1

PROJECT NETWORK INTERDEPENDENCY ALIGNMENT:

A NEW APPROACH TO ASSESSING PROJECT EFFECTIVENESS

Paul Chinowsky1, John E. Taylor2, and Melissa Di Marco3

ABSTRACT

The engineering and construction industry has evolved to a task-centric approach to evaluating

the effectiveness of projects. However, a narrow task-based view of project network logic

neglects the coordination of communication and knowledge exchanges across the project

organizational network. This paper departs from traditional approaches to introduce a new

approach to assessing project effectiveness that focuses on alignment of actual knowledge

exchanges with knowledge exchange requirements across task-organization network dyads. We

introduce a new modeling approach which we term as Project Network Interdependency

Alignment. Project Network Interdependency Alignment identifies potentially excessive or

insufficient communication and knowledge exchanges which can make projects ineffective. We

introduce the modeling approach and retrospectively validate it using a building renovation

construction project. The case study demonstrates that the approach can provide project

managers with the capacity to analyze task and organizational network interdependence on

projects and the critical capability to identify misalignments that impede project effectiveness.

Keywords: Interdependence; Organizational Issues; Project Networks; Social Network Analysis; Task

Networks 1 Associate Professor, University of Colorado at Boulder, Department of Civil, Environmental and Architectural Engineering, Boulder, CO 80309-0428; chinowsk@colorado.edu. 2 Assistant Professor, Columbia University, Civil Engineering and Engineering Mechanics Department, New York, NY 10027 USA; jt2411@columbia.edu. 3 Graduate Student, Dept of Civil Engineering and Engineering Mechanics, Columbia University, New York, NY 10027 USA; mkd2121@columbia.edu.

2

INTRODUCTION

The latter half of the 20th Century was witness to the expansive growth of information-centered

technologies. The development of the Internet, distributed databases, and n-dimensional graphics

are all manifestations of the information-centered society. The engineering-construction industry

energized its collective resources around this development to transform project information

processing to include web-based project extranets, 4-Dimensional CAD visualization and

modeling, and enhanced control processing. These information-centered project management

techniques emerged as advances to traditional critical path method (CPM) based control

strategies. Although these techniques were advances in information processing capacity, a

narrow task-based view of precedence logic may have diverted project managers from a central

element of project management, the coordination of knowledge in the project organizational

network. This paper departs from traditional approaches to introduce a new approach to

assessing project effectiveness that focuses on alignment of knowledge exchanges with

knowledge exchange requirements across task-organization dyads. We describe this new

approach as Project Network Interdependency Alignment.

Project Network Interdependency Alignment (PNIA) has its roots in contingency theory,

group dynamics, and network theory. The underlying emphasis of these approaches is to build

upon the capacity of the stakeholders within an organizational network to adapt to the demands

of a changing environment and adopt an interaction modality that is appropriate and fits the

environmental context. The fundamental assumption is that individuals or firms within the

network strive to achieve a collective outcome that benefits the whole, not the sub-optimization

of a few stakeholders. This assumption may hold true for projects involving one firm or a small

number of firms. However, in complex projects today there are often dozens of firms involved

which may have competing interests. Moreover, projects today are temporary organizations with

3

shifting participation from one project to the next. This relational instability makes achieving

collective alignment at the interdependent boundary between tasks and organizations extremely

challenging.

Project network interdependency includes both task interdependency within the project

task network and organization interdependency within the project participant network. At the

junction of these two interdependencies are the interstitial communication and knowledge

exchange requirements for effective project execution. This knowledge becomes particularly

critical as the industry becomes increasingly information-centered. For example, when architects

traditionally shared plans in the form of printed drawings the interstitial knowledge required to

enable the exchange was limited. An engineering or construction firm could readily interpret the

plans from the architect. However, as the sharing of electronic 2-D, and later 3-D, CAD models

became prevalent, engineers and contractors needed knowledge about how architects described

objects, defined layers, and they had to become capable of using a range of CAD tools. Working

with a variety of architects then became problematic as different architects used different

conventions and different CAD tools. As the degree of interdependency increases between tasks

or organizations, greater levels of communication and knowledge exchange are required from the

participants. The recognition of, management of, and adaptation to this interdependency requires

a new approach to assessing project effectiveness that diverges from a task-centric perspective to

a task-organization project network interdependency-centric perspective.

This paper introduces Project Network Interdependency Alignment as a means of

identifying omitted and potentially excessive or ineffective exchanges on projects. This network

interdependency focus is a significant departure from traditional project management in that we

introduce the impact of organizational dependence on achieving project effectiveness. The paper

4

begins with a description of the building blocks upon which this approach has been developed.

Given this background, we then describe the details of the PNIA approach and how it departs

from a traditional task-centric approach. We then illustrate and retrospectively validate the

PNIA approach on a building renovation construction project. The case study demonstrates that

PNIA can provide project managers with the capacity to analyze task and organizational network

interdependence on projects and, more importantly, the critical capability to identify

misalignments that create project vulnerability and impede project effectiveness.

BACKGROUND

Researchers in construction project management have achieved a relative consensus on the

importance of human factors or ‘people’ for successful project outcomes (Lechler, 2000). The

‘discovery’ that human and social capital is a key project success factor is of particular interest to

the current research focus based on its emphasis of network teams. Specifically, research into

the role that communication plays in project success is a precursor to the current research effort

into interdependency (Thomas, et al 1998). Additionally, research into the role that creating an

environment in which project participants can succeed as a foundation for the overall project

success provides a basis for analyzing the role that project networks play in project success

(Newell et al, 2002). The project network interdependency alignment approach builds upon this

focus on human capital by integrating two distinct research areas; social network analysis and

task network analysis.

5

Social Network Analysis

Social Network Analysis (SNA) has been an instrumental tool for researchers focusing on the

interactions of groups since the concept was introduced by Moreno in 1934 (Moreno 1960). At

the center of the concept is the basis that individuals or organizations exchange information

during the performance of any activity (Scott 1991; Haythornthwaite 1996). Given the premise

that any activity requires a transfer of information, the extension of this foundation is that these

exchanges can be mapped based on a graph format where actors and information exchange

become nodes and arcs within a graph (Wasserman and Faust 1994). The translation of these

interactions to a mathematical basis is the strength and validity of the network approach to

analyzing social interactions, communication and knowledge exchanges, and a range of

interactional phenomena. The ability to apply mathematical analysis to network information

exchange provides the researcher with established measurements for analyzing the effectiveness

and weaknesses of the group being studied (Alba 1982).

The social network concepts of cohesion, density, distances and relationships are

currently being applied by researchers in many diverse and distinct domains. Classic SNA

research focuses on sociological networks involving individuals in the workplace and their

exchange of information to complete tasks (Krebs 2004). Additional studies are focusing on

international relationships in areas such as research collaboration and international investment

(Krebs 2004). Construction engineering and management researchers have utilized network

analysis to examine issues such as the emergence of cultural boundary spanners in global

engineering services networks (Di Marco et al. 2009) and the structure and relationships within

project organizations (Chinowsky et al. 2008). Within these studies, the ability to map

participant relationships within a structure that can be visualized using network analysis software

6

is a significant benefit to network researchers. Specifically, work in network visualization

techniques is providing researchers with the ability to isolate relationships, visualize network

principles such as dominance, centrality, and egocentricity, and graphically present results that

were previously limited to mathematical matrices (Hanneman and Riddle 2005).

In the project management domain, the use of social network analysis has emphasized

project communications and the role communications can play in assisting coordination

functions (Pryke and Smyth 2006). However, communication is only one factor that can be

modeled with SNA tools and methods. As outlined in the Social Network Model for

Construction, human dynamics factors including reliance, trust, and values augment traditional

communication analysis when elevating the project analysis to include knowledge sharing and

high performance outcomes (Chinowsky, Diekmann, and Galotti 2008). Knowledge sharing has

significant importance to research into the multi-faceted collaboration requirements of complex

engineering design and construction work. In complex engineering task coordination,

communication may be insufficient. We need to understand the extent to which actual

knowledge is being exchanged at each task interdependent organizational dyad on the project.

The understanding of communication and knowledge exchange elements within a given project

network provide the capacity to identify coordination misalignments between organizations on

the project and their interdependent task assignment.

Task Network Analysis

Research on the analysis of tasks in a construction project as a network of activities began with

the development of arrow diagramming methods (Kelley 1961) and precedence diagramming

methods (Fondahl 1961) in the 1950s. These methods were developed as a response to the

7

increasing complexity of the projects that were being scheduled and introduced logic relating to

the dependency of a network of activities on each other (Archibald and Villoria 1967). The

calculations involved with each task network analysis method resulted in the identification of the

critical path. Much of the focus of research over the ensuing two decades focused on efforts to

develop heuristic methods to improve the accuracy and algorithms of critical path prediction

(Elmaghraby 1964), to include resources in task networks to understand and optimize time-cost

trade-offs (Clark 1961, Davis and Heidorn 1971), and to develop project control approaches

(Pilcher 1973).

Although research in these areas continues to draw attention from scholars, the last two

decades have been witness to a significant amount of research on project task network

scheduling using computerized systems. Research in these areas has been augmented with

information processing capacity to enable the application of fuzzy logic (Ayyub and Haldar

1984, Lorterapong and Moselhi 1996), integration with geographic information systems (Poku

and Arditi 2006) and three-dimensional computer aided design models (McKinney and Fischer

1998), and Monte Carlo simulation approaches (Lee 2005) in task network analysis research.

This research is critical to developing more accurate approaches to modeling and making

inferences from task networks. Yet, researchers have commented that an overly task-centric

approach to the management of projects neglects the important interface management function of

project management (Morris 1994). Managing interfaces becomes increasingly important as the

number of tasks in the project increase and researchers have shown that schedules managed by

construction project managers have larger numbers of activities than projects in other fields

(Liberatore et al. 2001). Researchers have examined the critical task network interfaces and

organizational network interfaces in construction projects. We need research that examines the

8

interfaces between task networks and organizational networks in order to develop a

comprehensive understanding of project networks and to effectively manage them.

PROJECT NETWORK INTERDEPENDENCY ALIGNMENT

The Project Network Interdependency Alignment approach entails three several steps. In this

section we first describe how we integrate task and organizational interdependence. We then

describe the first step of the modeling process which centers on the collection of communication

and knowledge exchange frequency data from project stakeholders. This technique is based on

the Social Network Model for construction (Chinowsky, Diekmann, and Galotti 2008). The

second modeling step focuses on evaluating the degree of task interdependency for each task

dyad in the network and attributing this level of knowledge exchange requirement to the project

stakeholders responsible for the task dyads. This integrated task and organizational network

diagram is referred to as the Project Network Interdependency Alignment model. The final step

involves evaluating the alignment between the Social Network Model of actual communication

and knowledge exchanges and the Project Network Interdependency Alignment model of

required communication and knowledge exchanges on the project. We present the Project

Network Interdependency Alignment approach in parallel with a description of the case project

we studied to validate the feasibility and accuracy of the approach.

Integrating Task and Organization Coordination

Project Network Interdependency Alignment is based on social network and task network

analytical perspectives. As described in the previous sections, these perspectives focus on the

criticality of coordination, knowledge sharing, and communication to effectively address the

9

complexity of task interdependency coordination requirements. Interdependency coordination

requirements are dependent on two factors, task coordination scope and organizational

coordination scope. Task coordination scope refers to the level of interdependency that exists

between each pair of tasks, or task dyad, in a project schedule. Organizational coordination

scope refers to the relationship between stakeholders executing each task dyad.

There are three levels of task interdependency which derive from the work of Thompson

(1967). The first of these levels, pooled interdependency, represents tasks that can be completed

at any time between the start and finish node of a project. These tasks are relatively rare in

complex design and construction projects as most tasks depend on other tasks. The next level,

sequential interdependence, indicates tasks that are dependent on other tasks being completed

before they can begin work. This allows coordination processes to occur such that a delay in a

precedent task can be communicated before the next task begins. The third, and most complex

level, is reciprocal interdependency where tasks are completed simultaneously and are dependent

on each other for both intermediate and final results. Many construction projects today are “fast-

tracked” which often involves forcing some activities to be executed concurrently. Collaboration

processes are significantly more difficult in this level as issues that arise during construction

related one task may adversely impact activities being concurrently executed.

Coordination requirements rise in complexity as the states evolve from pooled to

reciprocal interdependency. Coordination moves from having limited inter-task requirements in a

pooled interdependency state to sequential coordination, and finally to requiring concurrent

reciprocal coordination. At the core of this coordination requirement is the need for individuals

to extend beyond communications and information exchange to a focus on knowledge exchange.

Firms must collaborate to examine solution alternatives that effectively meet the demands of the

10

coordination requirements and ensure mutual benefits for each party. This level of exchange

requires the parties to exchange both explicit and tacit knowledge to explore solutions to inter-

task coordination issues as they arise. The difficulty with these exchanges is that in a relationally

unstable industry there is often insufficient past working interactions to support effective

collaborative task execution.

The PNIA approach is designed to determine whether the appropriate task-organization

alignment is in place to enable efficient exchanges and, hence, effective project execution.

Utilizing the underlying concepts described previously, PNIA determines if the appropriate level

of communication and knowledge exchange is occurring between the responsible participants in

tasks that are vulnerable to coordination misalignment. Utilizing an alignment measurement,

PNIA has the capacity to identify the potential areas where task-organization misalignment may

occur and where projects become vulnerable to collective coordination misalignment.

The Renovation Case Study

Coordinating the often divergent requirements of individual constituencies together with the

project plan is a core requirement of any project. However, renovation projects often bring an

additional set of coordination issues as new design requirements are integrated with existing

structures. Concurrent with these coordination issues is the need to balance schedule pressures

with the need to validate existing conditions within the structure. Because of the complex

coordination issues involved with a large renovation project, we selected such a case to examine

and retrospectively validate the PNIA concept. The case study was coordinated with a national

multi-sector builder (The Builder) who was given the responsibility to renovate four dormitories

on a university campus in a design-build delivery option. The total project budget was $58.3

11

million and each of the four dormitory renovations represented approximately one-quarter of the

total budget. Each of the four projects was scheduled to be completed in 14 months. The

specific context for the case study was the second of four dormitory renovation projects (The

Project).

The scope of The Project consisted of a complete renovation of the dormitory mechanical

systems and electrical systems as well as installing high speed internet lines. Additionally, the

project included a focus on sustainability where new windows, ventilation, and mechanical

controls were included in the project. A similar scope was established for each of the four

projects. The Builder created a core team of primary subcontractors and project management

staff that would complete and oversee all four projects. The first of the renovations served as a

trial project where the team focused on establishing efficient processes and team-building for the

entire set of four projects.

Given this context, we obtained a copy of the project schedule for The Project as well as

a list from the Builder of the primary personnel for each of the project stakeholders. The list of

participants for the project included 28 individuals and covered the design, construction, and

owner stakeholder teams. The combination of the personnel list and the project schedule

provided the basis for the PNIA analysis detailed in the following sections.

Collecting Data on Organization Network Exchanges

The first step in the PNIA approach is to collect organizational network data from project

stakeholders. For the case study project, this was enabled through the deployment of an

electronic, Web-based format of the Social Network Model survey to the project participants.

The survey contained questions that map to the levels in the Social Network Model. The intent

12

of the survey was to obtain data that corresponds to the perspectives of each individual in regards

to communication and knowledge exchange occurring between them and every other stakeholder

in the network. The 28 project personnel provided by the organization were each notified of the

survey and were given the opportunity to confidentially complete the survey. 26 of the 28

project personnel completed the survey in its entirety; the other 2 completed a portion of the

survey which was sufficient to include them in the PNIA case study and retrospective validation.

The study focused on two specific types of networks; communication exchange networks

(weekly, monthly and quarterly) (communication network is graphed in Figure 1) and knowledge

exchange networks (weekly, monthly and quarterly) (knowledge network is graphed in Figure 2).

These networks provide the insight into the continuity of exchanges. Continuous knowledge

exchange is a requirement for effective project execution. The 28 project participants were

analyzed to determine which of the other participants they interacted with on a weekly basis for

communication and knowledge exchange purposes. This analysis provided an overall network

and density measurement for each of the variables as well as an individual set of communication

dyads for use in the task interdependency analysis.

-----------------------------------------------------------------------------------

Insert Figure 1 & 2 here -----------------------------------------------------------------------------------

Assessing Organizational Network Exchange Requirements at Task Network Dyads

The goal of the second portion of the analysis was to develop a Project Network Interdependency

Alignment model. This involved determining the extent to which each organizational network

dyad, defined as two project team members who need to communicate and exchange knowledge

to execute the scheduled tasks they are responsible for, should be communicating and

13

exchanging knowledge based on their assigned tasks. We assumed that those organizational

network dyads responsible for tasks that are highly interdependent require the most significant

amount of communication and knowledge exchange. To determine the specific levels of

interdependence, a logical scheme was developed using network graphs.

First, the tasks in the case project were combined into all possible task dyad pairs,

defined as two tasks that have a defined relationship in the project schedule. A total of 428 task

dyads were identified. These were then each evaluated utilizing three task network

interdependency criteria: 1) their actual task interdependence derived from the schedule

precedence logic; 2) their time-space interdependence; and 3) their criticality determined from

whether the tasks were on the schedule critical path. Figure 3 shows the tasks in a network

diagram including the links indicating where task interdependency relationships existed. The

criticality of the tasks are illustrated by the size of nodes indicating their placement on the critical

path; the larger nodes are those tasks that if not completed would impact the completion of

subsequent tasks. For example, the plumbing rough-in and the fan coil installation tasks are

indicated with larger nodes as they both are on the critical path. The two nodes are also

connected with a link as they are indicated as having logical precedence dependency in the

project schedule.

For the first task network interdependency criterion we assumed a value of 0 for pooled

interdependence, 1 for sequential interdependence, and 2 for reciprocal interdependence for each

task dyad. This underestimates the additional coordination difficulty imposed by reciprocal

interdependence, however, for the purposes of assessing whether we could establish a project

network interdependency alignment model it is sufficient. For the second criterion, we attributed

a value of 1 if the activities would be completed in the same time and space since this requires

14

knowledge exchange even if the tasks themselves are not interdependent in terms of precedence

logic. If the activities did not occur at the same time and in the same space we attributed a value

of 0 to the task dyad on this criterion. The last criterion we considered was the criticality of the

tasks. A value of 1 was attributed if both of the activities were on the critical path for the project

and a value of 0 was attributed otherwise.

We then added up the values across the three criteria to arrive at the knowledge exchange

requirement for each task dyad. Returning to the plumbing rough-in to fan-coil example, the

tasks have reciprocal dependency, task-space dependency, and are both on the critical path, thus

giving this relationship the maximum score of 4 points. We then developed a task dyad matrix

which was used to create the task dyad communication and knowledge exchange requirement

network diagram contained in Figure 4. The communication and knowledge exchange

requirement level is indicated through the tie strength (line thickness) of the connecting links

between the task dyads. The thicker the lines, the more interdependent the tasks are. Once again,

the example is reflected with a strong tie between the plumbing rough-in and fan coil tasks in the

network.

-----------------------------------------------------------------------------------

Insert Figure 3 & 4 here -----------------------------------------------------------------------------------

Assessing Project Network Interdependency Alignment

In order to complete the project network interdependency analysis, we linked the organizational

network dyads with the task dyads to determine the level of knowledge exchange required. In

the plumbing rough-in to fan coil example, the responsible parties are p2 and p22 respectively.

For every task dyad with a knowledge exchange requirement value of at least 1, we determined

15

the organizational network dyad responsible for the two tasks. We then extracted the knowledge

exchange requirement values for that task dyad and associated it with the organizational network

dyad responsible for completing the interdependent tasks. However, in many cases the same

organizational network dyad was responsible for multiple task dyads. In this case, we summed

the knowledge requirement values in order to assess an overall level of knowledge exchange

required by the individuals in the organizational network dyad. For example, p2 and p22 are

responsible for multiple plumbing and fan coil tasks as well as other task relationships such as

plumbing and fire sprinkler installation. This multiple responsibility scenario results in a

cumulative effect on the knowledge exchange requirements.

We then graphed a Project Network Interdependency Alignment model (see Figure 5)

that normalizes and compares actual and required communication and knowledge exchange

patterns for the project network. To determine whether collaboration was appropriate, too little,

or too much, the model incorporates a quantitative process as follows:

1. The amount of actual coordination between each person dyad is normalized on a scale of

0 to 1 based on the amount of quarterly, monthly and weekly communication and

knowledge exchange that is identified in the SNA networks.

2. The amount of required coordination is normalized on a scale of 0 to 1 based on the

interdependency points described above.

3. The difference between the required and actual coordination amounts is then calculated

as the variance for each task-organization dyad.

4. The mean of the variances is then calculated together with the standard deviation to

determine the acceptable variation from the mean.

16

5. Acceptable coordination is then determined based on the rule that variance within one

standard deviation from the mean of the variances is considered appropriate, greater than

one standard deviation is excessive, and less than one standard deviation is too little.

-----------------------------------------------------------------------------------

Insert Figure 5 here -----------------------------------------------------------------------------------

Figures 5a, 5b, and 5c illustrate the results of this analysis for the case study. The thickness of

the link connecting each stakeholder in the network indicates the relative amount of actual versus

required communication and knowledge exchange. Figure 5a illustrates coordination

relationships that are appropriate for the project. These are the coordination links that are within

one standard deviation from the mean of the variances and are thus considered appropriate levels

of actual coordination in relation to the required coordination. Figure 5b illustrates the links

where there is relatively too much communication and knowledge exchange and, hence,

ineffective communication and knowledge exchange may be occurring. These links are those

where the actual minus the required was greater than one standard deviation from the mean of

the variances. Finally and most importantly, Figure 5c illustrates the linkages where there was

relatively too little communication and knowledge exchange. In this case the actual minus the

required was more than one standard deviation less than the mean of the variances. In other

words the actual communication and knowledge exchange was much less than was required by

the tasks for which the stakeholder pairs were responsible.

17

CASE STUDY RETROSPECTIVE VALIDATION

In order to retrospectively validate the above findings of the PNIA model and explore the

underlying influences behind the observed patterns of variance, we conducted interviews with

the Project Manager for The Builder. These interviews provided the opportunity to have an

external validation of the misalignment findings and obtain an insight as to why certain patterns

have taken shape in the coalition of the project. The initial question put to the Project Manager

concerned the overall performance of the Project. The Project Manager believed that the overall

project was going well, but two issues were of significant concern going forward for the next two

projects as identified in the following quotes:

• “We have not had a close relationship with the The Mechanical Engineer resulting in a

number of extra meetings to clarify work that was already decided upon,” and

• “The Owner representatives on the team are a significant bottleneck to progress. A

significant amount of time is spent updating the owner, but significant delays occur

waiting for responses to important issues.”

In terms of the former quote, this conflict is revealed in the PNIA network through the lack of

sufficient knowledge exchange between p5 (The Project Engineer) and p22 (The HVAC

Engineer). The networks indicate a strong need for exchange between these individuals, but

none is evidenced as occurring in the study. This coincides with the quote obtained from the

Project Manager.

In terms of the latter quote, the PNIA networks reveal that p20 (The Owner

representative for mechanical issues) and p14 (The Owner representative for the users) do not

have the required levels of coordination with the Project Engineer (p5), the HVAC Engineer

(p22), or the Architectural Designer (p10). This disconnect has a direct result in the project

18

experiencing both information delays and requirements for additional meetings by the Builder

with project stakeholders to provide the information that is not emerging from the owner

representatives. The Project Manager suggested that one reason for this may be the reluctance of

the owner to get involved with project issues due to contractual concerns. Additional issues

include the control of project execution and internal conflicts within the overall owner

organization.

These instances highlight the misalignment that can occur in task and organizational

network alignment. In each case, required interactions are missing due to misalignment between

network members in terms of; 1) actual and required communications, and 2) actual and required

knowledge exchanges. This misalignment is affecting individual tasks such as plumbing and

code approval since the organizational network dyads are not providing the knowledge required

to enable effective project execution. The PNIA model was able to accurately identify these key

communication and knowledge exchange misalignment issues that were occurring. Although

this is a single case study, the strong degree of agreement with the PNIA model results was

sufficient to retrospectively validate that a PNIA modeling approach can identify key project

network interdependency misalignments that affect project execution.

LIMITATIONS

Although the findings from the PNIA model were retrospectively validated with the renovation

case study, the approach does have several limitations that should be addressed in future

research. First, the PNIA modeling approach outlined in this paper includes communication and

knowledge exchanges between dyads in the network. The model identifies misalignments when

too little communication is occurring. However, in situations where organizational dyads have

19

significant amounts of trust or have worked together on similar tasks in the past, less

communication and knowledge exchanges may be needed to work together effectively. Future

research should consider the amount of trust and previous working experience at organizational

network dyads. Although not described in detail in the case validation section, there was one

organizational network pair the PNIA model indicated was coordinating less than the model

predicted was necessary and yet the tasks were completed effectively on the case project. The

project manager indicated that there was a long track record of the pair working together on

projects. Hence, examining the strength of trust and learning across organizations may be

required for to develop a comprehensive PNIA modeling approach.

Another limitation of the PNIA model is that it is based on normalized actual and

required collaboration and, as such, can only indicate whether there is relatively more or less

variance between actual and required. Future research should endeavor to develop scales for

actual and required communications such that the value derived from the variances is directly

interpretable. A third limitation that future research should address is the impact of an

organizational boundary being crossed at an organizational dyad. The current PNIA model does

not weight the amount of communication and knowledge exchange requirement to be greater

when the pair of stakeholders is from two separate organizations. Considering the additional

coordination required to work across organizational boundaries may represent an important

calibration for the PNIA model and should be considered in future research.

CONCLUSION

The PNIA approach to managing interdependency is an evolutionary step forward in project

management. The approach reintroduces the critical role that organizational networks play in

20

successful project execution. The PNIA approach provides a methodology to apply a

quantitative analysis of a project schedule and integrate the collaboration connections that exist

within the stakeholder network. This integration of communication and knowledge exchanges

with project task interdependencies addresses the central requirement that effective project

execution requires striking the right balance in communications and knowledge exchanges for

each organization and task pairing in the project network. This paper presents the PNIA

approach applied to a case study for a building renovation project. The paper illustrated how the

PNIA approach can be used to analyze a project schedule and identify potential disconnects

between project stakeholders in terms of collaboration. The case study illustrated how the lack

of appropriate coordination can result in potential project delays and miscommunications. As the

case study identified the same set of ineffective task-organization dyads as the PNIA approach,

we were able to use the case study to retrospectively validate the approach. Additional research

is required to refine, scale and further validate the PNIA modeling approach. However, the

initial evidence suggests that the project network interdependency alignment modeling approach

is a valuable line of inquiry to understand and improve project effectiveness.

ACKNOWLEDGMENTS

This material is based in part upon work supported by the National Science Foundation under Grant No.

0729253. Any opinions, findings, and conclusions or recommendations expressed in this material are

those of the authors and do not necessarily reflect the views of the National Science Foundation. The

authors also wish to thank the unnamed company and company representatives who gave up their

valuable time to participate in this research investigation.

21

REFERENCES

Alba, R. (1982). Taking stock of network analysis: A decade’s results. Research in the Sociology

of Organizations, 1, 39-74.

Archibald, R., and Villoria, R. (1967). Network-based Management Systems (PERT/CPM).

Wiley, New York.

Ayyub, B., and Haldar, A. (1984). “Project scheduling using fuzzy set concepts.” Journal of

Construction Engineering and Management, 110(2): 189-204.

Chinowsky, P., Diekmann, J., and Galotti, V. (2008). “The Social Network Model of

Construction,” Journal of Construction Engineering and Management, 134(10), 804-810.

Clark, E. (1961). “The optimum allocation of resources among the activities of a network,”

Journal of Industrial Engineering, Jan-Feb.

Davis, E., and Heidorn, G. (1971). “An algorithm for optimal project scheduling under multiple

resource constraints.” Management Science, 17(12): 803-816.

Di Marco, M., Taylor, J., and Alin, P. (2009). “Cultural boundary spanning in global project

networks.” LEAD 2009 Conference Proceedings, Lake Tahoe, CA.

Elmaghraby, S. (1964). “An algebra for the analysis of generalized activity networks.”

Management Science, 10(3): 494-514.

Fondahl, J. (1961). “Non-computer approach to the Critical Path Method for the Construction

Industry.” Technical Report No. 9, Stanford University, Stanford, CA.

Hanneman, R., and Riddle, M. (2005). Introduction to social network methods. Riverside, CA:

University of California, Riverside.

Haythornwaite, C. (1996). Social network analysis: an approach and technique for the study of

information exchange. Library and Information Science Research, 18(4), 323-342.

22

Kelley, J. (1961). “Critical path planning and scheduling: Mathematical basis.” Operations

Research, 9(3): 296-321.

Krebs, Valdis E. (2004). Managing the Connected Organization, Orgnet.com, www.orgnet.com.

Lechler, T. (2000). “Empirical evidence of people as project success factors.” In R. Lundin & F.

Hartman (Eds.), Projects as business constituents & guiding motives (pp. 217-227).

London: Kluwer Academic Publishers.

Lee, D. (2005). “Probability of project completion using stochastic project scheduling

simulation.” Journal of Construction Engineering and Management, 131(3): 310-318.

Liberatore, M., Pollack-Johnson, B., and Smith, C. (2001). “Project management in construction:

Software use and research directions.” Journal of Construction Engineering and

Management, 127(2): 101-107.

Lorterapong, P., and Moselhi, O. (1996). “Project-network analysis using fuzzy sets theory.”

Journal of Construction Engineering and Management, 122(4): 308-318.

McKinney, K., and Fischer, M. (1998). "Generating, evaluating and visualizing construction

schedules with 4D-CAD tools." Automation in Construction, 7(6): 433-447.

Moreno, Jacob L. (1960). The Sociometry Reader, Glencoe, IL: The Free Press.

Morris, P. (1994). The Management of Projects. Thomas Telford, London.

Newell, S., Scarbrough, H., Swan, J., Robertson, M., and Galliers, R. (2002). “The Importance of

Process Knowledge For Cross Project Learning: Evidence from a UK Hospital,”

Proceedings of the 35th Hawaii International Conference on Systems Sciences.

Pilcher, R. (1973). Appraisal and Control of Project Costs. McGraw Hill, London.

23

Poku, S., and Arditi, D. (2006). “Construction scheduling and progress control using

geographical information systems.” Journal of Computing in Civil Engineering, 20(50):

351-360.

Pryke, S. and Smythe, H. (2006). The Management of Complex Projects. Blackwell Publishing,

Ltd.

Scott, J. (1991). Social network Analysis: A Handbook, Sage: London.

Thomas, S., Tucker, R., and Kelly, W. (1998). “Critical Communications Variables.” Journal of

Construction Engineering and Management, ASCE, Jan/Feb 1998.

Thompson, J. (1967). Organizations in Action, McGraw Hill, New York.

Wasserman, S., and Faust, K. (1994). Social network analysis. Cambridge, MA: Cambridge

University Press.

24

Figure 1: Communication Exchange Network for Case Project

25

Figure 2: Knowledge Exchange Network for Case Project

26

Figure 3: Task Dyad Network for Case Project with Critical Path

27

Figure 4: Task Dyad Communication and Knowledge Exchange Requirements on Case Project

28

Figure 5: Project Network Interdependency Alignment Model

5a: Links indicating which relationships have satisfactory exchange levels.

5b: Links indicating which relationships have exchange levels that are too high compared to requirements.

5c: Links indicating which relationships have exchange levels that are too low compared to requirements.