Effect of Network Structure on Conflict and Project Value Creation

Abstract

This study explored the impact of network structure on conflict and project value creation. Network density and network centrality are two network structure dimensions. A survey was undertaken among professionals working in Chinese construction projects. A total of 308 surveys were analyzed using the structural equation model. The results revealed that network centrality has a negative impact on project value creation while network density has a positive impact. Network centrality has a negative impact on substantive conflicts but a positive impact on affective conflicts. The link between centrality and project value creation is weakened by substantive conflict but strengthened by affective conflict. This research gives a new direction for construction project governance and project value management. Furthermore, this research validates the constructive role of substantive conflicts, as well as the destructive impact of affective conflicts, thereby adding to the literature on conflict governance.

1. Introduction

With the increasing demand for infrastructure over the previous two decades, the quantity and scale of construction projects have increased [1]. Construction projects currently include the following characteristics: high investment, extended construction cycle, high complexity, dynamic environment, and multiple participating organizations (e.g., general contractors, subcontractors, design units, survey units, supervision units, suppliers) [2]. These organizations involved in the construction project vary in their professional expertise, skills, and capacities. No organization possesses all of the skills required to finish construction projects [3]. As a result, organizations build collaborative ties and a project network to complete construction projects. The project network is a vital route via which organizations obtain information, expertise, and resources [4]. However, different network structures (e.g., high-density network, high-centrality network) may have varied effects on construction project implementation. The structure of a project network, according to social capital theory, impacts whether and to what extent organizations embedded in the network gain resources (e.g., information, expertise, experience) [5]. Furthermore, the structure of the project network dictates how organizations incorporated in the network communicate with one another [6]. This may have an impact on organizational communication and cooperation, influencing the completion of construction project tasks and project value creation.

Construction projects are one-time, complex, uncertain, and characterized by incomplete contracts. Organizations engaged in the construction project network have varying levels of expertise, technology, and interest demands [7]. Therefore, organizational conflicts are common. In construction projects, interorganizational conflicts include substantive conflicts and affective conflicts. Affective conflict reflects organizational incompatibility, manifested as negative emotions between organizations, as well as explicit conflicting behaviors such as negative confrontation that arise when the organization feels uncoordinated or inconsistent with other organizations. Substantive conflict refers to different perspectives of an organization regarding the content of a project task, as well as different perspectives, contradictions, or disagreements of the organization regarding the specific process arrangements (e.g., schedule arrangement, project plan formulation, and resource allocation arrangement). If conflicts between organizations are not appropriately addressed or resolved in a timely manner, they may impede cooperation and the fulfillment of project duties [8]. Finally, this may have an impact on project value creation. Project value creation refers to achieving the goals of the project (i.e., quality, schedule, cost, and safety) and adding value to the owners and users of the construction project. According to social capital theory, the network structure (i.e., network density and network centrality) of construction projects may have an impact on interorganizational conflicts [5]. The high-centrality network contains organizations with significant power and authority. They may have control over the flow of resources (e.g., information, expertise, and experience) in the network [9]. However, this may distort information transfer and impede collaboration between organizations. This may lead to interorganizational disagreements during project implementation, resulting in interorganizational affective conflicts (i.e., interorganizational incompatibility manifested as negative emotions, as well as hostile behavior such as negative confrontation), which negatively affect project implementation and project value creation.

A high-density network is a highly interconnected communication network that promotes communication between organizations and increases the flexibility of their interactions, thus improving the efficiency of information dissemination and achieving task implementation coordination more quickly [10]. Simultaneously, a high-density network is typically more active, which aids in the sharing of knowledge and information between organizations, reduces interorganizational affective conflicts, and thus promotes the improvement of cooperation efficiency and effectiveness between organizations and project value creation [11]. The current research on network structure is mostly focused on soft elements in construction projects such as trust, communication, and leadership [11,12,13,14]. Few studies have been conducted to investigate whether network structure influences interorganizational conflicts and project value creation. This study investigates the impact of network structure on interorganizational conflicts and project value creation. This research splits network structure into two dimensions: network centrality and network density, and widens interorganizational conflicts into two types: substantive conflicts and affective conflicts. The research findings aid the project management team in better detecting the impact of network structure and in designing efficient project management strategies, ultimately promoting project value creation.

2. Literature Review

2.1. Network Structure

Over the last decade, social capital theory has been widely used to describe the relationship between organizational network structure (such as a trade network or a communication network) and organizational output (such as organizational performance) [15]. According to social capital theory, social capital is the whole of an organization’s actual and potential resources that are owned, available, and generated from its social interaction network [16]. It is widely assumed that social capital has three dimensions: (1) a structural dimension, which refers to the overall structure of the organization’s social relationship; (2) a relationship dimension, which refers to mutual respect and trust developed during the interaction between organizations; and (3) a cognitive dimension, which refers to the implicit resources that help organizations produce shared norms or collective goals [17]. Even with clarification of these three dimensions, social capital is a complex sociological structure, making it difficult to quantitatively study.

In the context of construction management, social capital theory has been applied in various aspects. Regarding project collaboration, social capital plays a pivotal role in facilitating cooperation among diverse stakeholders [15]. For instance, high levels of trust (a key element of relational social capital) between owners and contractors can lead to more open communication, reduced opportunistic behavior, and lower transaction costs. A study by Cheraghi et al. (2023) demonstrated that in construction projects where there was a strong foundation of trust among participants, disputes were resolved more amicably, and the overall project delivery was more efficient [2]. In terms of information sharing, the network structure provided by structural social capital enables the rapid dissemination of information [15]. Subcontractors with well-connected networks are more likely to access valuable technical knowledge and market information, which can enhance their operational capabilities and contribute to the project’s success. When it comes to resource integration, social capital allows construction firms to access external resources beyond their own organizational boundaries [9]. Through their social networks, firms can establish partnerships with research institutions to gain access to innovative construction technologies (cognitive social capital in the form of shared knowledge goals). This integration of resources not only improves the firm’s competitiveness but also has a positive impact on the overall project quality and sustainability.

The majority of extant network research focuses on assessing the structural dimension of social capital (such as network structure). Taking into account the characteristics of construction project networks [4,5,6,7], this study also focuses on the structural dimension of social capital to investigate the relationship between construction project network structure, project conflict, and project value creation. In reference to previous research on network structure, the two representative indicators of network centrality and network density were selected to reflect network structure [9,11,13]. Network density is one of the project network’s structural characteristic indicators [18]. Network density is commonly used to represent the closeness of relationships between stakeholders. In a construction project, stakeholders are linked and interact through cooperative relationships [5]. These collaboration links will progressively coalesce into a complex project network, influencing how organizations access resources and information [4]. The network density in the construction project network can be used to examine the degree of resource dissemination and measure the degree of contact of stakeholders in the project network [19]. Because of the homogeneity of stakeholders, it is advantageous for various stakeholders to create common aims and values through communication and cooperation. According to Zeng et al. (2022), high-density networks assist network members engage in more frequent activities [20]. Cao et al. (2017) proposed that increasing network density can better coordinate the behavior of network members [21]. Tang et al. (2019) argues that high-density communication networks aid in the achievement of organizational goals [22]. Therefore, increasing project network density may enhance the creation of trust between organizations and encourage stakeholders to access more information.

The degree to which an organization is placed at the core of the network is characterized as network centrality [23]. Network centrality is frequently used to express network participants’ status, influence, and reputation. High network centrality suggests that some actors are in the network’s focal point, controlling and coordinating information transmission and resource allocation [24]. Network centrality is typically divided into three categories: degree centrality, close centrality, and between centrality. Specifically, if a network actor is directly related to many other network members, the network actor has a high degree of degree centrality. If a network actor can connect to several other network actors via a short path, the network actor’s close centrality is high. Between centrality refers to the shortest path for one network actor relative to other network actors [25]. According to the three definitions of network centrality, network centrality primarily evaluates a network actor’s ability to access resources, control other actors, and avoid being controlled by other actors [26]. Because construction project resources are limited, highly centralized stakeholders play a controlling role in resource management. According to Rodan et al. (2004), network centrality influences managers’ management skills, which in turn influences project innovation performance [27]. Cappiello et al. (2020) investigated firm competitiveness in managing complicated projects using network centrality [28]. In general, network centrality is vital in coordinating and directing information transfer and knowledge sharing, and it has a significant impact on project value creation.

2.2. Social Network Analysis in Construction Project Management

Social network analysis (SNA) is an appropriate tool for examining network structure and network characteristics [4]. Existing research indicates that SNA is efficient in exploring the following areas: (1) interdependence between project-based organizations [6]; (2) interorganizational relationships [11]; (3) structural characteristics of project networks [12]; and (4) multi-level analysis of project networks [16]. Pryke (2004) emphasizes that from the perspective of stakeholders, the main focus of SNA research is to explore formal relationships between stakeholders and network structure characteristics [12]. This study focuses on the network density and network centrality of construction project networks. Network density is a fundamental indicator in SNA that quantifies the degree to which participants in a network are interconnected [18]. Mathematically, it is calculated as the ratio of the actual number of connections between participants to the maximum possible number of connections [19]. In high-density networks, participants have many direct connections, which facilitates the rapid dissemination of information, resources, and knowledge [4]. In the context of construction project networks, high-density networks can achieve seamless communication between stakeholders such as contractors, designers, and owners, potentially reducing information asymmetry and the occurrence of affective conflicts [20]. Network centrality focuses on the status and influence of individual actors in the network. It contains multiple dimensions [21]. Degree centrality calculates the number of direct connections that a participant has. A higher centrality degree indicates that participants are more embedded in the network. Betweenness centrality measures the ability of participants to control the flow of information between other participants, highlighting their role as a bridge in the network. Closeness centrality reflects the efficiency of a participant in a network reaching all other participants, typically measured by the average shortest path distance from the participant to other participants in the network [22].

Network density and network centrality are important components of SNA [20]. SNA provides a comprehensive perspective to examine the complex relationships and interactions within a construction project network. Network density provides an overview of the overall connectivity of a network, while network centrality explores the roles and influences of individual actors [9]. They help to understand how network structure shapes individual and collective behavior, which is closely related to the study of the impact of network structure on conflict and project value creation [10]. For example, high-density networks may generate more opportunities for collaboration, but if communication channels are not effectively managed, they can also increase the likelihood of conflicts. Stakeholders with high centrality typically have significant power in coordinating project activities, mediating disputes, and driving the overall value creation process. Stakeholders with high centrality can either serve as catalysts for resolving conflicts and enhancing project value creation or escalate conflicts due to their dominant positions [20].

The relationship between building information modeling (BIM), green building project management, and SNA is profound. BIM is a revolutionary digital approach that has transformed the traditional construction industry [17]. It is a digital representation of the physical and functional aspects of buildings, creating a comprehensive 3D model-based database that integrates architecture, engineering, and construction information [19]. SNA provides a powerful toolkit to explore the complex relationships and interactive networks among stakeholders in BIM-based projects and green building projects. In the context of BIM, SNA can be used to draw information flow networks between different participants. It helps identify key nodes or influential stakeholders who play a crucial role in disseminating BIM-related knowledge and coordinating BIM-based activities [21]. In green building project management, SNA can reveal the network structure of the stakeholders involved in sustainable practice implementation. It can identify leaders who drive the adoption of green strategies, as well as weak connections or communication barriers that may hinder the project from achieving green goals. Understanding these concepts and their interrelationships is crucial for exploring the impact of network structure on conflict and project value creation. The unique network structure in BIM integration projects and green building projects can have a significant impact on the occurrence of conflicts [17]. In BIM-based projects, conflicts may arise due to issues such as data ownership, differences in BIM proficiency among team members, or unclear communication channels within the BIM network [21]. Similarly, in green building projects, conflicts may arise from different interests and priorities among stakeholders, such as developers seeking to maximize profits while environmentalists prioritize sustainability. Meanwhile, in both cases, well-structured networks can promote value creation.

Integrated project delivery (IPD) is a collaborative project delivery method in which all major stakeholders, including owners, architects, engineers, contractors, and subcontractors, participate in early integrated collaboration [10]. It has the characteristics of sharing goals, risk sharing mechanisms, and jointly pursuing value optimization throughout the entire project lifecycle [7]. SNA is a powerful analytical tool that can be effectively applied to explore complex relational networks in IPD projects. In the IPD environment, multiple stakeholders such as owners, architects, engineers, and contractors work closely together, and SNA can be used to explore the information sharing network [10]. By applying indicators such as network density and centrality, researchers can determine which stakeholders are at the core of disseminating key project knowledge and which communication channels are most effective [19]. This helps ensure the smooth flow of information within the project team and reduces the possibility of misunderstandings that may lead to conflicts. In addition, SNA can detect potential bottlenecks or weak links in collaborative networks [22]. For example, it can identify stakeholders with limited connections to others whose isolation may hinder overall cooperation and slow down project progress [21]. By recognizing these issues, appropriate measures can be taken to strengthen the network and improve project implementation efficiency.

2.3. Project Conflict

Conflict is defined as incompatibility or confrontation [29]. Project conflicts primarily appear as interorganizational conflicts in construction projects. Interorganizational conflicts are defined as organizational contradictions or combative actions [30]. Construction projects often entail tens of millions of dollars in investment, a lengthy construction time, a variety of technology, and complex processes [4]. Furthermore, construction projects require a huge number of project participants, such as owners, contractors, design units, consulting units, supervisory units, and financial institutions [2]. These construction organizations have varying professional skills, core capabilities, and interest demands. These elements frequently cause project conflicts. Project conflicts can arise as a result of construction schedule and resource constraints, differing understandings of project plans, and disagreements in the priority of project objectives [31]. Furthermore, one-time construction processes, incomplete contracts, the project’s changing internal and external environment, and diverse cognition are all important variables that contribute to project conflicts [32].

In organizational management, conflict types often include affective conflicts involving relationships and emotions, as well as substantive conflicts involving tasks and monetary interests [20,33]. In construction projects, affective conflict reflects organizational incompatibility, manifested as negative emotions between organizations, as well as explicit conflicting behaviors such as negative confrontation, which arise when the organization feels uncoordinated or inconsistent with other organizations [34]. Substantive conflict refers to different perspectives of an organization regarding the content of a project task, as well as different perspectives, contradictions, or disagreements of the organization regarding the specific process arrangements (such as schedule arrangement, project plan formulation, and resource allocation arrangement) [35]. The impact of conflict on construction projects is multi-faceted. On the negative side, unresolved conflicts can significantly delay project schedules. Disputes over technical designs may require repeated revisions, extending the design phase and pushing back the start of construction [8]. Financially, conflicts often lead to cost overruns due to rework, legal disputes, and inefficiencies in resource allocation. In terms of quality, strained relationships among team members can result in a lack of attention to detail and corner-cutting, compromising the final product [5]. However, under certain circumstances, conflict can also have positive implications. Constructive task-related conflicts, for example, can stimulate innovation as different perspectives are debated, potentially leading to better-quality solutions.

2.4. Project Value Creation

The concept of project value creation is commonly used in fields such as economics, politics, and sociology [36]. The value creation phases in project management include the project implementation phase and the operating phase [37]. By removing non-value-added operations, a project’s value can be increased. Eliminating non-value-added operations lowers construction project investment and total lifecycle costs [38]. Construction project value creation can be achieved by optimizing project implementation procedures, improving project management models, and increasing task implementation efficiency [39]. The value creation must be viewed through the eyes of several project participants (such as owners, contractors, design units, supervision units, and users), rather than just the owners or contractors [36,37,38,39].

Three procedures are used to assess the value creation of a construction project. The first step is to provide value to the project, which involves meeting the project’s essential objectives (i.e., quality, duration, and cost) [40]. The second step is to add value to project stakeholders. The third step is to add value to the construction project’s users, such as raising user satisfaction [41]. Based on prior research and taking into account the features of the construction industry, this study defines project value creation in two dimensions: time dimension and participant dimension. In terms of time, the project’s value creation evaluation comprises both the project implementation and operation phases. Although different project participants may have different viewpoints on project value creation, their definition of project value creation is usually consistent during the implementation and operation stages of the construction project [42]. In this study, the evaluation indicators of construction project value creation include hard element and soft element dimensions. Hard elements include quality, cost, and schedule. Soft elements include cooperative effectiveness, trust, ability enhancement, and user satisfaction [43,44].

3. Research Hypothesis and Model

3.1. Research Hypothesis

3.1.1. Density and Conflicts

Network density reflects group cohesion. High network density promotes identity and trust among network actors, facilitating engagement and information exchange [19]. Previous research has shown that network density can influence the attitudes and behaviors of network actors, consequently affecting organizational performance [20,21,22]. A construction project is a temporary organizational cooperation network that is formed at the start of the project and disassembled at the end [2]. Unlike long-term cooperation networks (such as those in high-tech industries), organizations in transitory cooperation networks are frequently unfamiliar with one another, restricting interorganizational interaction [4].

A high-density cooperative network is a highly interconnected communication network, which promotes communication among organizations and increases the flexibility of interorganizational interactions, achieving the coordination required during project tasks and process implementation [20]. Furthermore, a high-density cooperative network is typically more active, which aids in the exchange of interorganizational knowledge and information, the reduction of affective conflicts, and the promotion of cooperation efficiency and effectiveness [22]. A high-density project network frequently possesses a wealth of knowledge and information. Due to differences in information and resources gained by project participants, viewpoints on project task layout may differ [31]. This may lead to discrepancies in task substance or contradictions in process arrangements amongst project participants, resulting in substantive conflicts [34].

According to social network theory, a higher network density indicates a greater number of connections and more frequent interactions among project stakeholders [17]. These intensive interactions facilitate the establishment of trust-based relationships, as repeated communication and cooperation help stakeholders understand each other’s values, working styles, and interests. In the realm of construction projects, when contractors, subcontractors, designers, and owners interact closely, they are more likely to resolve potential misunderstandings promptly [11]. For example, regular meetings and information exchanges can prevent minor disagreements from escalating into intense emotional confrontations. From the perspective of conflict management, positive relationships nurtured by high-density networks act as a buffer against affective conflicts [18]. When stakeholders trust and respect each other, they are more willing to adopt a collaborative attitude to address differences, rather than allowing emotions to dominate the situation [13]. In contrast, in a network with low density, limited interactions may lead to insufficient understanding, and when problems occur, stakeholders may lack the necessary emotional connection and communication channels to resolve them effectively, thus increasing the probability and intensity of affective conflicts [45].

H1a. 

Substantial conflicts are positively impacted by network density.

H1b. 

Affective conflicts are negatively impacted by network density.

3.1.2. Centrality and Conflicts

Network centrality describes the location of network actors, as well as their capacity to access and control network resources [15,23]. A highly centralized organization is positioned in the network’s core and has more opportunities to connect with other organizations [25]. A highly centralized organization means a strong project-participating organization that can promote task completion through their own influence and power [26]. Meanwhile, organizations with high centrality have more links, which might foster task resource integration and organizational cooperation [24]. This frequently decreases organizational inconsistencies about project tasks and leads to the development of consistent task resolution procedures, resulting in fewer interorganizational substantive conflicts.

Social network theory posits that entities with high network centrality possess greater control over resources, information, and decision-making power within the network [12]. In the context of construction projects, stakeholders with high centrality, such as main contractors or key design firms, are in a prime position to access critical project- related information and resources [20]. This advantage enables them to identify potential sources of substantial conflicts, such as disputes over task assignments, resource allocation, or technical specifications, at an early stage. From the perspective of conflict management, these central stakeholders can leverage their influential position and abundant resources to mediate conflicts effectively [8]. For instance, a general contractor with high network centrality can organize timely coordination meetings, share comprehensive project information among subcontractors, designers, and owners, and enforce unified standards and procedures [11]. By doing so, they can facilitate communication, align the interests of different parties, and promote consensus-building, thereby reducing the likelihood and severity of substantial conflicts. Construction projects typically include three types of deliverables: DB (i.e., design–build), DBB (i.e., design–bid–build), and IPD (i.e., integrated project delivery) [2,5,8]. Owners and contractors exhibit high centrality, but this pattern is contingent on the project delivery method [20]. Specifically, owners typically dominate design-phase centrality in DBB projects, contractors lead execution-phase centrality in DB projects, and both achieve balanced high centrality in IPD projects. Organizations in the construction project network have various interest demands [20]. The owner plans the project procedure in order to shorten the construction period. The contractor must, however, examine the sequencing of each task as well as the proper allocation of funds, personnel, and machinery [29]. The high centrality of owners and contractors can have an impact on information flow and resource sharing [25]. This frequently results in debates about project process arrangements and controversies over the project’s limited resource distribution, aggravating interorganizational affective conflicts. In the construction project network, although highly centralized organizations can play a role that crosses organizational boundaries, they may also create bottlenecks in information transmission and communication between organizations, because organizations with high centrality may distort information transmission and increase suspicion and distrust among other organizations [21,22]. During the construction project’s implementation, this will exacerbate affective conflicts between organizations [30].

H2a. 

Substantial conflicts are negatively impacted by network centrality.

H2b. 

Affective conflicts are positively impacted by network centrality.

3.1.3. Conflict and Project Value Creation

This study divides project conflicts into two types: substantive conflicts and affective conflicts [20]. Conflicts of various types have diverse effects on project value creation. Affective conflicts frequently have a negative impact on decision-making and interorganizational cooperation, eventually impeding project value creation [29]. There are three specific causes listed. First, affective conflicts might cause organizational members to conceal their genuine thoughts and beliefs, reducing communication and information exchange between organizations. Second, affective conflicts between organizations can lead to heated interactions and move the focus of the organization’s attention away from task content, ultimately breaking down collaboration between organizations. Third, affective conflict can diminish understanding between organizations, leading to an increase in confrontation [29]. Overall, affective conflicts impede efficient communication and cooperation among organizations, thus harming project value creation.

Substantive conflicts occur during the implementation of construction projects due to discrepancies in task content or procedure configurations [34]. A great number of fresh opinions and ideas sometimes emerge when addressing project tasks, resulting in substantive conflicts. It is widely assumed that substantive conflicts aid in the attainment of project objectives [30]. This is because substantive conflicts assist organizations in recognizing and exploring diverse perspectives and opinions, hence improving their understanding of project tasks and processes. Substantive conflicts can also improve the organization’s critical evaluation of task content, objectives, and execution techniques, hence driving task completion [32]. Furthermore, substantive conflicts can boost organizational cohesion, interorganizational linkages, and collaboration in the completion of difficult tasks [35]. This contributes to project value creation.

Organizational conflict theory indicates that under certain conditions, conflicts can stimulate constructive dialogue and innovation [20]. Substantive conflicts, which typically revolve around task-related issues such as technical approaches, resource allocation strategies, and project objectives, can bring diverse perspectives to the fore in construction projects [8]. For example, when different design teams hold conflicting views on the structural design of a building, one advocating for a more traditional and cost-effective approach while the other proposing an innovative but potentially more expensive solution, this conflict can trigger in-depth discussions [29]. Through such debates, stakeholders are forced to re-evaluate and justify their positions, leading to the discovery of new ideas that may integrate the advantages of both sides, such as a hybrid design that balances cost-effectiveness and innovation. Furthermore, substantive conflicts can also enhance the overall quality of project decision-making [32]. When conflicts arise, project teams are motivated to conduct more comprehensive research and analysis, involving multiple stakeholders in the decision-making process. This inclusive approach not only enriches the information base for decision-making but also increases the acceptance and commitment of stakeholders to the final decisions, reducing the likelihood of future disruptions and ensuring the smooth progress of the project towards value creation goals [20].

H3a. 

Project value creation is positively impacted by substantive conflict.

H3b. 

Project value creation is negatively impacted by affective conflict.

3.1.4. Network and Project Value Creation

A construction project is a temporary cooperative network consisting of numerous project participants [4]. An important topic in project management is the analysis of organizational output from a network perspective [20]. Construction projects are complex, uncertain, and multi-objective [8]. When making decisions, organizations must thoroughly comprehend the numerous risks, changes in the environment, and task complexity [5]. However, highly centralized organizations may restrict other organizations from expressing their thoughts and viewpoints, impacting interactions between organizations [21]. Previous research results have indicated that centrality has become a barrier to information transfer [24,35]. Organizations in the network’s core may also cause information distortion, influencing other organizations’ decisions and increasing their dissatisfaction. This will weaken interorganizational cohesion, reduce the efficiency and effectiveness of interorganizational cooperation, and have a negative influence on project value creation.

When compared to sparse networks, high-density project networks often have stronger cohesion, making it easier for organizations to create similar cognitive, behavioral, and normative standards [18]. These shared cognitive, behavioral, and normative criteria contribute to a more consistent basis of cooperation among organizations, facilitating project implementation and delivery [24]. Furthermore, high-density collaborative networks can reduce the opportunistic behavior of project participants to some extent [19]. This is because, in this highly interconnected cooperative network, an organization’s opportunistic behavior can swiftly spread throughout the network, harming the organization’s reputation. At the same time, other organizations in the network are more inclined to impose sanctions on the organization. As a result, opportunism is less likely to occur in a high-density project network [21]. It can be concluded that the advantages provided by high-density project networks are beneficial for project value creation.

H4a. 

Project value creation is positively impacted by network density.

H4b. 

Project value creation is negatively impacted by network centrality.

3.2. Research Model

The impact of project networks on conflicts and project value creation is still unknown. This study explores the impact of project networks on conflicts and project value creation using the input–mediator–output (IMO) model [20]. The input variable is network structure (i.e., network density and network centrality). From a theoretical perspective, network centrality and network density are two fundamental and widely-recognized indicators in network analysis [18,19]. Furthermore, these two indicators have been proven to be crucial in exploring the mechanisms of network-related phenomena across various fields, providing a solid theoretical basis for this study [11,17,20]. The mediating variables include two forms of conflict: affective conflict and substantive conflict. Project value creation is the output variable. Figure 1 depicts a theoretical model. The model specifically proposes that (1) network structure may lead to affective conflict and substantive conflict; (2) two types of conflict may affect project value creation; (3) network structure may affect project value creation; and (4) project conflict may act as a mediator between network structure and project value creation.

4. Methods

4.1. Questionnaire

A questionnaire was created to investigate the variables. Basic demographic information was analyzed, such as work experience and job position. The measurement items include network structure items, conflict items, and project value creation items. To develop measurement items, the three stages listed below are used. The first stage is to locate and cite items. The second stage is to improve items. Specifically, the items related to network structure were adapted from the research of Friedkin (1981), who conducted in-depth studies on network analysis in construction projects [18]. His research provides a solid theoretical foundation for measuring network centrality and density. For the measurement of project conflict, we referred to the scales developed by Gardiner & Simmons (1992) and Zeng et al. (2022), which have been widely used and validated in the field of project management [8,20]. These studies focus on distinguishing different types of conflict and their impact on project performance. As for project value creation, the items were inspired by the research of Li et al. (2020), Szatmari et al. (2021), and Huang et al. (2020), who proposed a comprehensive framework for evaluating project output from multiple perspectives [13,14,15]. We chose these studies because of their high relevance and authority in the respective research areas, and we made appropriate modifications to ensure the items were suitable for our specific research context.

The third stage involves discussions with experts in the construction industry [15]. The items were developed based on previous research. The goal of discussions with experts is to confirm the suitability of all items [19]. We solicited the opinions of 13 experts from owners, contractors, supervisory units, and design units. All experts had over 8 years of work experience in the construction project management field. They had participated in a wide range of projects, including large-scale infrastructure construction, commercial building projects, and residential development projects. In terms of education, 6 of them held a Ph.D. degree in construction management or related fields, while the remaining 7 had a master’s degree. Their educational backgrounds covered disciplines such as civil engineering, management science, and engineering economics. Regarding their positions, 5 experts were senior project managers in well-known construction enterprises, being responsible for overall project planning, execution, and control. Four experts served as professors at leading universities, engaged in both teaching and research in construction project management. The other 4 experts were consultants at professional construction consulting firms, providing strategic advice and technical support for various construction projects. This diverse group of experts ensured a comprehensive and professional perspective in the validation of our measurement items and research model. We came to an agreement on the items’ applicability after discussions. Except for the demographic information, all items in the questionnaire employed the Likert five-point scale (where 1 means “strongly disagree” and 5 means “strongly agree”).

To convert these qualitative opinions into numerical values suitable for structural equation modeling (SEM) analysis, we calculated the mean and standard deviation of the expert ratings for each path. The mean value was used as the initial estimate of the path coefficient [20]. This approach is in line with the common practice in the literature when incorporating subjective judgments into SEM. We used a two-stage approach in the SEM analysis. In the first stage, we ran the model with the initial path coefficient estimates from expert opinions as starting values. Then, in the second stage, we allowed the model to freely estimate the path coefficients based on the collected data, while still respecting the theoretical relationships indicated by the expert opinions [15]. This iterative process helped to optimize the path coefficients and ensure that they reflected both the expert knowledge and the empirical evidence from the data.

4.2. Pilot Test

Pilot tests were conducted in construction projects in Shanghai, Shandong Province, and Jiangsu Province. The interviewees included professionals from owners, contractors, design units, and supervisory units. A total of 230 questionnaires were distributed. After screening 147 collected questionnaires, 96 were judged to be valid. During the questionnaire screening process, we removed questionnaires that were not answered attentively or thoroughly. We then conducted normality tests on valid samples using a quantity–quantity plot. This is a common diagnostic technique for determining whether data has a normal distribution [20]. Figure 2 shows that the sample distribution is linear, thus satisfying a normal distribution.

In the pilot test, the corrected item total correlation (CITC) coefficient and Cronbach’s α coefficient were utilized to assess the reliability and validity of items [20,28]. The CITC value should not be less than 0.5. Cronbach’s α coefficient should not be less than 0.7. In addition, exploratory factor analysis (EFA) was performed. The results of EFA are presented in

Table 1. Results of EFA.

Variable	Item Code	Factor Loading	CITC Value	Cronbach’s Alpha Coefficient
Network Density	ND1	0.606	0.527	0.766
	ND2	0.752	0.588
	ND3	0.664	0.584
	ND4	0.760	0.592
	ND5	0.746	0.597
Network Centrality	NC1	0.668	0.551	0.813
	NC2	0.819	0.512
	NC3	0.727	0.602
	NC4	0.807	0.593
	NC5	0.652	0.676
Substantive Conflict	SC1	0.780	0.594	0.726
	SC2	0.710	0.597
	SC3	0.782	0.591
	SC4	0.779	0.585
	SC5	0.731	0.616
Affective Conflict	AC1	0.870	0.705	0.841
	AC2	0.814	0.744
	AC3	0.712	0.590
	AC4	0.791	0.612
	AC5	0.807	0.671
	AC6	0.729	0.631

, and the questionnaire is shown in

Table 2. Questionnaire.

Variables	Measurement Items
Network Centrality	1. Other organizations within this project rely on your organization for task coordination.
	2. Your organization is critical to the project’s success.
	3. Many organizations within this project have close relationships to yours.
	4. Your organization within this project frequently serves as a go-between.
	5. Among other organizations within this project, your organization has high influence.
Network Density	1. Your organization within this project has numerous direct relationships with other organizations.
	2. Organizations within this project are closely tied to each other through the project network.
	3. Many organizations within this project exhibit a high degree of cohesion.
	4. The project network is extremely active.
	5. The project network’s organizations within this project have an excellent cooperation relationship.
Affective conflict	1. There is a lot of emotional tension among organizations.
	2. There are several clashes among organizations.
	3. There is a lot of nonconformity among organizations.
	4. One party dislikes the other party’s style.
	5. Organizations have different cultures.
Substantive conflict	1. Organizations actively express their views.
	2. Work-related issues are frequently discussed among organizations.
	3. Partners present various perspectives on project task.
	4. Partners frequently provide suggestions for your work.
	5. Partners have differing viewpoints on process difficulties.
Project value creation	1. The quality, cost, and duration targets have been achieved.
	2. The resource use efficiency is high.
	3. This project has a favorable impact on users.
	4. The project process has received great marks from the partners.
	5. Interorganizational trust has improved.
	6. Partners look forward to future cooperation.

.

4.3. Data Collection

Nonprobability sampling was adopted in this study. This method is widely used in construction management studies [15]. The survey samples are from various construction projects in Shanghai, Jiangsu Province, and Shandong Province, and include technical personnel, as well as middle and senior management personnel from owners, contractors, supervisory units, and design units. We distributed 800 questionnaires by email and express delivery to technical and middle and senior management professionals, and 308 questionnaires were valid, yielding a response rate of 38.5% (308/800), which is consistent with the 30–40% benchmark used in most project management research [35].

Table 3. Sample information.

Characteristic	Category	Frequency	%
Project category	Highway project	83	26.9
	Railway project	91	29.7
	Industrial park project	134	43.4
Work position	Project engineer	115	37.3
	Department manager	79	25.6
	Professional manager	86	27.9
	Project manager	28	9.2
Job experience	<8 years	109	35.4
	8–18 years	151	49.0
	>18 years	48	15.6
Project participating organization	Owner	81	26.3
	Contractors	98	31.8
	Design units	80	26.0
	Supervision units	49	15.9

shows the sample information. Several factors contributed to the low response rate. From an objective perspective, the construction project domain poses unique challenges for sample collection. Construction projects are often large-scale, complex, and time-consuming endeavors, with a limited number of ongoing projects at any given time. Many enterprises are also highly sensitive about disclosing project-related information due to business confidentiality and competitive concerns, which significantly restricts our access to potential respondents. Subjectively, during the survey process, there might have been issues such as overly complex questionnaire design, which could have discouraged some participants from completing the survey. Additionally, the timing of the survey might not have been optimal for some busy project managers and stakeholders, leading to a lower willingness to participate.

4.4. Factor Analysis

Confirmatory factor analysis (CFA) can confirm the applicability of the items [46,47]. The CFA was calculated using AMOS 21.0. CFA created the construct reliability (CR) factors. The item that standardizes factor loads less than 0.6 was removed. The consistency between items is reflected in the CR. The CR value should be more than 0.5. The average variance extraction (AVE) method was employed to assess the convergence validity. If the AVE value is more than 0.5, the variable’s item has strong convergence validity. Fit metrics such as the ratio of chi square statistics to degrees of freedom (x²/df), the comparative fit index (CFI), and the incremental fit index (IFI) were used to test the model’s quality of fit. Specifically, the x²/df should be less than 3. The RMSEA should be less than 0.08. CFI and IFI should all be greater than 0.9 [20]. The results of CFA are presented in

Table 4. Results of CFA.

Variables	CR	AVE	Fit Indexes
Network density	0.81	0.71	$x^2/\mathrm{d}\mathrm{f}=2.63$; RMSEA = 0.063; GFI = 0.92; AGFI = 0.95; NFI = 0.91; IFI = 0.92; CFI = 0.94
Network centrality	0.82	0.66	$x^2/\mathrm{d}\mathrm{f}=1.90$; RMSEA = 0.072; GFI = 0.93; AGFI = 0.92; NFI = 0.93; IFI = 0.94; CFI = 0.91
Affective conflict	0.72	0.65	$x^2/\mathrm{d}\mathrm{f}=1.76$; RMSEA = 0.065; GFI = 0.92; AGFI = 0.93; NFI = 0.94; IFI = 0.96; CFI = 0.95
Substantive conflict	0.73	0.63	$x^2/\mathrm{d}\mathrm{f}=1.85$; RMSEA = 0.067; GFI = 0.92; AGFI = 0.91; NFI = 0.93; IFI = 0.91; CFI = 0.94
Project value creation	0.78	0.67	$x^2/\mathrm{d}\mathrm{f}=1.96$; RMSEA = 0.063; GFI = 0.94; AGFI = 0.92; NFI = 0.95; IFI = 0.91; CFI = 0.93

. While the above framework posits causal relationships between network structure, conflict, and project value creation, empirical validation is critical to confirm these hypotheses. The following section outlines the research design to test the theoretical model using SEM.

5. Model Test of the Theoretical Model

5.1. Structural Equation Model Test

SEM is the core analytical tool for testing the hypothesized relationships between network structure, conflict, and project value creation [34]. Bootstrap, a resampling technique, was employed to confirm the mediating role of conflict [20]. Unlike traditional Sobel tests, bootstrap does not assume normality of indirect effects, making it more reliable for our context [35]. These two methods played a crucial synergistic role in the analysis of this study. Specifically, SEM estimates the direct effects between latent variables, while bootstrap method is used to estimate indirect effects.

SEM is a popular method for investigating the correlations between variables in construction project management research [48]. The theoretical model was tested using SEM in this study. The fitting indications matched the standards. The x²/df ratio was 1.81. The RMSEA was 0.071, which is less than the criterion of 0.08. CFI and IFI are all greater than 0.9. Figure 3 and

Table 5. Results of SEM analysis.

Hypotheses	Coefficient	Critical Ratio	Standard Error	p Values
H1a	0.12 *	2.379	0.043	0.017
H1b	−0.16 ***	−4.078	0.040	0.000
H2a	−0.14 *	2.205	0.046	0.026
H2b	0.68 *	2.375	0.041	0.017
H3a	0.27 ***	5.793	0.040	0.000
H3b	−0.31 ***	−5.640	0.051	0.000
H4b	−0.12 **	−2.596	0.059	0.008
H4a	0.18 *	2.307	0.047	0.031

Notes: Abbreviations: ND, network density; SC, substantive conflict; AC, affective conflict; NC, net- work centrality; PVC, project value creation. * p < 0.05, ** p < 0.01, *** p < 0.001.

illustrate the model test results.

The SEM test results are shown in Table 5. First, network density positively impacts substantive conflict and project value creation, providing support for H1a and H4a, respectively (ND → SC, 0.12; ND → PVC, 0.18). Second, the impact of network centrality on substantive conflict and project value creation is negative, providing support for H2a and H4b, respectively (NC → SC, −0.14; NC → PVC, −0.12). Third, the impact of density on affective conflict is negative (ND → AC, −0.16), while the impact of centrality on affective conflict is positive (NC → AC, 0.68). This provides support for H1b and H2b, respectively. Fourth, the impact of substantive conflict and affective conflict on project value creation is positive and negative, respectively (SC → PVC, 0.27; AC → PVC, −0.31), which supports H3a and H3b. Furthermore, the direct, indirect, and total effects of indicators on project value creation are shown in

Table 6. Direct, indirect, and total effects of indicators on project value creation.

Indicator	Direct Effect	p-Value	Indirect Effect	p-Value	Total Effect	p-Value
Network Density	0.18 *	0.03	0.32 **	0.01	0.50 ***	0.000
Network Centrality	−0.12 **	0.008	−0.25 *	0.03	−0.37 **	0.001
Substantive Conflict	0.27 ***	0.000	—	—	0.27 ***	0.000
Affective Conflict	−0.31 ***	0.000	—	—	−0.31 ***	0.000

Notes: * p < 0.05, ** p < 0.01, *** p < 0.001.

.

5.2. Sub-Sample Analysis

This study stratified the sample into three groups based on stakeholder count (measured as the number of organizations involved): small projects (<10 stakeholders, n = 28); medium projects (10–20 stakeholders, n = 35); large projects (>20 stakeholders, n = 17). This study ran separate SEMs for each group. The results show that the path coefficients of each group are significant (see

Table 7. Sub-sample SEM results.

Path	Small Projects (n = 28)	Medium Projects (n = 35)	Large Projects (n = 17)	Overall Sample (n = 80)
Network Density → Affective Conflict	−0.12 **	−0.21 **	−0.25 *	−0.16 **
Network Centrality → Substantive Conflict	−0.22 *	−0.26 **	−0.18 *	−0.14 *
Affective Conflict → Project Value Creation	−0.35 **	−0.30 *	−0.28 *	−0.31 ***
Substantive Conflict → Project Value Creation	0.18 *	0.22 *	0.25 **	0.27 **
Network Density → Project Value Creation	0.22 **	0.11 *	0.16 *	0.18 *
Network Centrality → Project Value Creation	−0.17 *	−0.14 **	−0.18 *	−0.12 **

Notes: * p < 0.05, ** p < 0.01, *** p < 0.001. Stakeholder count groups: small = <10 stakeholders; medium = 10–20 stakeholders; large = >20 stakeholders.

).

5.3. Measurement Invariance Testing

This study conducted multi-group confirmatory factor analysis (MGCFA) to ensure measurement equivalence across project groups. The results show configural, metric, and scalar invariance (CFI > 0.90, RMSEA < 0.08), confirming that “network structure”, “conflict”, and “project value creation” are measured consistently across small, medium, and large projects (see

Table 8. Multi-group confirmatory factor analysis (MGCFA).

Invariance Type	CFI (Network Structure)	RMSEA (Network Structure)	CFI (Interorganizational Conflict)	RMSEA (Interorganizational Conflict)	CFI (Project Value Creation)	RMSEA (Project Value Creation)
Configural Invariance	0.92	0.04	0.93	0.06	0.94	0.05
Metric Invariance	0.91	0.04	0.90	0.03	0.92	0.02
Scalar Invariance	0.93	0.03	0.91	0.03	0.92	0.04

).

5.4. Mediation Test

After the SEM analysis, this study conducted a mediation effect analysis [49,50]. The bootstrap sample size was set to 3000 [51]. The results (

Table 9. Results of mediation test.

Source	Coefficients		CI
	Estimate	Boot Standard Error	Lower	Upper
Substantive conflict
Between network density and project value creation
	0.125	0.026	0.160	0.271
Between network centrality and project value creation
	−0.272	0.021	0.115	0.292
Affective conflict
Between network density and project value creation
	−0.430	0.019	0.102	0.241
Between network centrality and project value creation
	0.319	0.034	0.131	0.304

) show that (1) substantive conflict mediates the relationship between network structure and project value creation (0.125, CI = [0.160, 0.271]; −0.272, CI = [0.115, 0.292]); and (2) affective conflict mediates the relationship between network structure and project value creation (−0.430, CI = [0.102, 0.241]; 0.319, CI = [0.131, 0.304]). According to the above results, project conflict serves as a mediator.

6. Discussions

6.1. Network and Conflicts

According to the research results, network centrality promotes affective conflicts while decreasing substantive conflicts. On the contrary, network density increases substantive conflicts while decreasing affective conflicts. It can be seen that network structure has a diverse impact on conflicts. The finding that network centrality exacerbates affective conflicts is consistent with the research conclusion of Zhao and Jiang (2021), which suggests that high-centrality networks not only lead to power struggles, but also to information distortion [52]. This creates circumstances for the escalation of affective conflicts. High-centrality organizations have influence and authority in the construction project, not only affecting interactions between organizations but also managing information transfer, leading to distrust among different organizations throughout project implementation [53,54]. This results in affective conflicts throughout project implementation. The result that network centrality reduces substantive conflicts is similar to the research conclusion of Wang et al. (2015), implying that high-centrality networks encourage consistent activity among project participating organizations [55].

The findings of this study show that network density increases substantive conflicts while decreasing affective conflicts. This is because information and expertise are disseminated quickly and efficiently in high-density project networks, which promotes interorganizational interaction and the resolution of affective conflicts [56]. A densely connected network promotes information sharing inside the network and the establishment of common normative standards between organizations, hence boosting cooperation among project participants [57]. Furthermore, tightly connected project networks can improve network flexibility and reduce tension between organizations. This is due to the fact that densely connected networks can improve interorganizational interaction, communication, and trust, as well as promote the diversity of resources and information in the network [58]. A highly connected project network also aids in the generation of distinct network clusters. Different knowledge bases, professional competencies, and varying interest demands among network clusters might contribute to substantive conflicts.

To provide a deeper interpretation of the research findings, this study employed a qualitative interview design. Firstly, this study identified 11 interview participants, including project managers, contractors, designers, and subcontractors, as their different roles and perspectives in the project network can provide comprehensive insights [20]. Secondly, semi-structured interview questions were developed. These questions were open-ended and aimed to explore various dimensions related to the research question. For example, we asked participants to describe how they perceive the network density and centrality within their projects. We inquired about specific instances of substantial and affective conflicts they had experienced, especially in situations where network density was high or certain stakeholders had significant centrality. The questions also covered how the network structure influenced the emergence, development, and resolution of these conflicts. Additionally, we asked about their understanding of the mechanisms through which network density and centrality might impact different types of conflict.

Based on our preliminary analysis of the interview data, we can already propose some initial findings regarding the internal logic. It appears that high network density may suppress affective conflict primarily by enhancing trust between organizations. Frequent interactions in a dense network provide more opportunities for stakeholders to demonstrate reliability, fulfill commitments, and understand each other’s values and working styles, thereby fostering trust [13]. This increased trust, in turn, serves as a buffer against emotional tensions, reducing the likelihood of affective conflicts [31]. Based on our preliminary analysis of the interview data, several key findings emerged. When it came to network centrality, participants described situations where stakeholders with high centrality often had significant control over resources and decision-making. This power concentration sometimes led to dissatisfaction and resistance from other parties, triggering affective conflicts, such as disputes over power and influence. In terms of substantial conflicts, the dominance of centrally-positioned stakeholders could result in less diverse perspectives being considered, potentially leading to suboptimal decisions and a decrease in task-related conflicts. In conclusion, qualitative interviews provided us with valuable insights into the complex relationships between network structure, substantial conflicts, and affective conflicts.

6.2. The Impact of Project Conflicts

The findings of this study show that substantive conflicts have a positive impact on project value creation, while affective conflicts have a negative impact. These results are consistent with prior studies, which revealed that substantive conflict, as a constructive conflict, improves organizational performance [59,60]. In addition, this study’s findings highlight project conflict as a mediator between network structure and project output. Affective conflict has a stronger mediating effect than substantive conflict.

Affective conflicts commonly impede interorganizational cooperation and project task completion, limiting project value creation [61]. Furthermore, affective conflicts can shift an organization’s focus away from project tasks and toward interorganizational relationships. This can have an impact on organizational cognitive function, interorganizational cooperation, and task implementation, as well as cause interorganizational conflict and impede project task implementation, resulting in a negative impact on project value creation [62]. Substantive conflicts increase the generation of new ideas, which speeds up the settlement of project problems [63]. Substantive conflicts can also lead to the evaluation of multiple disciplinary areas, as well as in-depth thinking and creative ideas on project tasks. This can promote interorganizational cooperation, and thereby project value creation.

To provide a deeper interpretation of the research findings, this study employed a qualitative interview approach. Firstly, this study identified 11 interview participants, including project managers, contractors, designers, and subcontractors. Each of these roles plays a distinct part in project value creation and conflict management, and their varied perspectives provided a comprehensive understanding of the research topic [20]. Semi-structured interview questions were developed. These questions are open-ended and aim to explore various dimensions related to the research question. For example, we asked participants to describe specific instances of substantive conflict they had encountered, such as disputes over technical specifications or resource allocation, and how these conflicts were resolved. We inquired about the immediate and long-term impacts of such conflicts on the project, including whether they led to innovations, improved decision-making, or enhanced the overall quality of the project, thereby contributing to value creation. Regarding affective conflict, we explored how emotional tensions, personality clashes, or power struggles within the project team affected communication, collaboration, and ultimately, project performance and value. The questions covered stakeholders’ definitions and evaluations of project value creation and how they perceive the role of conflicts in this process.

Based on our preliminary analysis of the interview data, we found that substantive conflict often acts as a driver for project value creation. Stakeholders frequently reported that intense discussions and disagreements over technical or operational issues led to the discovery of more efficient construction methods, better-optimized resource allocation plans, and innovative design solutions [2]. These outcomes directly contributed to cost savings, improved project quality, and enhanced market competitiveness, thereby increasing project value. On the contrary, affective conflict was consistently described as a hindrance. The participants noted that emotional disputes damaged team morale, caused communication breakdowns, and led to delays in decision-making and task execution [11]. Such negative consequences ultimately eroded project value by increasing costs, compromising project schedules, and reducing stakeholder satisfaction [30]. In conclusion, the qualitative interviews enabled us to uncover the nuanced relationships between different types of conflict and project value creation.

6.3. Network and Project Value Creation

The research conclusion indicates that network centrality negatively impacts project value creation, lending credence to the notion that high network centrality is detrimental to organizational output. This result contradicts the findings of Li et al. (2020), who discovered that network centrality positively impacts project performance in interorganizational cooperation networks with high-tech backgrounds [64]. This is due to the fact that the high-tech industry is distinct from the construction industry. Cooperation between organizations in the high-tech industry is long-term, whereas collaboration in a construction project is short-term [20]. In a construction project, the collaborative connection starts at the outset and ends when the project is completed. Building emotional trust amongst organizations is challenging, since construction projects are transient and one-time in nature [20]. In this case, the stature and power that high-centrality organizations possess have the potential to sow mistrust among other organizations, endangering interorganizational collaboration [22]. Furthermore, information transmission can be distorted by high-centrality organizations, which can have a negative impact on interorganizational communication and information transmission [9]. These elements have a detrimental effect on project tasks, which in turn negatively impact project value creation.

The findings of this study show that network density positively impacts project value creation. This research conclusion is similar to previous research findings [65]. This is because construction projects are a temporary network composed of different project-participating organizations. A tightly connected construction project network makes it easier for organizations to share information and creates a unified action standard, which encourages collaboration across project-participating organizations [66]. Networks with strong connections can lessen tensions between organizations, boost organizational flexibility, and encourage interorganizational cooperation [67]. Thus, a tightly connected project network facilitates the efficient execution of construction projects and project value creation.

7. Theoretical Implications

This research has the following theoretical implications. First, the core theoretical implication of this study lies in the integrated framework that bridges network structure, conflict types, and project value creation—an under-explored linkage in the construction management literature. Unlike existing studies that either focus on network structure alone or conflict management in isolation, this study theoretically proposes and empirically validates the mediating role of conflict types (substantive conflict and affective conflict) in the network structure–value relationship. This is particularly novel in the high-conflict construction context, where prior research has overlooked how network density/centrality differentially trigger conflict subtypes, thereby affecting project value creation.

Second, prior studies investigating construction project networks have predominantly relied on bivariate regression or simple correlation analysis to explore direct relationships (e.g., between network density and project performance). These methods are ill-equipped to unpack mediating processes—a critical gap, as theoretical frameworks in project management increasingly emphasize that network structure influences outcomes through intermediate factors (e.g., conflict) rather than directly. In contrast, this study employs SEM, a multivariate technique uniquely suited for testing complex mediation models with multiple latent variables. By specifying a theoretical framework where network structure (density/centrality) predicts conflict types (substantive conflict/affective conflict), which in turn predict project value creation, this study formally validates the chain of effects hypothesized in our model. SEM allows us to estimate both direct effects and indirect effects (e.g., density → conflict → project value creation) simultaneously, using bootstrapping to confirm the significance of mediation paths. This approach addresses a key limitation of prior work, which often assumed but failed to statistically validate such mediating roles.

Third, this study demonstrates the detrimental effects of network centrality on project value creation, as well as the positive influence of network density on project value creation, by presenting evidence on the relationship between network structure and project value creation. These findings broaden our understanding of the construction project network. This study’s findings also highlighted conflict’s positive and destructive consequences. Furthermore, this study investigated the mediating role of conflicts in the setting of construction project networks. The findings revealed that substantive conflicts would mitigate the detrimental impact of project network structure on project value creation, while affective conflicts would increase it. As a result, a construction project network with a high degree of substantive conflict and a low level of affective conflict can give conditions for organizations to span divergent factions to boost project value creation.

8. Practical Implications

This study has the following practical significance. For policy: this study has practical significance for policy-making in the construction industry. Based on the results of this study, the current policies and regulations in the construction industry can be analyzed, and targeted improvement suggestions can be proposed. For example, it is suggested to establish industry standards for the design of network structures in construction projects. Developing specific guidelines for network density and centrality can help project participants avoid potential risks caused by inappropriate network structures. In addition, it is recommended that incentive policies be introduced to encourage construction companies to adopt effective conflict management strategies. For example, providing financial subsidies or preferential treatment to companies implementing comprehensive conflict resolution mechanisms in project bidding can promote the overall improvement of industry management practices. Integrating the results of this research into policy-making can help create a more scientific and regulatory friendly environment for the construction industry.

For society, construction projects play a crucial role in social development, and this study contributes to bringing positive social impacts. Optimizing the network structure of construction projects and effectively managing conflicts within them can help promote social harmony. By more effectively allocating resources within a well-organized project network, social resource waste can be reduced. By emphasizing these social benefits, this study aims to attract more social attention and support, and promote sustainable social development. In addition, project managers must understand that various types of project disputes have varying degrees of impact on project value creation. To promote the healthy development of the construction project network, the constructive impact of substantive conflicts should also be considered, and appropriate measures should be taken, such as brainstorming meetings that encourage different organizations to participate in discussions on project task implementation and process arrangements, thereby deepening understanding of task content and implementation methods and promoting smooth project implementation. In addition, project managers should be aware of the destructive effects of affective conflicts and take necessary actions, such as creating a pleasant project culture and learning environment, to mitigate their negative impact on society.

9. Conclusions

This study investigates the relationship between network structure, project conflicts, and project value creation in the construction project network. The results show that (1) network centrality is not conducive to resolving affective conflicts or promoting project value creation; (2) network density promotes the settlement of affective conflicts and project value creation; (3) substantive conflicts have a beneficial impact on project value creation, whereas affective conflicts have a negative impact; and (4) project conflict mediates the process of network structure influencing project value creation. This study serves as a reference for network structure governance, conflict governance, and project value management in construction projects. This study contributes to our existing understanding of the effect of project network structure on project value creation in the field of construction project management. This provides a new direction for construction project governance and project value management. Furthermore, this study validated the constructive role of substantive conflicts and the destructive impact of affective conflicts, contributing to the conflict governance literature.

10. Limitations

This study has the following limitations. First, different types of conflict might morph into each other under particular conditions. The transformation process between conflicts was not investigated in this study. Future research should investigate the metamorphosis of the two forms of conflict and their impact on project value creation. Second, the execution of construction projects is a dynamic process, and the project network structure may alter over time. As a result, future study should investigate the dynamic changes in project network topology and their impact on project value creation. Third, the sample data for this study is limited to Chinese construction projects. Future research should include more construction projects in order to broaden the research findings.