International Construction Specialty Conference of the Canadian Society for Civil Engineering (ICSC) (5th : 2015)

Exploring the relationship between project integration and safety performance Esmaeili, Behzad; Pellicer, Eugenio; Hasanzadeh, Sogand; Molenaar, Keith; Sanz, Amalia Jun 30, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52660-Esmaeili_B_et_al_ICSC15_311_Exploring_The_Relationship.pdf [ 347.23kB ]
52660-Esmaeili_B_et_al_ICSC15_311_Exploring_The_Relationship_slides.pdf [ 1.15MB ]
Metadata
JSON: 52660-1.0076374.json
JSON-LD: 52660-1.0076374-ld.json
RDF/XML (Pretty): 52660-1.0076374-rdf.xml
RDF/JSON: 52660-1.0076374-rdf.json
Turtle: 52660-1.0076374-turtle.txt
N-Triples: 52660-1.0076374-rdf-ntriples.txt
Original Record: 52660-1.0076374-source.json
Full Text
52660-1.0076374-fulltext.txt
Citation
52660-1.0076374.ris

Full Text

5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   EXPLORING THE RELATIONSHIP BETWEEN PROJECT INTEGRATION AND SAFETY PERFORMANCE Behzad Esmaeili1,4, Eugenio Pellicer2, Sogand Hasanzadeh1, Keith Molenaar3, and Amalia Sanz2  1 Durham School of Architectural Engineering and Construction, University of Nebraska–Lincoln, USA 2 Department of Construction Engineering, Universidad Politécnica de Valencia, Spain 3 Department of Civil, Environmental and Architectural Engineering, University of Colorado–Boulder, USA 4 Besmaeili2@unl.edu Abstract: While recent studies suggest that there is a relationship between indicators of project integration—such as early involvement of constructors—and the quality of project health and safety outcomes, there is no study that empirically investigates this relationship. To address this limitation in the current body of knowledge, the purpose of this paper is to explore the relationship between several characteristics of the integrated projects and construction safety performance. To achieve this objective, the following activities were conducted. First, to collect detailed information regarding recently completed building projects, a survey questionnaire was developed in which, safety performance metrics—such as the number of accidents—were considered as dependent variables, and project organization, team integration, and using emerging processes and technologies were included in the questionnaire as independent variables. Second, the questionnaire was distributed to reach a diverse set of respondents, and a database of 204 building projects was created.  The collected data was validated by conducting follow-up phone interviews with respondents. Finally, the database was analyzed using various statistical techniques to investigate the relationship between project integration indicators and safety performance. The results of this study provide preliminary evidence that early decisions of owners and contractors to move towards more integrated projects can impact the safety performance of projects.  1 INTRODUCTION Using more integrated project delivery systems are gaining traction in the construction industry. The integrated projects are usually characterized with usage of alternative delivery methods such as integrated project delivery method (IPD), team integration techniques such as partnering and col-location, and emerging processes and technologies such as building information modeling (BIM), prefabrication, and lean tools. There is an interest among both owners and contractors to understand the relationships among these new variables and typical definitions of project success such as cost, time, and safety outcomes (El-Asmar et al. 2013).    Considering its importance, several studies attempted to find empirical relationships between safety performance and other project related factors.  Most significant studies conducted by researchers focused on using empirical evidence provided in accident reports in national databases such as the Occupational Safety and Health Administration (OSHA) Integrated Management Information System (IMIS) (Hinze et al. 1998, Huang and Hinze 2003, Goh and Chua 2013, and Esmaeili et al. 2015a, Esmaeili et al. 2015b). 311-1 Although these studies provide significant contribution to educate project managers regarding factors that contribute to accident occurrence, they provide limited knowledge about the impact of project integration on safety performance. In fact, national databases provide a little information about the type of delivery method used, time of involvement of a contractor in the design process, existence of a partnering statement, or using building information modeling in a project.   To address this growing knowledge gap, the objective of this study is to determine the role of project integration in terms of project organization, team integration, and application of emerging processes and technologies on safety performance. To achieve this objective, a comprehensive questionnaire was developed and distributed into the U.S. construction industry to create a large empirical database of building projects. Then, the database was analyzed using various statistical analysis. This study departs from the current body of knowledge by empirically measuring the impact of project integration—such as project organization, team integration, and using emerging processes and technologies —on the safety performance of building projects. It is expected that the results of this study will provide a guideline for both owners and contractors to understand the impact of their early decisions on safety performance.   2 BACKGROUND  In this section, the research team aimed to identify a dependent variable that measures safety performance and to identify several independent variables that explain project integration. The notable results of the literature review and discussion of dependent and independent variables are provided below. 2.1 Dependent variables  The dependent variables are safety outcomes that reflect the measured safety performance or number of accidents in the project. However, as worker-hours increase, the possibility that workers will get exposed to hazards and get injured will also increase. Rather than collect the number of worker-hours, which is a difficult undertaking, to consider the impact of project size and worker-hours, the number of recordable incidents were divided by the area of the building. Thus, the safety performance was calculated using the following equation:        2.2 Independent Variables The independent variables are those factors that can impact performance outcomes. A comprehensive literature review was conducted to identify independent variables that define project integration. Four variables were selected to be included in the questionnaire for further analysis: (1) project complexity; (2) project organization; (3) team integration; and (4) using emerging processes and technologies. These variables are explained in more details in the following sections. 2.2.1 Project Complexity  One of the project characteristics that can influence project performance is complexity. Previous studies investigated the impact of complexity in design (Glavan and Tucker 1991) and scheduling (Nasser and Hegab 2006) of projects. In one of the recent studies, Puddicombe (2011) investigated the cost and schedule performance of 1,300 projects and found that complexity is an important characteristic of a project that has a crucial impact on project performance. However, there is no previous study that has empirically investigated the impact of project complexity on safety performance.   311-2 2.2.2 Project Organization  The construction industry is highly fragmented mainly due to the lack of communication among designers and contractors. The traditional delivery practices such as design-bid-build also exacerbate separation of design and construction process. Separation between different parties negatively impacts the potential for innovation as this division makes knowledge exchange more difficult.  There are several empirical studies that investigated the impact of project organization on project performance (Konchar and Sanvido 1998, Molenaar and Songer 1998, Ibbs et al. 2003, Hale et al. 2009). Most of these studies had consensus that as ones moves towards more integrated delivery methods, the possibility of cost and schedule growth decrease (Konchar and Sanvido 1998, Molenaar et al. 1999, Gransberg et al. 1999, El-Asmar et al. 2013). While the contribution of these studies are significant, none of them has considered the impact of different delivery method on safety performance of a project.   In addition to project delivery types, the project organization determines the time of involvement for different stakeholders. Song et al. (2009) studied the early involvement of contractors in design and its impact on construction schedule performance. They found that constructability input provided by contractor can improve drawing quality, material supply, information flow, and can indirectly impact construction schedule performance.  One of the major limitations of previous studies in this domain is that they did not consider the impact of early involvement of project stakeholders on safety performance.  2.2.3 Team Integration  Three major characteristics of team integration that are investigated in this study are: partnering, co-location, and collaborative goal setting. The complexity, uncertainty, and time pressure existed in construction industry has increased the need for cooperation, flexibility, integration (Eriksson and Pesamaa 2007). One of the concept that can affect project performance through improving team behavior of project participants is partnering. Partnering facilitates knowledge and resource sharing, and improves collaborative relationships that are necessary for successful project completion (Cacamis and El-Asmar 2014). Some studies measured the impact of partnering on project performance. In one of the seminal studies, Gransberg et al. (1999) found that partnering could help to control cost and time growth on projects greater than five million dollar. Nevertheless, there are limited number of studies looked at the relationship between project’s safety performances and partnering.  One of the most notable differences between integrated and less integrated delivery methods is co-location in an environment, which encourages informal collaboration and face-to-face communication among stakeholders (Raisbeck et al. 2010). Within an open floorplan office, known as the “Big Room,” a team of the contractors, architects, and owners work in a collaborative environment during the design and construction of a project. According to Hall et al. (2014), co-location facilitates collaborative decision making, innovation, and transparency. In this study, the research team decided to assess impact of using co-location on safety performance of a project.    The third characteristics of team integration that was investigated by the research team was collaborative goal setting. According to the OSHA act, the employers or contractors are responsible for the safety of their employees, and the standard form of contracts in the construction industry—such as AIA A201 and EJCDC E-500—explicitly transfers responsibility of safety to general contractors. However, if the contractor or builder is participating in goal settings, he or she can inform owners and designers regarding the impact of various decisions on safety performance. In other words, as far as the contractor is concerned regarding safety of his or her employees, the contractor can convince other project stakeholders to consider safety as one of the objectives. A minimal number of studies have investigated the impact of team integration on safety performance.  2.2.4 Using Emerging Processes and Technologies  Three emerging processes and technologies that seem to have a significant impact on enhancing safety performance are offsite prefabrication, using lean construction tools, and using building information modeling (BIM). As prefabrication provides several benefits to the project in terms of cost, time, and 311-3 productivity, its usage has been steadily increased in past decades (CII 2002, Eastman and Sacks 2008). Previous studies claimed that prefabrication has the potential to enhance safety performance by reducing the hazard levels of a task (Gambatese et al. 1997, Toole and Gambatese 2008). For example, to reduce risk of fall accidents, work can be shifted from a high elevation to the ground; to avoid the risk of cave-in, work can be shifted from inside an excavation to grade; or to reduce the risk of suffocation, work can be moved from inside a confined space to an open space (Toole and Gambatese 2008).  In addition, when work is shifted from the field to the factory, work can be conducted in an environment where the majority of risks are under control and equipment is in better condition.   Toole and Gambatese (2008) highlighted the importance of prefabrication in facilitating hazard prevention through design practices. They found that one of the main pathways that hazard prevention through design will progress in the future is in using prefabricated elements. Using prefabrication elements in design help designers to improve construction safety without fear of liability. In another study, Tanabe and Miyake (2010) proposed a safety design approach for an onshore modularization liquefied natural gas liquefaction plant. Their innovative design showed the feasibility of considering prefabrication elements in the design phase to enhance safety in downstream operations. While there is a common perception that prefabrication can enhance safety, there is no study that empirically supports this hypothesis.  Another independent variables that was identified from the literature is using lean construction tools. Lean production originated in the manufacturing and automotive industries, and after its successful contribution to the manufacturing industry, it has been used in the construction industry (Salem et al. 2006). Previous studies in the construction industry indicated that using lean tools has a positive impact of safety performance (Ohno 1988, Thomassen et al. 2003, Salem et al. 2006, Nahmens and Ikuma 2009, Wong et al. 2009, Womack et al. 2009). The research team decided to consider the impact of using lean construction tools beside other independent variables on safety performance.   Previous studies indicated that using building information modeling (BIM) has a great potential to improve safety performance in a project. Some of the potential applications are: (1) hazard identification; (2) design for safety; (3) site safety planning; and (4) training and education. First, BIM can be used to effectively enhance hazard identification by creating a 3D model of a project. Park and Kim (2013) developed a framework to integrate BIM, augmented reality, location tracking, and game engine technologies into a comprehensive site safety planning process. According to the framework, safety risk factors can be identified in the planning stage by visualizing 3D design and the required site safety facilities for project activities. Second, BIM enables designers to modify designs by considering safety constraints in early stages of a project (Rajendran and Clarke2011). To provide proof of concept, Qi et al. (2013) developed prevention through design (PtD) tools to provide feedback to designers about the hazards created by different design alternatives. Third, BIM-based site planning can improve safety performance by providing detailed spatial information of construction environments as wells as dynamic planning and 4D model (Sulankivi et al. 2009). For example, the visualization of a crane’s reach was carried out to examine the risk of load fall or the objects the crane’s rib could hit (Sulankivi et al. 2009). Finally, Rajendran and Clarke (2011) stated that 3D model of a building can be used in orientation sessions to enhance situational awareness of workers new to the working environment. Although previous studies stated that BIM can help to solve some of the safety issues on the site, a limited number of studies have empirically investigated the relationship between using BIM and safety performance.  3 RESEARCH METHODOLOGY The purpose of this study was to explore the impact project integration measures—such as delivery method, team selection, and other behavioral and technological factors—have on project safety performance. The objective of the study was achieved by conducting the following activities: (1) questionnaire development; (2) data collection; (3) data compilation and validation; and (4) data analysis. First, a survey questionnaire was developed to collect the required data from owners and contractors with an industry advisory board (Esmaeili et al. 2013, Franz et al. 2014). Recordable accidents per square foot of a project were collected as a dependent variable to measure safety performance, and several independent variables were collected including: project complexity; project delivery method (i.e., design-311-4 bid-build, construction manager at risk, design-build, integrated project delivery method); time of involvement of different parties; co-location; collaborative goal setting; and using offsite prefabrication, lean construction tools, and building information modeling.   Second, the survey was distributed by e-mail and postal mail to professional organizations in the U.S. construction industry. During the data collection process, more than 8500 project managers were contacted in the two years. A cover letter, set of instructions, and glossary of terms were attached to the survey in order to illustrate the project objectives and to explain the key terms. The diverse group of respondents was asked to answer the survey according to the most recent project that their organization completed (Franz 2014).   Third, the data collected via e-mail and postal mail were compiled and validated by conducting follow-up phone interviews. In total, 331 surveys were received and reviewed and compiled into a single database in Microsoft Access® with unique identification codes. The reviewing procedure included: verifying the project information to check for possible missing data, scrutinizing the consistency in given values, excluding out-of-scope projects, eliminating unverified projects, and resolving multiple response discrepancies. The reviewing procedure resulted in a database of 204 building projects (Franz 2014).   Forth, the descriptive statistics were examined for each variable and each distribution was checked for normality in the screening procedure. Then, correlations between variables were analyzed. Because all variables were non-normally distributed, correlations between them were analyzed using Spearman's rank-based correlation coefficient (Schumacker 2015). Correlation between recordable accidents were measured against other project performance measures such as cost, schedule, and delivery speed.  Kruskal-Wallis test was used for the group comparisons for recordable accidents per square foot area, because each of the response variables was significantly non-normal and the group sizes within could be relatively small for some categories. The Kruskal-Wallis test is a common nonparametric test that compares the overall population distribution for any number of groups. Since there are lots of "ties" in the data (observations with the same number of incidents) and in some cases quite a few groups, a chi-square approximation was used to calculate the p-value (Schumacker 2015). Post-hoc analyses were also conducted on significant groupings using pairwise Wilcox-Mann-Whitney tests with a Bonferroni adjustment. Since there were multiple comparisons in this data set, the Bonferroni adjustment kept the Type 1 error probability controlled. All analyses were done using R, which is an open-source statistical program (R Development Core Team 2011). 4 RESULTS AND DISCUSSION From 204 projects, 124 projects provided information about recordable accidents with a range of 62 and median of 1. Fifty-two percent of projects had at least one recordable accidents. These results indicate that, on average, safety performance of projects in the database was higher than the industry average. As mentioned earlier, in the questionnaire, several questions were asked related to independent variables such as project organization, team integration, and using emerging processes and technologies. The recordable incidents per square foot of a building for each of these independent variables are calculated and summarized in Table 1.  The results of the correlation analysis are summarized in Table 2. As is shown, only unit cost, project delivery speed, and construction delivery speed are significantly correlated with safety performance measures. This can be justified by considering that safety performance was calculated by dividing the number of accidents by area of the building; unit cost, project delivery speed, and construction delivery speed are also related to the total area of the building. Another project performance metric that is marginally correlated to the safety performance is construction cost growth.      311-5 Table 1: The recordable accident per square foot of a building for each independent variable   Variable Mean SD n  All Data 1.230 1.776 124  Project Complexity  Likert value 1~4 1.058 1.996 41  Likert value 5 1.068 1.387 41  Likert value 6 2.241 1.929 23 Project Organization Delivery Method  CM at risk 1.320 1.466 43 Design-bid-build 0.507 1.254 25 Design-build 1.433 2.112 54 IPD 2.821 1.606 2 Time of Builder Involvement  Pre-design 1.022 1.741 45 Conceptual 1.927 2.186 27 Schematic 1.388 1.452 14 Development 2.513 2.087 5 Documents 0.880 1.176 10 Bidding 0.594 1.329 23 Time of Mechanical, Electrical, and Plumbing (MEP) Contractor  Involvement Pre-design 0.738 0.942 12 Conceptual 2.007 2.339 11 Schematic 1.859 1.827 15 Development 1.732 2.109 12 Documents 1.220 1.259 18 Bidding 0.944 1.778 48 Time of Structural Contractor Involvement Pre-design 0.496 0.653 12 Conceptual 1.093 1.757 11 Schematic 1.596 1.598 9 Development 1.943 1.919 14 Documents 1.547 1.603 16 Bidding 1.158 1.937 54 Team Integration Using Partnering  Yes 1.244 1.732 32 No 1.224 1.800 92 Builder Participated in Co-Location Yes 1.588 1.771 19 No 1.083 1.766 61 MEP Contractor Participated in Co-Location  Yes 1.920 1.842 9 No 1.064 1.728 71 Structural Contractor Participated in Co-Location Yes 1.916 1.875 6 No 1.081 1.727 74 Builder Participated in Goal Setting Yes 1.447 1.895 92 No 0.629 1.243 28 Emerging Proc. & Tech.  Using BIM in a Project Yes 1.516 1.874 90 No 0.472 1.209 34 Using Prefabrication in a Project Low  1.037 1.747 62 High  1.441 1.817 59 Using Lean Scheduling Tools in a Project Yes 2.078 2.100 20 No 1.137 1.669 104 311-6 Table 2: Correlation between recordable accidents per square foot of area and other success measures Variable Correlation P-Value Project Cost Growth (%) -0.055 0.557 Project Schedule Growth (%) -0.044 0.627 Construction Cost Growth (%) 0.180 0.064 Construction Schedule Growth (%) -0.004 0.973 Unit Cost (log) 0.314 0.001 Project Delivery Speed  (sf/month of project duration; log)  0.298 0.008 Construction Delivery Speed  (sf/month of construction duration; log)  0.259 0.004 According to the results, the project complexity has a statistically significant impact on safety performance. This is understandable since in more complex projects, contractors might have to use unfamiliar means and methods that expose workers to unknown hazards. The post hoc analysis revealed that very complex projects (Likert value=6) have lower safety performance than projects with medium (Likert value=5, p-value=0.005) and low complexity (Likert value=1~4, p-value=0.037).  Table 3: Group comparison between independent variables using Kruskal-Wallis test Grouping variable Test stat (chi-square) p-value DF Project Complexity 15.56 0.001 3 Delivery Methods  9.75 0.021 3 Time of Builder Involvement 12.95 0.024 5 MEP Contractor Participated in Co-Location  6.84 0.009 1 Struct. Contractor Participated in Co-Location 5.49 0.019 1 Using BIM in a Project 13.58 0.000 1 Using Lean Scheduling Tools in Project 5.12 0.024 1 As far as project organization is concerned, the type of delivery method and time of builder involvement in a project design had significant impacts on safety performance. The post hoc analysis showed that projects delivered by design-bid-build method had a better safety performance than projects delivered by construction manager-at-risk (p-value=0.032). This is a surprise finding as it is a common belief that constructability input provided by a CMR agent in design phase can have a positive impact on safety performance. To understand the underlying reasons for such a difference, one needs to conduct follow up interviews with project personnel. Conducting post hoc analysis for the time of builder involvement showed marginally significant difference between the safety performances of projects that the builder involved in design development phase versus projects that the builder involved in bidding phase (p-value=0.096).  Because the number of projects in which the builders were involved in design development phase was only five and the results of Wilcox-Mann-Whitney test was marginally significant, the research team believes that the time of builder involvement should not be considered as a significant factor.   Although previous studies indicated that BIM can help to improve safety performance, the results of the study show that projects that did not use BIM had better safety performance. There are several other confounding factors that might have contributed to this finding. Empirical evidence provided in this study encourages further analysis to objectively measure impact of using various BIM functions on safety performance. Previous studies also claimed that using lean tools has a positive impact on safety performance; however, the findings of the current study imply that projects that used lean scheduling tools had more accident per square foot of a building (2.078) than projects that did not use any lean tools 311-7 (1.137). Further analysis of database revealed that projects that use lean tools, on average, had larger square foot area (335,431 SF) than the ones that did not use lean tools (151,457 SF). As discussed earlier, square foot area of a building plays a significant role in determining number of accidents.  The results of the study indicated that co-location of MEP and structural subcontractors have a significant impact on safety performance of projects; however, the post hoc analysis did not show any significant difference. The statistical analysis of results also could not find any significant relationship between safety performance and following independent variables: time of involvement of MEP and structural subcontractors; using partnering; builder participated in co-location; builder participated in goal setting; and using prefabrication in a project.  5 CONCLUSION This study assessed the factors contributing to safety performance by looking at project integration measures. The safety performance was defined as the recordable incidents per square foot of a building and several independent variables were considered related to project organization, team integration, and using emerging processes and technologies. The study has practical implications for both academia and practice. The large database of 204 building projects created in this study can be used to answer numerous questions defined by academia and the results of the study can also provide empirical evidence for both owners and contractors to understand the impact of their early decisions regarding project integration on safety performance.     While this study is unique in its kind and provides significant contributions to the body of knowledge, there are several limitations that need to be addressed in future studies. First, one of the limitations of the current study is to use lagging indicators (e.g., number of accidents) as safety performance metrics. One should note that recordable and lost time work incidents are rare events in projects and do not capture unsafe behaviors that do not lead to an incident. Further studies should be conducted to explore the impact of project delivery decisions on leading indicators such as near-misses. Second, we could not collect the number of worker-hours in projects, which inhibited us from calculating some of the standard safety measures such as total recordable incidence rate (TRIR) and days away, restricted or transferred (DART). Future studies should be conducted to address this limitation. Third, most of the independent variables in this study were simplified so we could measure them using a Likert scale. For example, complexity has several definitions in the construction industry and future studies should be conducted to investigate the relationship between complexity in design and construction activities and safety performance. Finally, as safety performance depends on people, more qualitative studies should be conducted to understand the drivers of unsafe behavior in worksites. Nevertheless, the research reported in this paper is a pioneer study in establishing the relationship between safety performance and independent variables that define integrated projects. Acknowledgements The Charles Pankow Foundation is thanked for supporting the research reported in this paper. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the writers and do not necessarily reflect the views of the Charles Pankow Foundation. The authors would also like to offer their sincere appreciation to Dr. Robert Leitch, Dr. John Messner, and Dr. Bryan Franz for their valuable support and collaborative efforts for this study. The contributions of the undergraduate research assistants who assisted with data collection are recognized, i.e. Kayleigh Arendt, Alexander Melle, Bryan Doyle, Shelby White, and Lars Anderson. We also like to thank those people who helped us in data collection. References Cacamis, M. E., and El-Asmar, M. 2013. Improving project performance through partnering and emotional intelligence. Practice Periodical on Structural Design and Construction, 19(1): 50-56. 311-8 Construction Industry Institute [CII]. 2002. Prefabrication, Preassembly, Modularization, and Offsite Fabrication in Industrial Construction: A Framework for Decision-Making. Construction Industry Institute, Research Summary 171-1 Austin, TX. Eastman, C. M. and Sacks, R. 2008. Relative Productivity in the AEC Industries in the United States for On-site and Off-site Activities. Journal of Construction Engineering and Management, ASCE, 134(7): 517–526. El-Asmar, M., Hanna, A., and Loh, W. (2013). ”Quantifying Performance for the Integrated Project Delivery System as Compared to Established Delivery Systems.” Journal of Construction Engineering and Management, ASCE, 139(11), 04013012. Eriksson, P. E., and Pesämaa, O. 2007. Modelling procurement effects on cooperation. Construction Management and Economics, 25(8): 893-901. Esmaeili, B., Franz, B., Molenaar, K. R., Leicht, R. M., and Messner, J. 2013. Construction Projects’ Performance Metrics and Critical Success Factors. CSCE Annual Conference (CSCE 2013), Montreal, Canada. Esmaeili, B., Hallowell, M., and Rajagopalan, B. 2015a. Attribute-Based Safety Risk Assessment. I: Analysis at the Fundamental Level. Journal of Construction Engineering and Management, ASCE,  04015021. Esmaeili, B., Hallowell, M., and Rajagopalan, B. 2015b. Attribute-Based Safety Risk Assessment. II: Predicting Safety Outcomes Using Generalized Linear Models. Journal of Construction Engineering and Management, ASCE, 04015022. Franz, B. W. 2014. Modeling the Role of Team Integration and Group Cohesion in construction Project Performance. Unpublished PhD dissertation, The Pennsylvania State University. University Park, PA. Franz, B., Esmaeili, B., Leicht, R. M., Molenaar, K. R., and Messner, J. 2014. Exploring the role of team environment in building project performance. Construction Research Congress (CRC2014), Atlanta, GA. Gambatese, J. A., Hinze, J., and Haas, C. T. 1997. Tool to Design for Construction Worker Safety. Journal of Architectural Engineering, ASCE, 3(1): 32−41. Glavan, J. and Tucker, R. 1991. Forecasting Design-related Problems— Case Study. Journal of Construction Engineering and Management, ASCE, 117(1): 47–65. Goh, Y. M., and Chua, D. K. H. 2013. Neural network analysis of construction safety management systems: A case study in Singapore. Construction Management and Economics, 31(5): 460–470. Gransberg, D.D., Dillon, W.D., Reynolds, L., and Boyd, J. 1999. Quantitative Analysis of Partnered Project Performance. Journal of Construction Engineering and Management, ASCE, 125(3): 161–166. Hale, D.R., Shrestha, P.P., Gibson, G.E. and Migliaccio, G.C. 2009. Empirical Comparison of Design/Build and Design/Bid/Build Project Delivery Methods. Journal of Construction Engineering and Management, ASCE, 135(7): 579–587  Hall, D., Algiers, A., Lehtinen, T., Levitt, R. E., Li, C., and Padachuri, P. 2014. The Role of Integrated Project Delivery Elements in Adoption of Integral Innovation. Working Paper Series, Proceedings of the Engineering Project Organization Conference (EPOC), Winter Park, CO. Hinze, J., Pedersen, C., and Fredley, J. 1998. Identifying Root Causes of Construction Injuries. Journal of Construction Engineering and Management, ASCE, 124(1): 67–71. Huang, X. and Hinze, J. 2003. Analysis of Construction Worker Fall Accidents. Journal of Construction Engineering and Management, ASCE, 129(3): 262–271. Ibbs, C. W., Kwak, Y. H., Ng, T., and Odabasi, A. M. 2003. Project Delivery Systems and Project Change: Quantitative Analysis. Journal of Construction Engineering and Management, ASCE, 129(4): 382–387.  Konchar M. and Sanvido V. 1998. Comparison of US project Delivery Systems. Journal of Construction Engineering and Management, ASCE, 124(6): 435–444. Molenaar, K. R. and Songer, A. D. 1998. Model for Public Sector Design-Build Project Selection. Journal of Construction Engineering and Management, ASCE, 124(6): 467-479.  Molenaar, K. R., Songer, A. D., and Barash, M. 1999. ‘Public-sector Design-Build Evolution and Performance. Journal of Management in Engineering, ASCE, 15(2): 54–62. Nahmens, I. and Ikuma, L. H. 2009. An Empirical Examination of the Relationship between Lean Construction and Safety in the Industrial Housing Industry. Lean Construction Journal, 1–12. Nasser, K. and Hegab, M. 2006. Developing a Complexity Measure for Project Schedules. Journal of Construction Engineering and Management, ASCE, 132(6): 554–561. Ohno, T. 1988. Toyota Production System, Productivity Press, Cambridge, MA. 311-9 Park, C. S. and Kim, H. J. 2013. A Framework for Construction Safety Management and Visualization System. Automation in Construction, 33: 95-103. Puddicombe, M. 2009. Why Contracts: Evidence. Journal of Construction Engineering and Management, ASCE, 135(8): 675–682. Puddicombe, M. 2011. The Contingencies of Project Management: A Factor Analytic Approach to Complexity and Novelty. International Journal of Construction Education and Research, 7(4): 259–274. Qi, J., Issa, R. R., Olbina, S., and Hinze, J. 2013. Use of BIM in Design to Prevent Construction Worker Falls. Journal of Computing in Civil Engineering, ASCE, 28(SPECIAL ISSUE: 2012 International Conference on Computing in Civil Engineering, A4014008). R Development Core Team. 2011. R: A language and environment for statistical computing. Rep. Prepared for the R Foundation for Statistical Computing, Vienna, Austria. Raisbeck, P, Millie, R and Maher, A. 2010. Assessing Integrated Project Delivery: a Comparative Analysis of IPD and Alliance Contracting Procurement Routes. Proceedings 26th Annual ARCOM Conference, Leeds, UK. Rajendran, S., & Clarke, B. 2011. Building Information Modeling: Safety Benefits & Opportunities. Professional Safety, 56(10): 44-51. Salem, O. M., Solomon, J., Genaidy, A., and Minkarah, I. 2006. Lean Construction: From Theory to Implementation. Journal of Management in Engineering, ASCE, 22(4): 168–175. Schumacker, R. E. 2015. Learning Statistics Using R. SAGE publications, USA. Song, L., Mohamed, Y., and AbouRizk, S. 2009. Early Contractor Involvement in Design and Its Impact on Construction Schedule Performance. Journal of Management in Engineering, ASCE, 25(1): 12–20. Sulankivi, K., Makela, T., and Kiviniemi, M. 2009. BIM-based Site Layout and Safety Planning. Proceedings of the First International Conference on Improving Construction and Use through Integrated Design Solutions, VTT Symposium 259, 125-140. Tanabe, M. and Miyake, A. 2010. Safety Design Approach for Onshore Modularized LNG Liquefaction Plant. Journal of Loss Prevention in the Process Industries, 23(4): 507-514. Thomassen, M. A., Sander, D., Barnes, K. A., and Nielsen, A. 2003. Experience and Results from Implementing Lean Construction in a Large Danish Contracting Firm. Proceedings of 11th Annual Conference on Lean Construction, Blacksburg, VA, 644-655. Toole, T. M. and Gambatese, J. 2008. The Trajectories of Prevention through Design in Construction. Journal of Safety Research, 39(2): 225–230. Womack, S. K., Armstrong, T. J., and Liker, J. K. 2009. Lean Job Design and Musculoskeletal Disorder Risk: A Two Plant Comparison. Human Factors and Ergonomics in Manufacturing, 19(4): 279–293. Wong, Y. C., Wong, K. Y., and Ali, A. 2009. Study on Lean Manufacturing Implementation in the Malaysian Electrical and Electronics Industry. European Journal of Scientific Research, 38(4): 521–535.   311-10  5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   EXPLORING THE RELATIONSHIP BETWEEN PROJECT INTEGRATION AND SAFETY PERFORMANCE Behzad Esmaeili1,4, Eugenio Pellicer2, Sogand Hasanzadeh1, Keith Molenaar3, and Amalia Sanz2  1 Durham School of Architectural Engineering and Construction, University of Nebraska–Lincoln, USA 2 Department of Construction Engineering, Universidad Politécnica de Valencia, Spain 3 Department of Civil, Environmental and Architectural Engineering, University of Colorado–Boulder, USA 4 Besmaeili2@unl.edu Abstract: While recent studies suggest that there is a relationship between indicators of project integration—such as early involvement of constructors—and the quality of project health and safety outcomes, there is no study that empirically investigates this relationship. To address this limitation in the current body of knowledge, the purpose of this paper is to explore the relationship between several characteristics of the integrated projects and construction safety performance. To achieve this objective, the following activities were conducted. First, to collect detailed information regarding recently completed building projects, a survey questionnaire was developed in which, safety performance metrics—such as the number of accidents—were considered as dependent variables, and project organization, team integration, and using emerging processes and technologies were included in the questionnaire as independent variables. Second, the questionnaire was distributed to reach a diverse set of respondents, and a database of 204 building projects was created.  The collected data was validated by conducting follow-up phone interviews with respondents. Finally, the database was analyzed using various statistical techniques to investigate the relationship between project integration indicators and safety performance. The results of this study provide preliminary evidence that early decisions of owners and contractors to move towards more integrated projects can impact the safety performance of projects.  1 INTRODUCTION Using more integrated project delivery systems are gaining traction in the construction industry. The integrated projects are usually characterized with usage of alternative delivery methods such as integrated project delivery method (IPD), team integration techniques such as partnering and col-location, and emerging processes and technologies such as building information modeling (BIM), prefabrication, and lean tools. There is an interest among both owners and contractors to understand the relationships among these new variables and typical definitions of project success such as cost, time, and safety outcomes (El-Asmar et al. 2013).    Considering its importance, several studies attempted to find empirical relationships between safety performance and other project related factors.  Most significant studies conducted by researchers focused on using empirical evidence provided in accident reports in national databases such as the Occupational Safety and Health Administration (OSHA) Integrated Management Information System (IMIS) (Hinze et al. 1998, Huang and Hinze 2003, Goh and Chua 2013, and Esmaeili et al. 2015a, Esmaeili et al. 2015b). 311-1 Although these studies provide significant contribution to educate project managers regarding factors that contribute to accident occurrence, they provide limited knowledge about the impact of project integration on safety performance. In fact, national databases provide a little information about the type of delivery method used, time of involvement of a contractor in the design process, existence of a partnering statement, or using building information modeling in a project.   To address this growing knowledge gap, the objective of this study is to determine the role of project integration in terms of project organization, team integration, and application of emerging processes and technologies on safety performance. To achieve this objective, a comprehensive questionnaire was developed and distributed into the U.S. construction industry to create a large empirical database of building projects. Then, the database was analyzed using various statistical analysis. This study departs from the current body of knowledge by empirically measuring the impact of project integration—such as project organization, team integration, and using emerging processes and technologies —on the safety performance of building projects. It is expected that the results of this study will provide a guideline for both owners and contractors to understand the impact of their early decisions on safety performance.   2 BACKGROUND  In this section, the research team aimed to identify a dependent variable that measures safety performance and to identify several independent variables that explain project integration. The notable results of the literature review and discussion of dependent and independent variables are provided below. 2.1 Dependent variables  The dependent variables are safety outcomes that reflect the measured safety performance or number of accidents in the project. However, as worker-hours increase, the possibility that workers will get exposed to hazards and get injured will also increase. Rather than collect the number of worker-hours, which is a difficult undertaking, to consider the impact of project size and worker-hours, the number of recordable incidents were divided by the area of the building. Thus, the safety performance was calculated using the following equation:        2.2 Independent Variables The independent variables are those factors that can impact performance outcomes. A comprehensive literature review was conducted to identify independent variables that define project integration. Four variables were selected to be included in the questionnaire for further analysis: (1) project complexity; (2) project organization; (3) team integration; and (4) using emerging processes and technologies. These variables are explained in more details in the following sections. 2.2.1 Project Complexity  One of the project characteristics that can influence project performance is complexity. Previous studies investigated the impact of complexity in design (Glavan and Tucker 1991) and scheduling (Nasser and Hegab 2006) of projects. In one of the recent studies, Puddicombe (2011) investigated the cost and schedule performance of 1,300 projects and found that complexity is an important characteristic of a project that has a crucial impact on project performance. However, there is no previous study that has empirically investigated the impact of project complexity on safety performance.   311-2 2.2.2 Project Organization  The construction industry is highly fragmented mainly due to the lack of communication among designers and contractors. The traditional delivery practices such as design-bid-build also exacerbate separation of design and construction process. Separation between different parties negatively impacts the potential for innovation as this division makes knowledge exchange more difficult.  There are several empirical studies that investigated the impact of project organization on project performance (Konchar and Sanvido 1998, Molenaar and Songer 1998, Ibbs et al. 2003, Hale et al. 2009). Most of these studies had consensus that as ones moves towards more integrated delivery methods, the possibility of cost and schedule growth decrease (Konchar and Sanvido 1998, Molenaar et al. 1999, Gransberg et al. 1999, El-Asmar et al. 2013). While the contribution of these studies are significant, none of them has considered the impact of different delivery method on safety performance of a project.   In addition to project delivery types, the project organization determines the time of involvement for different stakeholders. Song et al. (2009) studied the early involvement of contractors in design and its impact on construction schedule performance. They found that constructability input provided by contractor can improve drawing quality, material supply, information flow, and can indirectly impact construction schedule performance.  One of the major limitations of previous studies in this domain is that they did not consider the impact of early involvement of project stakeholders on safety performance.  2.2.3 Team Integration  Three major characteristics of team integration that are investigated in this study are: partnering, co-location, and collaborative goal setting. The complexity, uncertainty, and time pressure existed in construction industry has increased the need for cooperation, flexibility, integration (Eriksson and Pesamaa 2007). One of the concept that can affect project performance through improving team behavior of project participants is partnering. Partnering facilitates knowledge and resource sharing, and improves collaborative relationships that are necessary for successful project completion (Cacamis and El-Asmar 2014). Some studies measured the impact of partnering on project performance. In one of the seminal studies, Gransberg et al. (1999) found that partnering could help to control cost and time growth on projects greater than five million dollar. Nevertheless, there are limited number of studies looked at the relationship between project’s safety performances and partnering.  One of the most notable differences between integrated and less integrated delivery methods is co-location in an environment, which encourages informal collaboration and face-to-face communication among stakeholders (Raisbeck et al. 2010). Within an open floorplan office, known as the “Big Room,” a team of the contractors, architects, and owners work in a collaborative environment during the design and construction of a project. According to Hall et al. (2014), co-location facilitates collaborative decision making, innovation, and transparency. In this study, the research team decided to assess impact of using co-location on safety performance of a project.    The third characteristics of team integration that was investigated by the research team was collaborative goal setting. According to the OSHA act, the employers or contractors are responsible for the safety of their employees, and the standard form of contracts in the construction industry—such as AIA A201 and EJCDC E-500—explicitly transfers responsibility of safety to general contractors. However, if the contractor or builder is participating in goal settings, he or she can inform owners and designers regarding the impact of various decisions on safety performance. In other words, as far as the contractor is concerned regarding safety of his or her employees, the contractor can convince other project stakeholders to consider safety as one of the objectives. A minimal number of studies have investigated the impact of team integration on safety performance.  2.2.4 Using Emerging Processes and Technologies  Three emerging processes and technologies that seem to have a significant impact on enhancing safety performance are offsite prefabrication, using lean construction tools, and using building information modeling (BIM). As prefabrication provides several benefits to the project in terms of cost, time, and 311-3 productivity, its usage has been steadily increased in past decades (CII 2002, Eastman and Sacks 2008). Previous studies claimed that prefabrication has the potential to enhance safety performance by reducing the hazard levels of a task (Gambatese et al. 1997, Toole and Gambatese 2008). For example, to reduce risk of fall accidents, work can be shifted from a high elevation to the ground; to avoid the risk of cave-in, work can be shifted from inside an excavation to grade; or to reduce the risk of suffocation, work can be moved from inside a confined space to an open space (Toole and Gambatese 2008).  In addition, when work is shifted from the field to the factory, work can be conducted in an environment where the majority of risks are under control and equipment is in better condition.   Toole and Gambatese (2008) highlighted the importance of prefabrication in facilitating hazard prevention through design practices. They found that one of the main pathways that hazard prevention through design will progress in the future is in using prefabricated elements. Using prefabrication elements in design help designers to improve construction safety without fear of liability. In another study, Tanabe and Miyake (2010) proposed a safety design approach for an onshore modularization liquefied natural gas liquefaction plant. Their innovative design showed the feasibility of considering prefabrication elements in the design phase to enhance safety in downstream operations. While there is a common perception that prefabrication can enhance safety, there is no study that empirically supports this hypothesis.  Another independent variables that was identified from the literature is using lean construction tools. Lean production originated in the manufacturing and automotive industries, and after its successful contribution to the manufacturing industry, it has been used in the construction industry (Salem et al. 2006). Previous studies in the construction industry indicated that using lean tools has a positive impact of safety performance (Ohno 1988, Thomassen et al. 2003, Salem et al. 2006, Nahmens and Ikuma 2009, Wong et al. 2009, Womack et al. 2009). The research team decided to consider the impact of using lean construction tools beside other independent variables on safety performance.   Previous studies indicated that using building information modeling (BIM) has a great potential to improve safety performance in a project. Some of the potential applications are: (1) hazard identification; (2) design for safety; (3) site safety planning; and (4) training and education. First, BIM can be used to effectively enhance hazard identification by creating a 3D model of a project. Park and Kim (2013) developed a framework to integrate BIM, augmented reality, location tracking, and game engine technologies into a comprehensive site safety planning process. According to the framework, safety risk factors can be identified in the planning stage by visualizing 3D design and the required site safety facilities for project activities. Second, BIM enables designers to modify designs by considering safety constraints in early stages of a project (Rajendran and Clarke2011). To provide proof of concept, Qi et al. (2013) developed prevention through design (PtD) tools to provide feedback to designers about the hazards created by different design alternatives. Third, BIM-based site planning can improve safety performance by providing detailed spatial information of construction environments as wells as dynamic planning and 4D model (Sulankivi et al. 2009). For example, the visualization of a crane’s reach was carried out to examine the risk of load fall or the objects the crane’s rib could hit (Sulankivi et al. 2009). Finally, Rajendran and Clarke (2011) stated that 3D model of a building can be used in orientation sessions to enhance situational awareness of workers new to the working environment. Although previous studies stated that BIM can help to solve some of the safety issues on the site, a limited number of studies have empirically investigated the relationship between using BIM and safety performance.  3 RESEARCH METHODOLOGY The purpose of this study was to explore the impact project integration measures—such as delivery method, team selection, and other behavioral and technological factors—have on project safety performance. The objective of the study was achieved by conducting the following activities: (1) questionnaire development; (2) data collection; (3) data compilation and validation; and (4) data analysis. First, a survey questionnaire was developed to collect the required data from owners and contractors with an industry advisory board (Esmaeili et al. 2013, Franz et al. 2014). Recordable accidents per square foot of a project were collected as a dependent variable to measure safety performance, and several independent variables were collected including: project complexity; project delivery method (i.e., design-311-4 bid-build, construction manager at risk, design-build, integrated project delivery method); time of involvement of different parties; co-location; collaborative goal setting; and using offsite prefabrication, lean construction tools, and building information modeling.   Second, the survey was distributed by e-mail and postal mail to professional organizations in the U.S. construction industry. During the data collection process, more than 8500 project managers were contacted in the two years. A cover letter, set of instructions, and glossary of terms were attached to the survey in order to illustrate the project objectives and to explain the key terms. The diverse group of respondents was asked to answer the survey according to the most recent project that their organization completed (Franz 2014).   Third, the data collected via e-mail and postal mail were compiled and validated by conducting follow-up phone interviews. In total, 331 surveys were received and reviewed and compiled into a single database in Microsoft Access® with unique identification codes. The reviewing procedure included: verifying the project information to check for possible missing data, scrutinizing the consistency in given values, excluding out-of-scope projects, eliminating unverified projects, and resolving multiple response discrepancies. The reviewing procedure resulted in a database of 204 building projects (Franz 2014).   Forth, the descriptive statistics were examined for each variable and each distribution was checked for normality in the screening procedure. Then, correlations between variables were analyzed. Because all variables were non-normally distributed, correlations between them were analyzed using Spearman's rank-based correlation coefficient (Schumacker 2015). Correlation between recordable accidents were measured against other project performance measures such as cost, schedule, and delivery speed.  Kruskal-Wallis test was used for the group comparisons for recordable accidents per square foot area, because each of the response variables was significantly non-normal and the group sizes within could be relatively small for some categories. The Kruskal-Wallis test is a common nonparametric test that compares the overall population distribution for any number of groups. Since there are lots of "ties" in the data (observations with the same number of incidents) and in some cases quite a few groups, a chi-square approximation was used to calculate the p-value (Schumacker 2015). Post-hoc analyses were also conducted on significant groupings using pairwise Wilcox-Mann-Whitney tests with a Bonferroni adjustment. Since there were multiple comparisons in this data set, the Bonferroni adjustment kept the Type 1 error probability controlled. All analyses were done using R, which is an open-source statistical program (R Development Core Team 2011). 4 RESULTS AND DISCUSSION From 204 projects, 124 projects provided information about recordable accidents with a range of 62 and median of 1. Fifty-two percent of projects had at least one recordable accidents. These results indicate that, on average, safety performance of projects in the database was higher than the industry average. As mentioned earlier, in the questionnaire, several questions were asked related to independent variables such as project organization, team integration, and using emerging processes and technologies. The recordable incidents per square foot of a building for each of these independent variables are calculated and summarized in Table 1.  The results of the correlation analysis are summarized in Table 2. As is shown, only unit cost, project delivery speed, and construction delivery speed are significantly correlated with safety performance measures. This can be justified by considering that safety performance was calculated by dividing the number of accidents by area of the building; unit cost, project delivery speed, and construction delivery speed are also related to the total area of the building. Another project performance metric that is marginally correlated to the safety performance is construction cost growth.      311-5 Table 1: The recordable accident per square foot of a building for each independent variable   Variable Mean SD n  All Data 1.230 1.776 124  Project Complexity  Likert value 1~4 1.058 1.996 41  Likert value 5 1.068 1.387 41  Likert value 6 2.241 1.929 23 Project Organization Delivery Method  CM at risk 1.320 1.466 43 Design-bid-build 0.507 1.254 25 Design-build 1.433 2.112 54 IPD 2.821 1.606 2 Time of Builder Involvement  Pre-design 1.022 1.741 45 Conceptual 1.927 2.186 27 Schematic 1.388 1.452 14 Development 2.513 2.087 5 Documents 0.880 1.176 10 Bidding 0.594 1.329 23 Time of Mechanical, Electrical, and Plumbing (MEP) Contractor  Involvement Pre-design 0.738 0.942 12 Conceptual 2.007 2.339 11 Schematic 1.859 1.827 15 Development 1.732 2.109 12 Documents 1.220 1.259 18 Bidding 0.944 1.778 48 Time of Structural Contractor Involvement Pre-design 0.496 0.653 12 Conceptual 1.093 1.757 11 Schematic 1.596 1.598 9 Development 1.943 1.919 14 Documents 1.547 1.603 16 Bidding 1.158 1.937 54 Team Integration Using Partnering  Yes 1.244 1.732 32 No 1.224 1.800 92 Builder Participated in Co-Location Yes 1.588 1.771 19 No 1.083 1.766 61 MEP Contractor Participated in Co-Location  Yes 1.920 1.842 9 No 1.064 1.728 71 Structural Contractor Participated in Co-Location Yes 1.916 1.875 6 No 1.081 1.727 74 Builder Participated in Goal Setting Yes 1.447 1.895 92 No 0.629 1.243 28 Emerging Proc. & Tech.  Using BIM in a Project Yes 1.516 1.874 90 No 0.472 1.209 34 Using Prefabrication in a Project Low  1.037 1.747 62 High  1.441 1.817 59 Using Lean Scheduling Tools in a Project Yes 2.078 2.100 20 No 1.137 1.669 104 311-6 Table 2: Correlation between recordable accidents per square foot of area and other success measures Variable Correlation P-Value Project Cost Growth (%) -0.055 0.557 Project Schedule Growth (%) -0.044 0.627 Construction Cost Growth (%) 0.180 0.064 Construction Schedule Growth (%) -0.004 0.973 Unit Cost (log) 0.314 0.001 Project Delivery Speed  (sf/month of project duration; log)  0.298 0.008 Construction Delivery Speed  (sf/month of construction duration; log)  0.259 0.004 According to the results, the project complexity has a statistically significant impact on safety performance. This is understandable since in more complex projects, contractors might have to use unfamiliar means and methods that expose workers to unknown hazards. The post hoc analysis revealed that very complex projects (Likert value=6) have lower safety performance than projects with medium (Likert value=5, p-value=0.005) and low complexity (Likert value=1~4, p-value=0.037).  Table 3: Group comparison between independent variables using Kruskal-Wallis test Grouping variable Test stat (chi-square) p-value DF Project Complexity 15.56 0.001 3 Delivery Methods  9.75 0.021 3 Time of Builder Involvement 12.95 0.024 5 MEP Contractor Participated in Co-Location  6.84 0.009 1 Struct. Contractor Participated in Co-Location 5.49 0.019 1 Using BIM in a Project 13.58 0.000 1 Using Lean Scheduling Tools in Project 5.12 0.024 1 As far as project organization is concerned, the type of delivery method and time of builder involvement in a project design had significant impacts on safety performance. The post hoc analysis showed that projects delivered by design-bid-build method had a better safety performance than projects delivered by construction manager-at-risk (p-value=0.032). This is a surprise finding as it is a common belief that constructability input provided by a CMR agent in design phase can have a positive impact on safety performance. To understand the underlying reasons for such a difference, one needs to conduct follow up interviews with project personnel. Conducting post hoc analysis for the time of builder involvement showed marginally significant difference between the safety performances of projects that the builder involved in design development phase versus projects that the builder involved in bidding phase (p-value=0.096).  Because the number of projects in which the builders were involved in design development phase was only five and the results of Wilcox-Mann-Whitney test was marginally significant, the research team believes that the time of builder involvement should not be considered as a significant factor.   Although previous studies indicated that BIM can help to improve safety performance, the results of the study show that projects that did not use BIM had better safety performance. There are several other confounding factors that might have contributed to this finding. Empirical evidence provided in this study encourages further analysis to objectively measure impact of using various BIM functions on safety performance. Previous studies also claimed that using lean tools has a positive impact on safety performance; however, the findings of the current study imply that projects that used lean scheduling tools had more accident per square foot of a building (2.078) than projects that did not use any lean tools 311-7 (1.137). Further analysis of database revealed that projects that use lean tools, on average, had larger square foot area (335,431 SF) than the ones that did not use lean tools (151,457 SF). As discussed earlier, square foot area of a building plays a significant role in determining number of accidents.  The results of the study indicated that co-location of MEP and structural subcontractors have a significant impact on safety performance of projects; however, the post hoc analysis did not show any significant difference. The statistical analysis of results also could not find any significant relationship between safety performance and following independent variables: time of involvement of MEP and structural subcontractors; using partnering; builder participated in co-location; builder participated in goal setting; and using prefabrication in a project.  5 CONCLUSION This study assessed the factors contributing to safety performance by looking at project integration measures. The safety performance was defined as the recordable incidents per square foot of a building and several independent variables were considered related to project organization, team integration, and using emerging processes and technologies. The study has practical implications for both academia and practice. The large database of 204 building projects created in this study can be used to answer numerous questions defined by academia and the results of the study can also provide empirical evidence for both owners and contractors to understand the impact of their early decisions regarding project integration on safety performance.     While this study is unique in its kind and provides significant contributions to the body of knowledge, there are several limitations that need to be addressed in future studies. First, one of the limitations of the current study is to use lagging indicators (e.g., number of accidents) as safety performance metrics. One should note that recordable and lost time work incidents are rare events in projects and do not capture unsafe behaviors that do not lead to an incident. Further studies should be conducted to explore the impact of project delivery decisions on leading indicators such as near-misses. Second, we could not collect the number of worker-hours in projects, which inhibited us from calculating some of the standard safety measures such as total recordable incidence rate (TRIR) and days away, restricted or transferred (DART). Future studies should be conducted to address this limitation. Third, most of the independent variables in this study were simplified so we could measure them using a Likert scale. For example, complexity has several definitions in the construction industry and future studies should be conducted to investigate the relationship between complexity in design and construction activities and safety performance. Finally, as safety performance depends on people, more qualitative studies should be conducted to understand the drivers of unsafe behavior in worksites. Nevertheless, the research reported in this paper is a pioneer study in establishing the relationship between safety performance and independent variables that define integrated projects. Acknowledgements The Charles Pankow Foundation is thanked for supporting the research reported in this paper. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the writers and do not necessarily reflect the views of the Charles Pankow Foundation. The authors would also like to offer their sincere appreciation to Dr. Robert Leitch, Dr. John Messner, and Dr. Bryan Franz for their valuable support and collaborative efforts for this study. The contributions of the undergraduate research assistants who assisted with data collection are recognized, i.e. Kayleigh Arendt, Alexander Melle, Bryan Doyle, Shelby White, and Lars Anderson. We also like to thank those people who helped us in data collection. References Cacamis, M. E., and El-Asmar, M. 2013. Improving project performance through partnering and emotional intelligence. Practice Periodical on Structural Design and Construction, 19(1): 50-56. 311-8 Construction Industry Institute [CII]. 2002. Prefabrication, Preassembly, Modularization, and Offsite Fabrication in Industrial Construction: A Framework for Decision-Making. Construction Industry Institute, Research Summary 171-1 Austin, TX. Eastman, C. M. and Sacks, R. 2008. Relative Productivity in the AEC Industries in the United States for On-site and Off-site Activities. Journal of Construction Engineering and Management, ASCE, 134(7): 517–526. El-Asmar, M., Hanna, A., and Loh, W. (2013). ”Quantifying Performance for the Integrated Project Delivery System as Compared to Established Delivery Systems.” Journal of Construction Engineering and Management, ASCE, 139(11), 04013012. Eriksson, P. E., and Pesämaa, O. 2007. Modelling procurement effects on cooperation. Construction Management and Economics, 25(8): 893-901. Esmaeili, B., Franz, B., Molenaar, K. R., Leicht, R. M., and Messner, J. 2013. Construction Projects’ Performance Metrics and Critical Success Factors. CSCE Annual Conference (CSCE 2013), Montreal, Canada. Esmaeili, B., Hallowell, M., and Rajagopalan, B. 2015a. Attribute-Based Safety Risk Assessment. I: Analysis at the Fundamental Level. Journal of Construction Engineering and Management, ASCE,  04015021. Esmaeili, B., Hallowell, M., and Rajagopalan, B. 2015b. Attribute-Based Safety Risk Assessment. II: Predicting Safety Outcomes Using Generalized Linear Models. Journal of Construction Engineering and Management, ASCE, 04015022. Franz, B. W. 2014. Modeling the Role of Team Integration and Group Cohesion in construction Project Performance. Unpublished PhD dissertation, The Pennsylvania State University. University Park, PA. Franz, B., Esmaeili, B., Leicht, R. M., Molenaar, K. R., and Messner, J. 2014. Exploring the role of team environment in building project performance. Construction Research Congress (CRC2014), Atlanta, GA. Gambatese, J. A., Hinze, J., and Haas, C. T. 1997. Tool to Design for Construction Worker Safety. Journal of Architectural Engineering, ASCE, 3(1): 32−41. Glavan, J. and Tucker, R. 1991. Forecasting Design-related Problems— Case Study. Journal of Construction Engineering and Management, ASCE, 117(1): 47–65. Goh, Y. M., and Chua, D. K. H. 2013. Neural network analysis of construction safety management systems: A case study in Singapore. Construction Management and Economics, 31(5): 460–470. Gransberg, D.D., Dillon, W.D., Reynolds, L., and Boyd, J. 1999. Quantitative Analysis of Partnered Project Performance. Journal of Construction Engineering and Management, ASCE, 125(3): 161–166. Hale, D.R., Shrestha, P.P., Gibson, G.E. and Migliaccio, G.C. 2009. Empirical Comparison of Design/Build and Design/Bid/Build Project Delivery Methods. Journal of Construction Engineering and Management, ASCE, 135(7): 579–587  Hall, D., Algiers, A., Lehtinen, T., Levitt, R. E., Li, C., and Padachuri, P. 2014. The Role of Integrated Project Delivery Elements in Adoption of Integral Innovation. Working Paper Series, Proceedings of the Engineering Project Organization Conference (EPOC), Winter Park, CO. Hinze, J., Pedersen, C., and Fredley, J. 1998. Identifying Root Causes of Construction Injuries. Journal of Construction Engineering and Management, ASCE, 124(1): 67–71. Huang, X. and Hinze, J. 2003. Analysis of Construction Worker Fall Accidents. Journal of Construction Engineering and Management, ASCE, 129(3): 262–271. Ibbs, C. W., Kwak, Y. H., Ng, T., and Odabasi, A. M. 2003. Project Delivery Systems and Project Change: Quantitative Analysis. Journal of Construction Engineering and Management, ASCE, 129(4): 382–387.  Konchar M. and Sanvido V. 1998. Comparison of US project Delivery Systems. Journal of Construction Engineering and Management, ASCE, 124(6): 435–444. Molenaar, K. R. and Songer, A. D. 1998. Model for Public Sector Design-Build Project Selection. Journal of Construction Engineering and Management, ASCE, 124(6): 467-479.  Molenaar, K. R., Songer, A. D., and Barash, M. 1999. ‘Public-sector Design-Build Evolution and Performance. Journal of Management in Engineering, ASCE, 15(2): 54–62. Nahmens, I. and Ikuma, L. H. 2009. An Empirical Examination of the Relationship between Lean Construction and Safety in the Industrial Housing Industry. Lean Construction Journal, 1–12. Nasser, K. and Hegab, M. 2006. Developing a Complexity Measure for Project Schedules. Journal of Construction Engineering and Management, ASCE, 132(6): 554–561. Ohno, T. 1988. Toyota Production System, Productivity Press, Cambridge, MA. 311-9 Park, C. S. and Kim, H. J. 2013. A Framework for Construction Safety Management and Visualization System. Automation in Construction, 33: 95-103. Puddicombe, M. 2009. Why Contracts: Evidence. Journal of Construction Engineering and Management, ASCE, 135(8): 675–682. Puddicombe, M. 2011. The Contingencies of Project Management: A Factor Analytic Approach to Complexity and Novelty. International Journal of Construction Education and Research, 7(4): 259–274. Qi, J., Issa, R. R., Olbina, S., and Hinze, J. 2013. Use of BIM in Design to Prevent Construction Worker Falls. Journal of Computing in Civil Engineering, ASCE, 28(SPECIAL ISSUE: 2012 International Conference on Computing in Civil Engineering, A4014008). R Development Core Team. 2011. R: A language and environment for statistical computing. Rep. Prepared for the R Foundation for Statistical Computing, Vienna, Austria. Raisbeck, P, Millie, R and Maher, A. 2010. Assessing Integrated Project Delivery: a Comparative Analysis of IPD and Alliance Contracting Procurement Routes. Proceedings 26th Annual ARCOM Conference, Leeds, UK. Rajendran, S., & Clarke, B. 2011. Building Information Modeling: Safety Benefits & Opportunities. Professional Safety, 56(10): 44-51. Salem, O. M., Solomon, J., Genaidy, A., and Minkarah, I. 2006. Lean Construction: From Theory to Implementation. Journal of Management in Engineering, ASCE, 22(4): 168–175. Schumacker, R. E. 2015. Learning Statistics Using R. SAGE publications, USA. Song, L., Mohamed, Y., and AbouRizk, S. 2009. Early Contractor Involvement in Design and Its Impact on Construction Schedule Performance. Journal of Management in Engineering, ASCE, 25(1): 12–20. Sulankivi, K., Makela, T., and Kiviniemi, M. 2009. BIM-based Site Layout and Safety Planning. Proceedings of the First International Conference on Improving Construction and Use through Integrated Design Solutions, VTT Symposium 259, 125-140. Tanabe, M. and Miyake, A. 2010. Safety Design Approach for Onshore Modularized LNG Liquefaction Plant. Journal of Loss Prevention in the Process Industries, 23(4): 507-514. Thomassen, M. A., Sander, D., Barnes, K. A., and Nielsen, A. 2003. Experience and Results from Implementing Lean Construction in a Large Danish Contracting Firm. Proceedings of 11th Annual Conference on Lean Construction, Blacksburg, VA, 644-655. Toole, T. M. and Gambatese, J. 2008. The Trajectories of Prevention through Design in Construction. Journal of Safety Research, 39(2): 225–230. Womack, S. K., Armstrong, T. J., and Liker, J. K. 2009. Lean Job Design and Musculoskeletal Disorder Risk: A Two Plant Comparison. Human Factors and Ergonomics in Manufacturing, 19(4): 279–293. Wong, Y. C., Wong, K. Y., and Ali, A. 2009. Study on Lean Manufacturing Implementation in the Malaysian Electrical and Electronics Industry. European Journal of Scientific Research, 38(4): 521–535.   311-10  Presented by: Sogand Hasanzadeh; Behzad Esmaeili, PhD; Eugenio Pellicer; Amalia Sanz; Keith Molenaar, PhDJune 10th, 2015Exploring the Relationship between Project Integration and Safety PerformanceProject Motivation  Safety Performance: Implementing various injury prevention strategies;o Job hazard analysis; o Incentive programs; o Testing drug abuse;o … Workers demographic background;  Providing Training;   …Project Motivation  Small number of studies have EMPIRICALLYinvestigated the impact of following variables:  Project delivery method; Team selection;  Contract terms;  Time of involvement of different parties;  Using co-location;  Using partnering;    …Background2012-2014Objective: Provide objective data and empirical evidence to support how the performance is impacted by owner’s role, system integration, team behaviors, and various project delivery methods.”http://projectdelivery.weebly.com/Background Develop survey questionnaire Literature review Industry advisory boardResearch Methods Delivery method(Design-Bid-Build, CM at Risk, Design-Build, IPD)Contract terms(Lump Sum, Cost Plus, GMP)Team selection(Low Bid, Prequalified Bid, Negotiated)Rank Processes Technologies Behaviors1 Prequalification of team BIM uses Open book accounting2 BIM execution planning File sharing systems Shared risk and reward3 Partnering/team building Modularized designs Joint project management4 Co-location of team Communication latency Communication formality5 Lean decision-making tools File to fabrication Level of trust6 Risk management BIM ownership Clarity of leadership7 Process facilitator Facility management Contingency management8 Offsite prefabrication Last Planner Goal commitment9 Decision-making procedure Electronic design reviews Prior team relationship10 Design responsibility Visual management Multi-trade prefabricationProject Organization Project Execution Project Success Budget & schedule performance(Unit cost, Cost growth, Schedule growth, Delivery speed)Safety performance(Number of Incidents, Number of LWT, Worker Hours)Facility quality(System, Aesthetics, Functionality)Research Methods (Safety Metrics)  Recordable Incident Rate (RIR): Lost Time Case Rate (LTC): Days Away/Restricted or Job Transfer (DART): Severity Rate (SR):Research Methods (Safety Metrics) Number of worker-hours was missing for most of the projects.12• Develop Survey questionnaireResearch Methods • Pilot test (internal / external)• Collect data• Verify survey response dataValidation Call back contractor Call back owner Developing FAQ sheet Developing call-back guidelineExport “CPF-Working Database” to Excel and combine Access tables into a single sheet331 database entriesCombine multiple responses for the same project into a single case311 casesRemove cases that do not meet study inclusion criteriaRemove un-validated cases257 cases-74 casesUSABLE IN ANALYSIS204 cases-53 casesStep #1Step #2Step #3• Data analysisProject Data Characteristics124 Had Safety Data!CategoricalContinuousOrdinalNominalNaturally ordered as increasing or decreasingNon-orderedExamples: • 6pt rating scales • Level of LEED certification• Phase of involvementExamples: • Delivery method• Contract terms• Owner typeOrdered with meaningful intervalsExamples: • Cost growth• Schedule growth• Delivery speedResearch Methods (Analysis) Research Methods (Analysis) CorrelationTests for directional relationship between 2 variablesMay be positive or negativeGroup ComparisonTests for equal means, median, population distributionThe Kruskal-Wallis test was used because: The response variables was significantly non-normal; and  The group sizes within could be relatively small for some categories.  Since there are lots of "ties" in the data (observations with the same number of incidents) and in some cases quite a few groups, a chi-square approximation was used to calculate the p-value. All analyses were done using R.G#1 G#2 G#3Research Methods (Analysis) Post Hoc AnalysisPost-hoc analyses were also conducted on significant groupings using pairwise Wilcox-Mann-Whitney tests with a Bonferroni adjustment. Since there were multiple comparisons in this data set, the Bonferroniadjustment kept the Type 1 error probability controlled. G#1 G#2 G#3Results and Discussion (Recordable Accident/SQF)Variable Mean SD nRecordable Incidents 1.230 1.776 124Lost work Time 0.289 0.750 122Variable 0 >0Recordable Incidents 48% 52%Lost work Time 75% 25%Results and Discussion (Recordable Accident/SQF)Variable Mean SD nUsing BIM Yes 1.516 1.874 90No 0.472 1.209 34Using Prefabrication Low 1.037 1.747 62High 1.441 1.817 59Using Lean Scheduling Tools Yes 2.078 2.100 20No 1.137 1.669 104Variable Mean SD nDelivery Method CM at risk 1.320 1.466 43Design-bid-build 0.507 1.254 25Design-build 1.433 2.112 54IPD 2.821 1.606 2Results and Discussion (Recordable Accident/SQF)Variable Mean SD nTime of Builder Involvement Pre-design 1.022 1.741 45Conceptual 1.927 2.186 27Schematic 1.388 1.452 14Development 2.513 2.087 5Documents 0.880 1.176 10Bidding 0.594 1.329 23Time of Mechanical, Electrical, and Plumbing (MEP) Contractor  InvolvementPre-design 0.738 0.942 12Conceptual 2.007 2.339 11Schematic 1.859 1.827 15Development 1.732 2.109 12Documents 1.220 1.259 18Bidding 0.944 1.778 48Time of Structural Contractor InvolvementPre-design 0.496 0.653 12Conceptual 1.093 1.757 11Schematic 1.596 1.598 9Development 1.943 1.919 14Documents 1.547 1.603 16Bidding 1.158 1.937 54Results and Discussion (Recordable Accident/SQF)Variable Mean SD nProject Complexity 1~4  (Likert) 1.058 1.996 415      (Likert) 1.068 1.387 416      (Likert) 2.241 1.929 23Using Partnering Yes 1.244 1.732 32No 1.224 1.800 92Builder Participated in Co-LocationYes 1.588 1.771 19No 1.083 1.766 61MEP Contractor Participated in Co-Location Yes 1.920 1.842 9No 1.064 1.728 71Structural Contractor Participated in Co-LocationYes 1.916 1.875 6No 1.081 1.727 74Builder Participated in Goal Setting Yes 1.447 1.895 92No 0.629 1.243 28Results and DiscussionVariables Recordable Incidents/SQFLost Time Incidents/SQFCorrelation P-Value Correlation P-ValueProject Cost Growth (%) -0.055 0.557 -0.143 0.127Project Schedule Growth (%) -0.044 0.627 -0.044 0.631Construction Cost Growth (%) 0.180 0.064 0.013 0.894Construction Schedule Growth (%) -0.004 0.973 0.238 0.010Unit Cost (log) 0.314 0.001 0.289 0.001Project Delivery Speed (sf/month of project duration; log) 0.298 0.008 0.171 0.059Construction Delivery Speed (sf/month of construction duration; log) 0.259 0.004 0.216 0.017Correlation Between Safety Performance Metrics & other Performance MeasuresResults and DiscussionVariables Recordable Incidents/SQFLost Time Incidents/SQFCorrelation P-Value Correlation P-ValueProject Cost Growth (%) -0.055 0.557 -0.143 0.127Project Schedule Growth (%) -0.044 0.627 -0.044 0.631Construction Cost Growth (%) 0.180 0.064 0.013 0.894Construction Schedule Growth (%) -0.004 0.973 0.238 0.010Unit Cost (log) 0.314 0.001 0.289 0.001Project Delivery Speed (sf/month of project duration; log) 0.298 0.008 0.171 0.059Construction Delivery Speed (sf/month of construction duration; log) 0.259 0.004 0.216 0.017Correlation Between Safety Performance Metrics & other Performance MeasuresResults and DiscussionGrouping VariableRecordable Incidents/SQFLost Time Incidents/SQF DFGrouping Variable Test stat (chi-square) p-valueTest stat (chi-square) p-value DFProject Complexity 15.56 0.001 - - 3Delivery Methods 9.75 0.021 - - 3Time of Builder Involvement 12.95 0.024 11.73 0.039 5MEP Contractor Participated in Co-Location 6.84 0.009 - - 1Struct. Contractor Participated in Co-Location 5.49 0.019 - - 1Using BIM in a Project 13.58 0.000 3.87 0.049 1Using Lean Scheduling Tools in Project 5.12 0.024 - - 1Using Prefabrication - - 15.13 0.010 5Electronic File & Information Sharing - - 11.57 0.041 5Group Comparison Between Independent Variables Using Kruskal-wallis TestResults and Discussion (post hoc) Very complex projects (Likert value=6) have lower safety performance than projects with medium (Likert value=5, p-value=0.005) and low complexity (Likert value=1~4, p-value=0.037).  Projects delivered by design-bid-build (DBB) method had a better safety performance than projects delivered by construction manager-at-risk (p-value=0.032). Projects that did not use BIM had better safety performance. projects that used lean scheduling tools had more accident per square foot of a building (2.078) than projects that did not use any lean tools (1.137). SQF?  Public vs Private?  Type of project?  Claims?  Selection criteria?  ……+ Follow up phone interviews.Conclusions  This study assessed the factors contributing to safety performance by looking at project integration measures. The results of this study provide preliminary evidence that early decisions of owners and contractors can impact the safety performance of projects. Recordable incidents and lost work time incidents are lagging indicators to measure safety performance; leading indicators should also be considered. Safety performance is heavily impacted by foremen and supervisors attitude towards safety; considering only organizational variables does not help us to predict how workers behave on the site. Limitations Safety Performance Building Projects Database  Lagging indicators  Leading indicators Team SelectionDelivery MethodProcesses Contract Terms Behavior Future Works Technology Thank you for your time.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52660.1-0076374/manifest

Comment

Related Items