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

Motivation for interface management in contruction : a project complexity perspective Ahn, Seungjun; Shokri, Samin; Lee, SangHyun; Haas, Carl T.; Haas, Ralph C. G. 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-Ahn_S_et_al_ICSC15_197_Motivation_For_Interface.pdf [ 692.15kB ]
52660-Ahn_S_et_al_ICSC15_197_Motivation_For_Interface_slides.pdf [ 1.79MB ]
Metadata
JSON: 52660-1.0076318.json
JSON-LD: 52660-1.0076318-ld.json
RDF/XML (Pretty): 52660-1.0076318-rdf.xml
RDF/JSON: 52660-1.0076318-rdf.json
Turtle: 52660-1.0076318-turtle.txt
N-Triples: 52660-1.0076318-rdf-ntriples.txt
Original Record: 52660-1.0076318-source.json
Full Text
52660-1.0076318-fulltext.txt
Citation
52660-1.0076318.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   MOTIVATION FOR INTERFACE MANAGEMENT IN CONSTRUCTION: A PROJECT COMPLEXITY PERSPECTIVE    Seungjun Ahn1,4, Samin Shokri2, SangHyun Lee3, Carl T. Haas2, and Ralph C. G. Haas2 1 Department of Civil and Environmental Engineering, University of Alberta, Canada 2 Department of Civil and Environmental Engineering, University of Waterloo, Canada 3 Department of Civil and Environmental Engineering, University of Michigan, US 4 seungjun@ualberta.ca Abstract: Understanding project complexity is crucial for determining—or designing—the tools, methods, and skills required to effectively deal with interface issues in a construction project. However, understanding project complexity is not an easy undertaking; the concept of project complexity is composed of many interrelated sub-concepts, and thus is complex in itself. Given this background, this research aimed to define the dimensions of project complexity based on empirical data focusing on the variations in the project complexity factors between projects. To achieve this research objective, data for project complexity factors were collected via semi-structured interviews from 45 large-scale construction projects, and were analyzed using principal component analysis. As a result, 6 interpretable principal components were extracted from the dataset: ‘unclear scope of work for multiple stakeholders in the definition and design of projects with new technology’, ‘the uncertainty in boundaries and communication relative to other complexity factors’, ‘unfamiliarity with other project participants’, ‘the multiplicity of stakeholders relative to the amount of cost pressure and execution risks’, ‘the relative multitude of engineered items’, and ‘the high-level program/project administration’. These complexity dimensions can help with understanding sources of project complexity and determining the skills, tools, and systems to effectively cope with the sources of project complexity. Additionally, the analysis results hint that organizational interfaces should be effectively managed to prevent project failure in construction projects, and therefore support the need for advanced interface management in complex projects.   1 INTRODUCTION Interface management (IM) is an emerging practice in the construction industry. In a broad sense, IM is defined as “the management of communication, relationship, and deliverables among two or more interface stakeholders” (CII 2014). IM usually involves formalized ways of communication and coordination between parties involved in a project—e.g., formal procedures for interfacing between parties involved, designated interface managers/coordinators, and software systems for IM (Shokri et al. 2014). In traditional construction projects, interfacing between parties have mostly relied on informal and less organized means, such as verbal communication in face-to-face meetings, memos, phone conversations, and emails. However, these ways of dealing with interfaces become insufficient and ineffective for managing a complex project, and IM has recently emerged as an approach to better cope with complexity in construction projects (Shokri et al. 2014).  Therefore, understanding project complexity is crucial for determining—or designing—the tools, methods, and skills required to effectively deal with interface issues—i.e., level of systematization of practices for 197-1 IM—in a construction project. In other words, a better understanding of project complexity can give us an insight into how IM practices can help prevent—or, at least, mitigate—the adverse impact of complexity on performance in construction projects. In more general terms, understanding the sources of project complexity, the interrelationships between the sources, and to what extent the sources are salient and important for effectively managing a complex project might help with determining the tools, methods, and skills for not only IM but also for effectively dealing with project complexity in general (Remington et al. 2009; Maylor et al. 2008; Baccarini 1996). Project complexity has been known as one of the greatest factors of difficulty of project management (Remington et al. 2009; Baccarini 1996; Gidado 1996), and project failure (Parsons-Hann and Liu 2005), in the construction industry.  However, understanding project complexity is not an easy undertaking; the concept of project complexity is composed of many interrelated sub-concepts, and thus is complex in itself. Project complexity is multifaceted (Maylor et al. 2008), and it has many sources (i.e., dimensions)—“Not all projects are complex in the same way” (Remington et al. 2009). Due to this nature of project complexity, it is not clear which specific dimensions of project complexity can be addressed by IM, and how IM practices can help in dealing with those particular dimensions.    A number of researchers have attempted to define the dimensions of project complexity, and to assess the impacts that each of the project complexity dimensions has on project performance. Several research approaches have been proposed and used in this line of attempts, including framework development by aggregating the complexity factors found in the literature (Bosch-Rekveldt et al. 2011; Vidal and Marle 2008), Analytic Hierarchy Process (Vidal et al. 2011), workshops with industry practitioners (Maylor et al. 2008), surveys via interviews (Remington et al. 2009), and surveys and statistical analysis (Tatikonda and Rosenthal 2000 (factor analysis)).  In most of these proposed methods, complexity factors and dimensions were directly identified and assessed by research participants’ perceptions and evaluations. Therefore, the complexity dimensions that are defined by these methods are essentially a result of a human perceptual phenomenon (Liu 1999). This approach would be most useful in studying the subjective aspect of project complexity (e.g., perceived difficulty of project management). This view of project complexity is echoed in Fioretti and Visser’s (2004) statement that “complexity matters only because of the cognitive problems it gives rise to” (p.12), and Remington et al.’s (2009) statement that “[what matters is] how it is understood by the people who are affected….[therefore] complexity is most usefully conceptualised in cognitive terms.” However, this approach has limitations in producing objective knowledge about the interrelationships between project complexity factors and the levels of project complexity that may vary between different projects.        With this background in mind, this research aims to identify the dimensions of project complexity by looking at the main sources of variations in the project complexity factors in an empirical dataset. In other words, this research suggests that project complexity dimensions can be statistically defined based on empirical data for various project complexity factors. In the next section, a literature review on the definitions and factors of project complexity is provided. Then, the methodology and results of this research are presented in the following section. Subsequently, the implications of the results for IM as well as for general project management are discussed.   197-2 2 LITERATURE REVIEW ON PROJECT COMPLEXITY 2.1 Definitions of Project Complexity  Table 1 shows several selected definitions of project complexity found in the project management literature. As shown in this table, some disagreement and different—but, inter-related—perspectives exist in the definitions of project complexity in the literature (Bosch-Rekveldt et al. 2011; Vidal et al. 2011; Remington et al. 2009; Vidal and Marle 2008; Gidado 1996). One perspective to project complexity seems to be centered on managerial complexity (i.e., difficulty of management), which means that project complexity is the property of a project that makes managing the project difficult, and therefore requires more sophisticated managerial skills and systems. And, the other perspective seems to be more focused on the nature of the multitude of inter-related and interacting entities in projects (e.g., stakeholders, disciplines, trades, resources, processes, technologies etc.) and uncertainty. The concept of project complexity of interest in this research is in line with the latter view.  Table 1: Definitions of project complexity in the project management literature Authors Definition of Project Complexity Baccarini (1996) The degree to which a project consists of many varied interrelated parts with regard to any project dimension relevant to the project management process, such as organization, technology, environment, information, decision making and systems Gidado (1996) The measure of the difficulty of implementing a planned production work flow in relation to any one or a number of quantifiable managerial objectives  Vidal and Marle (2008); Vidal et al. (2011) The property of a project which makes it difficult to understand, foresee and keep under control its overall behaviour, even when given reasonably complete information about the project system Remington et al. (2009) The degree to which a project demonstrates a number of characteristics that makes it extremely difficult to predict project outcomes, to control or manage the project  2.2 Project Complexity Factors A large number of various sources—i.e., factors—of project complexity also have been discussed in the literature, and researchers have often attempted to categorize them. Table 2 shows several classifications of project complexity found in the project management literature. As shown in this table, many researchers have categorized and analyzed project complexity factors in different ways, while the most commonly used categories were organizational factors and technical factors. The project complexity factors that were looked at in this research can be categorized into these two groups from a broad perspective.      197-3 Table 2: Classifications of project complexity factors in the project management literature Authors (year) Classification of Project Complexity Factors Gidado (1996) ‘Within-tasks’ complexity factors (i.e. those that are inherent in the operation of individual tasks and originate from the resources or the environment) and ‘among-tasks’ complexity factors (i.e., those that originate from bringing different parts together to form a workflow)   Williams (1999) Structural uncertainty (i.e., the number and interdependence of elements) and uncertainty in goals and means Maylor (2003) Organizational factors (e.g., the number of people, departments, organizations, locations, nationalities, languages, and time zones involved in a project), technical factors (e.g., the level of novelty of any technology, system or interface), and scale/resource factors (e.g., the scale of the project and the size of the budget) Xia and Lee (2004)  Structural/ dynamic organizational factors (i.e., types of and number of relationships among hierarchical levels, formal organizational units, and specialization) and structural/dynamic technological factors (i.e., types of and number of relationships among inputs, outputs, tasks, and technologies)   Remington and Pollack (2007) Structural (i.e., interactions between many interconnected tasks), technical (i.e., unknown or untried design characteristics), directional (i.e., not agreed-upon goals and goal-paths), and temporal factors (i.e., volatility over the duration of the project) Geraldi and Adlbrecht (2007) Complexity of faith, complexity of fact, and complexity of interaction Vidal and Marle (2008) Technological factors and organizational factors, in terms of the size of project system, the variety of project system, independencies within project system, and context-dependence  3 RESEARCH METHODOLOGY 3.1 Data Collection A survey questionnaire was developed as a tool for collecting empirical data for project complexity factors (This survey questionnaire was developed as a part of a larger survey conducted by Construction Industry Institute Research Team 302 (CII RT 302). More information about this CII research project can be found in CII 2014.) Table 3 shows the project complexity factors included in the questionnaire. These factors were selected based on the result of the literature review and discussions with sixteen experienced project engineers/managers who participated in the questionnaire development process. The project characteristic factors that can be identified at a project’s onset were focused on in this research, and thus dynamic factors such as change orders and changes in the economic environment were excluded. As mentioned above, the complexity factors included in this research cover the organizational aspect as well as the technical aspect of project complexity, and are, especially, in line with the project complexity factors identified in Bosch-Rekveldt et al. 2011, Geraldi and Adlbrecht 2007, and Vidal et al. 2011.       197-4 Table 3: Project complexity factors included in the survey No. Project Complexity Factors  1 Cost Pressure  2 Schedule pressure 3 Extended/uncertain scope 4 Execution risks 5 Multiple owners (e.g., JV) 6 New technology 7 Large number of  suppliers/subcontractors 8 Multiple engineering centers 9 Government rules/regulations 10 Multiple general contractors/EPCs 11 Large number of engineered items 12 Multiple languages 13 Unfamiliar partners/collaborators 14 Not-aligned software/design standards between parties  15 Unclear geographical boundaries within project (“battery limit”) 16 Unclear requirements between involved parties 17 Unclear responsibilities between involved parties Then, the surveys were administered directly by the authors. The authors had either a face-to-face meeting or a web conference call with project engineers/managers/interface managers representing a project. The interviewees were asked to rate the project complexity factors on a 10-point scale by the question, “How much does this factor contribute to the complexity of your current project? (1 for the lowest and 10 for the highest)?” Also, the interviewees were asked to give responses solely based on their current project, or their most recently completed project if there was no current project to which they belonged.    The survey was conducted during 2013. In total, 46 projects owned or managed by US- or Canada-based organizations were studied in this survey, and the data that was valid for the analysis presented in this paper were collected for a total of 45 projects. The average rating of complexity factors ranges from 1.94 to 7.76 with a mean of 5.21 in the sample. This distribution of average complexity score shows that the sample included a wide range of projects in terms of project complexity. More detailed description of data collection method as well as the sample is available in Shokri et al. 2014.      3.2 Data Analysis The collected data were analyzed by principal component analysis (PCA). PCA is a widely used statistical method used for reducing the dimensionality of a dataset while retaining the variations in the dataset as much as possible. In PCA, the original data are transformed into a smaller number of new variables called principal components (PCs). Although there are several variants of the analysis method, one of the most common methods to define PCs is finding the eigenvectors of the correlation matrix of the original variables (Jolliffe 2002), which is the method used in this research. If the original variables are interrelated, the first few PCs—i.e., those with a large eigenvalue—retain most of the variations that were distributed over the original variables. If an appropriate number of PCs are selected, therefore, the PCs can retain most of the ‘information’ in the dataset, and that is why PCA can be used for reducing the number of dimensions in the dataset. In this research, PCs were used to define the major dimensions of project complexity extracted from the empirical dataset.   197-5 Table 4: Loadings for the first 6 PCs of project complexity*  Principal Components  1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 *Note. In bold type are the loadings whose absolute values are greater than half the maximum loading for the PC. The second PC has negative loadings for the variables such as ‘unclear responsibilities between involved parties’, ‘unclear requirements between involved parties’, ‘unclear geographical boundaries within project’, ‘not-aligned software/design standards between parties’, ‘multiple languages’, ‘unfamiliar partners/collaborators’, all of which are related to either the clarity of boundaries or communication between parties.  In the meantime, the second PC has positive loadings for many other variables such as ‘cost pressure’, ‘schedule pressure’, ‘execution risks’, ‘large number of suppliers/subcontractors’, ‘government rules/regulations’, and ‘large number of engineered items.’ Therefore, the second PC contrasts the uncertainty in boundaries and communication with the rest of the complexity factors, and implies that after the issues of the clarity of scope and of the newness of technology in the design have been accounted for, the main source of variation in the dataset is the uncertainty in boundaries/communication relative to the rest of the complexity factors.  The third PC has an outstanding loading for ‘multiple languages’, and ‘unfamiliar partners/collaborators’ follows in term of the size of loading. These two variables are thematically inter-related in general, and reflect that there are project participants with unfamiliar backgrounds. Therefore, this PC can be interpreted as ‘unfamiliarity with other project participants’. The fourth PC has negative loadings for the variables such as “multiple engineering centers’, ‘multiple general contractors/EPCs’, and ‘Large number of suppliers/subcontractors’, whereas it has positive loadings for ‘cost pressure’ and ‘execution risks’. Therefore, the fourth PC contrasts the multiplicity of stakeholders with cost pressure/risks, and implies that after the variances that are accounted for by the first three PCs, the main source of variations in the dataset is the multiplicity of stakeholders relative to the amount of cost pressure and execution risks.  197-7 The fifth PC has a relatively high level of negative loading for ‘large number of engineered items’ while having positive loadings for ‘cost pressure’, ‘multiple owners’, and ‘multiple general contractors/EPCs’. Therefore, the fifth PC contrasts the multitude of engineered items with cost pressure and the multiplicity of owners/general contractors, and implies that after the variances that are accounted for by the first four PCs, the main source of variations in the dataset is the multitude of engineered items relative to cost pressure and the multiplicity of owners/general contractors.  The sixth PC has the highest positive loadings for ‘government rules/regulations’ and ‘multiple general contractors/EPCs’, while having negative loadings for ‘extended/uncertain scope’, ‘large number of suppliers/subcontractors’, and ‘new technology’. Government rules/regulations and multiple general contractors/EPCs are related to the high-level program/project administration, and the sixth PC contrasts these factors with lower-level scope and technology factors. 5 DISCUSSION As shown in the previous section, PCA was used in this research for finding the main sources of variations in the dataset, which can be defined as major dimensions of project complexity. As a result of this method, 6 interpretable PCs were defined, and these PCs combined to explain more than 70% of the total variance in the dataset. In other words, the number (17) of original variables was reduced to the smaller number (6) of new variables by PCA, while 30% of the total variance was sacrificed. One should not take from this that the 30% of the total variance excluded in the 6 PCs are not important, but, given that the data were collected by surveys, and that the data are under the influence of random errors as well as the sample size, it seems acceptable to say that the 6 PCs capture the majority of the sources of variations in the dataset.  An interesting finding from this analysis is that the complexity factors that were highly rated for most of the projects, such as ‘cost pressure’, ‘schedule pressure’, and ‘execution risks’, did not have much variations in the dataset, and therefore contributed to the first few PCs relatively less than other factors. This shows that PCA is useful for distinguishing the complexity factors that actually vary between projects from the factors that are shared by most of the construction projects studied. Therefore, the results of PCA might be useful for specifically developing the managerial approaches to address the complexity factors that vary between projects and the managerial approaches to address the complexity factors that are common among projects.        The fact that the first PC is positively related with every original variable tells us that all of the original variables are inter-related to some extent, and the first PC can serve as a kind of composite indicator of the level of overall project complexity. This implies that some projects in the dataset are more complex in general than others, and that the score for the first PC can tell us overall how complex a project is. Therefore, this PC might be useful for determining the level of systematization of project management system required in a project. The project complexity dimensions defined by later PCs indicate more specific aspect (i.e., source) of project complexity and explain a smaller variance in the dataset. Noticeably, complexity dimensions defined by the PCs are related to organizational interactions (e.g., unclear scope of work for multiple stakeholders in the definition (1st PC), uncertainty in boundaries and communication (2nd PC), unfamiliarity with other project participants (3rd PC), and the multiplicity of stakeholders (4th PC)). These results hint that organizational interactions (i.e., interfaces between parties involved) are a major source of complexity in many projects. This implication is in line with observations/findings in previous research works. Vidal et al. (2009) reported that 11 project complexity drivers (i.e., factors) identified from their survey belong to the family of “project interdependencies”, and this takes up 61.1% of the entire project complexity drivers they identified. From this result, they also argued that their result helps justify the tools and methods that been recently developed to deal with interactions and interdependencies in projects.        Vidal and Marle (2008) stated that classical tools and methods of project management are not sufficient in dealing with the level of complexity in modern projects, and therefore, the current level of project complexity justifies the need for a new project management approach that can assist the existing ones, 197-8 such as IM. Similarly, Baccarini (1996) argued that project complexity should be treated by more efforts for integration such as coordination, communication, and control. Therefore, more advanced managerial skills and tools for interface management would be required to effectively manage complex projects, and in particular, it seems that such development of advanced skills and tools for project management should aim at those projects that have complex organizational structure and a large number of interfaces in it. An efficient implementation of managerial functions can influence the impact of project complexity on project performance (Gidado 1996), and therefore is expected to mitigate the adverse impact of complexity on performance in many construction projects. More research is strongly warranted in this area.    6 CONCLUSION Although many researchers have attempted to define project complexity and have investigated various factors that would affect complexity in construction projects, it is unclear what the dimensions of project complexity are and how they can be measured. This lack of knowledge about project complexity has been one of the main reasons that the construction industry has displayed great difficulty in effectively dealing with complexity in projects. Given this situation, this research aimed to define the dimensions of project complexity based on empirical data. To achieve this research objective, a questionnaire was developed  as a tool for collecting data for project complexity factors, surveys were administered in semi-structured interviews, and the collected data for 45 major construction projects were analyzed using PCA. As a result, 6 interpretable PCs were extracted from the dataset: ‘unclear scope of work  for multiple stakeholders in the definition and design of projects with new technology’ (1st PC), ‘the uncertainty in boundaries and communication relative to other complexity factors’ (2nd PC), ‘unfamiliarity with other project participants’ (3rd PC), ‘the multiplicity of stakeholders relative to the amount of cost pressure and execution risks’ (4th PC), ‘the relative multitude of engineered items’ (5th PC), and ‘the high-level program/project administration’ (6th PC). These dimensions of project complexity defined by the PCs tell us not only what the possible dimensions of project complexity are, but also which dimension explain the most variations in the dataset. Therefore, these complexity dimensions can help with assessing the level of complexity of a project at its onset and with determining the skills, tools, and systems that will be required to effectively cope with the sources of complexity in the project. Additionally, the analysis results hint that organizational interfaces should be effectively managed to prevent project failure in construction projects, and therefore support the need for IM in complex projects.  This research has several limitations that can be addressed in future research. First, the complexity factors included in this research are limited. More research efforts to investigate a broader set of project complexity factors are strongly warranted to gain a comprehensive view of various complexity factors that may play an important role in construction projects. Secondly, the sample size used in this research is modest. Since PCA is a data-driven method, a larger sample size may be required to gain a greater confidence with the PCs extracted in this research. Thirdly, the impacts of the project complexity dimensions identified in this research on project performances, such as cost and schedule, are yet to be investigated. Therefore, more future research efforts are strongly warranted to relate the levels of project complexity with project performance measures, and to investigate the effectiveness of approaches to address project complexity in this relationship.                Acknowledgements The research work presented in this conference paper was financially supported by the Construction Industry Institute, and is based on the results produced by the Construction Industry Institute Research Team 302 - Interface Management.      References Baccarini, D. 1996. The Concept of Project Complexity—A Review. International Journal of Project Management, 14(4): 201–204. 197-9 Bosch-Rekveldt, M., Jongkind, Y., Mooi, H., Bakker, H. and Verbraeck, A. 2011. Grasping Project Complexity in Large Engineering Projects: The TOE (Technical, Organizational and Environmental) framework. International Journal of Project Management, 29(6): 728–739. Chen, Q., Reichard, G., and Beliveau, Y. 2008. Multiperspective Approach to Exploring Comprehensive Cause Factors for Interface Issues.” Journal of Construction Engineering and Management, 134(6): 432–441. Construction Industry Institute. 2014. RS302-1 Interface Management, Construction Industry Institute, Austin, TX.   Fioretti, G. and Visser, B. 2004. A Cognitive Interpretation of Organizational Complexity. E:CO Special Double Issue. 6(1-2): 11-23. Geraldi, J.G. and Adlbrecht, G., 2007. On Faith, Fact, and Interaction in Projects. Project Management Journal, 38(1): 32–43. Gidado, K.I. 1996. Project Complexity: The Focal Point of Construction Production Planning. Construction Management and Economics, 14(3): 213–225.  Jolliffe, I. T. 2002. Principal Component Analysis, 2nd ed., New York, NY: Springer. Liu, A.M.M. 1999. A Research Model of Project Complexity and Goal Commitment Effects on Project Outcome. Engineering, Construction and Architectural Management, 6(2): 105–111. Maylor, H. 2003. Project Management, 3rd ed., Harlow, UK: FT Prentice Hall. Maylor, H., Vidgen, R., and Carver, S. 2008. Managerial Complexity in Project- Based Operations: A Grounded Model and Its Implications for Practice. Project Management Journal, 39: 15–26. Parsons-Hann, H. and Liu, K. 2005. Measuring Requirement Complexity to Increase the Probability of Project Success”, Proceedings of ICEIS 2005, Miami, 434–438. Remington, K. and Pollack, J. 2007. Tools for Complex Projects, Aldershott, UK: Gower Publishing company. Remington, K., Zolin, R. and Turner, R. 2009. A Model of Project Complexity: Distinguishing Dimensions of Complexity from Severity. Proceedings of the 9th International Research Network of Project Management, Berlin, 11–13. Shokri, S., Ahn, S., Lee, S., Haas, C. T., and Haas, R C. G. 2014. Current Status of Interface Management in Construction: Drivers and Impacts of Systematic Interface Management.” Manuscript submitted for publication. Tatikonda, M. V and Rosenthal, S.R. 2000. Technology Novelty, Project Complexity, and Product Development Project Execution Success: A Deeper Look at Task Uncertainty in Product Innovation. IEEE Transactions on Engineering Management, 47(1): 74–87. Vidal, L.A. and Marle, F. 2008. Understanding Project Complexity: Implications on Project Management. Kybernetes, 37: 1094–1110. Vidal, L.A., Marle, F. and Bocquet, J.C. 2011. Measuring Project Complexity using the Analytic Hierarchy Process. International Journal of Project Management, 29(6): 718–727. Williams, T.M. 1999. The Need for New Paradigms for Complex Projects. International Journal of Project Management, 17(5): 269–273. Xia, W. and Lee, G. 2004. Grasping the Complexity of IS Development Projects. Communications of the ACM, 47(5): 69–74.  197-10  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   MOTIVATION FOR INTERFACE MANAGEMENT IN CONSTRUCTION: A PROJECT COMPLEXITY PERSPECTIVE    Seungjun Ahn1,4, Samin Shokri2, SangHyun Lee3, Carl T. Haas2, and Ralph C. G. Haas2 1 Department of Civil and Environmental Engineering, University of Alberta, Canada 2 Department of Civil and Environmental Engineering, University of Waterloo, Canada 3 Department of Civil and Environmental Engineering, University of Michigan, US 4 seungjun@ualberta.ca Abstract: Understanding project complexity is crucial for determining—or designing—the tools, methods, and skills required to effectively deal with interface issues in a construction project. However, understanding project complexity is not an easy undertaking; the concept of project complexity is composed of many interrelated sub-concepts, and thus is complex in itself. Given this background, this research aimed to define the dimensions of project complexity based on empirical data focusing on the variations in the project complexity factors between projects. To achieve this research objective, data for project complexity factors were collected via semi-structured interviews from 45 large-scale construction projects, and were analyzed using principal component analysis. As a result, 6 interpretable principal components were extracted from the dataset: ‘unclear scope of work for multiple stakeholders in the definition and design of projects with new technology’, ‘the uncertainty in boundaries and communication relative to other complexity factors’, ‘unfamiliarity with other project participants’, ‘the multiplicity of stakeholders relative to the amount of cost pressure and execution risks’, ‘the relative multitude of engineered items’, and ‘the high-level program/project administration’. These complexity dimensions can help with understanding sources of project complexity and determining the skills, tools, and systems to effectively cope with the sources of project complexity. Additionally, the analysis results hint that organizational interfaces should be effectively managed to prevent project failure in construction projects, and therefore support the need for advanced interface management in complex projects.   1 INTRODUCTION Interface management (IM) is an emerging practice in the construction industry. In a broad sense, IM is defined as “the management of communication, relationship, and deliverables among two or more interface stakeholders” (CII 2014). IM usually involves formalized ways of communication and coordination between parties involved in a project—e.g., formal procedures for interfacing between parties involved, designated interface managers/coordinators, and software systems for IM (Shokri et al. 2014). In traditional construction projects, interfacing between parties have mostly relied on informal and less organized means, such as verbal communication in face-to-face meetings, memos, phone conversations, and emails. However, these ways of dealing with interfaces become insufficient and ineffective for managing a complex project, and IM has recently emerged as an approach to better cope with complexity in construction projects (Shokri et al. 2014).  Therefore, understanding project complexity is crucial for determining—or designing—the tools, methods, and skills required to effectively deal with interface issues—i.e., level of systematization of practices for 197-1 IM—in a construction project. In other words, a better understanding of project complexity can give us an insight into how IM practices can help prevent—or, at least, mitigate—the adverse impact of complexity on performance in construction projects. In more general terms, understanding the sources of project complexity, the interrelationships between the sources, and to what extent the sources are salient and important for effectively managing a complex project might help with determining the tools, methods, and skills for not only IM but also for effectively dealing with project complexity in general (Remington et al. 2009; Maylor et al. 2008; Baccarini 1996). Project complexity has been known as one of the greatest factors of difficulty of project management (Remington et al. 2009; Baccarini 1996; Gidado 1996), and project failure (Parsons-Hann and Liu 2005), in the construction industry.  However, understanding project complexity is not an easy undertaking; the concept of project complexity is composed of many interrelated sub-concepts, and thus is complex in itself. Project complexity is multifaceted (Maylor et al. 2008), and it has many sources (i.e., dimensions)—“Not all projects are complex in the same way” (Remington et al. 2009). Due to this nature of project complexity, it is not clear which specific dimensions of project complexity can be addressed by IM, and how IM practices can help in dealing with those particular dimensions.    A number of researchers have attempted to define the dimensions of project complexity, and to assess the impacts that each of the project complexity dimensions has on project performance. Several research approaches have been proposed and used in this line of attempts, including framework development by aggregating the complexity factors found in the literature (Bosch-Rekveldt et al. 2011; Vidal and Marle 2008), Analytic Hierarchy Process (Vidal et al. 2011), workshops with industry practitioners (Maylor et al. 2008), surveys via interviews (Remington et al. 2009), and surveys and statistical analysis (Tatikonda and Rosenthal 2000 (factor analysis)).  In most of these proposed methods, complexity factors and dimensions were directly identified and assessed by research participants’ perceptions and evaluations. Therefore, the complexity dimensions that are defined by these methods are essentially a result of a human perceptual phenomenon (Liu 1999). This approach would be most useful in studying the subjective aspect of project complexity (e.g., perceived difficulty of project management). This view of project complexity is echoed in Fioretti and Visser’s (2004) statement that “complexity matters only because of the cognitive problems it gives rise to” (p.12), and Remington et al.’s (2009) statement that “[what matters is] how it is understood by the people who are affected….[therefore] complexity is most usefully conceptualised in cognitive terms.” However, this approach has limitations in producing objective knowledge about the interrelationships between project complexity factors and the levels of project complexity that may vary between different projects.        With this background in mind, this research aims to identify the dimensions of project complexity by looking at the main sources of variations in the project complexity factors in an empirical dataset. In other words, this research suggests that project complexity dimensions can be statistically defined based on empirical data for various project complexity factors. In the next section, a literature review on the definitions and factors of project complexity is provided. Then, the methodology and results of this research are presented in the following section. Subsequently, the implications of the results for IM as well as for general project management are discussed.   197-2 2 LITERATURE REVIEW ON PROJECT COMPLEXITY 2.1 Definitions of Project Complexity  Table 1 shows several selected definitions of project complexity found in the project management literature. As shown in this table, some disagreement and different—but, inter-related—perspectives exist in the definitions of project complexity in the literature (Bosch-Rekveldt et al. 2011; Vidal et al. 2011; Remington et al. 2009; Vidal and Marle 2008; Gidado 1996). One perspective to project complexity seems to be centered on managerial complexity (i.e., difficulty of management), which means that project complexity is the property of a project that makes managing the project difficult, and therefore requires more sophisticated managerial skills and systems. And, the other perspective seems to be more focused on the nature of the multitude of inter-related and interacting entities in projects (e.g., stakeholders, disciplines, trades, resources, processes, technologies etc.) and uncertainty. The concept of project complexity of interest in this research is in line with the latter view.  Table 1: Definitions of project complexity in the project management literature Authors Definition of Project Complexity Baccarini (1996) The degree to which a project consists of many varied interrelated parts with regard to any project dimension relevant to the project management process, such as organization, technology, environment, information, decision making and systems Gidado (1996) The measure of the difficulty of implementing a planned production work flow in relation to any one or a number of quantifiable managerial objectives  Vidal and Marle (2008); Vidal et al. (2011) The property of a project which makes it difficult to understand, foresee and keep under control its overall behaviour, even when given reasonably complete information about the project system Remington et al. (2009) The degree to which a project demonstrates a number of characteristics that makes it extremely difficult to predict project outcomes, to control or manage the project  2.2 Project Complexity Factors A large number of various sources—i.e., factors—of project complexity also have been discussed in the literature, and researchers have often attempted to categorize them. Table 2 shows several classifications of project complexity found in the project management literature. As shown in this table, many researchers have categorized and analyzed project complexity factors in different ways, while the most commonly used categories were organizational factors and technical factors. The project complexity factors that were looked at in this research can be categorized into these two groups from a broad perspective.      197-3 Table 2: Classifications of project complexity factors in the project management literature Authors (year) Classification of Project Complexity Factors Gidado (1996) ‘Within-tasks’ complexity factors (i.e. those that are inherent in the operation of individual tasks and originate from the resources or the environment) and ‘among-tasks’ complexity factors (i.e., those that originate from bringing different parts together to form a workflow)   Williams (1999) Structural uncertainty (i.e., the number and interdependence of elements) and uncertainty in goals and means Maylor (2003) Organizational factors (e.g., the number of people, departments, organizations, locations, nationalities, languages, and time zones involved in a project), technical factors (e.g., the level of novelty of any technology, system or interface), and scale/resource factors (e.g., the scale of the project and the size of the budget) Xia and Lee (2004)  Structural/ dynamic organizational factors (i.e., types of and number of relationships among hierarchical levels, formal organizational units, and specialization) and structural/dynamic technological factors (i.e., types of and number of relationships among inputs, outputs, tasks, and technologies)   Remington and Pollack (2007) Structural (i.e., interactions between many interconnected tasks), technical (i.e., unknown or untried design characteristics), directional (i.e., not agreed-upon goals and goal-paths), and temporal factors (i.e., volatility over the duration of the project) Geraldi and Adlbrecht (2007) Complexity of faith, complexity of fact, and complexity of interaction Vidal and Marle (2008) Technological factors and organizational factors, in terms of the size of project system, the variety of project system, independencies within project system, and context-dependence  3 RESEARCH METHODOLOGY 3.1 Data Collection A survey questionnaire was developed as a tool for collecting empirical data for project complexity factors (This survey questionnaire was developed as a part of a larger survey conducted by Construction Industry Institute Research Team 302 (CII RT 302). More information about this CII research project can be found in CII 2014.) Table 3 shows the project complexity factors included in the questionnaire. These factors were selected based on the result of the literature review and discussions with sixteen experienced project engineers/managers who participated in the questionnaire development process. The project characteristic factors that can be identified at a project’s onset were focused on in this research, and thus dynamic factors such as change orders and changes in the economic environment were excluded. As mentioned above, the complexity factors included in this research cover the organizational aspect as well as the technical aspect of project complexity, and are, especially, in line with the project complexity factors identified in Bosch-Rekveldt et al. 2011, Geraldi and Adlbrecht 2007, and Vidal et al. 2011.       197-4 Table 3: Project complexity factors included in the survey No. Project Complexity Factors  1 Cost Pressure  2 Schedule pressure 3 Extended/uncertain scope 4 Execution risks 5 Multiple owners (e.g., JV) 6 New technology 7 Large number of  suppliers/subcontractors 8 Multiple engineering centers 9 Government rules/regulations 10 Multiple general contractors/EPCs 11 Large number of engineered items 12 Multiple languages 13 Unfamiliar partners/collaborators 14 Not-aligned software/design standards between parties  15 Unclear geographical boundaries within project (“battery limit”) 16 Unclear requirements between involved parties 17 Unclear responsibilities between involved parties Then, the surveys were administered directly by the authors. The authors had either a face-to-face meeting or a web conference call with project engineers/managers/interface managers representing a project. The interviewees were asked to rate the project complexity factors on a 10-point scale by the question, “How much does this factor contribute to the complexity of your current project? (1 for the lowest and 10 for the highest)?” Also, the interviewees were asked to give responses solely based on their current project, or their most recently completed project if there was no current project to which they belonged.    The survey was conducted during 2013. In total, 46 projects owned or managed by US- or Canada-based organizations were studied in this survey, and the data that was valid for the analysis presented in this paper were collected for a total of 45 projects. The average rating of complexity factors ranges from 1.94 to 7.76 with a mean of 5.21 in the sample. This distribution of average complexity score shows that the sample included a wide range of projects in terms of project complexity. More detailed description of data collection method as well as the sample is available in Shokri et al. 2014.      3.2 Data Analysis The collected data were analyzed by principal component analysis (PCA). PCA is a widely used statistical method used for reducing the dimensionality of a dataset while retaining the variations in the dataset as much as possible. In PCA, the original data are transformed into a smaller number of new variables called principal components (PCs). Although there are several variants of the analysis method, one of the most common methods to define PCs is finding the eigenvectors of the correlation matrix of the original variables (Jolliffe 2002), which is the method used in this research. If the original variables are interrelated, the first few PCs—i.e., those with a large eigenvalue—retain most of the variations that were distributed over the original variables. If an appropriate number of PCs are selected, therefore, the PCs can retain most of the ‘information’ in the dataset, and that is why PCA can be used for reducing the number of dimensions in the dataset. In this research, PCs were used to define the major dimensions of project complexity extracted from the empirical dataset.   197-5 Table 4: Loadings for the first 6 PCs of project complexity*  Principal Components  1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 *Note. In bold type are the loadings whose absolute values are greater than half the maximum loading for the PC. The second PC has negative loadings for the variables such as ‘unclear responsibilities between involved parties’, ‘unclear requirements between involved parties’, ‘unclear geographical boundaries within project’, ‘not-aligned software/design standards between parties’, ‘multiple languages’, ‘unfamiliar partners/collaborators’, all of which are related to either the clarity of boundaries or communication between parties.  In the meantime, the second PC has positive loadings for many other variables such as ‘cost pressure’, ‘schedule pressure’, ‘execution risks’, ‘large number of suppliers/subcontractors’, ‘government rules/regulations’, and ‘large number of engineered items.’ Therefore, the second PC contrasts the uncertainty in boundaries and communication with the rest of the complexity factors, and implies that after the issues of the clarity of scope and of the newness of technology in the design have been accounted for, the main source of variation in the dataset is the uncertainty in boundaries/communication relative to the rest of the complexity factors.  The third PC has an outstanding loading for ‘multiple languages’, and ‘unfamiliar partners/collaborators’ follows in term of the size of loading. These two variables are thematically inter-related in general, and reflect that there are project participants with unfamiliar backgrounds. Therefore, this PC can be interpreted as ‘unfamiliarity with other project participants’. The fourth PC has negative loadings for the variables such as “multiple engineering centers’, ‘multiple general contractors/EPCs’, and ‘Large number of suppliers/subcontractors’, whereas it has positive loadings for ‘cost pressure’ and ‘execution risks’. Therefore, the fourth PC contrasts the multiplicity of stakeholders with cost pressure/risks, and implies that after the variances that are accounted for by the first three PCs, the main source of variations in the dataset is the multiplicity of stakeholders relative to the amount of cost pressure and execution risks.  197-7 The fifth PC has a relatively high level of negative loading for ‘large number of engineered items’ while having positive loadings for ‘cost pressure’, ‘multiple owners’, and ‘multiple general contractors/EPCs’. Therefore, the fifth PC contrasts the multitude of engineered items with cost pressure and the multiplicity of owners/general contractors, and implies that after the variances that are accounted for by the first four PCs, the main source of variations in the dataset is the multitude of engineered items relative to cost pressure and the multiplicity of owners/general contractors.  The sixth PC has the highest positive loadings for ‘government rules/regulations’ and ‘multiple general contractors/EPCs’, while having negative loadings for ‘extended/uncertain scope’, ‘large number of suppliers/subcontractors’, and ‘new technology’. Government rules/regulations and multiple general contractors/EPCs are related to the high-level program/project administration, and the sixth PC contrasts these factors with lower-level scope and technology factors. 5 DISCUSSION As shown in the previous section, PCA was used in this research for finding the main sources of variations in the dataset, which can be defined as major dimensions of project complexity. As a result of this method, 6 interpretable PCs were defined, and these PCs combined to explain more than 70% of the total variance in the dataset. In other words, the number (17) of original variables was reduced to the smaller number (6) of new variables by PCA, while 30% of the total variance was sacrificed. One should not take from this that the 30% of the total variance excluded in the 6 PCs are not important, but, given that the data were collected by surveys, and that the data are under the influence of random errors as well as the sample size, it seems acceptable to say that the 6 PCs capture the majority of the sources of variations in the dataset.  An interesting finding from this analysis is that the complexity factors that were highly rated for most of the projects, such as ‘cost pressure’, ‘schedule pressure’, and ‘execution risks’, did not have much variations in the dataset, and therefore contributed to the first few PCs relatively less than other factors. This shows that PCA is useful for distinguishing the complexity factors that actually vary between projects from the factors that are shared by most of the construction projects studied. Therefore, the results of PCA might be useful for specifically developing the managerial approaches to address the complexity factors that vary between projects and the managerial approaches to address the complexity factors that are common among projects.        The fact that the first PC is positively related with every original variable tells us that all of the original variables are inter-related to some extent, and the first PC can serve as a kind of composite indicator of the level of overall project complexity. This implies that some projects in the dataset are more complex in general than others, and that the score for the first PC can tell us overall how complex a project is. Therefore, this PC might be useful for determining the level of systematization of project management system required in a project. The project complexity dimensions defined by later PCs indicate more specific aspect (i.e., source) of project complexity and explain a smaller variance in the dataset. Noticeably, complexity dimensions defined by the PCs are related to organizational interactions (e.g., unclear scope of work for multiple stakeholders in the definition (1st PC), uncertainty in boundaries and communication (2nd PC), unfamiliarity with other project participants (3rd PC), and the multiplicity of stakeholders (4th PC)). These results hint that organizational interactions (i.e., interfaces between parties involved) are a major source of complexity in many projects. This implication is in line with observations/findings in previous research works. Vidal et al. (2009) reported that 11 project complexity drivers (i.e., factors) identified from their survey belong to the family of “project interdependencies”, and this takes up 61.1% of the entire project complexity drivers they identified. From this result, they also argued that their result helps justify the tools and methods that been recently developed to deal with interactions and interdependencies in projects.        Vidal and Marle (2008) stated that classical tools and methods of project management are not sufficient in dealing with the level of complexity in modern projects, and therefore, the current level of project complexity justifies the need for a new project management approach that can assist the existing ones, 197-8 such as IM. Similarly, Baccarini (1996) argued that project complexity should be treated by more efforts for integration such as coordination, communication, and control. Therefore, more advanced managerial skills and tools for interface management would be required to effectively manage complex projects, and in particular, it seems that such development of advanced skills and tools for project management should aim at those projects that have complex organizational structure and a large number of interfaces in it. An efficient implementation of managerial functions can influence the impact of project complexity on project performance (Gidado 1996), and therefore is expected to mitigate the adverse impact of complexity on performance in many construction projects. More research is strongly warranted in this area.    6 CONCLUSION Although many researchers have attempted to define project complexity and have investigated various factors that would affect complexity in construction projects, it is unclear what the dimensions of project complexity are and how they can be measured. This lack of knowledge about project complexity has been one of the main reasons that the construction industry has displayed great difficulty in effectively dealing with complexity in projects. Given this situation, this research aimed to define the dimensions of project complexity based on empirical data. To achieve this research objective, a questionnaire was developed  as a tool for collecting data for project complexity factors, surveys were administered in semi-structured interviews, and the collected data for 45 major construction projects were analyzed using PCA. As a result, 6 interpretable PCs were extracted from the dataset: ‘unclear scope of work  for multiple stakeholders in the definition and design of projects with new technology’ (1st PC), ‘the uncertainty in boundaries and communication relative to other complexity factors’ (2nd PC), ‘unfamiliarity with other project participants’ (3rd PC), ‘the multiplicity of stakeholders relative to the amount of cost pressure and execution risks’ (4th PC), ‘the relative multitude of engineered items’ (5th PC), and ‘the high-level program/project administration’ (6th PC). These dimensions of project complexity defined by the PCs tell us not only what the possible dimensions of project complexity are, but also which dimension explain the most variations in the dataset. Therefore, these complexity dimensions can help with assessing the level of complexity of a project at its onset and with determining the skills, tools, and systems that will be required to effectively cope with the sources of complexity in the project. Additionally, the analysis results hint that organizational interfaces should be effectively managed to prevent project failure in construction projects, and therefore support the need for IM in complex projects.  This research has several limitations that can be addressed in future research. First, the complexity factors included in this research are limited. More research efforts to investigate a broader set of project complexity factors are strongly warranted to gain a comprehensive view of various complexity factors that may play an important role in construction projects. Secondly, the sample size used in this research is modest. Since PCA is a data-driven method, a larger sample size may be required to gain a greater confidence with the PCs extracted in this research. Thirdly, the impacts of the project complexity dimensions identified in this research on project performances, such as cost and schedule, are yet to be investigated. Therefore, more future research efforts are strongly warranted to relate the levels of project complexity with project performance measures, and to investigate the effectiveness of approaches to address project complexity in this relationship.                Acknowledgements The research work presented in this conference paper was financially supported by the Construction Industry Institute, and is based on the results produced by the Construction Industry Institute Research Team 302 - Interface Management.      References Baccarini, D. 1996. The Concept of Project Complexity—A Review. International Journal of Project Management, 14(4): 201–204. 197-9 Bosch-Rekveldt, M., Jongkind, Y., Mooi, H., Bakker, H. and Verbraeck, A. 2011. Grasping Project Complexity in Large Engineering Projects: The TOE (Technical, Organizational and Environmental) framework. International Journal of Project Management, 29(6): 728–739. Chen, Q., Reichard, G., and Beliveau, Y. 2008. Multiperspective Approach to Exploring Comprehensive Cause Factors for Interface Issues.” Journal of Construction Engineering and Management, 134(6): 432–441. Construction Industry Institute. 2014. RS302-1 Interface Management, Construction Industry Institute, Austin, TX.   Fioretti, G. and Visser, B. 2004. A Cognitive Interpretation of Organizational Complexity. E:CO Special Double Issue. 6(1-2): 11-23. Geraldi, J.G. and Adlbrecht, G., 2007. On Faith, Fact, and Interaction in Projects. Project Management Journal, 38(1): 32–43. Gidado, K.I. 1996. Project Complexity: The Focal Point of Construction Production Planning. Construction Management and Economics, 14(3): 213–225.  Jolliffe, I. T. 2002. Principal Component Analysis, 2nd ed., New York, NY: Springer. Liu, A.M.M. 1999. A Research Model of Project Complexity and Goal Commitment Effects on Project Outcome. Engineering, Construction and Architectural Management, 6(2): 105–111. Maylor, H. 2003. Project Management, 3rd ed., Harlow, UK: FT Prentice Hall. Maylor, H., Vidgen, R., and Carver, S. 2008. Managerial Complexity in Project- Based Operations: A Grounded Model and Its Implications for Practice. Project Management Journal, 39: 15–26. Parsons-Hann, H. and Liu, K. 2005. Measuring Requirement Complexity to Increase the Probability of Project Success”, Proceedings of ICEIS 2005, Miami, 434–438. Remington, K. and Pollack, J. 2007. Tools for Complex Projects, Aldershott, UK: Gower Publishing company. Remington, K., Zolin, R. and Turner, R. 2009. A Model of Project Complexity: Distinguishing Dimensions of Complexity from Severity. Proceedings of the 9th International Research Network of Project Management, Berlin, 11–13. Shokri, S., Ahn, S., Lee, S., Haas, C. T., and Haas, R C. G. 2014. Current Status of Interface Management in Construction: Drivers and Impacts of Systematic Interface Management.” Manuscript submitted for publication. Tatikonda, M. V and Rosenthal, S.R. 2000. Technology Novelty, Project Complexity, and Product Development Project Execution Success: A Deeper Look at Task Uncertainty in Product Innovation. IEEE Transactions on Engineering Management, 47(1): 74–87. Vidal, L.A. and Marle, F. 2008. Understanding Project Complexity: Implications on Project Management. Kybernetes, 37: 1094–1110. Vidal, L.A., Marle, F. and Bocquet, J.C. 2011. Measuring Project Complexity using the Analytic Hierarchy Process. International Journal of Project Management, 29(6): 718–727. Williams, T.M. 1999. The Need for New Paradigms for Complex Projects. International Journal of Project Management, 17(5): 269–273. Xia, W. and Lee, G. 2004. Grasping the Complexity of IS Development Projects. Communications of the ACM, 47(5): 69–74.  197-10  Motivation for Interface Management in Construction : A Project Complexity Perspective  Seungjun (Jun) Ahn, PhD  Postdoctoral Fellow University of Alberta ICSC’ 15, Vancouver, June 9, 2015  Samin Shokri, PhD  Postdoctoral Fellow University of Waterloo  SangHyun Lee, PhD  Associate Professor University of Michigan  Carl T. Haas, PhD  Professor University of Waterloo  Ralph C. G. Haas, PhD  Professor University of Waterloo 2 Background: What is Project Complexity? §  “Not all projects are complex in the same way.” (Remington et al. 2009)   §  Definitions of Project Complexity         Authors Definition of Project Complexity Baccarini (1996) The degree to which a project consists of many varied interrelated parts with regard to any project dimension relevant to the project management process, such as organization, technology, environment, information, decision making and systems Gidado (1996) The measure of the difficulty of implementing a planned production work flow in relation to any one or a number of quantifiable managerial objectives  Vidal and Marle (2008); Vidal et al. (2011) The property of a project which makes it difficult to understand, foresee and keep under control its overall behaviour, even when given reasonably complete information about the project system Remington et al. (2009) The degree to which a project demonstrates a number of characteristics that makes it extremely difficult to predict project outcomes, to control or manage the project 3 Background: What are the Factors of Complexity? §  Project Complexity Factors  Authors (year) Classification of Project Complexity Factors Gidado (1996) ‘Within-tasks’ complexity factors (i.e. those that are inherent in the operation of individual tasks and originate from the resources or the environment) and ‘among-tasks’ complexity factors (i.e., those that originate from bringing different parts together to form a workflow)   Williams (1999) Structural uncertainty (i.e., the number and interdependence of elements) and uncertainty in goals and means Maylor (2003) Organizational factors (e.g., the number of people, departments, organizations, locations, nationalities, languages, and time zones involved in a project), technical factors (e.g., the level of novelty of any technology, system or interface), and scale/resource factors (e.g., the scale of the project and the size of the budget) Xia and Lee (2004)  Structural/ dynamic organizational factors (i.e., types of and number of relationships among hierarchical levels, formal organizational units, and specialization) and structural/dynamic technological factors (i.e., types of and number of relationships among inputs, outputs, tasks, and technologies)   Remington and Pollack (2007) Structural (i.e., interactions between many interconnected tasks), technical (i.e., unknown or untried design characteristics), directional (i.e., not agreed-upon goals and goal-paths), and temporal factors (i.e., volatility over the duration of the project) Geraldi and Adlbrecht (2007) Complexity of faith, complexity of fact, and complexity of interaction Vidal and Marle (2008) Technological factors and organizational factors, in terms of the size of project system, the variety of project system, independencies within project system, and context-dependence  4 Background: Interfaces in Construction §  Interfaces and Complexity   : Fragmented information communicated via telecommunication  (phone, email, etc.)  Interface Stakeholders 5 Background: Interfaces in Construction §  Interfaces and Complexity   Fabrication (fabricação) Module Fabrication (모듈 조립) Engineering (ingénierie) Engineering (अिभय%i'की) Construction Site 5 : Fragmented information communicated via telecommunication (phone, email, etc.)  6 Background: Interface Management in Construction “Interface Management is the management of communications, relationships, and deliverables among two or more interface stakeholders in a construction project”    - Construction Industry Institute, 2014. : Interactions §  Project Interfaces without Interface Management  Background: Interface Management in Construction Systematic Approach to IM •  Interface managers/coordinators •  IM Procedure •  Interface Agreement •  Web-based IM software  §  Project Interfaces with Interface Management Background: Interface Management in Construction : Interactions 9 Background: Research Questions and Objectives §  Which dimensions of project complexity would most require the use of IM practices?  §  To what extent can the current IM practices mitigate the adverse impacts of project complexity on project management? 10    Methodology Start Literature Review Interview Questionnaire Development Interviews  (46 Large-Scale Projects) Data Analysis (PCA & PCR) Result Interpretation End 11 §  Project Complexity Items (on a 10-point Likert Scale)              Method: Questionnaire Items No. Project Complexity Factors  1 Cost Pressure  2 Schedule pressure 3 Extended/uncertain scope 4 Execution risks 5 Multiple owners (e.g., JV) 6 New technology 7 Large number of  suppliers/subcontractors 8 Multiple engineering centers 9 Government rules/regulations 10 Multiple general contractors/EPCs 11 Large number of engineered items 12 Multiple languages 13 Unfamiliar partners/collaborators 14 Not-aligned software/design standards between parties  15 Unclear geographical boundaries within project (“battery limit”) 16 Unclear requirements between involved parties 17 Unclear responsibilities between involved parties 12 §  Interface Management Practices (Yes/No)  –  Formal procedures for interfacing with other parties –  Designated interface managers/coordinators –  Designated information system for IM §  Satisfaction with IM Performance (on a 5-point Likert Scale) Method: Questionnaire Items 13 §  Principal Component Analysis (PCA) –  Reduce the dimensionality of a data set consisting of a large number of interrelated variables –  Transform to a new set of variables, the principal components (PCs)     Method: Analysis  (Figures adapted from Jolliffe, 2002) 14 §  Principal Component Regression (PCR) –  PCs instead of original variables as predictors –  Addresses the multicollinearity problem –  Relate the dependent variable with a reduced number of independent variables     Method: Analysis  1st PC 2nd PC Pth PC 15 §  Projects Types in the Sample          Method: Sample 11111112234511170 2 4 6 8 10 12 14 16 18Water/WastewaterLRTNuclearHospitalDamMiningStationStadiumChemical	  ManufacturingMetals	  refining/ProcessingNatural	  Gas	  ProcessingOil	  RefiningPower	  GenerationOil	  Exploration/ProductionNumber	  of	  ProjectsDistribution	  of	  Interviewed	  Projects	  by	  Types16 §  Project Sizes in the Sample          Method: Sample 114145110 2 4 6 8 10 12 14 16<$500M$500-­‐$1B$1B-­‐$5B$5B-­‐$10B>$10BNumber	  of	  projectsDistribution	  of	  Interviewed	  Projects	  by	  Size17 §  Project Locations in the Sample          Method: Sample 18 §  Total Variance and Cumulative Variance Explained by PCs          Results from PCA §  Loadings for the PCs          19 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 §  Loadings for the PCs          20 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 “The unclear scope of work for multiple stakeholders and involvement of new technology” §  Loadings for the PCs          21 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 “The uncertainty in boundaries and communication relative to other complexity factors” §  Loadings for the PCs          22 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 “The unfamiliarity with other project participants” §  Loadings for the PCs          23 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 “The multiplicity of stakeholders relative to the amount of cost pressure and execution risks” §  Loadings for the PCs          24 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 “The relative multitude of engineered items” §  Loadings for the PCs          25 Results from PCA Principal Components 1 2 3 4 5 6 Cost pressure  0.17 0.39 0.38 0.47 0.53 -0.10 Schedule pressure 0.58 0.39 0.36 0.20 -0.18 0.04 Extended/uncertain scope 0.65 0.15 -0.03 0.24 -0.24 -0.40 Execution risks 0.38 0.34 -0.30 0.39 0.03 0.20 Multiple owners (e.g., JV) 0.53 -0.07 -0.19 0.10 0.46 -0.01 New technology 0.66 0.07 -0.34 0.07 -0.01 -0.33 Large number of suppliers/subcontractors 0.56 0.38 -0.14 -0.36 0.08 -0.40 Multiple engineering centers 0.59 0.19 0.24 -0.57 0.16 0.02 Government rules/regulations 0.38 0.39 0.25 0.21 -0.14 0.54 Multiple general contractors/EPCs 0.44 0.27 -0.28 -0.41 0.30 0.44 Large number of engineered items 0.35 0.34 0.18 -0.26 -0.55 0.02 Multiple languages 0.30 -0.35 0.77 -0.06 -0.03 0.06 Unfamiliar partners/collaborators 0.54 -0.27 0.45 -0.09 0.13 -0.18 Not-aligned software/design standards between parties  0.47 -0.52 0.21 0.12 0.16 0.07 Unclear geographical boundaries within project  0.53 -0.32 -0.19 -0.34 0.06 0.14 Unclear requirements between involved parties 0.72 -0.32 -0.34 0.27 -0.14 0.12 Unclear responsibilities between involved parties 0.74 -0.46 -0.18 0.16 -0.23 0.14 “The high-level program/project administration” §  Hypothesis in PCR          26 Results from PCR §  For Entire Data Set (N=45)          27 Results from PCR Predictors Beta Standard Error Significance Level The 1st PC “The unclear scope of work for multiple stakeholders and involvement of new technology” -0.340 0.148 0.041 The 5th PC “The relative multitude of engineered items” 0.367 0.145 0.028 Note. The significance level of the final regression model = 0.009; R2 = 0.276        §  For Projects without IM practices (N=23) §  For Projects with IM practices (N=22) 28 Results from PCR Predictors Beta Standard Error Significance Level 1st PC “The unclear scope of work for multiple stakeholders and involvement of new technology” -0.556 0.192 0.048 Predictors Beta Standard Error Significance Level The 5th PC “The relative multitude of engineered items” 0.586 0.182 0.008 Note. The significance level of the final regression model = 0.048; R2 = 0.309        Note. The significance level of the final regression model = 0.008; R2 = 0.344       §  Project complexity factors and PCs §  PCR for entire dataset and for subsets §  What has to be addressed by more advanced IM practices in the future? §  Limitations –  Complexity factors included in the analysis –  Subjective ratings –  Sample size –  Representativeness of sample –  Effectiveness of each IM practice  29 Discussion and Conclusions 30 Acknowledgement §             Construction Industry Institute® §  CII RT-302 members §  Project managers involved in the data collection  31 Thank you! Any questions?  (seungjun@ualberta.ca) 

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.52660.1-0076318/manifest

Comment

Related Items