Collaborative relationship discovery in BIM project delivery: A social network analysis approach

Highlights

• Social network analysis is used to discover the evolution of stakeholders.

• Collaborative weight between stakeholders is quantitatively obtained from a network.

• A central stakeholder tends to be active and a broker at the same time.

• BIM modeler and BIM coordinator play a similar role in a BIM-based rail project.

• BIM coordinator is becoming increasingly important since the tender stage.

Abstract

A deeper understanding of collaboration among various stakeholders is imperative towards the success of the Building Information Modelling (BIM) enabled project delivery. A systematic research framework is proposed to properly analyze dynamic changes in stakeholders and their relationships by deploying social network analysis (SNA), assisting in stakeholder management. First, the work breakdown structure (WBS) is performed to organize a BIM project (i.e., rail project) into 4 levels: project name, stage, process, and activity. Second, different stakeholders are clarified based on the activity level over the whole project using interviews and surveys. Third, a project-level social network containing both the project ontology and stakeholders is established, from which cooperation information is then precisely extracted and stakeholder-level social networks with only stakeholders being nodes are created. Finally, SNA is carried out at both network- and node- levels. The results show that: (i) It is feasible to mine the collaboration information between different stakeholders through a project-level social network, (ii) The most active and central actor in each stage is continually changing, and the degree centrality and betweenness centrality are polarized between the stakeholders, (iii) There is a trend for a central stakeholder to be active and a broker at the same time, such as architect (#3), BIM coordinator (#4), BIM modeler (#6), civil & structural engineer (#9), and mechanical & electrical engineer (#18), and (iv) BIM coordinator (#4) should be given more attention, especially from the tender stage on. The research can contribute to both theoretical and practical development: (a) A new novel approach that could make the dynamic collaboration characteristics explicit was proposed with SNA, (b) A good knowledge of the dynamic collaboration attributes among different stakeholders in BIM-based rail projects could encourage a better project outcome.

Introduction

Construction projects are becoming increasingly complex in social, and technical perspectives for the sake of extreme urbanization [1]. This has greatly led to intricate relationships and serious arguments among various stakeholders in Architecture, Engineering & Construction industry, which in turn increases the needs in closer cooperation and communication among various stakeholders in the whole project lifecycle. These challenges shifted the motivation for Building Information Modelling (BIM) to the pursuit of construction project efficiency [2]. The benefits of BIM come about because of more effective collaboration among stakeholders due to a shared vision of the project. As a result, new types of roles have been introduced into the system, such as BIM modeler and BIM coordinator. Although construction projects are heavily dependent on domain knowledge of various interconnected stakeholders for a successful delivery and decision making [3,4], a lack of efficient guidelines for collaboration of various disciplines at all stages of a project has been highlighted [5]. Thus, the challenge rests on the development of a collaboration model to improve the communications between various stakeholders, promoting the adoption of BIM in construction projects further. There is a critical need for gaining a deep exploration and understanding of the interrelationships among different group members of a BIM-enabled project, where the stakeholders and their cooperation form a network.

Social network analysis (SNA) is usually the primary approach used in the complex cooperative system to estimate the performance of a project not only by identifying social structures in terms of connections and roles involved but also by finding the significance of relationships between information processing parties along with the critical level of individuals [6]. These can be achieved by measures including centrality analysis, community detection, and information diffusion analysis [7]. SNA has been greatly adopted by researchers for analysis and optimization of various organizational structures, such as networks for BIM-based design company that tried to improve the efficiency of a design company [4], smart living design that tried to calculate the financial impact [8], major construction projects that tried to increase the accuracy of stakeholder management [3], natural resource management that tried to find the most central roles [9], and Integrated Project Delivery (IPD) projects that tried to find the roles that are good at coordination work [10]. Overall, SNA grants project managers the strength to have a thorough comprehension of the social system of an organization, and thus achieve a better allocation of manpower, leading to performance improvement. However, the above-mentioned researches mainly focus on the static collaboration of stakeholders. As the composition of stakeholders and their interdependencies are continuously changing over time when the project moves forward, the collaboration between stakeholders cannot be conducted solely in stationary methods, but how their relations are changing should be appreciated as well [3,8].

To address this concern, some research efforts have been made to take into account the time factor during the process of collaboration analysis. For example, Li et al. [11] analyzed the involvement of stakeholders in three phases of a Near Zero Energy House (NZEH) but neglected the dynamic property of their relationships. Bouwman et al. [12] suggested a dynamic stakeholder analysis model where stakeholders in different phases of a design project were taken into consideration. Andrew et al. [13] revealed the changes in the level of involvement of various stakeholders within a Public-Private Partnerships (PPP) project in four different phases which were defined based on the financial perspective. Stephanie et al. [14] studied the dynamic nature of the relationships by using the Actor-Network theory. However, prior researches were primarily focused on the collaboration among a small group of stakeholders and limited to specific phases of a project, fewer studies have investigated the lifecycle of a project. In general, it is essential to consider the changes in stakeholder collaboration systematically.

Furthermore, to compactly consider the dynamism in collaboration among stakeholders, some factors should be precisely found to build a social network. The currently common practice is limited to the use of questionnaires for qualitatively acquiring information on the involvement of stakeholders (i.e., “How many stakeholders are involved in the project?”), their interactions (i.e., “Which two stakeholders have collaborations during the project?”), and the strength of the ties (i.e., “What do you think is the strength of their relationships?”) from respondents, which are used to build a network [15]. For example, Prell et al. [9] tried to gather the social network data from 147 stakeholders under 8 stakeholder categories to find how the stakeholders were related to each other and the tie strengths by telephone interviews. Hong et al. [11] collected the social network data about relationship types through sending surveys via emails. Marco et al. [16] defined values 1–5 to denote the weights of the relationships in social networks among stakeholders through interviews. However, this type of data collection has two disadvantages: 1) It is difficult to detect the evolution of stakeholders, and 2) The weights of the interrelationships among stakeholders may suffer from the unconscious bias due to the subjectivity of the respondents. The case is even worse when faced with complex networks, such as infrastructure projects [17,18]. Therefore, a novel method is proposed for reliable detection and discovery of relations between stakeholders and their intensity quantitatively. In the proposed method, the WBS of the construction project should be first developed, followed by surveys that ask for the stakeholder involved in each work package of the WBS, and by the establishment of the project-level network. As a result, the relationships between different stakeholders as well as their strengths can be automatically found from the project-level network, and stakeholder-level networks are built. Finally, SNA is conducted for further knowledge discovery.

In this research, the feasibility of the proposed methodology is examined. The main aim of this research is to answer the following questions: (1) How can the project-level social network of a BIM-enabled project be established based on the WBS, (2) How can the relationships and their intensities between different stakeholders in a stakeholder-level social network be found through a project-level network, and (3) How can the SNA be used to explore the collaborative attributes among different stakeholders in the lifecycle of a BIM-enabled project. This research contributes to (i) the state of the knowledge by presenting an innovative methodology that can extract the objective communications among various actors, and (ii) the state of the practice by revealing the collaborations among stakeholders in a BIM-enabled project and providing project managers with the suggestions on stakeholder management.

The remaining of the paper is organized as follows. Section 2 reviews the literature on knowledge discovery using SNA and its merits. Section 3 provides a cognization of the research methodologies in terms of work breakdown structure (WBS) and SNA. Section 4 presents the dynamics of collaborations between different stakeholders over time. In Section 5, discussions are given. In Section 6, the main conclusions and future works are drawn up.