S application tool .................................115 43 Safety effort of different stakeholders in a project lifecycle ............................................118 44 Model checking in a Microsoft Access ............................................................................119 45 Import computable rule in the Query window .................................................................119

PAGE 11

11 46 Model checking result in a Microsoft Access ..................................................................120 47 The architecture of current model server and potential extension ...................................121 48 Launch BIMserver and select project to be checked .......................................................122 49 User interface of Advance Query function of BIMServer ...............................................122 410 Checking area values of open ing elements with BIMserver by using computable rule written in direct approach ................................................................................................123 411 The relationship between net area value and opening element area value ......................124 412 Checking area values of opening elements with BIMserver by using computable rule written in direct approach ................................................................................................124 413 Checking layout of Solibri Model Checker .....................................................................125 414 Using the model tree navigates the subject building model ............................................126 415 Selecting desired constraint model / rule set....................................................................127 416 Checking results ...............................................................................................................128 417 Description of noncompliance in the Info View ............................................................128 418 The PSet_Revit properties in Info View ..........................................................................129 419 Result Details view ..........................................................................................................130 420 Neglecting the non compliance after selecting the Accepted Option ...........................130 421 Create report .....................................................................................................................131 422 Instance properties in BIM authoring tool .......................................................................132 423 New result of the check after changing the dimension of certain objects .......................133 424 Result of checking for slab openings ...............................................................................134 425 Designing alternative given in Info View ........................................................................134 51 IfcOpeningElement with an IfcArbitraryClosedProfileDef .............................................143 52 IfcA rbitraryClosedProfileDef defined by six points ........................................................143

PAGE 12

12 LIST OF ABBREVIATION S 3D Three Dimensions 4D Three Dimensions + Time 4D Safety Four Dimension al + Safety BIM Building Information Modeling BIM ASW BIM authoring softwar e CACIM Center for Advanced Construction Information Modeling CDM Construction Design and Management CII Construction Industry Institute CSI Construction Specifications Institute COBie Construction Operations Building Information Exchange ERDC Engineer Res earch and Development Center HSE Health and Safety Executive IAI International Alliance for Interoperability ICC International Code Council IFC Industry Foundation Class IPD Integrated Project Delivery KBS Knowledge Based System MCS Model Checking Software NBIMS National BIM standard NIOSH National Institute for Occupational Safety and Health OPS Onuma Planning System OSHA Occupational Safety and Health Administration PPE Personal Protective Equipment SQL Structured Query Language

PAGE 13

13 STEP Standard for the Exchange of Product model data XML Extensible Markup Language

PAGE 14

14 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 INTEGRATIO N OF CONSTRUCTION WORKER FALL SAFETY IN DESIGN THROUGH THE USE OF BUILDING INFORMATION MODELING By Jia Qi December 2011 Chair: Jimmie Hinze Cochair: R. Raymond Issa Major: Design, Construction and Planning The construction industry has incurred the most fatalities of any U.S. industry in the private sector in recent years. While many factors may contribute to this statistic, one likely cause is due to designers who often lack design for construction safety knowledge, which results in many safety hazards being built into project models. To improve the current situation, this research was undertaken to identify the possible influence of Building Information Modeling technology on construction safety. After identifying the extent of the positive impact of BI M technology on safety, the research entailed the development of a design for construction worker safety tool which efficiently makes designing for safety suggestions available to designers and constructors. Particular emphasis was placed on fall accidents since falls are the most frequently occurring cause of fatalities on construction sites. The Design Build delivery method is helpful for implementing this tool. The traditional Design Bid Build approach limits the abilities of contractors to contribute th eir knowledge to the project during the design phase, when they could add significant value to the project. With this designing for safety tool, it is possible for project participants to work together to optimize the building models/drawings, which provide valuable downstream benefits. Meanwhile, using the

PAGE 15

15 checking results from this tool, the constructors have the opportunity to take preliminary safety measures to address construction site hazards from the beginning of the project.

PAGE 16

16 CHAPTER 1 INTRODUCTION It is an alarming statistic that every work day nearly four workers are killed in the U.S. construction industry. In 2009, the construction industry fatalities represented 18.8% of all workrelated fatalities in the U.S. (Bureau of Labor Statistics 2009). The construction industry has incurred the most fatalities of any industry in the private sector in the past five years. Figure 1 1 shows the number and rate of fatal occupational injuries by industry sector in 2009. The number of fatalities sustained in the construction industry (816) far exceeds those of other industries. The fatality rate of the construction industry (9.7 fatalities per 100,000 workers) is exceeded by only three industries, namely mining; agriculture, forestry, fishing, and hunting; and transportation Figure 1 1. Number and rate of fatal occupational injuries by industry sector in 2009 (Source: Bureau of Labor Statistics) and warehousing. Because of the high injury and fatality rates and the high costs associated with these accidents, improving job site safety has become a major concern for many construction professionals (Gambatese 2008). Researchers have proposed many methods to reduce job site

PAGE 17

17 hazards. One technique is to incorporate safety considerations into the project during the design phase, or to involve designers in considering construction worker safety during the design process (Korman 2001). This approach is known as design for construction safety. While this approach is also known as prevention through design, especially am ong some federal agencies, the term design for construction safety will be used herein. Project History This research utilizes recently developed information technology to develop an enhanced software tool to facilitate the design for construction worker safety. Using a software tool to help designers implement the design for construction safety knowledge is not a new concept. In the 1990s, the Construction Industry Institute (CII) recognized the lack of designer involvement in construction worker safety due to their minimal education and experience in addressing safety on construction sites. The CII funded a research project to develop a software tool to assist designers in recognizing project specific hazards and in implementing design suggestions for co nsideration in the project design (Gambatese 1997). The design for safety suggestions were accumulated through research efforts that included input from designers, traditional construction contractors, and designbuild firms. These suggestions were incorporated into the Design For Construction Safety ToolBox (Gambatese et al. 1997). As new information technology emerged, CII expressed a need for a software tool which would replace the software tool created in 1996. To give design professionals the ability to more quickly and easily access design for construction safety suggestions, the Design for Construction Safety Toolbox, second edition, was developed in 2007 by Marini and Hinze, through the support of the CII. The second edition of the toolbox is a w eb based application that can also be operated via a compact disc. The database of this application consisted of a simple, external, text based (XML) file designed to easily accommodate the addition of future design for

PAGE 18

18 construction safety suggestions (Mar ini 2007). Besides the changes made to the database, other elements such as the application design, application navigation and software tour functions of the second edition include substantial improvements to make the application more deliverable and easy to use. This application is available from the CII. As Building Information Modeling (BIM) technology is being increasingly adopted by more design and construction companies, a large amount of work in the architecture, engineering and construction (AEC) domain is now completed by using three dimensional/object oriented computer software. In current usage, BIM does not automatically address construction worker safety. Instead, construction safety can currently be addressed by the inclusion of written stateme nts which is a cumbersome and inefficient procedure. The proposed research is to examine the possibility of developing a design for construction safety tool which could automatically address construction safety in electronic construction documents. Research Questions and Research Objectives Designers who lack design for construction safety knowledge have limited access to this type of information because the necessary channels of information flow do not exist. To solve this problem, researchers can use Buil ding Information Modeling (BIM) technology to deliver design for construction safety knowledge to designers. This research specifically addresses the following issues: (1) With advances in information technology, is it possible for a construction worker s afety tool to automatically check construction drawings and to provide design for safety suggestions during the design process? (2) Determine the approach or procedure that would be most suitable for automatically integrating design for construction safe ty suggestions into the design process without hampering designer creativity. (3) Computer aided critiquing technology has been used to conduct automated building code compliance assessments. This endeavor shows the possibility of using computers automa tically to check threedimensional (3D) models for certain rule sets. Building code

PAGE 19

19 compliance and design for construction safety knowledge are developed for different purposes. The differences between checking systems that use encoded building code infor mation as rule sets and checking systems that use the encoded design for construction safety knowledge as rule sets will be examined. (4) The supposed users of previous design for construction worker safety toolboxes are designers, most notably designers in designbuild firms. Constructors could also check drawings themselves before construction work is begun to develop plans to protect their employees. The objectives of this research are as follows: To understand how Building Information Modeling (BIM) a s a tool could be used to enhance construction worker safety, especially falls from elevation. To study the current practices of designers and make recommendations that will enhance construction work site safety and to review the current procedures of desi gn for construction worker safety to improve it by using BIM Technology. To develop a design for construction worker safety tool that would automatically check construction drawings and make design for safety knowledge available to designers during the de sign phase.

PAGE 20

20 CHAPTER 2 LITERATURE REVIEW This chapter describes the background of this research and summarizes the results of other related research work. There are three sections in this chapter. The first section introduces the concept of design for cons truction worker safety and the current level of implementation in the construction industry. Barriers to implementing this approach are examined and the legal implications are discussed in detail by comparing the implementation of designing for safety in t he U.K. and U.S. The impact of the UK regulatory effort is discussed. Various safety tools are examined to summarize past research work in this domain. The second section introduces the concept of Building Information Modeling/Virtual Design and Constructi on. A BIM map is used to describe the relationship between different applications. In particular, it focuses on code checking software which has many similarities to a potential automatic design for construction safety toolbox. Finally, the third section defines the computer aided critiquing system and introduces some prototypes which have been developed both in academia and in industry. Design for Construction Worker Safety The Concept of Design for Construction Safety In 1985, the International Labor Offi ce (ILO) suggested that design professionals should consider construction safety during the design process. Design for construction worker safety was not researched in the U.S. until Hinze (1992) conducted a survey to explore the relationship between safet y and design. That study showed that construction worker safety can be addressed during the design stage. Despite this, few design firms were identified that regarded the safety of construction workers as being within their scope of responsibility, primari ly because of designers fears of increased liability (Hinze and Wiegand 1992). In 1993, the Construction Industry Institute sponsored a research study for Hinze and Gambatese to develop a detailed

PAGE 21

21 understanding of the concept of designing for construction worker safety and to develop a design tool or toolbox (Gambatese et al. 1997). According to Szymberskis (1997) time/safety influence curve, the ideal time for construction safety to be a prime consideration in the life of a project is in the conceptual a nd preliminary design phases. Figure 21 shows that by including construction safety as a consideration earlier in the projects life cycle, project participants have a greater ability to positively influence construction worker safety. A significant abili ty to improve construction site safety is lost when it is considered later in the life cycle, such as during the construction phase. Figure 2 1. Time/safety influence curve ( Adapted from : Szymberski 1997) Hazard management theory also validates the valu e of design for construction safety. After identifying hazards to health and safety and assessing their potential to do harm, measures to eliminate or minimize them should be determined in accordance with the hierarchy of control (Manuele 1997; Andres 2002; Civil Construction Industry OHS Committee 2003).The hierarchy of control measures are as follows:

PAGE 22

22 Eliminate the hazard Substitute the hazard with a lesser risk Isolate the hazard Use engineering controls Use administrative controls Use personal protectiv e equipment Measures from the top of the hierarchy produce better results than the lower ones and should be considered wherever possible. According to this, control measures which make the workplace safe are more effective than measures which protect cons truction workers from a hazardous condition. Consider a practical example of using hierarchy of control measures to protect workers working at elevation. First, construction workers should avoid working at elevation whenever they can. This could be achieve d by designing out the need to work at elevation. Second, workers should use such measures a guardrails for fall protection when they cannot avoid working at elevation. Third, where workers cannot eliminate the fall risk, they should use personal protective equipment (PPEs) to minimize the distance and consequences of a fall if a fall accident would occur (Health and Safety Executive 2005). Figure 2 2 shows the hierarchy of control measures for fall protection. Studies of construction accidents and injuries confirm that many events have their origins upstream from the building process (Suraji 2001).A study by the European Foundation found that approximately 60% of construction accidents could have been avoided or reduced with more thought at the design stage (European Foundation 1991). Researchers in the U.K. contended that changes in the permanent design elements would have reduced the likelihood of 47% of the 100 construction accidents studied (Gibb e t al. 2004). Behm (2004) found that design was linked to accidents in approximately 22% of 226 construction injury incidents and 42% of 224 fatality incidents in Oregon, Washington, and California between 1990 and 2003. Later, Gambatese et al. (2007) used the Delphi method to validate Behms research findings.

PAGE 23

23 Figure 2 2. The hierarchy of control measures with practical examples ( Adapted from : U.K. HSE 2003) Implementing the Design for Safety Concept Even though the concept of design for construction safety is viewed as a viable intervention to improve safety, design firms do not commonly address construction worker safety in their designs (Gambatese et al. 2005a). From the results of a 1992 survey, Hinze found that less than one third of the design firms addressed construction worker safety in their designs, a nd less than one half of the independent constructability reviews conducted addressed construction worker safety (Hinze and Wiegand 1992). Hinze conducted studies that surveyed 377 project owners in the U.S. (Hinze 1994a; Hinze 1994b). The owners were aske d whether designers of their projects addressed construction safety in their designs. Only 16% of the owners in those studies indicated that they considered their employee safety in their designs (Figure 2 3). In a recent survey, 37% of the 19 respondents expressed that they were interested

PAGE 24

24 and willing to implement the concept, while 47% of the respondents gave a neutral response and 16% said they were not interested in it. Figure 2 3. Distribution of designer activity in addre ssing safety ( Adapted from : Hinze 2006). Research studies were conducted and relevant training was carried out to change this climate in the construction industry. With these efforts, researchers found that changes are slowly occurring in the construction industry. First, an increasing number of owners are demanding that all parties involved in their projects should be actively involved in safety management. Owners such as Intel, Inc. and the Southern Company tend to select designers who are safety consciou s. Second, some large designbuilders such as Jacobs and Fluor are also practicing design for safety in their projects. In the future, with the companies realizing the benefits they gain by adopting the design for safety concept, it is expected that more f irms will implement designing for safety concepts. Barriers t o Implementation While the evidence is clear that construction worker safety can be addressed during the design stage, the widespread implementation in practice throughout the U.S. is currently l acking May address safety in future 16% Worker safety is addressed 10% Worker safety not considered 45% Occasionally address safety for specific items 29% Distribution of designer activity in addressing safety

PAGE 25

25 (Hinze and Wiegand, 1992). There are several barriers that hinder the implementat ion (Gambatese 2005; Toole 2005): Designers fear of liability. Designers lack of safety expertise. Conflicts with the project delivery method. Additional costs incurr ed by designers. Time constraints. Lack of a regulatory mandate. In the U.S., there are liability fears on the part of designers for becoming involved in construction site safety. In the Gambatese et al. (2005) study in Oregon and Florida, 26% of the surv ey participants mentioned increased liability as a barrier. The construction documents define the design for competitive bidding, cost estimation, and conveying design intent, but the designers have no authority over or responsibility for the means and m ethods of construction. For instance, the American Institute of Architects (AIA) B 141 contract form seeks to limit the responsibility of the designer for actual construction issues, placing that responsibility instead on the contractor (AIA 2007). Legal c ounsels often advise designers to contractually and operationally limit their involvement in safety to minimize potential exposure to third party lawsuits (Behm 2005). Designers usually take their lack of familiarity with necessary methods as an excuse to evade the task of implementing health and safety in the design process (Hinze and Wiegand 1992). The partition between the design trade and the construction trade has caused cooperation problems in the construction industry for many years. Even though cons tructors and researchers have accumulated considerable design for construction safety knowledge, the dissemination of this information in the design community is difficult. Designers contend that no tools are available to help them implement the design for construction worker safety concept. As a result, various design for construction safety tools have been developed to solve these problems in

PAGE 26

26 recent years. Researchers suggest that 10 hour and 30hour OSHA courses for design professionals should be promote d. The process of procuring construction services is another barrier to implementing the design for safety concept. In the traditional methods of procuring construction via designbidbuild, contractors are typically not chosen until after the design is completed. Since contractors cannot be guaranteed the project award, even though they may have contributed their expertise during the design process, the needed communication about hazards between designers and builders usually does not occur. Meanwhile, the model contracts of the AIA and the Engineers Joint Construction Documents Committee (EJCDC) clearly state that designers have no site safety responsibilities. For example, the 2007 construction series of standard contract documents of EJCDC state that a contractor is responsible for initiating, maintaining and supervising all safety programs in connection with the work and inform the owner and its engineer of specific safety requirements that must be followed at the site and the corresponding obligation of owner and engineer to comply with such requirements (NSPE 2007). Design for construction safety will increase both direct and overhead costs for design firms. Considering safety during the design process requires more time to gain the needed training and to complete the design work. Unless owners are willing to pay higher design fees, designers will not be inclined to implement the design for safety concept to decrease total proje ct life cycle costs (Toole 2005 ). Time constraints are another concern of designers. Usually the tight project schedules established by owners discourage a thorough analysis of the safety issues in favor of satisfying other design requirements. Insufficient time is dedicated to the implementation of safety procedures during the design phase. For instance, when the designforsafety process was

PAGE 27

27 developed on one large capital project, the design team did not always welcome additional review steps and more opportunities for comments to which they had to respond, because the pace of the design was compromised (Weinstein et al. 2005; Hecker et al. 2005). Last, current U.S. regulations do not place the requirement on designers to design for construction safety. This further impedes the process of the design for construction worker safety, as it will be discussed in more detail in the next section. The above barriers to the implementation of design for construction safety are not insurmountable. Researchers have proposed some key changes needed for the implementation of the concept (Gam batese et al. 2005b; Toole 2005). In these key changes, design for construction worker safety tools could help designers acquire the necessary safety knowledge, as discussed in detail in the following section. The Legal Framework of Construction Health and Safety By analyzing the evolvement of legislation related to jobsite safety in the U.S., it is evident that safety standards and legislation are a result of a long history of change in the drive to reduce the number of worker injuries and fatalities (Gam batese 1996). However, legislation in the U.S. has lagged behind some other countries, while designing for construction safety has become increasingly more common in Europe and Australia. In 1992, Directive 92/57/EEC, also referred to as the Construction S ite Directive, was published, which was an initiative to improve occupational health and safety in the European construction industry (Bluff 2003). Great Britain responded by enacting the Construction (Design and Management) Regulations in 1994. Regulation s of the U.S. and the U.K. have different impacts on the implementation of the design for construction safety concept. Table 2 1 provides a comparison of the hierarchy of the safety requirements of the U.S. and the U.K.

PAGE 28

28 Table 2 1. Comparison of the hierar chy of safety requirements between U.S. and U.K. Hierarchy of Safety Requirements U.K. U.S. High Statutory Law European Union law: The Temporary or Mobile Construction Sites Directive 92/57/EEC UK statutory law: The Health and Safety at Work etc. Act 1974 Occupational Safety and Health Act of 1970 Medium Regulations Manual Handling Operations Regulations 1996 Construction (Design and Management) Regulations (CDM) 2007 The Management of Health and Safety at Work Regulations (MHSWR) 1999 Title 29, Part 192 6 of the Code of Federal Regulations (29 CFR 1926) Low Codes of practice Managing Health and Safety in Construction. CDM Regulations 2007. Approved Code of Practice ASCEs Policy Statement 350 (ASCE 2001) ASCE Code of Ethics (ASCE 1996) In the U.K., the health and safety related publications can be ranked into different levels based on their importance (Thope 2005). They are the Health and Safety at Work etc. Act (HSWA), statutory regulations, and codes of practice. The health and safety performance of the U.K. industry is guided and informed by the Health and Safety Commission (HSC) and the Health and Safety Executive (HSE) which were established in the mid 1970s by the Health and Safety at Work etc. Act 1974 (Howarth and Watson 2009). The Health and Sa fety Executive is also responsible for enforcing primary and secondary legislation. The primary legislation includes a number of Acts such as the Health and Safety at Work etc. Act 1974. The secondary legislation is made up of statutory instruments which a re also referred as regulations. The Health and Safety at Work etc. Act. 1974 (HSWA) is the primary piece of legislation covering occupational health and safety (HSE 2009). One of its effects is to secure the health, safety and welfare of persons at work (HSE 1974). The health and safety law of the U.K.

PAGE 29

29 construction industry is built on this act. This act provides the basis for the U.K. health and safety law and is further supported by sets of regulations (Howarth and Watson 2009). The secondary regulatio ns include regulations such as the Construction (Design and Management), Workplace (Health, Safety and Welfare) Regulations, etc. The Construction (Design and Management) regulations took effect in 1995 (HSE 2007), and place health and safety management du ties on all parties involved in a construction project. When it comes to the duties of designers, it explains: Every designer shall in preparing or modifying a design which may be used in construction work in Great Britain avoid foreseeable risks to the he alth and safety of any person carrying out construction work; maintaining the permanent fixtures and fittings of a structure (HSE 2007, p9) Approved codes of practice accompany the regulations. The creation of approved codes of practice is provided for by Section 16 of the Health and Safety at Work etc. Act (2007 HSE). They are not legally binding documents but serve to provide practical guidance for compliance with health and safety regulations and can be used in criminal court proceedings as a means to i nterpret whether the practical requirements of health and safety regulations have been met (Howarth and Watson 2009). In the U.S., industrial accidents rose significantly after World War II To ensure that employers provide employees with an environment free from recognized hazards, the Occupational Safety and Health Act (OSHAct) was enacted by Congress in 1970. The Act applies to employers in many industries such as construction and manufacturing. In section 5 of the OSHAct, the general du ty clause clearly mandates: Each employer shall furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees (U.S. Dept of Labor, 2009)

PAGE 30

30 The OSHAct created the Occupational Safety and Health Administration (OSHA), an agency of the Department of Labor. OSHA was given the authority to promulgate and enforce workplace health and safety standards. Title 29, Part 1926 of the Code of Federal Regulations (29 CFR 1926) is specifically for the construction industry. These standards apply to construction and maintenance work in all states. The federal OSHA standards focus on the safety management responsibilities of employers for the ir employees (Toole 2002). OSHA standards suggest that ignoring the hazards would not be a violation of the OSHA standards for an engineering firm since the engineer is neither a contractor nor a specialty contractor who requires a laborer to work in an unsafe environment. Furthermore, the federal OSHA standards rarely mention the possibility of the safety of workers being considered during the design of the project (Toole 2007). The likelihood of new or revised OSHA standards that specify the inclusion of architects and engineers might be low (Behm 2005). There is a controversial view as the OSHA regulation can be interpreted as applying whenever there is an employee/employer relationship (e.g. an engineer or architecture firm directs an employee to inspect work on a construction site). Some researchers suggest that the federal OSHA construction standards should be reexamined to require proactive hazard analysis to be initiated during the design stage, so contractors could do better to minimize probable cons truction hazards and provide a safe place to work for construction workers (Toole 2007). At the code of practice level, the site safety documents issued by major construction trade organizations contain significant differences in the site safety roles expected to be assumed by different project participants (Toole 2002). In general, the general contractor and specialty contractors are assigned the primary responsibility for construction worker safety. This is

PAGE 31

31 embodied through the policy statements or contr act terms of the ASCE, AIA, AGC and ABC. For instance, AIA A201 (2007) explicitly rejects the idea that the architect has a safety role: If the Contractor is then instructed to proceed with the required means, methods, techniques, sequences or procedures w ithout acceptance of changes proposed by the Contractor, the Owner shall be solely responsible for any loss or damage arising solely from those Owner required means, methods, techniques, sequences or procedures (AIA, 2007) The Architect will not have cont rol over, charge of, or responsibility for, the construction means, methods, techniques, sequences or procedures, or for the safety precautions and programs in connection with the work, since these are solely the Constructors rights and responsibilities under the Contract Documents, except as provided in Section 3.3.1. (AIA, 2007) Fortunately, the designers role in construction site safety is changing (Toole 2005 ). This can be detected from the alteration of the ASCE construction site safety policy. ASCE, in 1989, enacted its Construction Site Safety policy statement which stated that site safety was essentially the general contractors responsibility. The policy statement adopted in 1995 excluded the role of the engineer. Also, the revised 1998 statement omitted any reference to the engineer. The change emerged in the proposed new policy statement A 350 which states that construction site safety involves all the parties to some degree. All efforts to improve safety at construction sites are supported by AS CE and the most effective improvements can be achieved by a cooperative approach between all parties involved on a project (Muller 2002). The policy stated that engineers have the responsibility to: Recognize that safety and constructability are important considerations when preparing construction plans and specifications Both the frameworks of the U.S. and U.K are comprehensive and cohesive. Compared to the US legal system, the framework of the U.K. construction and safety regulations clearly define the d uties of the designers related to construction worker safety.

PAGE 32

32 Design for Safety Resources, Tools, and Processes Several approaches and tools have been developed to help designers identify design decisions that can significantly improve construction safety without compromising architectural form or function (Gambatese et al. 1997; Toole 2005; Gambatese et al. 2005a). Unfortunately, there is no existing classification of these approaches. To clarify the past development of the design for construction safety t echniques, a hierarchy will be presented of the design for safety tools, processes, and methods (Figure 2 4). The first effort is to improve the communication and coordination between the different stakeholders. Collaboration between designers and construc tion superintendents could be particularly effective to reduce construction worker injuries and fatalities. The second method is to form a systematic design process to eliminate or reduce design errors and increase the opportunity of incorporating the desi gn for construction safety knowledge into construction documents. If designers are not sufficiently experienced to continuously consider safety, outside design reviews should be conducted to improve the quality of the drawings. There are several stages in the design process where the design can be checked in some way. The third kind of endeavor is to conduct a thorough risk assessment of each design component. This is usually done by using various tools such as safety manuals, checklists, and software to he lp designers access design for construction worker safety knowledge. Furthermore, some tools can facilitate the checking process to save time and design fees. Project participants should first try to strengthen the collaboration between themselves. There should be ways through which information can be accurately and swiftly shared by different stakeholders. Secondly, designers could improve the design process by delivering

PAGE 33

33 Figure 2 4. Hierarchy of the designing for safety resources. safe construction documents for the constructors. Finally, various tools should be developed to meet specific needs. Some of these tools will be described. Though the improvement of safety performance is possible based on the effort of an individual discipline, communicatio n and cooperation among the project participants about the safety aspects of the project are important and necessary. It is believed that the major opportunity for improving the design and construction of facilities lies at the interface between discipline s (Fischer and Kunz 2004). It is especially when the traditional designbid build delivery method is used that a barrier to safety in design is most apparent. Active communication and collaboration could overcome the separation between design and construct ion and facilitate successful project completion. One proposed method to implement the design for safety is to increase the communication and coordination of work between designers and construction personnel, particularly those with excellent safety record s. Experienced superintendents could make substantial contributions to the designing for safety effort, provided that designers would

PAGE 34

34 recognize and harness their skills, site experience, and knowledge (Coble and Haupt 2000). In a practical project in which the design for safety process was developed and implemented, the lead designer insisted that he saw increased collaboration and dialogue with top construction managers and trade contractors as the greatest benefit of the whole design process (Weinstein et a l 2005; Hecker et a l 2005). Many endeavors have been made to make the design process better. Taylor (2007) claims that one approach is to study the design process and regard this process as a failure prone system. Thorpe (2005) suggests the best way is to turn safe construction practices into a standard part of the regular design process. Other research efforts have found the design and review process can influence the extent to which a design is successfully modified to impact safety. They think well planned and implemented designforsafety review processes could facilitate creating effective and efficient designs (Weinstein et al. 2005). After realizing the importance of the design process, another important issue is when the required site safety exp ertise should be provided. Ideally site safety expertise must be provided throughout the design process. However, implementing the design in this way may inhibit the designers creative process or hamper the usual design process. It is more practical to ha ve safety constructability knowledge provided through several progress reviews (Toole et al. 2006). Experience gained while developing past design for construction safety tools also demonstrated this problem. The Health and Safety Executive (HSE) of the U.K. developed a software prototype to provide designers with easy access to health and safety information by establishing a means of structuring HSE's information as a Knowledge Based System. After the Knowledge Based System (KBS) was completed, HSE sought feedback on the application from users. The general negative feedback of HSEs Knowledge Based System was the designers concerns

PAGE 35

35 about the timing to conduct the checks. Designers felt that they could pick up any health and safety risks during a risk analy sis after the main drawings had been completed. They preferred this approach, even though checking work early in the design phase could save time (HSE 2003). In Figure 2 5, Gambatese presented a design for construction safety process. The key component of this process is the incorporation of site safety knowledge into design decisions before the drawings are finally issued for construction (Design for Safety Workgroup 2008). Design for safety is a continuous improvement process in which specific design opti ons are assessed in terms of their positive or negative impact on safety at the end of each sub phase. Figure 2 5. Design for construction worker safety process ( Adapted from: Gambatese 1996 ) Hinze (2000) proposed a holistic approach to design that encompasses the designers consideration of the entire life cycle of the facility. This approach can be implemented by including safety considerations in constructability reviews (Gambatese 2000a). A thorough risk assessment of each component of the design should be conducted by designers (Hinze et al.

PAGE 36

36 1999). Such a safety review process has been implemented. In the U.S., Intel, Inc. developed a design for safety process which is called life cycle safety (LCS). The process involved many different parties thro ughout the course of programming, design, and construction. Consideration of design changes early in the project and input from the trade contractors are the key factors that contribute to success (Weinstein et al. 2005). In Australia, safety professionals created a tool called Construction Hazard Assessment Implication Review (CHAIR) which brings together all the key stakeholders involved in the design to help identify and eliminate inherent risks (WorkCover 2001). It provides a structured review process t hat incorporates focused reviews at different points in the design phase. The reviews provide a structured and systematic means to examine construction, maintenance, repair and demolition safety issues associated with a design. The elements of design and t he steps of the proposed construction tasks can be considered through different phases. In the Netherlands, Frijters and Swuste (2008) proposed a method which incorporates ten steps for choosing between alternative building elements to integrate safety and health into the design process. Two approaches are proposed based on the different timings for making revisions. The first one is the Rectilinear approach. If the design for safety software finds that a particular building element comprises a risk after the preliminary decision has been made, then the preliminary decision is asked to be reviewed and the design to be re drafted. The interactive method intends to minimize the risks by taking appropriate measures during the design phase. It is finalized b y comparing all alternative options and selecting an appropriate one. The two methods could be combined to generate a method which checks the documents both during and after preliminary decisions have been made (Frijters and Swuste 2008).

PAGE 37

37 Last, a thorough risk assessment of each design component could be conducted by using different design for safety tools which have been developed through past research studies. These techniques and tools can be sorted into the following categories: CAD tools with builtin checking Tool for drawing inspections Tool for design work through Checklist based design reviews for typical weaknesses Specification review Checklists are one of the most widely used tools in the construction industry. IDC, a design firm, developed a safety in design checklist, a database of design issues identified to address potential problems for construction or facility operations. IDC developed the checklist as an interactive and open ended tool for designers (Weinstein et al. 2005). Meanwhile des igners of this project expressed their concern about time. The pace of the design and construction on the project was fast, so the design team did not always welcome additional review steps and more opportunities for comments to which they had to respond. Construction specifications are another important tool for communicating between designers and constructors. Boukamp and Akinci (2007) proposed a constructionspecification processing approach which allows for automatic reasoning about construction specifi cations to determine their applicability to products in a project model and to extract requirements imposed by them. Then a system that facilitates construction safety improvement is developed based on this approach. The system could be used during the des ign phase to make designers aware of safety implications resulting from specific designs. Also, the system could support safety planning and evaluate the safety impacts of different construction methods (Wang and Boukamp 2007).

PAGE 38

38 Toole (2005) identified five tasks performed by designers in which they could potentially contribute to increasing construction worker safety. Toole and Gambatese (2008) suggested that design for construction safety is likely to evolve along four specific trajectories: increased pref abrication, increased use of less hazardous materials and systems, increased application of construction engineering, and increased spatial investigation and construction. The Construction Industry Institute (CII) sponsored a research study to develop a de sign tool titled Design for Construction Safety ToolBox that alerts designers of project specific construction safety hazards and provides suggestions on how a project can be designed to eliminate or reduce those hazards (Gambatese et al. 1997). A databa se of over 400 design practices was established through a review of construction industry publications and design manuals. Interviews with engineers, architects, constructors, and construction managers were also conducted to generate design for construction safety suggestions. The accumulated design practices can be implemented during the design phase to reduce or eliminate safety hazards during construction (Gambatese et al. 1997). The Design for Construction Worker Safety Toolbox, E dition 2, was developed in 2007 (Marini 2007) to meet the needs of CII. More than one hundred additional suggestions were collected to enrich the knowledge base of the toolbox. Furthermore, the toolbox is based on new authoring software which makes the tool accessible via the w eb. The database is much easier to update because of the use of an XML file. Other functionalities such as navigation and print have been designed for ease of use. Researchers in Australia developed ToolSHeD (Tool for safety and health in design) to help designers to integrate the management of occupational health and safety risks into the design process. ToolSHeD provides access to information related to protection against falls from

PAGE 39

39 heights during maintenance work on building roofs (Cooke 2008). It is a w eb based system and uses argument trees to effectively represent design safety risk knowledge. Knowledge based expert systems could also be used to facilitate the access of designers to safety knowledge. In 2003, a prototype was developed in the U.K to s tructure HSEs information as a knowledge based system for use by designers. More details about this system will be introduced in the section of the computer aided critiquing system. To sum up, due to the current lack of legislation enforcing designers to fulfill design for safety in the US construction industry, the decision to implement safety knowledge is ultimately made by the design professional. This implementation can be facilitated considerably through the efforts of facility owners and contractors. Through the past decades of research effort, the improved design processes which allow for the consideration of worker safety during the design phase of the project have been created. Design for safety knowledge and design tools that facilitate designers being a part of addressing hazards that could contribute to construction site injuries and fatalities have also been accumulated and developed. All of these approaches and tools will significantly benefit the entire construction industry. Building Informat ion Modeling This section introduces the role and potential impact of information technology on the construction industry. It also defines the concept of Building Information Modeling (BIM). Various BIM applications are explored. The connection between construction worker safety and the delivery method is also examined and a new delivery method (Integrated Project Delivery Method) is introduced. Adoption of Advanced Information Technologies in Construction Industry Information technology (IT) facilitates th e architecture, engineering and construction (AEC) industry to model, analyze, simulate, and predict a projects performance. Methods and

PAGE 40

40 approaches are proposed to integrate project information and leverage information across disciplines and phases to ena ble better work processes and to make more reasonable project decisions (Fischer and Kunz 2004; El Mashaleh et al. 2006). Information technology also makes it easier and more cost effective for AEC managers to manage the safety of their employees. In the p ast decades, information technology including MS Word made it easier and cost effective to process the documents which helped promulgate OSHA standards. Now, more powerful information technology makes possible the coordination and integration of information across disciplines and throughout several phases. New tools and approaches that support the concept of optimization of a projects design from the perspective of multiple disciplines are emerging (Fischer and Kunz 2004). More research studies on the effi cient means for providing designers with information needed to perform design for construction safety are needed. The construction industry is usually slow in adopting and utilizing new technologies because of existing barriers (Mitropoulos and Tatum 2000) Barriers keeping contractors from using the latest technology include fear, initial investment costs, the time to learn how to use the technology, and the lack of support from the senior company leadership (AGC 2006; Toole 1998). There are also driving f orces that prod the construction industry to adopt new information technologies. The driving forces include competitive advantage, process problems, technological opportunity and institutional requirements (Mitropoulos and Tatum, 2000). When the AEC indust ry finds that certain information technology could provide tremendous opportunities, that technology will be widely adopted. Building Information Modeling (BIM) is a good example, as it has been swiftly adopted because project stakeholders perceive the sig nificant benefit of implementing BIM technology.

PAGE 41

41 Concepts of Building Information Modeling Building Information Modeling (BIM) will be described and two important characteristics of it will be introduced. The legal issue related to the use of Building Info rmation Modeling will also be explained. Definition of Building Information Modeling In recent years, Building Information Modeling (BIM) has gained considerable professional and industrial attention. The earliest concept was published in an AIA Journal by Professor Eastman in 1975 (Eastman et al. 2008). After more than three decades of development, it is now regarded as one of the most promising developments in the AEC industry. There are multiple definitions of Building Information Modeling. In the BIM ha ndbook, Eastman (2008) defines BIM as a modeling technology and an associated set of processes to produce, communicate, and analyze building models. The AGC defines BIM as the development and use of a computer software model to simulate the construction a nd operation of a facility. The resulting model, a Building Information Model, is a data rich, object oriented, intelligent and parametric digital representation of the facility, from which views and data appropriate to various users needs can be extracted and analyzed to generate information that can be used to make decisions and improve the process of delivering the facility (AGC 2006) The National Building Information Standard Project Committee defines BIM as a digital representation of physical and functional characteristics of a facility. A BIM is a shared knowledge resource for information about a facility forming a reliable basis for decisions during its life cycle from inception onward. (NBIMS 2006) This third definition is deemed most applicabl e to this research study.

PAGE 42

42 A related term is Virtual Design and Construction (VDC) which is proposed by the Center for Integrated Facility Engineering (CIFE). VDC is the use of integrated multidisciplinary performance models of design construction project s to support explicit and public business objectives (Kunz and Fischer 2009) It forms an integrated framework and set of methods to manage the construction project, including those aspects of the project that must and can be designed and managed. The bui lding, the designconstruction process and the organizations follow the processes to design, build and operate the building (Kunz and Fischer 2009), while Building Information Modeling (BIM) focuses on the building elements of the VDC model. BIM can be vie wed as a subpart of VDC or VDC can be viewed as the ultimate state or end game or BIM automation (Eastman et al. 2008). In practice, VDC and BIM are usually interchangeable, but BIM will be used primarily in this research study. In the past 50 years, CAD ( Computer Aided Drafting) has been widely used in the architectural profession. As a new information technology, BIM as a design tool is different from CAD in several ways (Eastman et al. 2008; Greenwold and Driver 2007). First, BIM offers higher level geom etry operations than CAD does, so designers can represent entire objects rather than drawing sets of lines and points. Designers can define much more complex structures of object families and the relationships that exist between them. Second, the BIM desig n tool entails a great deal of domain specific knowledge. Third, BIM model users can define object families in their own way without resorting to computer programming. Last, as a single unified description of a building, BIM can facilitate coordination bet ween project participants. Characteristics of Building Information Modeling Parametric modeling is a critical capability of BIM which makes the automatic update features possible, and significantly enhances the models productivity. To some extent, BIM cou ld be viewed as an object based parametric model with a set of object families. Standard

PAGE 43

43 practices and codes can be programmed and embedded within it to define object behaviors (Eastman et al. 2008). Users could use both predefined and customized object fa milies. The predefined object families include both vendor provided families and web based downloadable families. A custom object family can be created when a needed object family does not exist. Interoperability is another characteristic of BIM. As differ ent application software is needed to support all of the tasks associated with building design and production, interoperability should be considered when passing data between different software. It identifies the need to pass data between applications, and for multiple applications to jointly contribute to the work at hand (Eastman et al. 2008). There are four main ways to exchange construction data between two applications. One of them is the public level exchange format which uses an openstandard building model such as Industry Foundation Class (IFC). IFC as a data model was developed by the International Alliance for Interoperability (IAI) to create a large set of consistent data representations of building information for exchange between AEC software a pplications. One of the applications of IFC format is to check the preliminary concept design against the specific projects programmatic spatial requirements (Eastman et al. 2008). Many organizations such as Singaporean CORENET and the International Code Council (ICC) in the U.S. are working on this. Legal i ssues Legal issues such as who owns the construction databases, who pays for them, and who is responsible for their accuracy are challenges faced by BIM users. Professional groups have developed guideli nes for contractual language to cover issues raised by the use of BIM technology (Eastman et al. 2008). ConsensusDOCS 301 BIM Addendum is a standard document to globally address the legal uncertainties associated with utilizing BIM. It provides a tool to utilize BIM from start to finish, so contractors could closely integrate project delivery with

PAGE 44

44 owners and design professionals (AGC 2008). The AIA E2022008 establishes the procedures and protocols the parties agree to follow with respect to the development and management of a Building Information Model throughout the course of the project. Meanwhile, it is not a standalone document, but it is intended to be attached as an exhibit to an existing agreement for design services or construction of a project (AI A 2008). As suggested by the AGC, while the use of BIM may well change the ways that projects are conceived, designed, communicated, and built, it will not change the core responsibilities of the members of the project team (AGC, 2006). Contractors and construction managers will still need to organize and lead the onsite construction effort; BIM technology vendors must ensure that their solutions facilitate the building process and these relationships as they exist rather than attempt to shift the responsibilities of the project team members into a contrived software work flow process (AGC 2006; Khemlani 2006). A broader view of the legal issues can be found in other reports (Larson and Golden 2008; AGC 2006). Building Information Modeling Software IFC i nt erface Various BIM architectural design tools (e.g. Autodesks Revit and Bentleys Architecture) grew out of the object based parametric modeling capabilities developed for mechanical systems design. Object based parametric modeling represents objects by parameters and rules that determine the geometric and nongeometric properties and features (Eastman et al. 2008). Often a building model cannot be directly exchanged between different applications because different BIM design tools rely on different definitions of their base objects. Considering that no single computer application has the ability to conduct all kinds of work, the building information needs to be passed between different applications. In the right portion of Figure 2 6, there are diffe rent types of downstream applications which are developed

PAGE 45

45 for specific usage. The interoperability must be realized on a project where different teams are using different software. To solve this problem, the IFC data model has been adopted as the international standard for data exchange and it will be used in this research to transfer data from a BIM architectural design tool to a design for construction safety toolbox. Figure 2 6. IFC Interfaces ( Adopted from : Eastman 2008) Classification of BIM s oftwar e BIM software can be classified into several categories: architecture, structural engineering, MEP, construction, planning and visualization, codes and specifications, facility and asset management, etc. Figure 2 7 shows a graph developed by the Quarry Gr oup, which depicts the classification of current BIM software. It shows that various software applications are divided

PAGE 46

46 into different disciplinary categories. It also illustrates how data are transferred between different software (Quarry Group 2009). Fi gure 2 7. BIM r oadmap ( Adapted from : Quarry Group 2009) Pre design tools such as Onuma Planning System (OPS) are used to facilitate information gathering or decision making processes in the early stages of a project (Smith and Tardif 2009). Authoring tools, such as Autodesk Revit, Bentley Architecture, and Graphisoft ArchiCAD are developed for building design. Design firms also use audit and analysis tools which can only be used to analyze building information models created by authoring tools such

PAGE 47

47 as Revit Architecture (Smith and Tardif 2009). Audit and analysis tools are used to do specific work such as clash detection, sustainable design analysis, code compliance, and cost estimating. BIM can be used by constructors to conduct construction cost estim ating, constructability analysis, and construction sequencing. Owners can utilize BIM to do facility management and maintenance. Code c hecking s oftware One software application that warrants a detailed analysis is one designed for code checking. In civil e ngineering, most of the expert knowledge has taken the form of codes and regulations. The purpose of a building code is to establish minimum requirements necessary to protect public health, safety and welfare in the built environment (Wix and Liebich 2009) It is different from the design for construction safety knowledge because the principle of design for construction worker safety is to protect construction workers during the process of construction. Another difference between them is that there are curr ently no administrative processes to enforce designers and constructors in fulfilling design for construction safety responsibilities or to ensure that projects are in compliance. The traditional manual code checking method is time consuming and error prone. The vision of automatically checking building plans against codes emerged as early as the 1970s. Before the 21st century, researchers had made some progress in this area. However, it was not until recent years that the new generation of software technology and standards made automated checking much easier (Wix and Liebich 2009). In the U.S., automated code checking was championed by the International Code Council (ICC). The ICC is a membership association dedicated to building safety and fire prevention by developing the codes used to construct residential and commercial buildings. Many U.S. cities, counties and states adopt codes developed by the ICC ( 2009). Since 2004, the ICC tried to use

PAGE 48

48 object based technology for representing their codes and creat ed the SMARTcodes project. The ICC is currently cooperating with corporations such as Solibri and AEC3 to develop the platform for its automated code checking. Solibri, Inc is the pioneer in providing out of box software that automates the checking and a nalysis of the BIM model (Solibri 2009). Solibri Model CheckerTM (SMC) is an IFC model checker. SMC imports models created in other software packages and checks them for compliance with rules. These rules can be a comprehensive set of general purpose BIM checking rule sets or user defined specific rules. This tool has great potential for owners and code officials to quickly verify if building models are in compliance (Solibri 2009; Quarry Group 2009). COMcheck and REScheck were developed by the U.S. Departm ent of Energy (DOE) to improve the energy efficiency of the nations buildings by promoting building energy codes. These software tools simplify energy code compliance by offering a flexible computer based alternative to manual calculations (DOE 2009). Mor e information can be obtained from the DOEs website for the building energy codes program. BIM has been defined and several types of software tools related to this research have been introduced. Even though BIM technology offers considerable benefits to t he construction industry, adopting BIM technology alone will not eliminate all the construction worker injuries and/or guarantee a successful project. Construction work is so dynamic that many other factors will influence the outcome of the project. Withou t excellent management and safety oriented stakeholders, a zero injury objective cannot be achieved. Delivery Method Influence on Safety Performance Designers often do not have the expertise to make the design drawings sufficiently safe for construction workers to finish the construction work without injury. The inadequate interoperability partly causes this problem. Interoperability identifies the need to pass data

PAGE 49

49 between applications, and for multiple applications to jointly contribute to the work at hand (Eastman et al. 2008). Studies show that the lack of interoperability can cause tremendous inefficiencies and waste in the construction industry. For example, a report of the National Institute of Standards and Technology (NIST) determined that the AEC s oftware had inadequate interoperability that totaled costs of $15.8 billion per year (NIST 2004). One major reason for the present situation is the current project delivery methods do not support cooperation and data exchange between project participants. The longstanding separation of the roles of architects and contractors makes it difficult for team members to share knowledge and common project data. In the most prevalent delivery method of DesignBid Build (Figure 2 8) for example, the designer develops a design based on the owners requirements, then a constructor is selected to build it. Figure 2 8. Participant relationships in the design bidbuild method With this procedure, the project is designed with little expertise from the constructor who act ually constructs the project. As a result, many constructability and safety issues are not

PAGE 50

50 considered until the construction phase. Furthermore, public works projects often dictate that open bidding must be used in government construction projects, so substantive early involvement of the actual constructor is essentially prohibited (AIA 2007). The early period of the design phase is important for the project. As Figure 29 shows, the project teams ability to affect project variables such as cost, schedul e and constructability decrease as the project progresses. Meanwhile, the cost of making changes dramatically increases. Figure 2 9. Teams ability to affect project variables ( Adapted from : Eastman 2008) The type of project delivery method will impact the extent to which safety can be addressed in the design. The forms of project delivery alter the roles played by the different parties and the allocation of their responsibilities. A 1992 study found that designbuild firms addressed safety in their proj ect designs more often than design only firms (Hinze and Wiegand 1992). Gambatese (2005) criticized the traditional general contract (designbidbuild) approach as keeping the parties isolated, with no payback apparent for the designers to address construc tion worker safety. Alternative project delivery methods can be used to access the

PAGE 51

51 constructors knowledge to find safety hazards and to facilitate the implementation of design modifications (Gambatese et al. 2005b). Toole (2007) also confirms that both the fee structure and model contract terms of a designbuild project could induce design engineers to consider construction safety during the design phase. Ash (2000) contends that Great Britains Construction Design and Management (CDM) regulations which pl ace a duty on designers to ensure design for safety yields some success when designers and constructors work together more closely, such as in designbuild and construction management companies. Integrated Project Delivery method To solve the current probl em, the Integrated Project Delivery (IPD) method has been introduced. The AIA (2007) defined IPD as a project delivery approach that integrates people, systems, business structures and practices into a process that collaboratively harnesses the talents and insights of all participants to optimize project results, increase value to the owner, reduce waste, and maximize efficiency through all phases of design, fabrication, and construction. Integrated projects are uniquely distinguished by highly effective c ollaboration among all participants to maximize value for the owner, commencing at early design. Figure 2 10 shows that project flow in an integrated project differs significantly from a nonintegrated project. Design decisions are moved upstream as far as possible to make the process more effective and less costly (AIA 2007). Specifically, IPD allows constructors to contribute their expertise in construction techniques early in the design process resulting in improved project quality and financial performa nce during the construction phase. IPD leverages early contributions of knowledge and expertise through the utilization of new technologies. The IPD allows the designer to benefit from the early contribution of the constructors expertise during the design phase. Designers can fully understand the ramifications of their decisions at the time the decisions are made. The close

PAGE 52

52 collaboration eliminates a great deal of waste in the design, and allows data sharing directly between the design and construction tea m, thereby eliminating a large barrier to increased productivity in construction. Figure 2 10. Traditional and redefined project phases ( Adapted from : CURT 2004; AIA 2007) Since project participants share expertise and risks, Figure 2 8 could be modifie d as shown in Figure 2 11 which demonstrates the situation under the IPD method. All the stakeholders of a project will own one database which incorporates all the information about the project. Even though architectural and structural engineers will have their own drawings, their models will be derived from one database. Different stakeholders will have different authorities to access or revise the database. After the project is finished, the owner will have ownership of the database.

PAGE 53

53 Figure 2 11. Parti cipant relationships in a project under the IPD method ( Adapted from: MacLeamy 2009) Contracts that define new business terms that facilitate collaboration are essential for adopting the integrated delivery method. They could help designers find opportunit ies in nontraditional areas throughout the life cycle of projects (CURT 2005). Unlike traditional construction contracts, the discrete responsibilities of the designers and constructors are more intertwined. For the benefit of the project participants, eve ry participants work scope should be clarified in the IPD agreement after negotiating the risk allocation on the project. Both the AIA and AGC have launched agreements supporting IPD. The ConsensusDOCS 300 brings owner, design professional and constructor together as a core team to act on behalf of the project (Perlberg 2009). The AIA has introduced two types of agreement packages for the emerging process. The Transitional Agreements are a first step to IPD, providing for an early collaboration of projec t participants in an arrangement modeled after existing construction manager agreements (AIA 2009). When the project participants are fully prepared, Single Purpose Entity could be used. It allows a complete sharing of risk and reward. With this type of

PAGE 54

54 agreement, the participants work together to design and construct the project with mutually agreed upon cost goals. Both types of agreements encourage project participants to implement BIM to realize the full potential benefits. Integrated Project Delivery with Building Information Modeling The IPD process unlocks the power of BIM (Yoders 2009; AIA 2007). The full potential benefits of both IPD and BIM can be achieved only when they are used together. The IPD project team reaches an understanding regarding how the building model will be developed, accessed, and used, and how information can be exchanged between different participants. Without such a clear understanding, the model may be used incorrectly or for an unintended purpose (AIA 2007). Some projects have successfully used IPD and BIM together to deliver projects to owners. These initial practices show how BIM and IPD could be used together to deliver a project by reaching criteria such as design aesthetics, materials selection, budget, schedule, and sustainability (Yoders 2009). Some key elements were summarized to help succeeding project stakeholders. In summary, the change in delivery methods provides project participants with new opportunities to succeed. Project participants must take on new roles and competencies to achieve it. More benefits could be realized by using IPD and BIM together. Computer Aided Critiquing System Computer Aided Critiquing System in Design and Construction K nowledge based expert system s are a subfield of Artificial Intelligence (AI). The fragmentation of the U.S. construction industry causes tremendous inefficiencies in all phases of a construction project. In the past decades, several knowledge based expert systems with different functions have been developed in design and construction domains to reduce these

PAGE 55

55 adverse impacts of fragmentation (Fischer 1991). Many different problems which need specific expertise now could be solved at the field level. When the knowledge based expert system technology is used in the design process to support designers in different domains, it can be called a computer aided critiquing system or an automated critiquing system. According to Oh et al. (2008), the definition of a computer aided critiquing system used for design purposes is: A design critiquing system is a tool that analyzes a work in progress and provides feedback to help a designer improve the solution. It may ask relevant questions, point out errors, suggest alternatives, offer argumentation and rationale, or (in simple and obvious cases) automatically correct errors. A critiquing system not only alerts designers to their errors, but also helps designers improve their drawings with constructive feedback (Oh et al. 2008). Notable critiquing systems include Singapores CORENET syste m and Solibri Model Checker. Summary of Related Work Researchers have developed a few automated critiquing systems to support designers when making design decisions. In this subsection, the mechanism of some of these systems will be analyzed. Three criteri a should be considered when selecting the model system. First, the system should be developed to assist the architects/engineers in making informed decisions by providing them with machine generated feedback. Second, the critiquing system should have the knowledge based reasoning function which could automatically check inputted drawings based on encoded rules. A knowledge based reasoning structure consists of a set of rules with logic equations that use the object attributes in the 3D model to assist the user in making informed decisions (Korman and Tatum 2006). Last, the data for the buildings which will be input into the checking system should be created in object oriented threedimensional (3D) models, in which

PAGE 56

56 the basic components of drawings are buildi ng elements such as walls, doors, and so on, rather than geometric elements in the traditional CAD software. Fischer (1991) developed the Construction Knowledge Expert (COKE) in the early 1990s, which indentifies designrelevant constructability knowledge to help designers of concrete structures check for constructability issues during the initial design. First, COKE builds a symbolic geometrical and topological project model based on the information in the CAD system. That object based CAD system was speci fically developed to create the project data for constructability feedback because CAD systems at that time could not effectively link to an expert system. The system uses high level application heuristics to determine whether a construction method is appl icable of not. Lastly, the system compares the project data in the symbolic model with specific constraints about the layout and dimensions of structural elements. If any conflict is identified, the system will alert the designer (Fischer 1991). A more rec ent effort of CIFE was to promote the coordination of the MEP trades. Korman and Tatum (2006) proposed a knowledge based system that is able to provide valuable insight to engineers and construction personnel, to assist them in resolving coordination probl ems for multiple MEP systems. The system uses objectoriented 3D models and a knowledge based reasoning structure to assist the MEP coordination process by linking components of an object with a particular set of knowledge. In an earlier research study, th ey developed a knowledge framework and reasoning structure for the system. The model based reasoning (MBR) method and the heuristics method are used as the reasoning structure to perform diagnostic tasks. MBR could abstract geometric and topological data f rom the geometric model and then determine the spatial relationships between the components in the model. After classifying the types of interferences, heuristic reasoning is used to determine and resolve coordination conflicts.

PAGE 57

57 CORENET (Construction and R eal Estate NETwork) was an IT initiative conducted by the Singapore government to re engineer the business processes of the construction industry. One task of this project was to develop an automated national building code compliance checking system called e Plan Check. Owners could submit building models in IFC format to the Singapore regulatory agency for automated code checking by this system. One of the important projects in CORENET is the Integrated Building Plan and Building Services System (IBP/IBS). This part of the project was developed through a coordinated effort with Jotne EPM Technology AS, Norway, by using EPMs software applications. The applications of EPM Technology are based on the EXPRESS Data ManagerTM (EDM) utilizing ISO standards, parti cularly ISO 10303, which is a STEP or Standard for the Exchange of Product model data. One application in this suite of products is EDMmodelCheckerTM which provides validation of the dataset being used based on the EXPRESS data modeling language (Jotne EPM Technology 2009). This module has been used to implement the e plan Check system. With the passage of the Construction (Design and Management) Regulations 1994, the U.K. Health and Safety Executive (HSE) was concerned that safety should be as much a key a spect in design as it is during construction and operation (AEC3 2009). The belief was held that the poor health and safety record of the construction industry could be improved by encouraging designers to give more consideration to health and safety issue s during the design stage (HSE 2003). NNC Limited, a U.K company, cooperated with AEC3 and IAI UK to develop a prototype system in which health and safety information is structured as a knowledge based system that can be delivered to designers. The whole project was split into five phases, with three being directed to the development of the system. These three phases include: data gathering, data structuring and data delivery. The data are primarily concerned with the hazards while working

PAGE 58

58 at height, and ac cidents due to falling objects. The major sources of safety and health information included an invited health and safety consultant, documentations available from the HSE and experiences of designers from both internal and external to NNC Limited. These da ta are classified and structured into three areas: (1) textual health and safety data; (2) health and safety rules; and (3) object properties. A software application should be selected as the platform to deliver the information. After investigating the via bility of the system that could carry out the checking function, the HSE prototype used software that was developed by Singapore CORENET as the design checking mechanism because the building regulations compliance checking is analogous to the checking of d esigns against health and safety risks (HSE, 2003). Last, an object based CAD system exports design data in the IFC format to an EDM database provided by EPM Technology. Design data are tested against health and safety requirements that are graded accordin g to levels of risk. The checking results are reported through graphic and rule browsing software (AEC 2009; HSE 2003). As introduced in the Code Check Software section, in 2004 the International Code Council (ICC) began to develop object based technology to represent their codes and to test submitted construction documents. The key elements are a model checking application and SMARTcodes. The following paragraphs will explain the project by analyzing two major tasks: encoding building code requirements int o a rule base and developing model checking software. The International Code Council (ICC) defines the SMARTcodes as I Codes in an IFC (Industry Foundation Classes) compliant, interoperable format that can be used by Model Checking Software (MCS) as a rul e set of limits or constraints from the code and applied by the BIM to show where conflicts occur (ICC 2009). To encode the SMARTcodes, the obstacles are the interpretation and presentation of the building code and the encoding of that presentation in a

PAGE 59

59 r ule base (Conover 2009; Nisbet 2008). A protocol and software program were used to create tagged representations of building codes that have a tagging schema that reflects the logic and requirements of the codes from the text of the codes (Conover 2009). A n online version of the Solibri software application or the AEC3 Xabio webbased test bed could be adopted in the model checking application. Solibri Model CheckerTM (SMC) is a product of Solibri, Inc. This application is designed to find potential problem s, conflicts, or design code violations in a building model through three steps. First, the user creates a building model in an IFC compliant application and saves it in the IFC format. Next, the user selects and loads the specific rule sets that will be u sed for checking the model. Last, the application automatically conducts the checking and produces the results. Users could export a report from the applications for further analysis. The key point to successfully use this application is to develop rule se ts which can be used for projects of different types and in different locations. The application provides a number of default rule sets and the user also can develop customized rule sets (Khemlani 2009). In the SMARTcodes project, a customized version of S olibri Model Checker is adopted as the model checking application. Figure 2 12 depicts a block diagram of a system for checking a building information model against SMARTcodesTM in a trusted entity. The BIM authoring software is software that creates and maintains a building model. Besides Solibri checker, AEC3 XABIO also can be used as the checking engine in ICCs project by sharing a dictionary of testable concepts which are defined both in plain language and in buildingSMART terms with SMARTcodes (Nisb et et al. 2008). It uses EPM and Octaga technology.

PAGE 60

60 Figure 2 12. Block diagram of a system for checking a building information model against SMARTcodesTM ( Adapted from: Conover 2009) AEC3 XABIO can check a whole regulation or an individual clause and then generate a full explanation. It is also web based in cooperation with the Apache Software Foundation and Octaga. On the website of ICC, videos demonstrate the trial system which uses Solibri or AEC3 XABIO checking engines. The e Plan Check is also used as the checking engine to show the interoperability of the SMARTcodes (AEC3 2009). Figure 2 13 shows the relationship between three different checking engines. Checking systems based on these checking engines are also demonstrated. The SMARTcodesTM project started from energy codes which are relatively easy to define and codes in other domains were gradually formalized into rule sets. Furthermore, this system currently does not promise full automation since some codes that are nuanced and subject to interpr etation will have to be checked manually (Khemlani 2009).

PAGE 61

61 Figure 2 13. Predominant checking systems House Designer is a graphical tool for product and house design developed by Selvaag BlueThink. Several rule sets, such as geometry rules, functional rul es and building systems can be defined in the system. This software can generate a building from rules, not just passively checking if building models satisfy certain rule sets. The user sketches the functional intent and layout, and the House Designer applies the rules and automatically works out the various elements and decisions that constitute a complete building design. The software immediately warns the user when conflicts appear and rules are violated (Selvaag, 2009). This section described some comp uter applications related to this research. Table 2 2 summarizes some computer applications associated with this study

PAGE 62

62 Table 2 2. Computer aided critiquing systems Name Purpose Knowledge Bases Developer Date COKE Provide constructability feedback fo r the preliminary design of reinforced concrete building structure Constructability knowledge CIFE 1989 Prototype Mechanical, Engineering, and Plumbing Coordination Tool To assist in MEP coordination during the design stage Design criteria, construction, operations, and maintenance CIFE 2003 2006 CORENET e Plan Check Check design for regulatory compliance through the internet gateway Authorities Regulations Singapore BCA Since2000 Knowledge Based Expert system Deliver health and safety information to des igners Information concerning hazards of working at heights UK HSE & NNC Limited 2003 Solibri Model Checker Provide commercial model checker Rule sets Solibri, Inc 1999 AEC3 XABIO Check regulations or an individual clause Rule sets AEC3 2006 BlueThink H ouse Designer Product and house design Rule sets Selvaag 2007

PAGE 63

63 CHAPTER 3 METHODOLOGY The initial objective of this research was to explore how BIM technology can be used to support construction safety efforts. This was dominated by developing a design f or construction safety (D4S) software tool which can automatically check building models to identify potential hazards on construction site. Only after identifying the safety hazards can the engineers and constructors assess and mitigate identified hazards prior to or during the construction phase. Specifically, this research study was focused on identifying the fall hazards, as falls are the most frequently occurring types of accidents resulting in fatalities, which account for forty percent of all constru ction worker fatalities. With 36 collected best practices related to falls, particular emphasis was placed on openings located either i n floor slab s or in wall s in that falls are often associated with workers on roofs and floors with openings. Hinze and Wi lliam found that f alls from roofs/ floor edges and falls through roof/floor openings accounted for 40% of fall fatalities (2011). The falls associated with scaffolds and ladders are not discussed at the current stage because most of these are temporary str uctures and are difficult to associate with specific building locations. The research methodology for this research consists of six sections: The first section discusses the possible influences of BIM technology on construction safety. Then the opportunit y of using a computer aided critiquing system that provides design for safety knowledge to designers is assessed. The second section analyzes a typical design process to identify the appropriate timing at which to conduct design for construction safety che cking to effectively deliver the safety knowledge to designers under the environment of computer aided critiquing systems. The third section presents the work of collecting existing design for construction safety suggestions from past research results and efforts to develop new suggestions

PAGE 64

64 to enrich the knowledge base. These collected suggestions are classified into five categories as falls, struck by, caught in/between, electrical shock and others. For this research study the software tool is exclusively f ocused on fall hazards. The fourth section explains the ideal architecture and functionalities of the tool. It also explains other features of the tool, such as the software interface and interactive method between the user and the model checking software. The fifth section introduces the process of selecting a viable authoring environment as the platform to develop model checking software. These software platforms can be classified into different categories based on the programming languages used to compil e the computable rules and an assessment of how friendly the user interfaces are. The sixth section differentiates between the semantic query and spatial query. The usage of a library to constrain the terminologies is introduced. It is the technical founda tion to compile the computable rules. The seventh section describes the methods of encoding collected fall protection safety suggestions into computable rules which can be used as rule sets in different model checking software. The eighth section discusses the preparation of building models which are used to test the compiled computable rules. Finding the influence of BIM technology on Construction Safety As described in the introduction, the first objective of this research is to delve into the potential impact of BIM technology on enhancing construction worker safety. To get an idea of how BIM technology as a tool could address construction worker safety, the research effort concentrated on two resources. Knowledge was acquired through a literature search to find the potential positive impacts of BIM technology on construction worker safety. During this process, various textbooks and journal papers were reviewed. The Internet was also found to provide valuable knowledge related to this research.

PAGE 65

65 After evalu ating the current usage of BIM technology in the construction industry, it was concluded that design for construction worker safety is an appropriate area where BIM could be used to enhance construction safety. That is, while the theory of design for const ruction worker safety has been gradually established through past research studies, the implementation of this concept has not been widely carried out in project designs. There is a compelling need for tools that can put the safety in design concept into practice. Thus, the second objective of this research is to discover a means by which design for safety knowledge can be implemented in practice by using BIM as a tool. Based on previous research, it was recognized that the development of a model checking s ystem is a viable approach to deliver valuable design for construction worker safety suggestions to design professionals. The various advantages of BIM technology made a computer aided critiquing system/model checking system be an advanced application to a ddress construction worker safety during the design process. After identifying design for construction worker safety as the specific topic of this research and a model checking system as the tool to address this issue, the primary effort was to analyze the current design process to identify the appropriate time to conduct design for construction safety checking work by using such a tool. Design for Construction Safety Process There are many small details that need to be taken into account during the design phase. As a result, only the best designers consider design for construction worker safety as an integral part of the main design process. Potential construction hazards can be easily designed into building models during the design process, so design revie ws should be conducted to detect noncompliance and make designs safer. One of research questions is to determine when to conduct a design for construction worker safety building model review.

PAGE 66

66 Ideally, a computer aided critiquing system performs two major tasks to help designers provide high quality building models in a mature design process. One is to minimize noncompliance during the main design phase by informing and training designers about alternative design options and the consequence reducing practice s. Another method is to conduct noncompliance detection and correction after the main design work has been completed by carrying out safety checks. By using these two methods, a design for construction safety tool could be able to improve designs by providing appropriate knowledge or checking the final building models. Past research studies ha ve recognized, however, that a limited number of progress reviews for safety may be more favorable. That is, because of the pressures on designers to meet deadlines an d budgets and to solve problems, design for construction safety is usually ignored in the design process. One merit of a computer aided critiquing system is that it can facilitate the access by designers to the design for construction safety (D4S) knowledge during the design process or assist designers in making decisions which are more constructable for the project life cycle. Because of the limitation of the current model checking software, the D4S software tool i s mainly focused on checking for noncompl iance of the finished building model. This tool also provides other functions such as model navigation, result presentation, and report generation. The issues about when and how to conduct a safety review have been discussed in the previous two sections. T he following sections are focused on developing a model checking system which can be used to automatically check building models for noncompliance. Defining the Rule Source To develop the model checking system, the design for safety suggestions that were i ncorporated in the database must first be collected. This section describes the process of accumulating existing and new design suggestions. The existing suggestions came primarily

PAGE 67

67 from two previous editions of a safety toolbox. New design suggestions incl uded provisions from the OSHA regulations, publications of the National Institute for Occupational Safety and Health (NIOSH) and knowledge from the HSE of the U.K. Existing Design for Construction Worker Safety Suggestions Previous research had already dev eloped and compiled numerous designing for safety suggestions. The first edition toolbox developed by Hinze and Gambatese (1997) included 430 design for safety suggestions. Marini ( 2007) in the second edition of the toolbox added more than one hundred ne w suggestions into the database. These suggestions were collected from safety design manuals and checklists, ideas generated by the researchers and CII research team members, and interviews with industry personnel Through these previous two research effor ts, nearly all the major potential problem areas had been examined and corresponding design for safety suggestions had been developed. In the second edition of the toolbox, these existing suggestions were classified into 20 categories. The following list s hows these categories: Administrative: Layout Administrative: Planning Administrative: Design Sitework: Layout Sitework: Roads and Paving Sitework: Earthwork Foundations Roofing Structural: Steel Structural: Concrete Structural: Masonry Structural: Timber and Wood Finishes: General Finishes: Stairs and Railings Finishes: Ladders Doors and Windows Mechanical and HVAC Electrical Industrial Piping

PAGE 68

68 Tanks and Vessel This sorting method was reasonable and proved to be effective for the second edition of the toolbox. Other classifying methods that are more suited to the object oriented software were considered and will be discussed in the following subsections; however, before the model checking software was finally confirmed, the above sorting method was used to classify and compile new suggestions. New Design for Construction Worker Safety Suggestions There were two major sources of the new suggestions. The first one was from the U.K. HSE. Other new suggestions were gained from examining the OSHA regulations. Th e passage of Construction Design and Management (CDM) has tremendously promoted the implementation of design for construction worker safety in the U.K. The Health and Safety Executive (HSE) as the official in charge of safety has accumulated considerable s afety knowledge which can be used in this research. The design for safety suggestions which were collected during the project Knowledge Based System To Deliver Health And Safety Information To Designers especially focused on how to avoid risks while wor king at height. The possible differences between the U.S. and U.K. construction industries should be considered. Meanwhile, additional suggestions were collected by further examining the OSHA regulatory provisions. As the employer of construction workers, the constructor is given the primary responsibility for worker safety. Also, the OSHA standards and policies for the construction industry are directed at the constructor (Toole and Gambatese 2002). The voluntary participation of designers in the OSHA trai ning and education can potentially increase construction site safety.

PAGE 69

69 Some OSHA provisions involve architects and engineers or include several instances where the services of a professional engineer are required for analysis and design of temporary constr uction structures. Meanwhile, designers should possess at least a limited degree of expertise in construction safety to contribute to construction worker safety. In another words, they should be familiar with the OSHA regulations. Unfortunately, most desig ners know little about the OSHA regulations. In the survey conducted by Gambatese (2003), none of the 36 surveyed civil engineering departments offered a course strictly on construction safety. Another survey conducted with design firms by Toole and Marqui s (2004) found that less than one fourth of the U.S. participants believed that employees in their design firms were capable of identifying site hazards to which construction workers are exposed. Some researchers have suggested that 10hour and 30hour OSH A courses for design professionals should be developed and promoted. The OSHA regulations (Code 1994) were used to develop design suggestions in the research conducted in 1996. Later, Gambatese et al. (2003) introduced more information about engineering m andates stipulated in OSHA regulations. In this research, the OSHA regulations were further analyzed and the literature (such as the Designers Guide to OSHA) was a helpful resource to devise additional design for safety suggestions. The collected suggest ions were compiled and classified according to twenty categories previously listed. This helped ensure that no duplications of existing suggestions were included in the database. Classification of Design Suggestions Through previous research, a large numbe r of design for construction worker safety suggestions had been collected. All suggestions were sorted into certain categories for future use. The first edition of the safety toolbox used three ways or paths to sort suggestions: project component, const ruction site hazards, and project systems. These three ways of accessing design for safety suggestions were still valuable for this research study. In this phase, the

PAGE 70

70 feasibilities of these three methods were evaluated based on the function and architectur e of the new D4S software tool. Every way had its own merit when it came to manage knowledge. When using project component s the application tool would present a list of components typically found on construction sites. Considering object oriented BIM appl ications also based on various components of a building, this method could be a viable choice. Rather than examining a particular component constituting a building, the designer could focus on specific jobsite hazards. This can help designers access knowle dge related to certain types of hazards. Major types of causes of injuries and fatalities such as falls, cavein s and electronic shock could be controlled through this method. This was also the approach taken by HSE to develop their knowledge based system in the U.K. Classifying suggestions by project system was the third way, which reflected the standard Construction Specifications Institute (CSI) format and numbering system. The advantage of this method was that the International Code Council (ICC) also adopted the CSI format to manage their building codes. Using a dictionary was a major character of ICCs SMARTcodes project. The advantage of the dictionary and classification based aspect of the work was that it enabled the codes to be searched to identify only those that were relevant to a particular topic and to deliver these exclusive of all other, non relevant codes. The dictionary was managed by the CSI in cooperation with ICC. While using either of these category methods could incorporate all the sugg estion into the toolboxs database, the first two methods were jointly adopted to develop the D4S software tool. Past research studies disclosed that fall accidents account for a large portion of construction injuries and fatalities. For this research stud y, the rule sets of the D4S tool were mainly focused on the potential fall hazards in the building models. That was done because most D4S suggestions related to falls are directly associated with permanent building structures, which

PAGE 71

71 makes them easier to be complied into computable rules. The designing for construction safety best practices were reviewed to identify those provisions that deal with fall protection. More than thirty provisions on fall protection were identified. These provisions were further c lassified into different groups based on their target building objects. The list of the design suggestions is provided in the Appendix A. The classification of D4S best practices can be clarified by using a tree shape diagram which is shown in the Figure 31. D4S suggestions were classified into five categories, including falls, struck by, caught in/between, electrical shock and others. Falls can be further sorted into five categories associated with different building components where accidents happen. F igure 3 1. Classification of design for construction worker safety best practices The D4S safety suggestions were classified. The architecture and functionality of the potential model checking system will be discussed next

PAGE 72

72 Architecture and Functionality of the desired System After D4S suggestions had been collected and classified, the next step was to develop the proposed Design for Construction Safety (D4S) software tool -a construction safety checking system. The purpose of this system was to automatica lly check imported building models which are in IFC format or other document formats to alert users to opportunities for improving construction safety. The system should provide design for construction safety knowledge quickly, easily and economically. Sof tware platforms were selected based on the requirements discussed in this section to effectively deliver safety knowledge to users. According to the technical requirements of these software platforms, suggestions were formalized and encoded to rule sets, w hich also can be referred to as constraint model or computable rules. The architecture of the software tool was first set up to define the scope of this tool. Several issues were considered as important. Researchers had identified some factors which should be considered when developing a computer aided system (Oh et al. 2008). According to experience from former research efforts, three features of the D4S tool were closely considered when deciding when and how to intervene in the design process by the desig ners of this software tool. The first feature concerned the conditions under which the software tool would activate the intervention function. Systems taking an active critiquing strategy would continuously monitor designs as they evolve and offer feedback while a passive critiquing strategy would only give feedback when designers specifically ask for it. Because the timing of providing design for construction safety knowledge had been decided in the second section, th e system developed in this study adopt ed the passive critiquing strategy to avoid distracting the designers as designs evolve, although the active critiquing strategy may be a better solution for some approaches.

PAGE 73

73 The second feature concerned the type of feedback. The applications developed for code checking work usually point out the specific errors in building models. The D4S software tool is different in that it suggests the opportunities of integrating safety into the building models even though the documents may be errorless. That is becaus e currently the design for safety knowledge consists of suggestions and there are no compulsory legal clauses to enforce their fulfillment. So the software tool should be developed to not only offer negative evaluations but also provide positive knowledge. The third feature concerned the form in which the feedback will be reported to users. This D4S software tool includes two types of feedback: text messages and graphical markers to indicate the locations where improvements are possible. Text messages give detailed explanations to designers to help them to better understand the design alternatives. By colorfully marking the problematic components in the proposed three dimensional drawings, the D4S software tool explicitly shows designers the locations that n eed to be addressed. After determining the course of action for the above three issues, the architecture of this tool was defined. Figure 32 shows the desired architecture of the D4S software tool. The x axis represents the project process from the beginn ing of the design to the deliver y of the documents to the constructors. This process begins on the left with the design development period when the designers draft the initial building models by using BIM authoring software. It then evolves into the design review phase and the checking for code compliance period. Compliance checking of a building model is conducted in these two phases This culminates in the construction phase. The design process is an iterative one. Users could submit building models and c heck against noncompliance by using the D4S tool After the report identifies the problematic building components, the us ers can revise their drawings by returning to the architectural design tools.

PAGE 74

74 The core of the entire process is the model checking sof tware which is supported by a dictionary and a design for construction safety rule set. After the design for construction safety knowledge has been incorporated into the construction documents, shop drawing s can be used by the constructors for further cons truction work. Figure 3 2. Architecture of Design for Construction Safety tool After the architecture of the D4S software tool had been determined, the next step was to define the functionalities of the D4S software tool. The software tool was developed to have two main functions. The first function consists of checking building models against the design for construction safety rule sets. The second function of the D4S tool was to provide safety information related to certain building components. This wa s based on both the characteristics of the design for construction safety knowledge and the reasoning process of the software tool. One of the differences between building codes and design for construction safety knowledge i s that a large number of design suggestions are in the textual form at without any

PAGE 75

75 parametric information. Many of these suggestions were very difficult to be encoded into rule sets that can be compared with the properties of building components and can be used to restrict noncompliance. Consequently, it is better to keep their original form and show them to the user in text format while most of the building codes are connected to attributes that can be physically measured. This point is very similar to delivering constructability knowle dge to designers during the preliminary design phase. In consider ation of this point, th is research needed to find an appropriate way to deliver safety knowledge to designers. Three possible ways were examined to find the appropriate methods for delivering design for construction safety information to the D4S software tool users. The first type was health and safety rules which could be checked by the model checking software for non compliance. Most suggestions related to objects which have clear design parameters that could be checked in this way. Parameters could be set as rule requirement s in the system. If the value of an object in the drawing violates the parameter, then the system reports the noncompliance to the users. The second type was safety kn owledge in textual form. When a suggestion was too difficult to be directly checked, the system should provide knowledge in textual form. Information in this form can be general (e.g. locate rooftop mechanical/HVAC equipment away from roof openings), or re lated to individual construction objects (e.g. locate roof opening away from the edge of the structure). The third method was to deliver information by using an object properties setting. Designers could recode and deliver relevant health and safety inform ation within their drawings. For example, if a designer specifies the construction method for an object as unusual, the designer could suggest to the constructors that this is an important point of concern during construction. After assessing the entire mechanism of the D4S software tool, this method was not

PAGE 76

76 considered in this D4S application tool. One reason was that a large portion of the information of this type is related to the construction method, which makes them inappropriate to be included in the D4S knowledge database. Another reason was that designers need to do additional work to input such properties, which makes this step more time consuming. All of these made this functionality an unattractive one to designers. This application tool used the first two methods to deliver design for construction safety knowledge to users. Based on the software tool developed from Solibri Model Checker, the interface of this D4S software tool consists of four subviews: Model tree. It is a treeview of all model objects. Rule sets. Users c an select rules that will be used to check the model. Result view. Detailed explanation will be shown in this window. 3dimensional virtual reality model browser. After the interface and the method to deliver information to us ers have been determined, the basic functionalities of the D4S tool can be shown, as in Figure 3 6. The process of checking a building model for non compliances includes the following steps. First, the user imports the building model into the rule checker. Then the 3D view can be shown on the right hand side of this application tool. The navigation functions usually include Zoom, Spin and Walkthrough. On the left hand side there are checkboxes which are used to select objects and rule sets. The user could get detailed properties of any object by selecting an object tab. The user also can access all design for construction safety suggestions by selecting them on rule sets. A detailed explanation of every suggested design provision will be provided and some gr aphs will also be given to illustrate complex issues. Next, the user can select the rules that will be used to check specific objects against After running the checking function, two sets of results will be produced. One is a list of all non compliance is sues identified in the drawings, along with suggestions about how to eliminate or mitigate these issues. The user could print the report in PDF or other forms.

PAGE 77

77 Another set of results will be shown on the right hand side in the form of a 3D view. Red circle s will show all the components which violate certain rule sets. After getting the report from the model checker, the user can change drawings in the architectural designing tools or keep the original design ideas if other requirements need to be met. Desig ners will be advised to keep a record of their decisions for future use. Figure 33 demonstrates basic functionalities of a D4S software tool. Figure 3 3. Functionality of the D4S software tool Platform Chosen After the discussion of the architecture an d functionality of the potential model checking system, appropriate software platforms were selected. The computable rules/rule sets were then developed based on these platforms or building model servers.

PAGE 78

78 Several software tools were examined to find the ap propriate ones to be used as the fundamental platforms of the D4S software tool. Only a few of them meet the requirements discussed in the last section. The major difficulty was that the 3D model checking technology was still a very advanced research area with a limited number of commercial software applications specifically developed for this purpose. In addition, the few successful software platforms were developed and owned by industry companies. These companies applied for patents of their technologies and hardly any publications were released for reference. Fortunately, some free BIM servers possessed the model query function, which uses programming language to filter building models to retrieve useful information. Four BIM servers were used in this res earch study as the software platform. P rogramming languages have to be used to effectively query information from building models stored in different kinds of database format s. For the purposes of this research study, the software platforms were classified into three types based on how friendly they were in regard to their user interfaces used to compile D4S suggestions into computable rules. In the first type of software platform, the database can be queried by directly using a programming language. The o bvious advantage of this kind of platform i s its flexibility. Each suggestion/best practice can be compiled into computable rules by manually encoding programming codes. Many engineers in the AEC domain do not have the expertise to work with the programmin g languages because the language syntax usually is not immediately recognizable to nonprogrammers. Therefore, a large amount of training is needed to prepare engineers to be competent computable rule writers or users. The open source software tool BIMserver belongs to this type. The computable rule compilers need to understand two programming languages --

PAGE 79

79 EXPRESS and Java schema to effectively query build models. This cross disciplinary collaboration is often required to successfully complete such a projec t. The second kind of software platform was developed to solve this problem by simplifying communications for users with limited programming knowledge. These platforms adopt a user friendly code builder or code editor, which reduces the amount of unfamilia r syntax the engineers have to address. The Solibri Model Checker and bimServices belong to this type. BIMserver was also developing a generic sequencefinder to facilitate the code programming process. An attempt was made to access the code builder of bim Services, but they could not provide any information. The third type of the platform adopted the natural language interface, which means that the user could enter the constraints in plain conversational language directly into the system. The constraint par sing process is much more difficult to fulfill. Currently, only a small portion of the rules can be successfully transferred in this way. No software of this kind was used as the platform for a D4S tool in this research study. Figure 3 4 shows these three types of platforms and their subtypes. The classification of these subtypes is based on the schema of the exported building model and the programming language used to query these data. The building model actually can be treated as a database. It stores all the information of a 3D building. A model checking software uses model query language to retrieve certain information in which the user is interested. The information of a building model can be exported into many different kinds of formats. Two kinds of database were used in this study. The first one is a relational database such as Microsoft ACCESS and Microsoft Excel. Structured Query Language (SQL) was used to query this kind of database. The second one is in IFC/IFCXML schema.

PAGE 80

80 Figure 3 4. Querying B IM model/database with different programming languages and derived platforms/servers Data Saved in a Relational Database The first software considered as a potential platform was M S Access, because it is very popular software and is easy to use for most of engineers. The Autodesk Revit series provides several exporting mechanisms, and the relational database is one of them. Other formats include IFC, Relational Database, DWG, etc. Some of them were found not suitable for storing geometric information, which means that they cannot be used for a querying task. For instance, the DWG and gbXML are two document formats that do not function very well with the semantic query.

PAGE 81

81 To export a building model from the Revit for model checking purpose, the first viable opt ion was exporting the information into a relational database, such as M S Access, by using Microsoft Open Database Connectivity (ODBC). The relational database management system (RDBMS) is a database management system based on relational models. The data ar e stored in the form of tables and the relationships among the data are also stored in the form of tables. It is the most basic and popular database system in the software industry. Many software applications, either commercial or open source, are develope d based on the relational database model. In this research study, a 3 D building model was prepared and exported into an MS Access document to examine the validity of information stored in such a database. Figure 35 shows that the exported file from Revit Architecture contains a set of tables. Many basic building components (as well as their properties) of a typical building model a re included in these tables. Figure 3 5. Building features in a relational database In a single table labeled Windows, the information for all windows and their ID, Level and SillHeight were listed. The bottom elevation of a wall opening can be identified by checking the height of its window sill. Before glass is put into a window opening, this opening presents a serious haza rd on a construction site, especially when window sill is less than 39 inches in height. Window sills at this height will act as guardrails during construction. Then the Structured Query Language (SQL) was adopted to query this database. SQL is the most ex tensively used programming language to manipulate the MS Access database.

PAGE 82

82 The second method adopted to export attributes from a building model to a relational database was using software tools such as IFC File Analyzer (IFA). The IFA used IFCsvr to parse an IFC file and stored all the information in a M S E xcel document. The IFA generates a spreadsheet from an IFC file. A worksheet in the spreadsheet i s created for each type of IFC entity in the file. Every row in the worksheet contain s the attributes of a n IFC entity. Figure 36 shows a summary worksheet created by the IFA software tool. However, this method was not adopted in this study because of the severe information loss during the format transfer process. Figure 3 6. Worksheet created by the IFA D ata in IFC /Ifcxml Schema The second option i s to export a building model into IFC format. The IFC specification was defined by Standard for the Exchange of Product model data (STEP) technologies and was published as an open ISO 10303 standard. Compared to the other exporting schema, the IFC file contain s more spatial relationship information; however, it also ha s the most complex schema.

PAGE 83

83 Properties a re often not directly attached to building components, but are related through long indirect identification r eferences (IDRef). Consider checking the height of a window sill as an example. To find the sill_height of the subject window, the relationship between this window and the attached wall must be clarified. The IDRef of four related entities needs to be trac ed to link this IfcWindow to an IfcWall. Figure 37 shows the linkage between the IfcWindow to IfcWall with the HelloWall example, which was developed by buildingSMART for tutoring purposes The HelloWall example can be found at the website of buildingSMAR T at http://buildingsmarttech.org/implementation/getstarted/hello world/example 1. To clarify the connections between different IFC entities, a software tool IfcQuickBrowser was used to display IFC files in a tree structure. The function of this software tool was that, when selecting an item in the top window, all entities referring to this item (inverse references) were displayed in the bottom window. The path in Figure 3 7 can be simplified into IfcWindow IfcOpeningElement had an IfcRelVoidsElement relationship to the wall, indicating that the opening was subtracted from the wall. The IfcWindow had an IfcRelFillsElement relationship to the opening, indicating that the opening was to be filled with the window. Even through IdRef, not all entities properties and special relationships can be directly found in an IFC file. There were a la rge number of properties and relationships that were not explicitly attached to entities. Th is information must be derived by analyzing other explicit IFC

PAGE 84

84 relationships. Figure 3 7. Linking the IfcWindow to IfcWall in HelloWall example F or example, after linking the IfcWindow to IfcWall, it was found that the sill height does not exist in some IFC files or is not supported by the dictionary of certain model checking software. To get the value of this property, the term internal_sill_height needed to be calculated by using the following equation: get_sillheight(aWindow) get_floorheight(get_floor(aWindow)).

PAGE 85

85 To compile any D4S best practice into a computable rule, the first step was analyzing how objects were linked with different attribute s and relationships. A solid understanding of the IFC schema and its hierarchy was necessary to do this work and to successfully extract information from building models. This task was often extremely difficult and time consuming because of the large size and complex schema of IFC. The complexity of the EXPRESS language and IFC schema makes deriving component properties directly from an IFC file a formidable task even for many programmers. Therefore some software developers turned to the more popular XML sc hema and developed software prototypes based on ifcXML. The ifcXML schema is a derivation from the Express schema which is richer. It is the corresponding Extensible Markup Language (XML) schema of the IFC schema. Several software platforms based on differ ent types of XML parsers were examined to validate their function for code checking. BimServices, one of the platforms used in this research study belongs to this type. It was originally developed by AEC3 for the United States Army Corps of Engineers ERDC. It was a suite of command line utilities using the TNO IfcEngine, which was also adopted as the core of BIMserver. Data in Other XML Schema Because of the implicitness of many spatial relationships, the original ifcXML file is still too complicated to que ry even for the most basic building components. For instance, it needs to relate five elements through IDRefs to identify a single opening relationship between a door and a wall, while four elements to attach simple properties such as a dimension or is ex ternal to a wall. This is very similar to the IFC schema because they are closely related to each other. To solve this problem, the ifcXML file can be further simplified by transferring it to other XML related formats. Like transferring information from I FC to ifcXML, some geometric and topological information is lost during the transfer of ifcXML into another XML format. The loss

PAGE 86

86 of data when they are transferred between different schemas is Information Lost, which is illustrated by Figure 3 8. Sometime s this information loss is acceptable, because the purpose and the targeted query objects decide whether the transfer processes are suitable or not. If all the lost information is redundant and not related to the subsequent compliance checking work, this t ransfer is tolerable since all the useful information is kept. On the other hand, the data transfer is unacceptable if any information of interest is lost during the transfer process. Figure 3 8. Information lost during data transfer process From the ab ove figure, it is obvious that IFC schema provides the most information that is necessary to derive construction features. It includes all objects and most of their properties, relationships, and location information. Sometime ifcXML is used to substitute IFC in some circumstances because of the complexity of STEP technology and EXPRESS language which define the IFC schema. The mapping to the XML schema definition necessarily loses some constraints including rules, inverse relationships and derived attribut es, so certain special relationships cannot be queried during the following period. Transferring ifcXML into other XML schema usually causes further loss of information. The software platforms in this kind

PAGE 87

87 were closely examined and found that they were not suitable for developing the D4S software tool because of the severe information loss. Structuring the Rule Source In the previous section, three kinds of software platforms that can be used as model checking software were discussed. After selecting the appropriate software tools as the hosting platforms, the design for construction safety suggestions were compiled into computable rules based on the requirement of these platforms. To create computable rules, either spatial based query or semantic based que ry can be used. While Borrmann and Rank proposed the spatial query in 2010, most of the computable rules were based on the semantic query method. The spatial query concerns the topological and geometrical properties of building models and comprise of metric, directional and topological operators. These properties or relationships are explicitly available in the building models, so the 3D model must be analyzed to acquire the necessary data. The semantic query, however, uses information predefined in the bui lding models, which includes properties of BIM entities and/or relationships between them. Semantic ambiguity was a common problem for the semantic query. The different terms may relate to the same entities. On the other hand, several entities may share th e same terminology. The usage of dictionary, taxonomy, and ontology significantly improve s the situation. A Dictionary define s vocabularies, terms and definitions. The Dictionary can make sure that the property is always assigned the same meaning and uni t of measurement. It i s also helpful to avoid uncontrolled expansion of the concepts. There were two options to establish a dictionary. One was to constrain the terms used from a dictionary based on the buildingSMART IFD (ISO 120063) efforts. Another opti on i s to constrain the terms used from a dictionary based on RDF/OWL ontology. The IFD library is an open library. The terms and concepts are defined, semantically described and given a unique identification number (GUID).The IFD works with

PAGE 88

88 Industry Founda tion Classes (IFC) and makes information exchange possible. The IFC constrains the format for information exchange, and IFD defines a standardized understanding of commonly used information. The relationship between IFD and IFC is shown in Figure 39 toget her with a third component, IDM, which is used to form a completely interoperable information system. The IDM is a specification of which information to exchange and when to exchange th at information. It works with IFC and IFD to fulfill the information ex change. Figure 3 9. Interoperability through ISO standards Among the software platforms selected for developing the D4S software tool, the dictionaries of BIMserver and BimServices were easily accessible to users or software developers because the BIMserver is an open source software tool. BimServices also allowed for the addition of new terminologies into its dictionary in support of this research. It was found that the dictionary of BIMserver defines nearly 1,000 IFC terms. The specification of every t erm is saved in a separate java format document. These specifications were originally defined by buildingSMART International. The developer of BIMserver compiled them to fit the usage of a BIM server. When executing the querying function, the server needs to first import all the terms

PAGE 89

89 by using the code import org.bimserver.ifc.emf.Ifc2x3.*. Even though this dictionary was already very comprehensive, the users were still allowed to modify or even add new specifications into the dictionary based on their ne eds. Creation of Rule Sets The IFC schema and IFD library provide the foundations for exchanging and sharing information between different software applications. In this stage, the design for construction safety suggestions were compiled into computable rules based on the four platforms selected in the previous section. Rule Sets Based on Relational Database To manipulate and extract the data stored in a relational database such as M S Access 2007, the Structured Query Language (SQL) was used. For example, c onsider the D4S suggestion Design window sills to be 42 inches minimum above the floor level. The noncompliances can be found through the following steps: Step 1 Exported a building model into M S Access 2007 by using the Microsoft Open Database Connect ivity (ODBC) option which was provided by Revit architecture software. The exported data include project parameters that have been assigned to different element categories in the original building models. For each element category, Revit architecture expor ts a database table for model types and another table for model instances with assigned values. For example, Figure 3 10 shows a table listing all window types and there are 17 types of windows indicated in the column TypeMark. The first column Id list s the primary keys of all 17 types of windows. Another table listing all window instances is shown in Figure 31 3. Note that ODBC exports presents metric units only. The Revit architecture adds relationships to the data tables using primary keys and refe rence values. A primary key in each element table is the column labeled Id. The number is the unique identifier for each instance or occurrence of a window.

PAGE 90

90 Figure 3 10. Table listing window types Step 2. In the Floors view, the Floors were listed based on Level values. In this case, floor 20877 had the lowest elevation level value of 30, so it was recognized as being the lowest or the first floor. This can be verified by the levels table which is shown in Figure 311. There are five levels of the subject building model. The first level has Id value 30. After identifying the first floor, the floor numbers were assigned to all other floors based on their level value as shown in Figure 312. Figure 3 11. Table listing levels Figure 3 12. Sequencing floor levels based on Level value in Floors spreadsheet

PAGE 91

91 Step 3. In the Windows view, all Windows which were located above the first floor a re isolated. In this case, windows with level value over 30 were screened out. Figure 3 13. All windows above the first floor in Windows Spreadsheet Step 4. In the Windows view, comparing windows SillHeight to the minimum requirement, 42 inches or 1.0668 meter. Figure 3 14. Filtering windows by SillHeight Step 5. In the Windows view, windows IDs of all noncompliant windows and their floor numbers were returned as checking results. The retrieved windows IDs were sequenced by floor level as shown in Figure 315. Figure 3 15. The returned result of noncompliance Rule Sets Based on BIM server An IFC model server is a database management system that centrally stores the building information model and manages all access to it. Various parties involved in a construction project can share information through it. The URL bimserver.org is link ed to an open source IFC model server project and plays a role as an information hub that allows users to merge, filter, query, or even conduct clash detection. From the collected falls prevention best practices, the following provision was selected to loc ate the method of how to compile a computable rule based on BIMserver: Locate the

PAGE 92

92 floor/roof opening which area value is larger than 4 square feet. Openings in structures on construction sites are a major causation for falls from elevation. During constr uction, elevator shafts, skylights and stairs can all appear as openings in the floor slab. Windows and doors are also closely related to openings in that they are normally inserted in to an IfcOpeningElement using the IfcRelFillsElement relationship. The opening element represents a void within any element that has a physical manifestation. It stands for opening, recess or chase, and can be inserted into walls, slabs, beams, columns, or other elements. The IFC specification provides two entities for opening elements. IfcOpeningStandardCase is used for all openings that have a constant profile along a linear extrusion. Another entity, IfcOpeningElement, is used for all other occurrences of openings and in particular also for niches or recesses. The second en tity, IfcOpeningElement, i s used in this research study. In this research study, two methods were proposed to query an IFC file through the Advanced Queries function of BIMserver. The purpose of this direct method was to retrieve the area value using the semantic information already stored in the model. The indirect method searches for the geometric parameters of the opening elements, and obtains the area value through further calculations. Direct approach The quantities relating to an IfcOpeningElement a re defined by the IfcElementQuantity and are attached to the IfcRelDefinesByProperties One quantity defined by the IFC specification is the NominalArea, area of an opening as viewed by an elevation view for wall openings or as viewed by a ground floor vie w for floor openings. First, the NominalArea i s used to query the area value of an IfcOpeningElement. The hierarchy graph shown in Figure 316 shows the relationship between an IfcOpeningElement and its area value. Figure 316 shows that the

PAGE 93

93 IfcSlab and If cOpeningElement are two separate entities. These two entities have to be first linked to locate each IfcOpeningElement in the floor slab. Then the relationship between IfcOpeningElement and its area value is found. This was one of the most difficult steps because the IfcQuantityArea i s separately stored from the IfcOpeningElement. At this point the specifications of all the IFC 23 terms defined by the BIMserver dictionary become critical resources to understand and establish these relationships. Figure 316. IFC hierarchy for opening a rea v alue r etrieval After validating the linkage between an IfcOpeningElement and its area value, the following approach was proposed to query the area value of openings : Step 1. Load the building stories Step 2. For ev ery storey, get the IfcProduct and check the instance as IfcSlab.

PAGE 94

94 Step 3. Using RelVoidsElement, get the related IfcFeatureElementSubtraction for the IfcSlabs Step 4. Use the information from step 3 and the IfcRelDefines to obtain the IfcPropertySetDefint ion collection. Step 5. Check for the IfcElementQuantity instance in the collection from S tep 4. Step 6. Using the IfcElementQuantity in step 5, get the IfcPhysicalQuantity Step 7. Typecast the IfcPhysicalQuantity to IfcQuanityArea to get the required Area Value for consideratio n. This approach directly queries the area value of an IfcOpeningElement, which means that the quantities are not obtained by calculating the geometric information of the opening element. In direct approach Besides directly accessing the area value by using IfcQuantityArea, the area value of an IfcOpeningElement can be calculated by using its dimension parameters. The second approach was proposed to indirectly get the area value of an IfcOpeningElement. As shown in Figure 317, the If cRectangleProfileDef contains the YDim and XDim of a rectangular opening. The IfcRectangleProfileDef.YDim is the opening width, and the IfcRectangleProfileDef.XDim is the opening height. IfcAxis2Placement3D is used to locate and originate an object in three dimensional space and to define a placement coordinate system. Figure 3 17. Dimension parameters of an IfcOpeningElement The IFC R2.0 Object Diagram developed by the Building Lifecycle Interoperable Software Project (BLIS Project) was a helpful tool a t the beginning stage of this process It shows that the

PAGE 95

95 area value can be obtained by using the geometric dimension of an entity. With the progress of the project, the IFC R2.0 Object Diagram IfcOpeningElement was found to not represent the correct relationships between an IfcOpeningElement and its IfcRectangleProfileDef because the IFC schema had been updated from the old IFC R2.0 to the new IFC 23. A new diagram based on IFC 23was then developed to replace the old one. Figure 318 shows a partial hie rarchy and attributes of an opening element. The full object diagram was more complicated, so only relationships that were useful for querying purposes are demonstrated here. The geometric relationships between an opening element and its dimension paramete rs can be derived from this diagram. The definitions of all these terms were confirmed by documentations of International Alliance for Interoperability (IAI), which can be obtained from the buildingSMART alliance website http://www.buildingsmarttech.org/ifc/IFC2x3. In Figure 3 18, the IfcLocalPlacement defines the local coordinate system, which is referenced by all geometric representations. The relative placement of an opening element to the IfcWa ll and IfcSlab is recorded through it. The IfcProductDefinitionShape defines all shape relevant information about an IfcProduct. It allows for multiple geometric shape representations of the same product. The IfcShapeRepresentation represents the concept o f a particular geometric representation of a product or a product component within a specific geometric representation context. It has an inherited attribute RepresentationType to define the geometric model used for the shape representation. The swept area solid is a predefined type of RepresentationType. It can be created through either an extrusion or a revolution. The IfcExtrudedAreaSolid is defined by sweeping a bounded planar surface. It defines the extrusion of a 2D area by using the direction and depth. A 2D area is given by a profile definition. The

PAGE 96

96 opening element extrusion segments may have any profile. Three profiles are supported by IFC 2. They are IfcRectangleProfileDef, IfcCircleProfileDef and IfcArbitraryClosedProfileDef. Figure 3 18. IFC 2 object diagram IfcOpeningElemen t The IFC rectangle profile is the most common profile of an IfcOpeningElement. It defines a rectangle as the profile definition used by the swept surface geometry or the swept area solid. In IFC 2, rectangles are de fined centric, which means the placement location has to be set with IfcCartesianPoint (XDim/2, YDim/2) as shown in Figure 319.

PAGE 97

97 Figure 3 19. Parameters of rectangle profile definition The indirect approach can be realized through the following steps: S tep 1. Load the building stories. Step 2. For every storey, get the IfcProduct and check the instance as IfcSlab. Step 3. If the instance is an IfcSlab, get its generalized parent object IfcElement. Step 4. Using IfcElement, check to determine if the objec t has any opening elements using getHasOpenings() which will be a relationship IfcRelVoidsElement. Step 5. Collect the IfcOpeningElement into a set from the relationship class IfcRelVoidsElement using getRelatedOpeningElement(). Step 6. The IfcOpeningEleme nts searched for are rectangular opening elements along with their respective area values. Step 7. Iterate the collection of IfcOpeningElements, to get the X Dimension and Y Dimension values to calculate the area as follows.

PAGE 98

98 Step 8. The o bjective is to find the path: IfcOpeningElement IfcRectangleProfileDef for the methods getXDim() & getYDim(). Step 9. Get the IfcRepresentation and iterate over them to find the IfcShapeRepresentation which will return the items of IfcExtrudedAre aSolid. Step 10. IfcExtrudedAreaSolid has the profile definition among which IfcRectangleProfileDef is defined. Step 11. Check for the instance of IfcRectangleProfileDef to invoke the getXDim() and getYDim(). Step 12. Calculate the area as a product of the return values of the above dimension methods. Step 13. F iltering the IfcOpeningElement using the search area criteria. Rule Sets Based on Solibri Model Checker A large number of default building codes have already been included in the Ruleset Folders and Ruleset Library as part of Solibri Model Checker (SMC). The users can adjust rule parameters to make these preset rules fit their needs. These rule sets are compiled in java using the SMC application programming interface (API), which is not publicly avai lable. With the Ruleset Manager, the user still can create and edit some new rules. For example, the computable rule which checks the window sill height was compiled with General Purpose Property Rule template. To compile this rule, the user interface wa s switched to the Ruleset Manager, which is shown in the Figure 3 20 The Ruleset Manager has five Views. They are Rule Set Folders, Libraries, Info, Workspace, and Parameters. The Ruleset Folders View on the left side of the window shows all the available rule sets saved on the hard disk. These rules sets are kept in several Ruleset Folders based on their disciplines. The user can upload any of these rule sets and conduct corresponding rule checks. On the right side of the window, the Libraries View show s all basic computable rule templates which can be edited based on the users need.

PAGE 99

99 Figure 3 20. The Rule Set Manager i nterface for browsing and editing rules In the Libraries View, Figure 3 21, the General Purpose Property Rule was selected. The description of this rule can be found in the Info View. This rule was designed to check any combination of properties of a given component type. Then the users can add this template into the Workspace View and configure the rule parameters in the Parameter View. In the Parameter View, which is shown in the Figure 323, one discipline was selected to make sure that only components in this specified discipline will be checked. In the Property Value Rules table, the Window was selected as the component type to b e checked. The property to be checked was the Bottom Elevation. Usually there are several numerical properties that can be checked, which depend on the selected component type. The operator selected here was At least, and the value was set to 42 inches By using the Type column, certain construction types can be selected to limit the model checking. For instance, when

PAGE 100

100 checking the openings size in a building model, the user may only want to check the openings located on the building roof, so the roof can be set in the type column to limit the checking scope. After configuring Property Value Rules, the user can impress the output issues with the Categorization Order table. For the Space Checking Options, Check Only Spaces was used. Finally, the user can save the configured template as a new rule set in the Designing for Construction Worker Safety folder. Figure 3 21. The Libraries View Figure 3 22. Workspace View

PAGE 101

101 Figure 3 23. Parameter View Rule Sets Based on Bimservices To use bimServices as the model checking software, the development of computable rules is characterized by soft coding which follows a pre defined mark up methodology (Nisbet, 2010b). The software programming for this can be done automatically or semi automatically based on predefined procedures. Figure 3 24 shows a general coding process based on BimServices. The paper based Normative document can be encoded into IFC/ifcXML format computable rule through a set of transformations. The first step i s to transfer original D4S suggestions into computer readable Baseline Electronic Suggestions which a re in HXML format. Then, the logical relationships between different terms and term properties a re tagged by marking them with different colors, which a re corresponding to the SM ARTcode

PAGE 102

102 Figure 3 24. Encoding D4S suggestions into safety Constraint Model protocol. Then the markedup text is edited into structured HTXML either manually or automatically by using a code editor/builder. Finally, an XSLT format configuration is used t o convert marked up XHTML into single D4S computable rule. The Transform1 function of bimServices use s the output setting within the XSLT file to anticipate the general output format required. Rather than manually tagging and marking up a baseline electr onic suggestion, using a code editor is more time and cost effective. Furthermore, this enables a person skilled in the AEC

PAGE 103

103 domain to develop applicable rules without the support of programmers. The XSLT document is still under development at th is point in time As mentioned above, a color system was adopted to tag baseline electronic suggestions to make the four logistic concepts apparent. Table 3 1 shows these four different markup colors and their meanings. Table 3 1. Four different m ark up c olors Colo r Meaning Function Purple Selection What different situations does this apply to? Green Applicability What is the scope of the check? Blue Requirement What is required? Orange Exceptions What exceptions are there? The checking of the window sill heig ht was used as an example to show how a semantic concept can be implemented into a computable rule and be applicable to model checking. The logistic relationships were: Selection: Topic=window Applicability: Topic=window Property=isexternal Requirement: To pic=window Property=Internal_Sill_height Comparator=greater than Value=42 Unit=inch The Internal_Sill_height was calculated by using the equation: get_sillheight(aWindow) get_floorheight(get_floor(aWindow)). It should also be noticed that there was one Exception in this provision. The windows located on the first floor were exempted from checking because the differences between the altitude levels of window sills and exterior earth (ground) are normally within a safe range. This discussion is an assumption because computable rules based on bimServices have not been developed. The topographic information may not be accurately portrayed in a BIM model so a manual check is advised.

PAGE 104

104 The Figure 3 25 shows the computable rule encoding process. At the top of the diagram is the D4S suggestion in a normative document format. After coloring, marking up and mapping, the normative document can be transferred into a computer readable rule. Figure 3 25. Sample r ule of bimServices Building Model Preparation Beside s compiling the computable rules, appropriate building models needed to be prepared to test the effect of the tentative rules. With the rapid adoption of BIM technology in the AEC industry, it was not difficult to find example test cases to explore the inf ormation typically captured in IFC databases. For example, the IAI provided several links to various examples at http://www.ifcwiki.org/index.php/Examples The object properties and other information of interest might be not included in a random building model. In this case, the model checking software cannot conduct the check and return a value or term such as UNKNOWN to

PAGE 105

105 let the user know that certain information does not exist in the subject model. This causes an uncheckable situation. Therefore, certain rules need to be followed to prepare the test models. The basic requirement of testing building model was that the model should look like a normal building, and all targeted information must be included in it. For this research study, the targeted information was the window sill height and the size of the floor openings. The existence of the information can be verified by using an IFC viewer such as IfcStoreyView, which was developed by the Kar lsruhe Institute for Technology. This software tool can be used to verify the attribute value s of most of the building components. Figure 326 shows the properties of an IFC opening element reviewed by using IftStoreyView. It can be easily found from the f igure that this opening is located on the third floor and that t he area value of this opening element is 78.5 square f eet. Figure 3 26. Reviewing IFC opening e lement properties in IfcStoreyView

PAGE 106

106 The purpose of this research was to identify and assess pot ential fall hazards. Two specific D4S provisions were the focus of this research. The first provision is Design window sills to be 42 inches minimum above the floor level. While a 39inch window sill height would satisfy the OSHA requirement of fall prot ection, a more conservative value of 42 inches was selected as a safer approach. The object of compliance checking is the height of the window sill. The second provision is Locate the floor/roof opening which has its smallest dimension greater than 6 inch es. The objects of compliance checking are floor and roof openings. A sample model 4351.ifc was developed by using the Revit A rchitecture BIM authoring tool Building model 4351 was a three story building There were three windows and four floor openings i n this model. This can be verified by using an IFC viewer. Figure 327 is the front elevation of this building model. From this graph, the three windows that are located on different stories in the building can be easily located. Figure 3 27. Three wind ows in building model 4351.ifc

PAGE 107

107 Figure 3 28 shows all the IfcOpeningElements in the model. The two large floor openings were created by stairs. The two smaller ones were created by an elevator shaft which penetrates the level 2 and level 3 floor slabs. F igure 3 28. Opening elements in building model 4351.ifc

PAGE 108

108 CHAPTER 4 RESULTS Two major results are discussed in this chapter. The first one is the positive impact that BIM technology can potentially have on construction site safety. This consists of the changes that will be realized by constructors that adopt BIM applications. New construction methods and new applications could help constructors reduce injury and fatality rates by increasing the opportunity for prefabrication, conducting clash detection, and providing virtual reality to let construction workers become more familiar with the project. The second result is a D4S software tool which could help designers implement the design for construction safety concept in practice. This software tool was deve loped by using BIM technology to set up a design for safety database and adopting a model checking software which will automatically check developed building models against design for construction safety rule sets. Particular emphasis is placed on fall haz ards since falls are the most frequently occurring causation of fatalities on construction sites. BIM technologys impact on construction safety BIM technology, especially BIM applications, could help constructors reduce injury and fatality rates in the fo llowing work areas: Safety Planning Work sequencing Construction schedule Clash detection Improve the construction documents quality: jointly review design decisions Communicate design intent between trades and design discipline Safety Planning The develo pment of project specific safety plans is important to the safety management for construction projects. The safety plans assist management personnel in establishing a healthy and

PAGE 109

109 safe working environment which benefits all construction workers. The traditi onal safety plans are mainly developed based on 2D drawings. With the prevalence of BIM technology, 3D and 4D safety planning have begun to emerge. The 3D safety plan s use virtual reality applications and tools to help project participants get familiar with the construction site environment. 4D safety also becomes a necessary method because poor planning and scheduling usually cause s chaos on construction sites. A well arranged construction schedule can positively influence construction site safety. If the schedule of a project gets improved, the safety performance will be improved. This can be realized by identifying risks and devising better planning control. When project participants also consider schedule issues when they plan 3D safety, the 3 D saf ety planning evolves into the 4D safety planning. 3D/Virtual Reality Designs are becoming more and more complex, and few people on construction sites can readily read and understand the complex drawings. Many construction requirements rely on personto pe rson instruction. During this process, errors occur when reading drawings, interpreting them and passing on the information. By using BIM applications, constructors could pre build the project in model form far in advance of actual construction. Constructi on workers could become familiar with the proposed building structures by viewing 3D walkthroughs which simulate the view from the perspective of a human who is walking through the facility. With the development of the virtual reality (VR) technology in re cent years, VR laboratoties have been developed which can be used as VR safety training systems. Figure 4 1 is a photograph of a BIM computer lab at the Center for Advanced Construction Information Modeling at the University of Florida. These advanced 3 D virtual reality devices were set up for tutorial and training purposes. Students and AEC professionals can learn or assess the building structure and

PAGE 110

110 internal mechanical, plumbing, electrical and fire protection (MEP/FP) systems in 3 D view through both of the major projectors and seperately estalished workstations. Figure 4 1 Photograph of BIM computer lab at Center for Advanced Construction Information Modeling at the University of Florida (Photo courtesy of Jia Qi) Schedule/4D The project schedule u ndoubtedly impacts construction site safety. For instance, construction workers tend to neglect surrounding hazards when an expedited schedule is associated with their tasks, which compromises safety performance. In contrast, a well organized project reduces the pressure on construction workers and correspondingly diminishes the occurrence of accidents. Past research studies have discussed the relationship between the work schedule and construction site safety. It was found that safe projects and success in scheduling are jointly achievable (Hinze 2006; Yi and Langford 2006). BIM applications enable "what if" scenarios to be examined by visual risk management and time based workflow planning when

PAGE 111

111 linked with clash detection. The what if analysis is a structured brainstorming method which is usually used to uncover hidden hazards. 4D models link components in 3D CAD models with construction schedules. The 4D model allows project participants to view the planned construction of a facility over time and review the status of a project in the context of a 3D CAD model. In practice, constructors use 4D models to study different design and schedule alternatives and educate workers about what would be happening during each stage of construction. Although the coordination between scheduling and safety planning has been identified as an important factor to a successful project, this question has not been fully discussed under a 4 D environment. BIM technology is being substituted for 2D CAD technology With the rapid progress of BIM technology, it is necessary to propose the concept of four dimensional safety (4 D Safety). 4 D Safety is the technology that uses 3 D software to detect the location of construction site hazards and meanwhile uses scheduling software to identify corresponding high risk time periods. There are various construction hazards on construction sites during different time periods. Since the construction site conditions vary according to the progress of the building, the construction hazards that ex ist during the construction process of a building structure will disappear with the completion of that building structure. The corresponding safety measures must be applied for the appropriate conditions and at the right timing. This is another difference between design for construction worker safety suggestions and building codes. B uilding codes are used to protect the public and end users who will take occupancy of the completed building. M odel checking software (MCS) developed for code checking just needs to check the as build model to ensure that the building fulfills all the code requirements. Design for construction worker safety protects construction workers during the entire project life cycle.

PAGE 112

112 M odel checking software developed for this purpose shoul d be able to check the safety conditions of a building information model at any point in time during the construction phase. By coordinating the safety planning with scheduling in a 4D environment, the project participants could locate potential construct ion site hazards in advance and then take reasonable measures to reduce or eliminate the hazards. Conducting C lash D etection Construction managers can check distances between different building components and arrange workflow planning for time and space co ordination by using applications such as NavisWorks. On a construction site, two kinds of conflicts usually exist that can cause accidents: the static ones and the dynamic ones. The static ones consist of conflicts between in place structures and other fea tures such as idle equipment. Construction planning should be done during the design phase to identify and to address the two kinds of undesirable conflicts. This requires planning for sufficient space to allow construction workers to build and maintain the building components and equipment. For example, in a boiler room, the circulator pumps are generally designed high up near the ceiling. A worker would have to stand on a ladder to replace the pumps, and would have to work between various piping systems t o accomplish the tasks. In this situation, maintenance workers could readily fall off the ladder, drop the pump, or suffer some other type of injury when making contact with adjacent piping. A maintenance worker may need to shut off a valve to work safely. Planning is also needed so that the work of different construction specialists can be coordinated when activities must run in parallel in order to meet schedules. Dynamic conflicts consist of the conflicts between different work crews in the field. This kind of conflict is usually costly in that both the project budget and the project schedule are affected. In the past, the coordination ability of project managers was crucial to solving this kind

PAGE 113

113 of problem. The project with a competent project manager and sound management system usually could reduce the occurrence of conflicts in the field and make projects more profitable. Better construction site planning can be arranged by using BIM applications. Past research validates this by showing that 83 percent o f contractors as BIM application users achieve reduced conflicts. Increasing P refabrication Prefabrication could reduce injuries in two ways. First, fewer workers are needed on a construction site, which facilitates construction site management. It is unde rstandable that fewer injuries will occur because more work is being performed in a more controlled environment. Second, less work will be done at high places or at elevation, which reduces the chance of falling injuries. While prefabrication can improve c onstruction site safety, inconsistency, inaccuracy, and uncertainty in design could make it difficult to fabricate materials offsite. When using 2D drawings, architects often choose to include fewer details in their drawings. Many errors in construction documents are produced during the design phase. These errors accumulate and are not found until prefabricated parts are delivered to the construction site. In a manual assembly, if a structural element is a half inch off alignment, it may not be noticed. Com ponents that are added later may not fit and have to be trimmed or shimmed. The prevailing BIM architectural design models will increase the percentage of prefabrication by producing more accurate construction documents. Seventy seven percent of the contra ctors predicted that model driven prefabrication will be the dominant value five years from now in the survey conducted by McGraw Hill Construction ( 2009) Other case studies (Khanzode et al., 2008) also support the above points. The Camino Medical Office Building, for example, is one of the first project studies that quantitatively measured the benefits of using BIM tools for MEP coordination. The labor savings ranged from

PAGE 114

114 20% to 30% for all MEP specialty contractors. In total, 203,448 field work hours wer e spent on the construction project and only one recordable injury occurred (Khanzode et al., 2008). The RIR of the MEP specialty contractors was nearly 1.02, which is far less than the industry average. These achievements are attributed to the improved wo rkflow due to the use of 3D/4D models which resulted in more off site prefabrication and efficient field installations. When reviewing past BIM surveys and case studies, it was noticed that past research studies could be improved. First, the lack of a comprehensive and systematic method of defining goals and choosing among design options impedes the production of design drawings which incorporate adequate safety considerations. Gane (2008) pointed out that architectural criteria (e.g., aesthetics, area eff iciency, site views) prevail over engineering performance criteria (e.g., energy efficiency, structural performance). The same is true of construction safety. Safety issues often are not an important criterion when designers make decisions. The designers also seldom conduct safety reviews after the conceptual design to ensure that construction knowledge is incorporated in the design phase and safety issues are minimized. To change this situation, the concept of making zero accidents a reality should be p ropagandized in the construction industry. Secondly, when calculating the return on investment (ROI) of BIM, the savings related to improved construction safety also should be considered. Furthermore, past research studies usually have overlook ed the savin gs caused by decreased indirect costs of construction accidents, because it is difficult to identify and quantify the hidden costs of injuries. For instance, the ratio of field indirect costs to direct costs is 0.85 for medical case injuries. The reduction of both direct and indirect costs of injuries should be considered when calculating the savings attributed to BIM.

PAGE 115

115 Design for Construction Worker Safety Software Tool In the last section, the application of BIM technology in the AEC domain and its benefit to construction worker safety were discussed. In this section, the results from different model checking software are shown The design process gets improved by using this Life Cycle Safety process. There are several stages in the design process where t he building model can be checked in some way. With the help of a D4S application tool which c an check the level of compliance with specific safety requirements, the design review process c an be further improved and construction site safety c an be enhanced. Figure 4 2 shows the construction project life cycle after the D4S application tool has been adopted. It also demonstrates the data flow throughout the whole project lifecycle. Designer BIM Software Designer Owner IFC Files Inspector Designer ContractorChecking System OSHA CII UFKnowledge Based Expert System Subs Contractor OSHA New Onsite Fatality/Injuries Inspector Designer ContractorReport Phase 1 Phase 3 Phase 2 Phase 4 Phase 5 Phase 6 Phase 7 Figure 4 2. Construction project life cycle safety with a D4S application tool In Phase 1, drawings are generated by various BIM design applications such as Revit and ArchiCAD which support the export of multiple versions of IFC files. Then the IFC format file which consists of geometric information of building components and the relationships between them is abstracted from these design applications. In Phase 2, design for construction worker safety suggestions are collected, classified and compiled into computable rules in a model checking software.

PAGE 116

116 In Phase 3, a D4S application tool is developed. After the computable rules are imported to the checking system, IFC format files c an be imported into the checking system to check for compliance, and designers c an access design for construction safety knowle dge of various building components. In Phase 4, the checking system generates both checking reports and visual graphs. Reports list detected design errors which violate specific rules in the knowledge base. Meanwhile graphs explicitly show building structu res or building areas where design alternatives are available to avoid construction worker injuries. These reports and graphs c an help all project stakeholders by providing them with valuable D4S information. Designers c an acquire useful construction safet y knowledge and revise their drawings before delivering them to the constructors. C ode officials c an evaluate the projects drawings for code compliance, cite deficiencies and request revisions to resolve potential problems. The constructors also c an check their building models before or during construction, and use temporary structures or PPE to diminish their employees exposure to unsafe working environments. In Phase 5, constructors finish construction work after the building models have been checked. T wo potential problems may still cause injuries and fatalities on construction sites. The first problem is the lack of an administrative program which will ensure compliance. The second problem is that construction workers may not adhere to safety practices OSHA will collect information related to new accidents and constructors also will learn considerably from those incidents. In Phase 6, researchers will receive feedback from industry professionals to enrich the design for safety database.

PAGE 117

117 In Phase 7, mor e stakeholders of a project will be concerned about construction worker safety after the industry safety culture has changed. This checking system will be embedded in the BIM software as a plug in, which makes it more accessible to users. The entire proces s is a learning loop. The project participants should enrich the knowledge based expert system with new best practices and innovative techniques as the design is a process in which more information is to be added as more information is gained through proje ct successes and failures. This D4S application tool also helps various stakeholders involved in the work to protect construction workers. The current situation is that the general contractor is usually expected to review shop drawings for safety while the architect and engineer are not responsible for any safety reviews of the shop drawings. To correct design problems before they get frozen into the fabricated product, designers, constructors and official authorities need to be involved in the process. By analyzing a typical project lifecycle, two groups of stakeholders are identified as the main users of a design for construction safety tool. They are the designers and contractors. Owners also could check designs for safety by using this tool. Figure 4 3 s hows when various project participants could address safety issues during different phases of a project lifecycle. The black areas are phases when stakeholders should consider construction safety issues. Designers should address safety from the proposal st age to the construction document stage. Constructors should consider safety in construction and maintain phases. It is beneficial that owners consider safety issues throughout the project lifecycle. The gray area is the period when constructors could ideal ly cooperate with designers and incorporate construction safety knowledge into the building design.

PAGE 118

118 Figure 4 3. Safety effort of different stakeholders in a project lifecycle Checking Results with a Relational Database As discussed in Chapter 3, a building model can be exported and saved in a relational database such as M S Access. The Structured Query Language (SQL) could then be used to query this database. Here the detailed checking process using M S Access 2007 is given. The D4S rule used is Design wi ndow sills to be 42 inches minimum above the floor level. Step 1. Launch MS Access 2007and open the database that will be queried. In this case, model 4321.ifc is used. Step 2. Select Query Design in the Create window to c re a te a new object query Then se lect SQL View in Design View to display the Query window. The user interface is shown in Figure 4 4. Step 3. Type the compiled SQL statement in the Query window. This step is shown in Figure 4 5. Step 4. Click the Run button to execute the SQL statement, a nd the result is shown in Figure 4 6.

PAGE 119

119 Figure 4 4. Model checking in a Microsoft Access Figure 4 5. Import computable rule in the Query window

PAGE 120

120 Figure 4 6. Model checking result in a Microsoft Access Besides querying a relational database directly with SQL and VBA, more complex model servers can be developed based on either a relational database or other kinds of storage methods. These model servers can provide query access to the model using either SOAP (Simple Object Access Protocol), XML, or Express language. Figure 4 7 shows the simplified architecture of a multilayered model server. At the beginning, the IFC files are imported into the database. The database can be a relational database or other database such as Oracle Berkley DB engine which is used in BIMserver.org project. There are different layers in an IFC model server. These layers process data by making a simple query to the model using common language. Kiviniemi (2005) and Beetz (2010) reviewed the model servers currently in use and th e model servers that are under development.

PAGE 121

121 Figure 4 7. The a rchitecture of c urrent m odel s erver and potential e xtension Checking Results with BIMserver In Chapter 3, the computable rule based on BIMserver was written using Java language and the IFC hie rarchy. The computable rules are enclosed in APPENDIX B and APPENDIX C. To use BIMserver as the platform for compliance checking, the user first needs to login into the server and select the project that will be checked. Here the project 4351 is selected t o be under review as shown in Figure 48.

PAGE 122

122 Figure 4 8. Launch BIMserver and select project to be checked After selecting the project which the user wants to check, the BIMserver provides six potential functions which can be executed. The Query function i s used for a compliance check. The server further provides two options: Simple Query and Advanced Query. Here the user should select the Advanced Query option and import the computable rule in the compile window. This user interface is shown in Figure 49. Figure 4 9. User interface of Advance Query function of BIMServer

PAGE 123

123 There are two buttons at the bottom right of the user interface. The Compile function automatically checks the validity of the computable rule. It reports the grammar errors if the imported rule is wrong. The Compile & Run function checks errors as Compile function and executes code checking if the imported rule is well written. Here the computable rule is compiled using the direct approach is first imported and executed. The resul t of the check is shown in Figure 4 10. Figure 4 10. Checking area values of opening elements with BIMserver by using computable rule written in direct approach Figure 4 10 shows that there are a total of four opening elements in project 4351. This matches the number of IFC opening elements in the original building model. The retrieved area values mismatch the true values in the original model. After further validation, it was found that the retrieved values were the net area value of the slabs to which opening elements are attached. The relationship between the net area value, slab area value and opening element area value can be linked using the following equation: = Net area value of slab This relationship is shown in Figure 4 11. Area A is the slab area value, and area B in white color is a slab opening element. The net area value of the slab is marked in orange color. Even though the retrieved values are not the area values of opening elements, this does not mea n the direct approach that uses semantic information in the model is wrong. The direct method is

PAGE 124

124 still a generic method to query an IFC building model when using BIMserver. It may not work in a few specific circumstances, such as the unusual storage of cer tain semantic information. Figure 4 11. The relationship between net area value and opening element area value The validity of a computable rule written in the indirect approach is checked. The d irect method retrieves the area value using the semantic i nformation already stored in the model. The indirect method tried to find geometric parameters of opening elements, and obtained the area value through further calculations. The rule is imported into the query window in the same way as the rule is compiled in the direct approach. After running the check, the result is shown in Figure 4 12. The X dimensions, Y dimensions, and area values of all three opening elements in the rectangular shape are listed. These values match the area value in the origional buil ding model. Figure 4 12. Checking area values of opening elements with BIMserver by using computable rule written in direct approach

PAGE 125

125 Checking Results with Solibri Model Checker The building model is exported into an IFC file after being created in a BIM authoring software such as Autodesk architecture. The model is opened in the Solibri Model Checker (SMC), ready to be explored and checked. The model checking layout of SMC consists of four functional subwindows. Necessary operations to conduct the model checking are realized through them. They are: Rule sets. When the developed computable rules are loaded into the model checking software, they are called rules sets in SMC. Users could select rules that will be used to check the model. Results View. Non compliance will be shown in this window. 3Dimensional virtual reality model browser. Info View. This description window shows the detailed explanations of the selected safety suggestion. Figure 4 13 shows functional and informational view s which compose t he user interface. Figure 4 13. Checking layout of Solibri Model Checker

PAGE 126

126 Rule execution/Rule check reporting Three major steps Navigation, Rule execution, and Rule check reporting -are needed when using this system to check for noncompliance. The Navigation function can be activated by selecting the Model Tree tab, which can be expanded or collapsed to show the components of the subject building. Users can get familiar with the building model by exploring it. This progress is shown in Figure 414. Figure 4 14. Using the model tree navigates the subject building model After the users are acquainted with the structure of the subject building, they can load the design for construction safety rule sets to start checking the building model. The avai lable rule sets/constraint model is listed at the left side of the interface. Figure 4 15 shows the interface after adding the rule sets.

PAGE 127

127 Figure 4 15. Selecting desired constraint model / rule set After loading the desired rule sets/constraint model, th e user can initiate the checking process by clicking on the Check icon. All noncompliances are reported as checking results. The checking results will be displayed in the results view, showing whether a specific safety rule has been satisfied or not. The user could click a single checking result to locate the problem in the right side 3D view. Meanwhile, the detailed explanation of violated rules can be seen in the description window. Figure 416 shows the user interface after running the checking function. From the picture there are two windows for which the sill height is lower than the pre set value.

PAGE 128

128 Figure 4 16. Checking r esults The user can select any checking result in the Result View for a further review as shown in Figure 4 17. A detailed descrip tion of the specific violated rule will be given in the Info View. Figure 4 17. Description of noncompliance in the Info View

PAGE 129

129 The Info View can show the user more comprehensive information of the problematic entity as shown in Figure 418. Here the Sil l Height of the window is 3 feet, and the window is located on the second floor. Other PSet_Revit values also can be found from the Info View. Figure 4 18. The PSet_Revit properties in Info View Sometimes even though an object violates the design for sa fety rules, the user might consider whether some action is needed. It is possible that the violation does not pose a serious hazard on the construction site, or that other more important rules overpower this one, so the user might elect to overlook the noncompliance. The Checking Decision setting in the Result Details view makes this function possible, which is shown in Figure 419. For every identified noncompliance, the user can decide whether to accept or reject the checking result. If the user sel ects the Accepted option, the model checking tool will neglect this violation as shown in Figure 4 20. If the Rejected option is selected, this violation will be ear marked for future reference.

PAGE 130

130 Figure 4 19. Result Details view Figure 4 20. Negle cting the noncompliance after selecting the Accepted Option

PAGE 131

131 A noncompliance report can be created by exporting checking results in PDF, XLS or XML format. This function makes it is much easier for users to save the checking result for future usage or s hare it with other project participants. Figure 4 21 shows the Create Report setting of the software tool. The report content and report type can be set through this dialog box. Figure 4 21. Create report After finding the noncompliance elements in t he subject building model, the project participants need to decide the appropriate actions, if any, to be taken to address the problems. The designers could change the building model to make it safer. For example, F igure 4 20, the safety checking system reports that a window does not conform to the loaded Rule Set. The Info View also gives a D4S suggestion: Design window sills to be 42 inches minimum above the floor level. The designer could go back to the BIM authoring tool to correct the height of the w indow sill. In Figure 422 it is evident that, in the authoring tool, the current height of window sill is 12 inches, which does not meet the safety requirement. The designer should make a change to the building model to satisfy the minimum requirement.

PAGE 132

132 Th e constructor also could take steps on the construction site to eliminate fall hazards. After finding that the opening in the exterior wall is a potential hazard on the construction site, the constructor can either negotiate with architects to change the p arameter (window sill height) or install temporary fall protection. Figure 4 22. Instance properties in BIM authoring tool After changing the sill height of the window on the second floor in the building design tool, the user can run the model checking software to check the building model for validation. Figure 4 23 shows the window on the first floor still does not meet the D4S requirement. This might be acceptable if the elevation difference between the interior window sill and the exterior topography does not seriously violate the requirement.

PAGE 133

133 Figure 4 23. N ew result of the check after changing the dimension of certain objects The same procedure can be used to find all floor openings for which the area values violate the minimum rule of being less than 4 square feet. For model 5351.ifc, it is found that all four floor openings do not satisfy the requirement. This result of the D4S check is shown in Figure 424. In the result view, four noncompliant openings and their area values are given. In the na vigation view, these noncompliances are clearly marked with red color. Other structural parts in the same building are shaded to half transparent, so that the user can better recognize the relative positions. The user can also select any non compliance in the result view, and a detailed description is then given in the bottom info view. Meanwhile, the 3D view zooms in to the selected non compliance, so the user can have a closer review as shown in Figure 425.

PAGE 134

134 Figure 4 24. Result of checking for slab o penings Figure 4 25. Designing alternative given in Info V iew

PAGE 135

135 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS Conclusions As designs increase in complexity, it is expected that construction worker safety will become more and more important during the design phase. Through the review of the literature, it is evident that Building Information Modeling (BIM) as a tool can be used to address construction safety. By adopting this new computer technology, safety performance is expected to improve by increasing the opportunity for prefabrication, conducting clash detection, and providing virtual reality to help train construction workers to be more familiar with the project. D4S computable rules are developed based on different software platforms. These application tools can be used to automatically check for fall hazards in 3D building models and provide design alternatives to users. They can be used by architects/engineers during the design process or by constructors before commencing construction work. For instanc e, engineers can use the D4S application tool to check project models to identify the opportunities for integrating safety into the building models during the design process, obviating the need for constructors to address certain safety measures on site. I f engineers failed to do that, constructors could use the software to check for potential hazards which were not eliminated when designing the permanent structure. The constructors would know when and where they should initiate measures to address hazards by using certain temporary structures to protect the construction workers. Four application tools are proposed as platforms for compliance checking. They are M S Access, BIMserver, Solibri Model Checker, and BimServices. The research results showed that a r elational database and its related software, such as M S Access and M S Excel, can be used for simple code checking. The Construction Operations Building Information Exchange (COBie) project, which was undertaken by the Engineer Research and Development Cent er (ERDC), also

PAGE 136

136 verified that building structural information can be exported from an IFC document to an MS Excel document. Compliance checking can be conducted based on this COBie format. Therefore, the construction worker safety issue can be addressed wi th the same approach. The benefit of using an IFC File Analyzer or other software tools to export building information into a relational database is that the user can view all component attributes at once rather than drilling down to values for an individual entity. Sometimes this functionality is more effective and convenient. A current challenge is that there is no specific software tool that has been developed to extract construction safety related structural information out of a building model. This limits the effect of code querying with a relational database. The COBie format has been developed primarily for Facility Management (FM) purposes. It is difficult to propagate compliance checking using a relational database until a similar application tool is developed to export safety related information out of the building models. From the results of the D4S check shown in the C hapter 4, it was found that the three other software platforms are all powerful tools for code checking. Even though all of them c an conduct compliance checking very well, each of them still has its own characteristics when it comes to computable rule compiling. The advantages and disadvantages for rule writing are listed in Table 51. For Solibri Model Check, the Ruleset Manager is used for writing the computable rule. The Ruleset manager has a user friendly interface. The default rule library also provides several rule templates. If a D4S suggestion could fit with any provided template, the rule compiling process becomes fairly e asy. Otherwise, a Solibri consultant can be contacted for additional

PAGE 137

137 assistance. A new computer language can be developed for rule transferring under the guidance or permission of Solibri. For bimServices, the computable rule can be written by either autom atically editing with a SMARTcode builder or manually adding protocols into a normative document. The compil ation of XLST files which are later used for document mapping are technically demanding. For BIMserver, the computable rule is written in Java lan guage with the IFC hierarchy. Though Java language is a popular computer language, only a few researchers/engineers are acquainted with the IFC hierarchy prescribed under the BIMserver environment. Table 5 1. Comparison computable rule compiling process based on three software platforms Model Checking Software (Platform) Advantage Disadvantage Solibri Model Checker Friendly user interface Not an open source software bimServices Using XML and XLST mapping makes standardized computable rule possible Th e SMARTcode builder/editor will be released soon Experienced programmer is needed BIMserver Easier coding work: directly using Java language to query the database Limited documentation is available Relational Database Easily allow a user to access this software Information loss It should be noticed that these advantages and disadvantages are all about the computable writing process. It has nothing to do with the functionalities of these platforms. All of these platforms provide multiple functions whic h are listed in Table 5 2. Besides, the above judgment is based on the expertise of selected programmers at the University of Florida. Other entities that want to develop their own computable rules might have different viewpoints based on the coding experi ence of their programmers and resources.

PAGE 138

138 Table 5 2. Comparison of functionalities between three software platforms ( Adapted from Bimservices, BIMserver, and Solibri) With a software tool that could automatically identify safety hazards in building models, constructors and designers need to cooperate with each other to effectively improve construction worker safety. Cur rently it is mainly contractors who initiate, maintain and supervise all safety precautions and programs because construction methods are within their realm of expertise. In the era of BIM technology, more and more information is transferred between projec t BimServices Transform: To interoperate between IFC, ifcXML and other representations suc h as COBIE2 and HTML. Filter: To reduce an IFC model by filtering out selected objects and relationships. Compare: To compare two IFC models from the project downwards. Compliance: To check an IFC model for compliance against regulations or code requ irements. BIMserver Merging : To merge sub models into a base model. Revision management: To get revisions by sub projects and main project. Change finder : To let the software find changes that were made to the model. Checkout & update warnings : To get a big warning when model inconsistencies seem to appear. Filter & Query: To enlist a specific IFC object in a model. Rules and advanced Query: To use Java code to make advanced queries or rules. Solibri Model Checker Model checking. Merge: To m erger multiple models together in a single, compressed SMC file. Clash detection: To conduct rule based clash detection. Change finder: To compare two versions of a model and see the changes. Quantity takeoffs: To generate quantity takeoffs that can subsequently be fed to estimating applications.

PAGE 139

139 participants. This requires that constructors develop procedure s for validating information received from the other stakeholders. With the new D4S application tool, constructors can check their building models before undertaking the construction works. T hey can erect warning signs and adopt reasonable protection methods during the construction process. But from either the construction safety or the building information perspectives, the constructors cannot substitute for designers to incorporate D4S knowl edge into designs. First, the architect/engineers scope of work is to design permanent structures. If the safety hazards on a construction site have been built in the project models after the initial design work, it will take considerable effort for the constructor to remedy the deficits by using the temporary structures. From the information life cycle perspective, in the traditional design bidbuild delivery process, the construction documents reach the highest level of information maturity at the bid stage. After the contract is awarded, the maturity level of the construction documents gradually declines as no more comprehensive compilation is conducted. This is the information content decay theory. Before there is legislation which requires designer s to consider the construction site safety during the design phase, in the short run, cultivating an environment of information stewardship seems an opportunity to promote the idea of D4S Smith and Tardif (2009) defined i nformation stewardship as a mean s that building industry professionals should regard the information they create with an attitude of stewardship rather than ownership. As it is impossible for everyone in the building industry to understand other peoples business process and to anticipat e how the information they create will be used by other people, project participants should recognize the information they create is just a part of the whole task in the project lifecycle. They should create and organize the information in an integrated an d structured manner to ensure the information is valuable to others.

PAGE 140

140 In the long run, architectural criteria should include construction worker safety as a goal of the design process. Currently, the goals of designers are not comprehensive or systematic. A rchitectural criteria (e.g., aesthetics, area efficiency, site views) prevail over engineering performance criteria (e.g., energy efficiency, natural ventilation, structural performance). Designers seldom consider construction worker safety during the desi gn process. Only when the construction worker safety issue is clearly set as a goal of the design process will designers deliberately generate and chose design options which make the construction site safe. The adoption of the designbuild method will most likely benefit from the cooperation. Even when using the traditional designbidbuild method, the risk for both designers and constructors will be reduced and the adversarial climate in the construction industry will change. Designers might ask for increased design fees considering that they are taking more responsibilities. The business arrangements should compensate the designers for their extra effort to create a better building information model. By adopting this process, both designers and constructor s will contribute their efforts to incorporate designing for construction safety information into construction documents and cooperate to minimize or eliminate non compliance in drawings. On one hand, it will be easier for designers to consider construction worker safety during the design phase because this application tool provides accessibility to related safety knowledge. On other hand, constructors as the data recipients will also have the ability to check building models and take corresponding precauti ons. The safety culture will be changed -the designers and constructors will work collaboratively to identify potential hazards early and correct them before accidents occur. The uniqueness of each construction project makes it difficult to evaluate the benefits of BIM. With the large number of construction projects that adopted BIM and that have been

PAGE 141

141 successfully finished in recent years, the evidence shows that BIM technology could improve construction worker safety. Finally, it should be realized that no one method is capable of identifying all types of safety hazards on a construction site. Multiple methods or application tools should be used to address the full range of design for safety issues. Drawbacks The collected D4S suggestions can be classified into two types: quantitative and qualitative. The first kinds of suggestions are constrained either by precise parameters or by certain materials. An example of a quantitative constraint is Design window sills to be 42 inches minimum above the floor leve l. Window sills at this height will act as guardrails during construction. Another kind of suggestion is more descriptive and difficult to be checked. A concept can be uncheckable because a building model may never have the information, or because the inf ormation will exist only on site in the actual building, or in the mind of an inspector. For example, one best practice is Design appropriate and permanent fall protection systems for roofs to be used for construction and maintenance purposes. Consider pe rmanent anchorage points, lifeline attachments, and/or holes in the perimeter for guardrail attachment. OSHA standards also correspondingly require that (Appendix C to Subpart M fall protection): (h) Tie off considerations. (1) One of the most importa nt aspects of personal fall protection systems is fully planning the system before it is put into use. Probably the most overlooked component is planning for suitable anchorage points. Such planning should ideally be done before the structure or building i s constructed so that anchorage points can be incorporated during construction for use later for window cleaning or other building maintenance. If properly planned, these anchorage points may be used during construction, as well as afterwards. However, mos t building models do not contain such detailed structural objects. The second drawback of the current safety tool is the lack of a systematic classification of the results of the D4S check based on its severity. Although the knowledge base of the D4S

PAGE 142

142 appli cation tool comprehensively and systematically includes a large number of designing for construction worker safety suggestions, the information about the probability and severity of violating certain provisions has not been incorporated in the current soft ware tool. This is mainly due to the lack of related data. As shown in the C hapter 4, Solibri Model Check has the function to classify non compliances based on preset severity. Every computable rule could be assigned a risk rate. Users could set a toleranc e level before conducting compliance checking. Only the noncompliances with a rate higher than the preset risk value would be reported to the user. The third drawback is that the construction schedule has not been added into the system as a factor. The si tuation on construction sites keeps changing day by day. The safety hazards on site one day might be gone during the next few days as progress is made on the project. Thus, just checking the final building model will miss many hazards which emerge and subs equently disappear on a construction site. The project schedule also should be considered when conducting the code checking. In Chapter 3, a computable rule based on BIMserver is developed to query the area value of an IfcOpeningElement. As mentioned in that chapter, the opening element extrusion segments may have many profiles, and three kinds of profiles are defined in IFC 23. They are IfcRectangleProfileDef, IfcCircleProfileDef and IfcArbitraryClosedProfileDef. The IFC rectangle profile is the most common profile of the IfcOpeningElement. The rule developed in Chapter 3 is based on the IfcRectangleProfileDef. However, the other two profiles may also exist in building models. For instance, Figure 51 demonstrates a paragraph of IFC code which defines an I fcOpeningElement with an arbitrary closed profile. The IfcArbitraryClosedProfileDef defines an arbitrary 2 D profile. From a list of points stored in IfcPolyLine, the outer boundary of this

PAGE 143

143 surface or solid can be constructed, which is shown in Figure 52. It is relatively difficult to compute the area value of an opening element with an IfcArbitraryClosedProfileDef. Figure 5 1. IfcOpeningElement with an IfcArbitraryClosedProfileDef Figure 5 2. IfcArbitraryClosedProfileDef defined by six points Recomm endations for Future Research Since D4S knowledge can be incorporated into designs in various ways, research studies should be conducted to assess different processes and their impact in addressing construction

PAGE 144

144 worker safety. Research is needed to develop evaluation metrics, identify benchmarks for safety and health performance, and assess the performance relative to benchmarks. To support designers in incorporating safety into the design, it is important to know how designers think and work. The knowledge of the mental process and communication patterns of designers or design teams is limited. This D4S application tool is currently designed for general use, which means there are no specific rule sets in this application for certain kinds of buildings. The f unction of the building, such as commercial, residential or industrial, will affect the kinds of provisions that are applicable to a specific construction project. With the improvement of the database, users such as health centers and education al instituti ons could accumulate best practices for particular types and develop their own rule sets. Furthermore, some provisions in the D4S database are not applicable to object oriented compliance checking. They have to be checked through a manual process. Last, t he design for construction safety application mainly uses IFC to check noncompliance in finished building drawings, so the features of the current IFC are important to the performance of code checking applications. The function of the checking system can be improved in two ways by enhancing the interoperability of IFC and BIM applications. First, the function of the IFC schema needs to be improved to support the D4S rules checking system to conduct more types of check s Currently, the code checking systems are restricted in their applications because of a restricted range of objects and parameters for encoding building codes and domain knowledge. The good news is that buildingSMART International keeps on updating the IFC schema. The IFC 24 was introduced in 2010 and it is be ing adopted by more and more software tools.

PAGE 145

145 The interoperability between software applications should be further improved. For participants of a project, software which w ould enhance the ability of individual firms to effectively comm unicate with other firms should be selected. On the other hand, wh at is more important is that software producers should guarantee the reliability of the exchanged information and enhance the integrity of the exchanged data. Although all the BIM software u sed in this research study were certified as compliant with IFC release 2, a large amount of applications in the AEC industry still do not support the IFC data format to exchange project data.

PAGE 146

146 APPENDIX A FALL PROTECTION BEST PRACTICS Section 1: F LOOR O PENINGS 1.1 Group floor openings together to create one larger opening rather than many smaller openings. 1.2 Locate floor openings away from passageways, work areas, and the structural perimeter. 1.3 Eliminate tripping hazards around floor openings (Avoid minor cha nges in floor elevations and raised thresholds). 1.4 When design features such as ventilation systems, trash chutes, chimneys, elevators, skylights, etc. cause floor openings to occur during construction, provide a warning in the plans and specifications for construction, and design in permanent/temporary guardrail systems and sequence them early in the construction process for use by all contractors. Section 2: R OOF O PENINGS 2.1 Locate roof openings away from the edge of the structure. 2.2 Group roof openings toget her to create one larger opening rather than many smaller openings. 2.3 Provide permanent/temporary guardrails around roof openings. 2.4 Eliminate tripping hazards (raised areas or other encumbrances) around roof openings. 2.5 Locate rooftop mechanical/HVAC equipme nt away from roof edge and roof openings. 2.6 Locate skylights on flat areas of the roof and away from the roof edges. 2.7 Place skylights on a raised curb (1012 inches). 2.8 Provide permanent guardrails around skylights. 2.9 Design domed, rather than flat, skylights with shatterproof glass or add strengthening wires. Section 3: R OOF F ALL P ROTECTION 3.1 Design the parapet to be 42 inches tall. A parapet of this height will provide immediate guardrail protection and eliminate the need to construct a guardrail during const ruction or for future roof maintenance.

PAGE 147

147 3.2 Minimize the roof pitch to reduce the chance of workers slipping off the roof. 3.3 Provide a guardrail around roof accesses and roof work areas. 3.4 Design appropriate and permanent fall protection systems for roofs to be used for construction and maintenance purposes. Consider permanent anchorage points, lifeline attachments, and/or holes in perimeter for guardrail attachment. 3.5 Design in a means of attaching a railing and safety lines for roofing operations. 3.6 Design and s chedule eye bolts or other connections used for window maintenance so that they can be constructed as early as possible and used during construction. 3.7 When specifying roofing materials which are not suitable for walking, such as corrugated fiberglass panel s, ensure they are distinguishable from safe, secure walking surfaces on the roof, or install guardrails around surfaces not suitable for walking. 3.8 Provide a covering, or extend the roof line over exterior stairs, ramps, and walkways. 3.9 Before demolishing a nd renovating any roof structure which is damaged, ensure that an engineering survey is performed by a competent person to determine the condition of the roof, trusses, purlins, and the structure itself to evaluate the possibility of the structure and its components failing during the work, and to evaluate how fall protection devices will be incorporated into the structure being demolished/renovated. 3.10 Avoid the design of elevated exterior structures, equipment, etc. next to roof edges. Section 4: STRUCTURA L: STEEL 4.1 In multistory buildings, design each floor plan to have a smaller area than the story below to prevent objects and workers from falling more than one story. 4.2 To minimize the risk of falling, minimize the number of offsets, and make the offsets a consistent size and as large as possible. 4.3 Prefabricating steel to accommodate fall protection. 4.4 Design special attachments or holes in members at elevated work areas to provide permanent, stable connections for supports, lifelines, guardrails, and scaffol ding. 4.5 For tower type structures, design a cable type lifeline system into the structure that allows workers to be hooked onto the structure and allows for their movement up and down the structure. Section 5: STRUCTURAL: CONCRETE and MASONRY

PAGE 148

148 5.1 Design scaffo lding tie off points into exterior walls of buildings for construction purposes. 5.2 For precast concrete members, provide inserts or other devices to attach fall protection lines. 5.3 Design perimeter beams and beams above floor openings to support lifelines (m inimum dead load of 5400 lbs.). Design connection points along the beams for the lifelines. Note on the contract drawings which beams are designed to support lifelines, how many lifelines, and at what locations along the beams. Section 6: W INDOWS 6.1 Design w indow sills to be 42 inches minimum above the floor level. Window sills at this height will act as guardrails during construction. Section 7: W ALKWAYS and F LOORS 7.1 Protect exterior walkways and platforms from the weather by providing a covering, extending t he roof line, or locating them on the sheltered side of the structure. 7.2 Locate exterior walkways and platforms away from the north side of the structure to prevent the buildup of moss and ice due to lack of sun. 7.3 Using natural lighting for stairways and ac cess areas. 7.4 Provide nonslip walking surfaces on floors, walkways and platforms adjacent to open water or exposed to the weather.

PAGE 149

149 APPENDIX B SAMPLE CODES: NET AR EA VALUE OF IFCSLAB /* Author: ======= Name : Simon Qi Department : Building Constr uction, University of Florida, USA Work : Design for Construction Worker Safety Advised by : Professor Raymond Issa, Jimmie Hinze Query : Locate the floor/roof opening Description : Find all the query object of IfcOpeningElement which are attached to either a floor/ roof IfcSlab. Approach: 1. Load the Building stories 2. For every stories, get the IfcProduct and check the instance as IfcSlab. 3. Using RelVoidsElement, get related IfcFeatureElementSubtraction for the IfcSlabs 4. Use the collection in step 3 and the IfcRelDefines, get the IfcPropertySetDefintion collection. 5. Check for the IfcElementQuantity instance in the collection at step 4. 6. Using the IfcElementQuantity in step 5, get the IfcPhysicalQuantity 7. Typecast the IfcPhysicalQuanti ty to IfcQuanityArea to get the required Area Value for consideration. */ package org.bimserver.querycompiler; import java.io.PrintWriter; import org.bimserver.ifc.database.IfcDatabase; import java.util.*; import java.util.ArrayList; import java.ut il.Iterator; import org.bimserver.ifc.emf.Ifc2x3.*; public class Query implements QueryInterface { private IfcDatabase model; private PrintWriter out; @Override public void query(IfcDatabase model, PrintWriter out) { /* Checking how many eleme nt quantity is present in the given model. To ensure the model has element quantity*/

PAGE 150

150 List qty = model.getAll(IfcElementQuantity.class); out.println("Total Element quantity in the model are "+qty.size()); List stories = model.getAll(IfcBuildingStorey.class); List Opening_subjectToArea = new ArrayList(); if(!(stories.isEmpty())){ for (IfcBuildingStorey storey : stories) { for (IfcRelContai nedInSpatialStructure rel : storey.getContainsElements()) { for (IfcProduct product : rel.getRelatedElements()) { if (product instanceof IfcSlab) { IfcSlab tempSlab = (IfcSlab)product; IfcElement ifcslabElement = (IfcElement)tempSlab; for(IfcRelVoidsElement relVoids : ifcslabElement.getHasOpenings()){ Opening_subjectToArea.add(relVoids.getRelatedOpeningElement()); } } } } } } else{out.println("Building stories not available... ");} if (!(Opening_subjectToArea.isEmpty())) { out.println("Total Opening Element in the model which are associated to slabs are + Opening_subjectToArea.size()); Iterator OE_it = Opening_subjectToArea.iterator(); while (OE_it.hasNext()) { IfcOpeningElement openingElement = (IfcOpeningElement) OE_it.next(); for (IfcRelDefines ifcRelDefines : openingElement.getIsDefinedBy()) { if (ifcRelDefines instanceof IfcRelDefinesByProperties) { IfcRelDefinesByProperties ifcRelDefinesB yProperties = (IfcRelDefinesByProperties) ifcRelDefines; IfcPropertySetDefinition relatingPropertyDefinition = ifcRelDefinesByProperties.getRelatingPropertyDefinition(); if (relatingPropertyDefinition instanceof IfcPropertySet) {

PAGE 151

151 IfcProp ertySet ifcPropertySet = (IfcPropertySet)relatingPropertyDefinition; for (IfcProperty ifcProperty : ifcPropertySet.getHasProperties()) { if (ifcProperty instanceof IfcPropertySingleValue) { IfcPropertySingleValue ifcPropertySingleVa lue = (IfcPropertySingleValue)ifcProperty; if (ifcPropertySingleValue.getNominalValue() instanceof IfcAreaMeasure) { IfcAreaMeasure ifcAreaMeasure = (IfcAreaMeasure)ifcPropertySingleValue.getNominalValue(); out.println("Area fo r + openingElement + ": + ifcAreaMeasure.getWrappedValue()); } } } } } } } } else { out.println("No match for Opening element incorporated to slab");} } }

PAGE 152

152 APPENDIX C SAMPLE CODES: IFCOPE NINGELEMENT AR EA VALUE RETRIVEAL /* Author: ======= Name : Simon Qi Department : Building Construction, University of Florida, USA Work : Design for Construction Worker Safety Advised by : Professor Raymond Issa, Jimmie Hinze Query : Locate the floor/roof opening Description : Find all the query object of IfcOpeningElement which are attached to either a floor/ roof IfcSlab. Approach: 1. Load the Building stories 2. For every story, get the IfcProduct and check the instance as IfcSlab. 3. If the instance is an IfcSlab, get its generalized parent object IfcElement. 4. Using IfcElement, check the object has any opening elements using getHasOpenings() which will be a relationship IfcRelVoidsElement 5. Collect the IfcOpeningElement into a collection from the relationship class IfcRelVoidsElement using getRelatedOpeningElement() 6. Interested IfcOpeningElements are rectangle opening elements and its pertaining area. 7. Iterate the collection of IfcOpeningElements, to get the X Dimension and Y Dimension values to calculate area as follows. 8. Objective is to find the path: IfcOpeningElement > IfcExtrudedAreaSolid > IfcRectangleProfileDef for the methods getXDim() & getYDim() 9. Get the IfcRepresentation and iterate over them to find the IfcShapeRepresentation which will return the items of IfcExtrudedAreaSolid 10. IfcExtrudedAreaSolid has the profile defintion among which somewhere IfcRectangleProfileDef. 11. Check for the instance of IfcRectangleProfileDef to invoke the getXDim() and getYDim() 12. Calculat e the area as a product of the return values of the above dimension methods. 13. Use the area value for suitable filtering of IfcOpeningElement under subjective search area criteria. */ package org.bimserver.querycompiler; import java.io.PrintWriter; import org.bimserver.ifc.database.IfcDatabase; import java.util.*; import java.util.ArrayList; import java.util.Iterator;

PAGE 153

153 import org.bimserver.ifc.emf.Ifc2x3.*; public class Query implements QueryInterface { private IfcDatabase model; private Prin tWriter out; @Override public void query(IfcDatabase model, PrintWriter out) { List stories = model.getAll(IfcBuildingStorey.class); List Opening_subjectToArea = new ArrayList(); if(!(stories.isEmpty())){ for (IfcBuildingStorey storey : stories) { for (IfcRelContainedInSpatialStructure rel : storey.getContainsElements()) { for (IfcProduct product : rel.getRelatedElements()) { if (product instanceof IfcSlab) { IfcSlab tempSlab = (IfcSlab)product; IfcElement ifcslabElement = (IfcElement)tempSlab; for(IfcRelVoidsElement relVoids : ifcslabElement.getHasOpenings()){ Opening_subjectToArea.add(relVoids.getRelatedOpeningElement()); } } } } } } else{out.println("Building stories not available...");} if (!(Opening_subjectToArea.isEmpty())) { out.println("Total Opening Element in the model which are associated to slabs are + Opening_subjectToArea.size()); Iterator OE_it = Opening_subjectToArea.iterator(); while (OE_it.hasNext()) { IfcOpeningElement openingElement = (IfcOpeningElement) OE_it.next(); IfcProductRepresentation representation = openingElement.getRepresentation();

PAGE 154

154 IfcProductDefinitionShape ifcProductDefinitionShape = (IfcProductDefinitionShape)representation; for (IfcRepresentation ifcRepresentation : ifcProductDefinitionShape.getRepresentations()) { if (ifcRepresentation instanceof IfcShapeRepresentation) { IfcShapeRepresentation ifcShapeRepresentation = (IfcShapeRepresentation)ifcRepresentation; for (IfcRepresentationItem item : ifcShapeRepresentation.getItems()) { if (item instanceof IfcExtrudedAreaSolid) { IfcExtr udedAreaSolid extrudedAreaSolid = (IfcExtrudedAreaSolid)item; IfcProfileDef sweptArea = extrudedAreaSolid.getSweptArea(); if (sweptArea instanceof IfcRectangleProfileDef) { IfcRectangleProfileDef rectangleProfileDef = (IfcRectangleProfileDef)sweptArea; float area = rectangleProfileDef.getXDim() rectangleProfileDef.getYDim(); out.println("X dimension:"+rectangleProfileDef.getXDim()); out.println("Y dimension:"+rectangleProfileDef.getYDim()); out.println("Area calculated from geometry: + area); } } } } } } } else { out.println("No match for Opening element incorporated to slab");} } }

PAGE 155

155 APPENDIX D SAMPLE CODES: QUERYI NG RELATIONAL DATABASE WIT H SQL SELECT Windows.[Id], Windows.[Level], Windows.SillHeight FROM Windows WHERE Windows.SillHeight < 1.0668 AND Windows.Level > (SELECT MIN(Floors.Level) FROM Floors) ORDER BY Windows.Level ASC

PAGE 156

156 LIST OF REFERENCES Adachi, Y. (2002) IFC Model Server development project official webpage VTT building and SECOM, Finland, < http://cic.vtt.fi/projects/ifcsvr/index_exc.html> (Mar 28, 2011) American Institute of Architects, Inc. (AIA). (1997). Standard general conditions of the construction contract. AIA A201 Washington, D.C. American Institute of Architects, Inc. (AIA). (2008). E2022008 Building Information Modeling Protocol Exhibit (May 8, 2011). American Institute of Architects, Inc. AIA. (2009). Two Types of IPD Agreements Which is Right for You? < http://www.aiacontractdocuments.org/ipd/agreements.cfm > ( Oct 31, 2011) Andres, R. N. (2002). Risk assessment & reduction: a look at the impact of ANSI B11.TR3. Professional Safety 47(1), 2026. Ash, R. (2000). CDM and design: where are we now and where should we go? A personal view. Proceedings of the design for safety and health conference. London, 151158. Associated General Contractors of America (AGC). (2008). New ConsensusDOCS contract first to address BIM. < http://www.agc.org/cs/news_media/press_room/press_release?pressrelease.id=192 > (Oct 30, 2011). Beetz, J., V an Berlo, L, d e Latt, R. and van den Helm, P. (2010). BIMserver.org An open source IFC model server. Proc. of the CIB W78 Conference Cairo, 18. Behm, M. (2004). Establishing the link between construction fatalities and disabling injuries and the design for construction safety concept. PhD dissertation, Oregon State Univ., Corvallis, Ore. Behm, M. (2005). Linking construction fatalities to the design for construction safety concept. Safety Science, 43(8), 589611. Bell Hvard., Bjrkhaug, Lars., and Hjelseth, Eilif. (2009) Standardized Computable Rules < http://www.standard.no/Global/PDF/Bygg,%20anlegg%20og%20eiendom/Rule_Checkin g_Report%20web%20stor%20fil.pdf > ( Oct 26, 2011) Bluff, L. (2003), Regulating safe design and planning construction works. National Centre for Occupational Health and Safety Regulation, Australian National University, Canberra, Working Paper 19.

PAGE 157

157 Borrmann, A., Rank, E. (2010) Handbook of Research on Building Infor mation Modeling and Construction Informatics: Concepts and Technologies Query Support for BIMs using Semantic and Spatial Conditions Information Science Reference, Hershey, PA Boukamp, F. and Akinci, B. (2007). Automated processing of construction specifications to support inspection and quality control. Automation in Construction, 17(1). 90106. Bureau of Labor. (2009). Number and rate of fatal occupational injuries, by industry sector. < http://ww w.bls.gov/iif/oshcfoi1.htm > (Oct. 4, 2011) Chachere, J. and Haymaker, J. (2008). Framework for measuring rationale clarity of AEC design decisions. Center for Integrated Facility Engineering (CIFE), Stanford Univ., Stanford, Calif. Civil Construction Industry OHS Committee. (2003). The South Australian building and construction industry common site safety induction course pocket book. Coble, R.J. and Haupt, T.C. (2000). "Potential contribution of construction foremen in designing for safety." Proceedings of the Designing for Safety and Health Conference, C.I.B. Working Commission W99 and the European Construction Institute (ECI) London, England. 175 180. Conover, D. (2009). Method and apparatus for automatically determining compliance with buildi ng regulations. < http://www.faqs.org/patents/app/20090125283> (July 13, 2009). Cooke, T., Lingard, H. and Blismas, N. (2008). The development and evaluation of a decision support tool for health and safety in construction design. Engineering, Construction and Architectural Management 15(4), 336351. Cox, R., and Issa, R. (2006). Top ten Florida residential building code violations. University of Florida, Gainesville, Fla. Cubides, M. (2006). Residential builders perspectives of building code inspections in certain Florida counties. Master thesis, University of Florida, Gainesville, Fla. Design for Safety Workgroup. Design for construction safety 2 4 hour course, June, 2008 OSHA Allia nce Program Construction Roundtable. Dym, C. L., and Levit, R. E. (1991). Knowledge Based Systems in Engineering, McGraw Hill, New York. Eastman, Chuck., Lee, Jae min., Jeong, Yeon suk., and Lee, Jinkook., (2009). Automatic rule based checking of building designs. Automation in Construction, 18, 10111033.

PAGE 158

158 Eastman, Chuck., Teicholz, Paul. Sacks, Rafael., Liston, Kathleen. BIM Handbook: A Guide to Building Information Modeling for Owners, Managers, Designers, Engineers and Contractors. (2008). John Wiley & Sons, Hoboken, N.J. El Mashaleh, M., O'Brien, W., and Minchin, E., (2006). Firm performance and information t echnology utilization in the construction industry. Journal of Construction Engineering and Management 132(5), 499507. European Foundation. (1991). From drawing board to building site: working conditions, quality, economic performance European Foundation for the Improvement of Living and Working Conditions HMSO Books, London. Fischer, G. (1989). Humancomputer interaction software: lessons learned, challenges ahead." IEEE Software 6(1), 44 52. Fischer, M. (1991). Constructability Input to Pre liminary Design of Reinforced Concrete Structures. Center for Integrated Facility Engineering (CIFE), Stanford Univ., Stanford, Calif. Fischer, M., and Kunz, J. (2004). The scope and role of information technology in construction. Center for Integrated Facility Engineering (CIFE), Stanford Univ., Stanford, Calif. Frijters, A. C. P., and Swuste, P. H. J. J. (2008). Safety assessment in design and preparation phase. Safety Science, 46(2), 272281. Gallaher, M., et al. (2004) Cost analysis of inadequa te interoperability in the U.S. capital facilities industry. National Institute of Standards and Technology, U.S. Department of Commerce, Galthersburg, Md. Gambatese, J.A. (1996). Addressing construction worker safety in the project design. Ph.D. Disse rtation, University of Washington, Seattle, WA. Gambatese, J.A. (1998). Liability in designing for construction worker safety. Journal of Architectural Engineering. 4(3), 107112. Gambatese, J.A. (2000). Safety constructability: designer involvement in construction site safety. Proceedings of the American Society of Civil Engineers (ASCE) Cons truction Congress VI Orlando, Florida. 650660. Gambatese, J. A. (2003). Safety emphasis in university engineering and construction programs. Int. e Journal of Construction. < http://www.bcn.ufl.edu/cgi bin/cm/receiver.pl?trsi_01=iejc;trsi_02=main > (Oct. 3, 2011)

PAGE 159

159 Gambatese, J. A ., Behm, M. and Hinze, J., (2003). Engineering mandates stipulated in OSHA regulations. Proceedings of the 2003 Construction Research Congress ASCE, Honolulu, HI, March 1921, 2003. Gambatese, J. A., Behm, M. a nd Rajendran, S. (2007). Designs role in construction accident causality and prevention: perspectives from an expert panel. Safety Science, 46(4), 675691. Gambatese, J. A., Hinze, J. W ., and Behm, M. (2005). Investigation of viability of designing for safety. The Center to Protect Workers Rights (CPWR), Silver Spring, MD. Gambatese, J. A., Hinze, J.W., and Haas, C.T. (1997). Tool to design for construction worker safety. Journal of Architectural Engineering, ASCE 3 (1), 3241. Gambatese, J. A. (2008). Research issues in prevention through design. Journal of Safety Research 39(2). 153156 Gane, V. and Haymaker, J. (2008). Benchmarking conceptual highrise design process. Center for Integrated Facility Engineering (CIFE), Stanford Univ., Stanford, Calif. Gibb, A. G. (2004). Designing for safety in health in construction: A European/UK perspective. Designing for safety and health in construction, S. Hecker, J. Gambatese, and M. Weinstein, eds., UO Press, Eugene, Ore., 4456. Greenwold, S., and Driver, D. (2007). Building Information Modeling with Revit Architecture. < www.autodesk.com/revitarchitecture > (Mar 12, 2009). Hale, A., Kirwan, B., and Kjell n, U. (2007). Safe by design: where are we now? Safety Science, 45(12), 305 327. Health and Safety Executive (HSE). (1974). Health and Safety at Work etc Act 1974. < http://www.hse.gov.uk/legislation/hswa.pdf > (April 23, 2009). Health and Safety Executive (HSE) (2003). The development of a knowledge based system to deliver health and safety information to designers in the construction industry HSE Books, Sudbury. Health and Safety Executive (HSE). (2005). Question and answer brief for the construction indust ry on the work at height regulation 2005. Health and Safety Executive (HSE), United Kingdom. Health and Safety Executive (HSE). (2007). The Construction (Design and Management) Regulations 2007. < http://www.opsi.gov.uk/si/si2007/uksi_20070320_en_1> ( Oct 23, 2011).

PAGE 160

160 Health and Safety Executive (HSE). (2009). Health and Safety at Work etc Act 1974. < http://www.hse.gov.uk/legislatio n/hswa.htm > ( Oct 23, 2011). Hecker, S., Gambatese, J., and Weinstein, M. (2005). Designing for worker safety: moving the construction safety process upstream. Professional Safety 50(9), 32 Hinze, J. W. (1994a). A study of the construction activity projections for 1994. Rosslyn, VA: Associated Builders and Contractors (ABC), January. Hinze, J.W. (1994b). A study of the construction activity projections for 1995. Rosslyn, VA: Associated Builders and Contractors (ABC), December. Hinze, J.W. (2006). Construction Safety. Prentice Hall. Hinze, J.W. (2000). Designing for the life cycle safety of facilit ies. Proceedings of the Designing for Safety and Health Conference, C.I.B. Working Commission W99 and the European Construction Institute (ECI) London, England.121 127. Hinze, J. W. and Wiegand, F. (1992). Role of designers in construction worker safety. Journal of Construction Engineering and Management 118(4), 677684. Hinze, J.W., Coble, R.J., and Elliott, B.R. (1999). Integrating construction worker protection into project design. Proceeding of the Second International Conference of CIB Working Commission W99, Honolulu, Hawaii. 395401. Hopf, P. S. (1982). Desig ner's guide to OSHA: a practical design guide to the Occupational Safety and Health Act for architects, engineers, and builders McGraw Hill, New York. Howarth, T., and Watson, P. (2009). Construction safety management. Wiley Blackwell, Chichester, U.K .; Malden, MA. Innovaya. (2009). Innovaya Visual 4D Simulation < http://www.innovaya.com/prod_vs.htm > ( Oct 31, 2011). International Labour Office (ILO). (1985). Safety and health in building and civil engineering work. International Labour Office, Geneva. Issa, R. R. A. (2003). Virtual Reality: a solution to seamless technology integration in the AEC industry. 4D CAD and visualization in construction: development and applications. (2003). Issa, R. R. A., and Anumba, C. (2007). Computing and information technology (IT) research in civil engineering self fulfilling or industry transforming? J. Comp. in Civ. Engrg., 21, 301302.

PAGE 161

161 Jotne EPM Technology. (2009). EDMmodelChecker. < http://www.epmtech.jotne.com/index.php?id=512200>. (July 11, 2009) Khanzode, A., Fischer, M. and Reed, D. (2008). Benefits and lessons learned of implementing building Virtual Design and Cons truction (VDC) technologies for coordination of mechanical, electrical, and plumbing (MEP) systems on a large healthcare project. Electronic Journal of Information Technology in Construction, 13, 324342. Khemlani, L. (2006). The AGC's BIM Initiatives a nd the Contractor's Guide to BIM. < http://www.aecbytes.com/buildingthefuture/2006/AGC_BIM.html> (July 27, 2008) Khemlani, L. (2009). Solibri Model Checker. < http://www.aecbytes.com/review/2009/SolibriModelChecker.html > (July12, 2009) Kiviniemi, A., Fischer, M. and Bazjanac, V. (2005). Integration of multiple product models: IFC model servers as a potential solution. Proc. of the 22nd CIB W78 Conference Dresden, Germany. 37 40. Korman, R. (2001) Wanted: new ideas. Panel ponders ways to end accidents and health hazards. Engineering News Record McGraw Hill, December 31, 2001, 2629. Korman, T. M., and Tatum, C. B. (2006). Prototype tool for mechanical, electrical, and plumbing coordination. Journal of Computing in Civil Engineering, 20(1), 3848. Kunz, J., and Fischer, M. (2009). Virtual design and construction: themes, case studies and implementation suggestions. Center for Integrated Facility Engineering (CIFE), Stanford Univ., Stanford, Calif. Larson, D. A., and Golden, K. A. (2008). Entering the brave new world: an introduction to contracting for BIM. William Mitchell Law Review 34. L iebing, R. (1987). Construction regulations handbook Wiley, New York. Loulakis, M. C., and Santiago, S. J. (2000). Do OSHA construction standards apply to designers? Civ il Eng innering, 70(1), 72. MacLeamy, P. (2009). Sharing Risk & Reward. < http://www.building connections.info/news/4%20Patrick%20MacLeamy.pdf > ( Sep 29, 2011) Manuele, F. A. (1997). On the Practice of Safety John Wiley and Sons, New York. Marini, J. (2007). Design for construction worker safety: a software tool for designers. MSBC thesis, Gainesville, University of Florida. McGraw Hill Construction ( 2009) Prefabrication and modularization: Increasing productivity in the construction industry. Bedford, MA.

PAGE 162

162 Mitropoulos, P., and Tatum, C. B. (2000). Forces driving adoption of new information technologies. Journal of Construction Engineering and Management 126(5), 340348. Muller, F. (2002). ASCE site safety policy: introduction. Practice Periodical on Structural Design and Construction, 7(2), 52. National BIM Standard. (2006). Executive c ommittee m eeting < http://www.nibs.org/BIM/NBIMS_ChairExecCmtePres.pdf > (June 25, 2009) National Society of Professional Engineers. (2007). Highlights of the 2007 EJCDC Construction Series. < http://www.nspe.org/resources/documents/pepp/EJCDC2007Summary.PDF >. (Oct. 2, 2011). Nepal, M., StaubFrench, S., Zhang, J., Lawrence, M. and Pottinger, R. (2008). Deriving construction features from an IFC model. Proc., CSCE 2008 Annual Congress, Quebec City, Canada. Niemeijer, R. A., de Vries, B., and Beetz, J. (2010 ). Designing with constraints: Towards mass customization in housing industry. 10th International Conference on Design & Decision Support Systems Netherland Nisbet, N. and Liebich, T. (2007). ifcXML implementation guide. IAI buildingSMART report U. K Nisbet, N., Wix, J. and Conover, D. (2008). The future of virtual construction and regulation checking. Virtual Futures for Design, Construction & Procurement, Brandon, P. and Kocaturk, T., Wiley Blackwell. Oh, Y., Gross, M. D. and Do, Ellen Yi Luen (2008). Computer aided critiquing systems: lessons learned and new research directions. Computer Aided Architectural Design and Research in Asia (CAADRIA) Chiang Mai, Thailand, 161 167. Perlberg, B. (2009) ConsensusDOCS: Contracts Built by Consens us for the Projects Best Interest. Construction Litigation Reporter 30(1). Quarry Group. (2009). Quarry BIM Roadmap. < http://www.quarry group.com/QG3/ > (Mar 18, 2009). Selvaag Bluethink. (2009). Bl uethink House Designer Technical White Paper. < http://www.selvaag.no/SiteCollectionDocuments/Bluethink/House_Designer_WP.pdf > (Jan, 2011) Smallwood, J.J. (1996) The influence of designers on occupational safety and health. Proceedings of the First International Conference of CIB Working Commission W99, Implementation of Safety and Health on Construction Sites, Lisbon, Portugal. 203213.

PAGE 163

163 Smith, K.D., and Tardi f, M. (2009). Building Information Modeling: a strategic implementation guide for architects, engineers, constructors, and real estate asset managers. Wiley, NJ. Suraji, A., Duff, A. R., and Peckitt, S. J. (2001). Development of causal model of construc tion accident causation. J. Constr. Eng. Manage., 127(4), 337344. Szymberski, R., (1997). Construction project safety planning. TAPPI Journal 80 (11), 6974. Taylor, J. R. (2007). Understanding and combating design error in process plant design. Safety Science, 45(12), 75 105. The Construction Users Roundtable. (2004). Collaboration, integrated information and the project lifecycle in building design, construction and operation. The Construction Users Roundtable (CURT). The Construction Users Roundtable. (2005). Optimizing the construction process: An implementation strategy. The Construction Users Roundtable (CURT). Thope, B. (2005). Health and safety in construction design. Gower Publishing Company, Burlington, VT. Toole, T. M. (1998). Uncertainty and home builders adoption of technological innovations. Journal of Construction Engineering and Management .124(4). 323332. Toole, T. M. (2002). Comparison of site safety policies of construction industry trade groups. Practice Periodical on Structural Design and Construction, 7(2), 9095. Toole, T. M. (2005). Increasing engineers role in construction safety: opportunities and barr iers. Journal of Professional Issues in Engineering Education and Practice 131(3), 199207. Toole, T. M., (2007). Design engineers responses to safety situations. Journal of Professional Issues in Engineering Education and Practice 133(2), 126131. Toole, T. M., and Gambatese, J. A. (2002). A primer on the federal OSHA standards. Practice Periodical on Structural Design and Construction, 7(2), 5660. Toole, T. M., and Gambatese, J. A. (2008). The Trajectories of Prevention through Design in Construction. Journal of Safety Research, 39(2), 225230. Toole, T. M., and Marquis, S. (2004). Site safety attitudes of U.S. and U.K. design engineers. Proc., Designing for Safety and Health in Construction: A Research and Practice Symposium Portland, Ore. Toole, T. M., Hervol, N., Hallowell, M. (2006). Designing steel for construction safety North American Steel Construction Conference San Antonio, TX, Februar y 8 11, 2006.

PAGE 164

164 U.S. Dept. of Energy. (2009.) Building Energy Codes Program. < http://www.energycodes.gov/ > (July 3, 2009) U.S. Dept. of L abor. (2000). Occupational safety and health standards for the construction industry. 29 CFR 1926, Occupational Safety and Health Administration, U.S. Government Printing Office, Washington, D.C. U.S. Dept. of Labor. (2009). OSHA Act of 1970. < http://www.osha.gov/pls/oshaweb/owasrch.search_form?p_doc_type=oshact > (April 23, 2009) Wang, H.H., and Boukamp, Frank. (2007). Leveraging project models for automated identification of co nstruction safety requirements. 2007 ASCE International Workshop on Computing in Civil Engineering, July 24 27, 2007, Pittsburgh, Pennsylvania. William C. (2011). Analysis of the causes of construction fatalities from 2007 to 2009. MSBC thesis, Gainesv ille, University of Florida. Wix, J., and Liebich, T. (2009). Projects. < http://www.aec3.com > (July 3, 2009). WorkCover, (2001). Construction hazard assessment implication review (CHAIR): a safety in design tool. WorkCover and New South Wales, Sydney. Yoders, J. (2009). BIM + IPD: three success stories. Building Design and Construction, April 1, 2009. Zhang, Jiemin.(2008) Evaluations on XML Standards for Actual Applications. Master Thesis University of British Columbia Vancouver Canada

PAGE 165

165 BIOGRAPHICAL SKETCH Jia Qi was born i n 1982 in Zhangjiakou, China. He received his Bachelor of Management degree in Construction Engineering Management at the Shijiazhuang Tiedao University. In 2005 he started his stu dies in Management Science and Engineering at the Tianjin University. Here, he got involved in scientific research study as graduate assistant to Professor Jingmin Zha. He finished his study with a master s thesis on the topic of corporate social responsibility in the construction industry in 2007. In the same year he joined the M.E. Rinker, Sr. School of Building Construction at the University of Florida. He received his Ph.D. from the University of Florida in the fall of 2011. His research interests are i n the fields of construction worker safety and Building Information Modeling. He has published several conference papers.