Social Relation Cognitive Model on Virtual Prototyping Technology in Construction Project

—It is valuable to study social relation cognition of participants about virtual prototyping technology in construction projects to promote its acceptance and application. Building information modeling (BIM) is a transformative virtual prototyping technology for construction industry. The analysis of social relation is closely related to BIM application to stakeholders, which involves many cognitive indicators, i.e. organizations, spaces and behaviors. Based on social network analysis, a cognitive model for actors’ social relation to BIM knowledge is proposed�Using survey data, this paper explores the cognitive ability between participants. Through statistical analysis of indictors in the model, it is found that significant difference of virtual pro-totyping technology in social relation cognition exists.


I. INTRODUCTION
Virtual prototyping (VP) is a computer-aided design process concerned with the construction of digital models and realistic graphical simulation [1]. VP technology has been extensively applied to the automobile and aerospace industries. Several studies have used VP technology to develop effective dynamic project design and construction, such as the Virtual Design and Construction (VDC) [2], Virtual Construction Environment (VCE) [3], Virtual Facility Prototyping (VFP) [4], and DIVERCITY [5]. However, VP technology in construction industry only played a limited role until the application of building information modeling (BIM) was widely accepted by stakeholders.
BIM is regarded as an important technology of the last decade in the construction industry (Eastman et al., 2008;Jernigan, 2007;Guo et al., 2010). BIM is one of the important areas in currently Virtual Reality (VR) research and is expected to envision efficient collaboration, improved data integrity, distributed access and retrieval of building data and high quality project outcome through enhanced performance analysis, as well as multidisciplinary planning and coordination [5] [6]. It is necessary to explore the cognitive model on BIM to understand the different cognitive abilities among actors in construction projects.
Recently, the network analysis approach has been receiving attention within the construction industry, where concepts such as communication between project participants are receiving significant attention (Morton et al. 2006;Katsanis 2006). Social cognitive theory provides a conceptual framework in which the determinants and mechanism of such effect can be examined. People's behavior is shaped and controlled by either environmental influence or internal dispositions. Social relation cognitive modeling belongs to social computing domain and it studies the regular pattern of agents' social behavior via computer technology and solve various problems with inter-group communication and collaboration.
The motivation for this research centers on the social relation and indicators in cognitive model on virtual prototyping technology (especially BIM) beyond the traditional construction perspective on project management.

II. LITERATURE REVIEW OF SOCIAL RELATION COGNITIVE
Following the successful implementation of virtual prototyping in the manufacturing industry, many studies in the construction industry put their efforts on using VP to produce an effective dynamic construction project. BIM has been the direction of the future development in construction industry, but many factors affect the BIM cognition model, especially in aspects of the interoperability between software, BIM value and inter- In recent years, some organizations and academics developed the related research on social cognitive model of VP technology. In the surveys to North America, Middle East and Europe, McGraw-Hill (2009, 2010 reported that users' cognitive differences were existed in Return on Investment (ROI), potential value, internal revenue, expected value and other cognitive aspects. Social relation cognitive model also depends on the environment factors [7] [8]. AWCI (2009) analyzed BIM application in curtain wall industry and found that the agents have disagreement with BIM social relation and cognition [9]. Ning GU & Kerry London (2010) focused on the VP technology perception, cognition and interest between stakeholders and found that the architects and vendors are the most active organizations [10]. Shady Attia (2010) made two questionnaires and analyzed the cognitive experiences and behaviors of architects and engineers. There are significant differences in the aspects of simulation tools, information management, knowledge integration and models sharing [11]. J Won, G Lee (2010) studied the success factors of BIM application provided by 61 international experts [12]. Gilligan and Kunz (2007) and Damian (2008) explored the values and barriers of BIM in order to examine the understanding of social relations [13][14].
In an inter-organization network, cognition of VP technology is exchanged and therefore organizations in the

I. SOCIAL RELATION COGNITIVE MODEL
The studies above about cognitive model on VP technology use questionnaires and more qualitative analysis, mainly focused on the level of understanding of VP technology, value perception and judgment, barriers, influencing factors, etc. Studies generally agree that the stakeholders of government, owner, contractor, architect, investigator and vendors participating subjects on VP technology were in different positions, and less quantitative analysis methods were used.

A. Social relation
According to Wasserman and Faust (1997), a social network is a social structure made of actors (nodes) that are connected by one or more specific type of relations (ties), such as friendship, firm alliance, or international trade. The study utilizes UCINET software to explore the network diagrams, the individual networks and cohesion subgroups.
Density and centrality are the most common indicators of a network's connectivity. They represent the extent of how densely and cohesively nodes in a network are interconnected (Pryke 2004), as shown in the following equation: where l = number of existent lines and n = number of existent nodes.
where ! !" ! number of degree that a node i receives from a node j and n= number of existent nodes.
where ! ! !! ! ! number of shortest paths from node s to node t that pass through node i.

C. Evaluation method and mathematical model
Principal Component Analysis (PCA) is a mathematical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components.PCA may get the largest contribution of variance to datasets by keeping low level principal components and ignoring high level ones. a. Data collection and standardization of original indicators: According to the data demand, the paper establishes the sample matrix and standardizes transformation as following.
c. Solution of eigenvector equation: where cumulative contribution rate ! !
Transform indicator variables into principal components: where ! ! named m th principal component, and ! ! ! is the unit eigenvectors for each ! ! !! ! !! !! ! ! !!. e. Comprehensive evaluation to m principal components: We can get the overall score to each principal component weighted.

A. Background of the project
The Shanghai Center is the tallest building under construction in Shanghai, China, which was started in Nov 2008. It will be 632 meters in height and have a total of 121 stories. The investment will reach 14.8 billion Yuan. The project tries to follow sustainable development concepts in its full life cycle of construction and achieve high efficiency of material consumption. As a result, many VP technologies, especial BIM, are widely used in order to ensure the targets of the project are successfully achieved.

B. Outline of the Survey
By interview with certain VP experts, we determine the survey to explore the social relations of VP technology (this paper taking BIM as an example) based on the above researches. The questionnaire consists of three parts, as PAPER SOCIAL RELATION COGNITIVE MODEL ON VIRTUAL PROTOTYPING TECHNOLOGY IN CONSTRUCTION PROJECT follows: firstly, with whom you interact the BIM cognition in your position; secondly, the richness of the amount of knowledge you offered (divided into the level of low, medium and high); thirdly, your opinion of the indicators in BIM communication (Table I), using mensuration of Liker scale 5 levels to indicate the degree of satisfaction to interactive knowledge of VP technology.
In this project, 42 people were involved in VP technology including the owner, contractors, designers, operators and some others (Table II). All actors responded and 100% of the questionnaires were valid.

C. Social relation network
For this study, UCINET 6 is provided for mathematical measurements. As illustrated in Fig.1, structural concerns within the project network are identified. Except the two isolated nodes of S5 and S6, most nodes have significant relations in VP technology communication with others. The node O6 (BIM manager) is in the core of the network, which exchanges VP technology cognition with many nodes in the network and promotes the two-way flow of knowledge. The nodes of D1 and C5 also occupy important positions.
Using degree centrality of SNA, we can make quantitative analysis of the inputs or outputs of each agent's cognition and find out the key network members in organizational cognition. The density of the network is 0.387(N of Ties is 144 and Avg Degree is 3.93), and network Centralization is 25.85%. This shows that the link between agents of the network is not very close. Some nodes rely on only a few ones (C5, O6, D1), which means cognitive transformation and exchange are less frequent.
As shown in Fig 2, Betweenness Centrality represents the control degree to knowledge transfer channels of a node. The actor of BIM manager (O6) has the highest Betweenness, about 46.377%, and the nodes of D1, C5, D8, O1 and C1 also reach more than 10%. It indicates that these members in the network occupy the positions of structural holes in the exchange of knowledge within the organization and play important roles as bridges in delivery processes. They control the flow of knowledge effectively to others.   Outcomes of cohesive subgroup analysis are shown in Figure 3 and Figure 4. The organization's network can be divided into three major subgroups, which is consistent with the project's organizational structure based on VP technology. Some key nodes play "bridge" roles.

D. Variables and Model calculations
Using Matlab software on influencing factors of the Principal Component Analysis, the extraction principle is the principal component which has eigenvalues greater than 1, cumulative contribution rate! !"#. The cumulative contribution rate of the first eight principal components is! ! !"!!" ! !"#. Therefore, these eight new variables could evaluate the social relation cognizance instead of the original 27 variables, and total score models obtained are as follows: In the above equation, weighting factors of each variable reflect the degree of importance between indicators. Negative values indicate the degree less than the average level while positive values mean above average.

E. Model Analysis and Evaluation
We make the average value of each indicator (satisfaction, table III) as the horizontal axis and the weight of each indicator (Significance, Eq(10)) as the vertical axis, then draw the Quadrant Map in Figure 5. The key indicators of the BIM cognition of the project and the performance status of each indicator are clearly shown. The indicators of X 8 ,X 9 ,X 10 , X 11 ,X 13 ,X 21 and X 25 are also important to cognitive model. Most agents think we should take more concerns to understand the values, contracts, skill of engineers, responsibilities, incentive mechanism and organizational structure. Those factors must be improved to maximize the efficiency of VP technology in construction projects.

c. Quadrant III (insignificant Area)
Just only three indicators of X 3 ,X 6 and X 7 belong to the insignificant ones with which agents of the Shanghai Center Project have less interest and less satisfaction in the currently stage. The limitation of their cognition indicates that "myopia insights" is still the most harmful obstacles to BIM implementation in China. The context of civil software, standards and norms, information security and property rights affect the national application in construction industry.

d. Quadrant IV (Maintain Area)
The indicators of X 4 , X 5 , X 12 ,X 16 , and X 17 are not very important to some stakeholders in the project. With the huge investment and ROI expected, they care more about the cost of hardware than that of software. Because BIM maturity of VP technology is still at a low level, using information technology is still under the complex characters in China. It is a long way for the government to establish laws and standards to control the development of VP technology successfully.

IV. CONCLUSION
The knowledge and cognition of VP technology in construction projects need participants to work together in a team. It depends on the social relation in the organizational network. This study analyses the social relation cognitive model of different actors in Shanghai Center in China. As the most popular VP technology in construction industry, Building Information Modeling is gaining attention gradually by most stakeholders. In the social network of the project, social relation, density, centrality and cohesive subgroup are explored to find the crucial actors, such as BIM manager of owner, BIM engineer of contractor and BIM designer. The analysis of cognitive indicators show that different agents have significant different cognition in the evaluation of important and satisfaction. There is still a long way to improve the performance of indictors and it also needs guidance to be established at the level of the national strategy for BIM.