**Identifying Significant Risks and Analyzing Risk Relationship for Construction PPP Projects in China Using Integrated FISM-MICMAC Approach**

## **Xiaoyan Jiang 1,\*, Kun Lu 1, Bo Xia 2, Yong Liu <sup>3</sup> and Caiyun Cui <sup>4</sup>**


Received: 30 August 2019; Accepted: 17 September 2019; Published: 23 September 2019

**Abstract:** To meet the growing demand for public facilities and services, many developing countries, including China, have adopted the concept of public–private partnership (PPP). However, there are many risks in PPP projects. Furthermore, these risks affect each other, which may lead to project failure. However, the existing research on the PPP risk relationship has not gone into sufficient detail. Therefore, in order to fill this literature gap, this study proposes a procedural method to analyze the correlation between PPP risks. Firstly, this study, identifies the risks of construction PPP projects in China by combining the literature review with a case study and interviews. Then, fuzzy interpretative structural modeling (FISM) is used to reflect the relationships between these risks and reveal the failure mechanisms of PPP projects. In addition, based on matrix impact cross-reference multiplication applied to a classification (MICMAC) analysis, the risk is divided into four clusters, according to the driving and dependence power, to show the relationship level of the risk. Finally, the paper compares and discusses the research results with other studies and puts forward some suggestions on PPP risks. The FISM-MICMAC method adopted in this study considers the fuzzy of the PPP risk relationship and improves upon previous studies. In addition, the method of FISM-MICMAC can provide a new risk assessment tool for risk management strategies in the field of construction engineering and management.

**Keywords:** Public–private partnership (PPP); risk identification; risk relationship; triangular fuzzy number; ISM; MICMAC

## **1. Introduction**

A PPP (public–private partnership) refers to a partnership between the government and private investors to provide public infrastructure projects, public goods, and services [1,2]. PPP originated in the UK [3]. As it can deliver high-quality results within the concession period and budget [4], PPP has attracted extensive attention from the public sector and has been adopted in a number of countries [5–7].

However, PPP projects are characterized by large scale investments, long contract concession periods, and complex technologies, which give rise to many potential risk factors in the implementation process, which could also lead to the failure of PPP projects [8–10]. According to the World Bank, 279 PPP projects have been "Cancelled" since 1990 [5].

In the wake of the 2007–2008 global financial crisis, the Chinese government has become increasingly interested in PPP as a way for local governments to transfer their debts [10]. However, compared with developed countries, the growth of PPP in China is still in its infancy [11]. Furthermore, some PPP projects cannot be successful in China [12,13]. As a result, on 16 November 2017, China's Ministry of Finance announced the "Notice of standardizing project library in PPP integrated information platform" (hereafter referred to as "No. 92") [14], designed to correct the current problem of deviation and variation in the course of PPP project implementation, further improve the quality of the project library storage, as well as project effectiveness and information given to the public, and improved social supervision. Since the release of No. 92, from 16 November 2017 to 16 August 2019, the number of "withdrawal projects" reduced by the PPP project library of the Ministry of Finance has reached 6955, and there are 12,561 existing PPP projects in the PPP project library [15].

PPP project failures and cost overruns are not rare, and they pose a threat to sustainable development [16]. The failure of a PPP is largely due to PPP risks [10,17]. There are many risks in PPP projects which, furthermore, are correlated [18,19]. According to accident causal chain theory, accidents are usually caused by a chain of events in the system and the sequential occurrence of causality [20,21]. The failure of PPP projects is usually not caused by a single risk factor, but by a single risk inducing other risks, forming a chain of PPP project risks and, finally, leading to the failure of the PPP [18,19].

Existing studies have mainly focused on the key success factors of PPP [10], risk factors [18,22], risk allocation [17,23], and risk assessment [11,24], etc. However, no in-depth studies on the relationships between PPP risks have been carried out. It is crucial to know what relationships exist between PPP risks, and which risks play a leading role in triggering project failure, in order to help PPP managers to better understand the impact of the relationship between risks on project failure, and to distinguish the key points in future risk management and control.

This study identifies the key risk factors of PPP projects by literature survey, case study, and interviews. Then, fuzzy interpretative structural modeling (FISM), which combines fuzzy theory and interpretative structural modeling (ISM), is used to reflect the relationships between these risks and reveal the failure mechanisms of PPP projects. In addition, key risk factors are classified with the MICMAC (matrix impact cross-reference multiplication applied to a classification) method, and the driving effect of each risk factor on the failure of PPP project is analyzed. Finally, the similarities and differences between this study and other studies are analyzed, and suggestions for managing key risk factors of PPP projects are given.

#### **2. Literature Review**

## *2.1. General Review of PPP Risks*

The impact of risks on the completion of a PPP project is usually significant [11]. In PPP projects, various risks exist, not only due to the complexity of the financial and organizational structures of the project, but also due to the large amount of investment, long operation period, sophisticated technical know-how of the project, political impact, and government involvement [25].

The existing studies on PPP risks have mainly been focused on risk factors, risk allocation, risk assessment, and risk identification, etc. Only a limited number of studies have investigated the relationships of PPP risks. Some researchers have sought to identify the risk factors associated with PPPs in specific projects or in specific countries [26,27], and have generally categorized them in terms of being equally shared by both parties or mostly allocated either to the public or private partners [28]. In addition, many scholars have also studied proper allocation of risk factors between the project participants [1,29,30]. Jin combined fuzzy logic with artificial neural network techniques to design a neuro-fuzzy model to help the effective allocation of risks in PPP projects [31]. Ameyaw and Chan's approach based on fuzzy set theory outlined the risk allocation principle, explained the fuzziness inherent in the human cognitive process, and made a case study of risk allocation in a PPP water supply project in Ghana [32]. According to the findings of Jayasuriya et al., there has often been disagreement

between PPP regulators (public partner) and operators (private partner) about the preferred risk allocation [28]. For PPP risk assessment, efforts have also been made to develop models to evaluate PPP risk values [11,24,33,34]. Li and Zou applied a fuzzy analytic hierarchy process (FAHP) to the risk assessment of PPP projects [24]. Mazher et al. proposed a PPP infrastructure project risk assessment method based on a fuzzy measure and a non-additive fuzzy integral [35]. Thomas et al. proposed a risk probability and impact assessment framework for build–operate–transfer (BOT) roads, based on a fuzzy-fault tree and the Delphi method [36]. Bai et al., based on the methods of fuzzy comprehensive evaluation model and failure mode, conducted an effects and criticality analysis for evaluating the sustainability risk level of PPP projects [37].

PPP risks identification and relationships among PPP risks will be described in the following subsections.

## *2.2. Identification Methods of PPP Risks*

As mentioned, some scholars have systematically reviewed the risk factors. PPP risks are specific for different types of projects. For sponge city projects [38], water projects [29], highway projects [39], waste-to-energy projects [26], marine projects [40], and other project types, the inherent PPP risks are different. In addition, for different countries, such as Indonesia [41], China [26], the United States [42], Iran [9], and Malaysia [43], the PPP risks are also different.

Due to the specificity of PPP project risks, specific risks should be extracted for different types of PPP projects in different countries. Some scholars and risk managers have attempted to introduce a variety of different risk identification methods into PPP projects, among which the three most frequently used risk identification methods are listed in Table 1.


**Table 1.** Common methods for identifying public–private partnership (PPP) risks.

#### *2.3. Relationship among PPP Risks*

From the above analysis, it is evident that there are relationships between PPP risks that lead to the mutual influence of risks [18,19]. In analyzing risks at different project stages, it is important to consider their interrelationships, because an understanding of the relationships between the risks can facilitate a more holistic risk identification and assessment. The existence of risk at an earlier stage may contribute to increased risk manifesting in later stages, or it may not. Sometimes a less important risk can affect a more significant risk, both in terms of likelihood and severity [49].

For example, Valipour, A. et al. identify PPP shared risks using an approach in the form of a hybrid Fuzzy method and Cybernetic Analytic Network Process (CANP) model [50]. Wang et al. proposed a risk model, named the Alien Eye's Risk Model, to show the hierarchical levels of the risks and the influential relationship among the risks in a risk influence matrix [51]. Alada ˘g and I¸sık used FAHP to determine the priority factors of design and construction risks in build–operate–transfer (BOT) type mega transportation projects [52]. Furthermore, a fuzzy analytic network process (FANP)

method was applied for overcoming the problems of interdependencies and feedback among different risk-ranking alternatives in freeway PPP projects [53]. Multiple-regression analysis was also proposed to estimate and quantify risk interrelationship [54].

## *2.4. Research Gap and Contribution*

These studies have used techniques, such as fuzzy synthetic evaluation [11], fuzzy analytic hierarchy processes [24,33], 2-dimensional linguistic information [34], ISM-MICMAC method [19], hybrid fuzzy cybernetic analytic network process model [50] and multiple-regression analysis [54] etc., to analyze PPP risks and their relationships. However, studies on these relationships and influences based on failed PPP cases in China is lacking. Moreover, it is common to utilize a single approach, but less common to utilize a fuzzy-ISM integrated MICMAC method comprehensively in PPP risks relationship analysis.

The main contribution of the study is to present the application of the fuzzy-ISM technique integrated MICMAC analysis on the relationship among significant PPP risks, which affect the implementation of PPP projects in China, and also to analyze which risks play a leading role in triggering project failure. The 28 cases in this study also make an important contribution to PPP case studies, because failed PPP projects in China can be used as important references for other countries. In addition, this study makes a contribution to relevant literature on common methods for identifying PPP risks and to our understanding of failed PPP projects by summarizing some abnormal phenomena, such as government buyback, project predicament or no operations, severe losses, management right transfers, failure to realize value for money (VFM), and contract cancellations.

## **3. Research Methods**

This study draws on literature, case studies and expert interviews to collect data, and identifies the significant PPP risks in China. Then, the fuzzy ISM (FISM) approach is adopted to clarify the interaction relationships among the PPP risks and to establish a hierarchical structure of these risks. Finally, the MICMAC integrated with fuzzy-ISM approach is employed to determine the significant risks from the perspective of their interaction status. There are six analysis steps in this study, as shown in Figure 1. The details of these six steps will be described in the following subsections.

**Figure 1.** Research framework of the study.

#### *3.1. Step 1: Collecting Data and Identifying Risk Factors*

The preliminary factor list is extracted by case analysis and literature review, following which, the factor list is modified and simplified by expert interview.

The 28 failed PPP projects are from the World Bank website [55] and China Public–Private Partnerships Center [56]. In order to comply with Chinese national conditions, all of these cases and the literature were based on the actual situation of China. But we cannot find a unified definition on "failed project". Based on prior studies [57,58], this study deems a PPP "failure" as when the following phenomena occur: government buyback, project predicament or no operations, severe losses, management right transfers, failure to realize VFM, and contract cancellations. The cases were chosen as typical examples of the failed PPP projects, with their unique characteristics and failure causes were an excellent match to the critical PPP risk identification. Details of the 28 cases can be found in Appendix A. We have adopted a broadly similar format where possible, in each case drawing out the key risk factors relevant to failure or problem description.

## *3.2. Step 2: Using Triangular Fuzzy Number to Build Fuzzy Direct Relation Matrix*

This study utilizes triangular fuzzy number to build fuzzy direct relation matrix. It has been found that fuzzy research, as applied in the construction management discipline, can be divided into two broad fields—fuzzy sets and hybrid fuzzy techniques [59]. The triangular fuzzy number method belongs to the former category [60].

Since the publication of the seminal work "Fuzzy sets" by Zadeh [61], defuzzification by using the fuzzy approach has led to many successful practical applications. For example, Paek applied a fuzzy set approach to price to analyze construction risk [62]. Zhang and Zou used a fuzzy analytic hierarchy process (FAHP) for the appraisal of the risk environment pertaining to joint ventures, in order to support the rational decision-making of project stakeholders [63]. Abdelgawad and Fayek used Fuzzy fault-tree analysis to quantitatively assess risk events in the construction industry [64].

Zadeh proposed the concept of fuzzy sets in order to solve problems described as semi-structured or ill-structured, which allows us to process and transform imprecise information effectively and flexibly [57]. Because of the uncertainty of objective things and the fuzziness of human thought, fuzzy decision-making based on fuzzy sets theory has become a basic method for decision-making, meanwhile triangular fuzzy numbers have been extensively applied in fuzzy control and fuzzy decision-making [65–67].

In this study, the fuzzy triangle number of the relationship between various factors is obtained by expert scoring. A triangular fuzzy number is usually represented by three letters: *l*, *m*, and *r*. These three parameters, respectively, represent the minimum possible value, the median value, and the maximum possible value, (i.e., *l* ≤ *m* ≤ *r*) [49]. Referring to the value table of language operators and triangular fuzzy numbers given by Li [68] (as Table 2), we let experts judge the strength of the relationship between the two risk factors.

**Table 2.** Triangular fuzzy number corresponding to language operator.


According to the results of the questionnaire, a triangular fuzzy relation matrix is established to judge the strength of the relationship between failure risk factors. In triangular fuzzy relation matrix *<sup>D</sup>*,*k*, the triangular fuzzy number *<sup>d</sup>* ,*k ij* = *l k ij*, *mk ij*,*r<sup>k</sup> ij* is used to represent the judgment result of the influence degree of the risk factor *Ri* on *Rj*, given by the expert (as follows).

$$
\widetilde{D}^{k} = \begin{bmatrix}
0 & \overline{d}\_{12}^{k} & \cdots & \overline{d}\_{1n}^{k} \\
\overline{d}\_{21}^{k} & 0 & \cdots & \overline{d}\_{2n}^{k} \\
\vdots & \vdots & \vdots & \vdots \\
\overline{d}\_{n1}^{k} & \overline{d}\_{n2}^{k} & \cdots & 0
\end{bmatrix} \tag{1}
$$

Then, the fuzzy direct relation matrix *D*. of failure risk factors can be obtained by converting fuzzy data into crisp scores (CFCS) method [69]. The specific steps are shown below:

1. Standardize the triangular fuzzy number

$$\begin{aligned} a\_{ij}^k &= \left(l\_{ij}^k - \min l\_{ij}^k\right) / \Delta\_{\text{min}}^{\text{max}}\\ b\_{ij}^k &= \left(m\_{ij}^k - \min m\_{ij}^k\right) / \Delta\_{\text{min}}^{\text{max}}\\ c\_{ij}^k &= \left(r\_{ij}^k - \min r\_{ij}^k\right) / \Delta\_{\text{min}}^{\text{max}}\\ \Delta\_{\text{min}}^{\text{max}} &= \max r\_{ij}^k - \min l\_{ij}^k \end{aligned} \tag{2}$$

2. Calculate the left and right limits of the standardized values *u<sup>k</sup> ij* and *vk ij*

$$\begin{aligned} u\_{ij}^k &= b\_{ij}^k / \left( 1 + b\_{ij}^k - a\_{ij}^k \right) \\ v\_{ij}^k &= c\_{ij}^k / \left( 1 + c\_{ij}^k - b\_{ij}^k \right) \end{aligned} \tag{3}$$

3. Calculate the total value of the standardized values

$$w\_{ij}^k = \frac{u\_{ij}^k \left(1 - u\_{ij}^k\right) + \left(v\_{ij}^k\right)^2}{1 - u\_{ij}^k + v\_{ij}^k} \tag{4}$$

4. Calculate the exact value of expert triangle fuzzy judgment *d<sup>k</sup> ij*

$$d\_{ij}^k = \min l\_{ij}^k + w\_{ij}^k \cdot \Delta\_{\text{min}}^{\text{max}} \tag{5}$$

5. Calculate the standardized accurate value *dij*, as evaluated by experts

$$d\_{ij} = \frac{1}{p} \cdot \sum\_{k=1}^{p} d\_{ij}^{k} \tag{6}$$

6. Determine the fuzzy direct relation matrix *D* of key risk factors

$$D = \begin{bmatrix} 0 & d\_{12} & \cdots & d\_{1n} \\ d\_{21} & 0 & \cdots & d\_{2n} \\ \vdots & \vdots & \ddots & \vdots \\ d\_{n1} & d\_{n2} & \cdots & 0 \end{bmatrix} \tag{7}$$

## *3.3. Step 3: Using the Intercept Coe*ffi*cient to Obtain the Skeleton Matrix*

By using appropriate intercept coefficients α, the fuzzy direct relation matrix is transformed into a skeleton matrix S, which can be directly used for Boolean operation. When the elements in the fuzzy direct relation matrix are greater than or equal to α, the corresponding position of the elements in the skeleton matrix is set to 1; otherwise it is set to 0.

$$s\_{ij} = \begin{cases} 1 \left( d\_{ij} \ge \alpha \right) \\ 0 \left( d\_{ij} < \alpha \right) \end{cases} \tag{8}$$

## *3.4. Step 4: Calculating Reachability Matrix and Reconfigurable Reachability Matrix*

ISM is an effective model that can clearly define inter-relationships between multiple elements of a problem [70]. In order to study the correlation between evaluation objects, Mandal and Deshmukh introduced interpretative structural modeling (ISM) [71]. ISM has become an analytical tool widely used by scholars to study the interdependence and interaction of factors, and is widely used in policy analysis, supply chains and in other fields [72,73]. For example, Tseng et al. used ISM to construct a systematic model of interconnected natural disaster risks [74]. Yanmei et al. identified power overload risk factors and established ISM to analyze the relationships between these factors [75]. Li et al. also used ISM to assess the risks of India's thermal power plants [76].

The advantage of ISM is that it requires fewer questionnaires than other methods, such as structural equation modeling and the Delphi technique, and is able to extract a clear structural view from unstructured models [77,78]. However, in ISM, only the existence of an influencing relationship between two factors is investigated; if there is a relationship between factors, it is denoted by "1", otherwise, it is denoted by "0" [78]. However, the association between the factors is fuzzy and cannot be easily divided into "related" or "unrelated" [79]. To solve this problem, many scholars have applied fuzzy set theory to ISM, in order to consider the degree of influence between factors, in an approach called fuzzy interpretative structural modeling (FISM) [78,80,81].

As described in the previous section, this study uses the triangular fuzzy number method integrated with ISM in this paper. The process is as follows.

First, the Boolean operation of the skeleton matrix and the identity matrix is performed, where the operation rules are as follows:

$$0 \cdot 0 + 0 = 0, \ 0 + 1 = 1, \ 1 + 1 = 1, \ 0 \ast 0 = 0, \ 0 \ast 1 = 0, \ 1 \ast 0 = 0, \ \text{and} \ 1 \ast 1 = 1$$

Meanwhile, it must follow the shift law characteristic of matrix operations; that is, when *a* directly reaches *b* through a path of length 1 and *b* directly reaches *c* through a path of length 1, then *a* must reach *c* through a path of length 2.

Then, the reachability matrix R can be obtained through a Boolean operation of the skeleton matrix S. When all products are equal, the reachability matrix R can be obtained.

$$(S+I) \neq (S+I)^2 \neq \dots \neq (S+I)^{r-1} \neq (S+I)^k = (S+I)^{k+1} = R \tag{9}$$

Lastly, based on the reachability matrix, the risk factors of failure are rearranged according to the value of driving power and the reconstituted reachability matrix *R*∗ is obtained.

## *3.5. Step 5: Using ISM to Hierarchize Risk Factors*

According to the interaction relation of failure risk factors, the hierarchical results can be obtained by referring to the values of elements in the reconstructed reachability matrix. Connecting all the factors with arrow lines, the ISM diagram of the interaction of failure risk factors will be achieved.

#### *3.6. Step 6: Using MICMAC to Classify Risk Factors*

After ISM, MICMAC (matrix impact cross-reference multiplication applied to a classification) is often used as a complementary method to analyze the driver power and dependence power of the risks [82]. The driver power of a risk means the total number of risks it can influence, whereas the dependence power of a risk means the total number of risks which can influence it [83]. According to the driver and dependence powers of the factors, they can be divided into four groups: autonomous, dependent, linkage, and independent. Issues having weak driving and dependence powers are in the autonomous group, whereas issues having strong dependent and driving powers are in the linkage group [78]. Issues which have strong driving and weak dependence powers are known as independent issues, whereas issues which have strong dependence power and weak driving power are known as dependent issues [78].

Many scholars have applied ISM-MICMAC to PPP risk analysis. Iyer et al. analyzed the hierarchical structure of PPP risks of Indian roads for the first time by combining ISM and MICMAC [19]. Han et al. used ISM-MICMAC to analyze PPP risks in brownfield remediation projects in China [83]. Li and Wang used a fuzzy analytic network process and ISM-MICMAC for PPP risk assessment [18].

Although these analyses based on ISM-MICMAC have achieved some beneficial results, they have all ignored the notion that the impact degree of PPP risks is vague and cannot be judged simply by the impact or no impact. Based on this, FISM is adopted in this paper to consider the fuzziness of the relationships between PPP risks, replacing the traditional "0 or 1" judgment with a fuzzy number, in order to reduce the subjectivity of expert judgment.

In this study, the categories of risk factors are determined by FISM integrated MICMAC method. Based on the reconstituted reachability matrix Rˆ\*, each factor is plotted as a point in the four quadrants of the rectangular coordinate system, according to the driving power and dependency power of each failure risk factor, and the key risk factors are classified.

## **4. Results and Findings**

By following Step 1 in Section 3, a literature review and case study analysis was conducted to elicit the preliminary list of risk factors. A total of 29 preliminary risk factors were extracted to the preliminary risk list.

To mitigate the deficiency of the literature review and case study, we invited five experts for two rounds of interviews to validate the preliminary list. We chose experts who had sufficient time available for being interviewed and for summarizing these risk factors. Other experts that were considered, but who were too busy, were excluded from the risk identification stage. These experts included one lawyer involved in PPP consulting, one manager of a PPP project company, two managers of a PPP consulting company and one PPP research scholar, all of whom have participated in the management of PPP projects and have a rich knowledge of PPP theory and practice.

After the first round of interviews, based on the opinions of various experts and further confirmation in the second round of interviews, the risk factors that led to the failure of PPP projects were summarized into 20 risks (see Table 3). The literature and case sources of these 20 risks can be seen in Tables 4 and 5.

These 20 risks are: government decision-making approval risk (R1); policy and regulatory change risk (R2); government credit risk (R3); government regulatory risk (R4); planning and design risk (R5); bidding risk (R6); contract risk (R7); facility matching risk (R8); financing risk (R9); economic risk (R10); project change risk (R11); construction risk (R12); project income risk (R13); parallel project competitive risk (R14); operation and maintenance risk (R15); force majeure risk (R16); public opposition risk (R17); organizational coordination risk (R18); environmental risk (R19); and project company violates laws and regulations (R20).

In this study, in addition to the five original experts who participated in expert interviews, another 10 experts were invited to conduct a questionnaire survey on the relationship between PPP risk factors by means of e-mail, WeChat and on-site survey (see Appendix B). All 15 experts have rich knowledge of PPP theory and practice, and their background information is shown in the Table 6. Through comparative analysis of questionnaire data, the high impact of other risks on the force majeure risk of R16 was taken as an invalid judgment standard to eliminate the invalid questionnaires. In total, there were 9 valid questionnaires.


**Table 3.** Preliminary risk list and final risk list.

**4.** Literature statistics for PPP risk factors.

**Table** 


$$\mathbf{S} \text{ causes statistics for PPTP } \mathbf{r} \& \mathbf{k} \& \text{ for trees.}$$

**Table** 



The triangular fuzzy number of 9 valid questionnaires were filled in the triangular fuzzy relation matrix. Then, we converted triangular fuzzy relation matrix to fuzzy direct relation matrix through converting fuzzy data into crisp scores (CFCS) method. The final fuzzy direct relation matrix is shown in Table 7.

**Table 7.** Fuzzy direct relation matrix of the PPP key risk factors.


After obtaining the fuzzy direct relation matrix, the intercept coefficient α was adopted to convert the fuzzy direct relation matrix into a skeleton matrix, which was directly used for Boolean operation, as shown in Table 8.


**Table 8.** Skeleton matrix of the PPP key risk factors.

Reachability matrix was obtained after Boolean operation of the skeleton matrix, and on the basis of the reachability matrix, all failure risk factors were rearranged according to the value of driving power to get the reconfigurable reachability matrix (see in Table 9). In order to reduce the amount of calculation, we used Python programming in the operation process.

**Table 9.** Reconfigured reachability matrix of the PPP key risk factors.


Risk factors were hierarchized by using the ISM diagram. In the ISM diagram, the classification of a risk factor hierarchy can be determined by the driving power of a risk factor; that is, the risk factor with smaller driving power sits on the higher level in the risk system of ISM diagram, and the risk factor with higher driving power sits on the bottom position in the ISM diagram. We obtained the hierarchical partition results and ISM diagram according to the driving power of each risk in reconfigured reachability matrix. According to the positions of the risk factors in the ISM diagram, they were divided into three areas: upper, middle, and bottom, as shown in Figure 2. In Figure 2, we can see that:


**Figure 2.** Interpretative structural modeling (ISM) diagram of the PPP key risk factors.

After obtaining the ISM diagram that represents the mutual influence of key risk factors in PPP, it is necessary to refer to the MICMAC method to determine the categories of risk factors, so as to provide corresponding prevention suggestions for different types of risk in a more targeted way.

Based on the reconfigurable reachability matrix, each factor was plotted as a point in the conventional X-Y coordinate system according to the values of driving power and dependence power. Thus, the 20 key risk factors were divided into four categories and distributed in four quadrants, as shown in Figure 3.

**Figure 3.** The driving–dependence power diagram of the PPP key risk factors.

Each quadrant contains risk factors as follows: Quadrant I (Autonomous risks) includes: R2 (policies and regulations change risk), R3 (government credit risk), R6 (bidding risk), R8 (facility matching risk), R10 (economic risk), R11 (project change risk), R14 (parallel project competitive risk), R16 (force majeure risk), R18 (organizational coordination risk). The driving power and dependence power of the risk factors are relatively low, which means they are not easily influenced by other risk factors, and also do not easy to lead to the occurrence of other risk factors;

Quadrant II (Dependent risks) includes: R1 (government decision-making approval risk), R5 (planning and design risk), R4 (government regulatory risk), R17 (public opposition risk), R19 (environmental risk), R20 (project company violates laws and regulations). They have a powerful driving power and weak dependence power, which are key risk factors and basic conditions in the system;

Quadrant III (Linkage risks): there is no any risk. It means that there is no risk that is susceptible to, and affects other, risks

Quadrant IV (Independent risks) includes: R7 (contract risk), R9 (financing risk), R12 (construction risk), R13 (project income risk), and R15 (operation and maintenance risk). They all have stronger dependence, meaning that the occurrence of these risk factors is largely due to the change or accumulation of other risk factors.

## **5. Analysis and Discussion**

## *5.1. Discussion of Method*

In this study, the FISM is adopted to understand the magnitude of the relationships between the risks [80]. This takes the fuzzy of relationships between PPP risks into account, which has a greater advantage than the traditional ISM method. It also helps in identifying the magnitude of cascading relationships between lower-level issues to higher-level issues [78]. In addition, the method of FISM-MICMAC can provide a new risk assessment tool for risk management strategies in the field of construction engineering and management.

FISM combines fuzzy theory and uses fuzzy language conversion to convert the subjective risks that are difficult to quantify, which is caused by too many uncertain factors in PPP, so as to realize the process of risk evaluation from subjective to objective, and to provide a more scientific theoretical basis for risk response. Compared with traditional ISM, FISM has shown greater advantages [78,80,81,88].

In this study, the reachability matrix is obtained by using the intercept coefficient to fuzzy direct relation matrix. This method shows that risk managers can obtain different reachability matrices through different intercept coefficients, according to their own requirements of the risk correlation degree.

## *5.2. Risk Structure Analysis*

(1) The bottom risk factors of Figure 2 can be classified into two categories. One is likely to occur at any stage in the life cycle of a PPP project, including R4 (government regulatory risk), R20 (project company violates laws and regulations), R19 (environmental risk), and R17 (public opposition risk); whereas R1 (government decision-making approval risk) and R5 (planning and design risk) are pre-project risks. Once these occur, they may cause the formation of a risk chain. Therefore, in the early stages of a PPP project, attention should be paid to the government's decision-making and the planning and design of the PPP project. In the life cycle of PPP projects, there should be focus on government regulation, delinquency of private, environmental pollution and public opposition.

An empirical study in 2011 showed that the three most important risk factors for Chinese PPP projects are government intervention, government corruption, and poor public decision-making processes [89], which is consistent with our research. In addition, we also need to make a special description of R19 (environmental risk), as China imposed an environmental tax in January 2011 [90]. With increasingly strict environmental protection policies, the price of raw materials in the construction industry has been generally rising, which leads to the rising cost of many construction projects which, finally, leads to the failure of projects. From the above analysis, we can see that the fundamental risks affecting PPP projects are not invariable, but rather, are closely related to the actual situation.

(2) From Figures 2 and 3, we found that the R7 (contract risk), R9 (financing risk), R12 (construction risk), R13 (project income risk), and R15 (operation and maintenance risk) have higher dependency values (R15 is in the middle of Figure 2, but in the Quadrant IV of Figure 3), indicating that they are easily affected by other risks. The occurrence of these risks is often caused by the accumulation of other risks and, when these risks occur, they can easily and directly cause the failure of the project, in agreement with the study of Li and Wang [18]. We can prevent such risk factors by controlling the occurrence of the other risks that have an impact on them. In addition, it should be realized that, as these risk factors are easily influenced by other risk factors, their own performance can reflect whether the risks affecting them have been well-controlled or not; that is to say, they can also serve as an indicator of the effect of risk management.

(3) In addition, it can be seen, from Figure 3, that no risk factors are located in Quadrant III, which reflects that there is no PPP risk that can be affected by many factors and can affect many factors, which is consistent with the research of Iyer et al. [19]. This is demonstrates that the key risk factors usually do not lead to project failure through a singular role, but rather, increase the impact on the project through a correlation between risks, cumulatively contributing to the failure of a PPP project.

## *5.3. Comparison with Other Studies*

By comparing Iyer's study on the Indian highway PPP [19], Han's study on the Chinese brownfield remediation projects PPP [79], and Li's ISM study on Chinese PPP [18], we find that:


a government with weak supervision and ability, a private party with inability and irregularities, and the lack of organization and coordination between them, are projected at the bottom of the risk. These risks can be controlled, and the managers and decision-makers must work to control these risks.


## *5.4. Suggestions to the Government*


period for soliciting public opinion, so as to improve public participation, give full play to the role of public supervision, and ensure the normal operations of the project.

*5.5. Limitations of this Study*

This study, has some limitations:


## **6. Conclusions and Future Research**

There is a close interaction between PPP project risks, which has been neglected in relevant literature. Furthermore, the existing research on the interaction of PPP risks lacked consideration of complexity and ambiguity among risks. Based on this, this study has adopted the FISM-MICMAC method to discuss the risk factors and their hierarchical relationship of PPP projects. In addition, the method of FISM-MICMAC can provide a new risk assessment tool for risk management strategies in the field of construction management.

This study started by identifying key risk factors that lead to the failure of PPP projects, using a combination of various risk identification methods, such as case study, literature survey, and expert interviews, in order to identify 20 key risk factors that have an important impact on the success or failure of PPP projects. Then, FISM was used to discuss the PPP project risk factors and their hierarchical relationship, where the hierarchical structure chart of the key risk factors of PPP projects was constructed. Moreover, MICMAC was used to classify the driving power and dependence power of the risk factors, so as to solve the complex problems of the interaction and feedback relationships between key risk factors in PPP projects. The final sequence of factors, which have the obvious characteristics of network, was obtained as a result of an analysis of the dependencies between risk factors.

On the basis of this study, further research can be carried out in the future: (1) Although we have ascertained relationships of influence between risks, the degree of influence, impact speed, and impact mechanisms remain to be further studied; (2) The coupling effect between risk factors needs to be further analyzed, and the overall risk of the project under the coupling effect needs to be evaluated; (3) In future studies, risk factors for the failure of PPP projects can be combined with liability sharing to establish a liability sharing model under the interaction of risks, which can provide a decision-making basis for clearly defining responsibility and compensation for failure.

**Author Contributions:** Conceptualization, X.J.; methodology, X.J. and K.L.; software, K.L.; formal analysis, X.J. and B.X.; investigation, Y.L. and C.C.; data collection, K.L. and Y.L.; writing—original draft preparation, X.J. and K.L.; writing—review and editing, B.X.; funding acquisition, X.J. and Y.L.; supervision, X.J.

**Funding:** The research was supported by the National Natural Science Foundation of China (71672180), the Soft Science Research Plan of Department of Housing and Urban–Rural Development of Anhui Province in China (No. JS2016AHST0011), Innovation Research Plan of Anhui Construction Engineering Group in China (No. W2018JSZX0002).

**Conflicts of Interest:** There is no conflict of interest.






**TableA1.***Cont*.

suspended for a long time.


 risk;

### **Table A1.** *Cont*.

## *Sustainability* **2019**, *11*, 5206

 risk.



## **B. Questionnaire on the Risk Relationship of PPP Projects**

Dear experts,

Hello! Research Group of Dr. Jiang, School of Civil Engineering, Hefei University of Technology, is carrying out research on " the risk relationship of PPP projects ". Our research group sincerely invites you to give your opinion on this survey based on your past experience and relevant knowledge in PPP projects. Your opinion is very valuable and will play an important role in our research. The data and information collected in this questionnaire will only be used for academic research and will not negatively affect your daily work and life. Thank you for your understanding and support!

Jiang Group

Table 1 shows the list of key risk factors of PPP and their meaning explanation. This survey gives a list of risk factors according to the past practice in the 28 failure PPP project cases, related literature survey, and expert interview. After familiarizing the risk factors in Table 1, please complete Table 3 according to the reminders in Table 2.


**Table 3.** Risk factors and their meaning explanation for PPP projects.


**Table 3.** *Cont*.

According to the scoring in Table 2, please judge the mutual relationship of two factors, and fill the evaluation results into the corresponding positions in Table 3.


**Table 4.** Language operator, triangular fuzzy number, and scoring.

**Example**: when you consider the impact of the risk factor R1 on R2 to be very low impact, you should fill the "1" in the second row and third column of Table 3.


**Table 5.** Mutual relationship intensity between two risk factors.

Note: the risk factors affecting are column and the risk factors bearing the influence are row; there is an asymmetric relationship between risk factors. Please complete this questionnaire as soon as possible. Thank you again for your patience.

## **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Development of a Bridge Management System Based on the Building Information Modeling Technology**

**Chunfeng Wan 1,\*, Zhenwei Zhou 1, Siyuan Li 1, Youliang Ding 1, Zhao Xu 1, Zegang Yang 2, Yefei Xia <sup>2</sup> and Fangzhou Yin <sup>1</sup>**


Received: 20 July 2019; Accepted: 20 August 2019; Published: 23 August 2019

**Abstract:** With the development of the Chinese transportation industry, the number of bridges has increased significantly, but this results in high pressure of structural maintenance and management. Bridge management system (BMS) is critical for efficient maintenance and ensured safety of bridge structures during long-term operation. Building information modeling (BIM) is an emerging technology with powerful visualization and informatization capability, making it an ideal tool for developing modern management systems. This paper introduces the development of a bridge management system based on the BIM technology. Industry Foundation Classes (IFC) and International Framework for Dictionaries (IFD) standards are studied and extended, and coding rules are proposed for the Chinese bridge industry. Also, a standard structural modeling method is proposed to fast build the bridge BIM model. Web-BIM oriented bridge management is proposed, and portable devices are introduced into the system. Collaborative management is realized for different users. The BIM-based maintenance management system is designed. Finally, a practical BIM-based BMS is established for a long-span cable-stayed bridge in China. This system integrates the BIM with the geographic information system (GIS) and contains information management, inspection management, technical condition evaluation, and enables users to cooperate with each other. Such a BMS could help to improve the management efficiency and ensure its normal operation, providing a useful platform for the maintenance of massive bridges in China.

**Keywords:** BIM; information model; bridge; maintenance; management system

## **1. Introduction**

In recent years, with the rapid development of the economy and the advancement of urbanization, China's bridge construction has reached its crescendo. Bridges play an important role in the transportation system by supporting economic and social development and are therefore regarded as one of the most important and indispensable infrastructures. According to the statistics, there are currently more than 800,000 road bridges in China, ranking first in the world [1], which gives severe pressure to their maintenance. However, at the present stage in China, people still mainly focus on construction but not maintenance. During the long-time operation of the bridges, bridge structures suffer different levels of damages caused by loads and environmental effects, which then reduces the structures' reliability and shortens their expected service life-time. For example, severe rust and breakage of the boom can result in bridge collapse, such as the destruction of the Yi Bin south gate bridge in China. There are also many similar cases of bridge defect or even destruction accidents, and the main cause is the lack of maintenance [2,3].

Bridge management system (BMS) is an integrated computer system to provide decision support throughout design, construction, operation, and maintenance phases. BMS can improve the efficiency of management and decrease redundant costs in dealing with infrastructure management issues [4]. Many countries invested a great deal of resources and efforts to develop efficient BMSs, including Finland [5], Denmark [6], Germany [7], and Japan [8]. In China, Chinese Bridge Management System (CMBS) was developed by the Ministry of Transport of China (MOT) in 1993 and has been widely used in China after continuous improvement. The main functions of CBMS are to master the basic data of highway bridges, supervise the regular inspection, and provide technical support for maintenance [9]. However, there are still many problems during the application of CBMS. Firstly, the information stored in the system is fragmented and isolated. Secondly, it is hard to realize visualization when the maintenance information is separated from the bridge model and makes it difficult to handle for users. Additionally, bridge engineers and managers find it difficult to collaborate with each other through the system.

In recent years, building information modeling (BIM) has become the most flourishing technology in the building industry, and it has been extended to infrastructure engineering [10]. This technology is a new approach and can be used for design, construction, and facilities management of structures, wherein a digital representation of the building process is used to facilitate the exchange and the interoperability of information [11]. The application of BIM brings about cost reduction, quality control, and efficiency improvement throughout the life cycle of the project. A survey by Stanford University in the United States pointed out that BIM can eliminate 40% of extra changes, reduce the contract price by 10% by discovering and resolving conflicts, and shorten the project duration by 7% [12]. With these advantages, BIM has also been adopted by commercial software, such as Autodesk Revit, ArchiCAD, and Allplan [13]. Furthermore, BIM has been successfully implemented in many bridges' designs [14,15] and constructions [16–18]. However, the application in the maintenance phase started relatively late. McGuire and Atadero [19] utilized the BIM to manage the inspection and evaluate information. Abudayyeh and Al-Battaineh [20] adopted the as-built bridge information model for maintenance and management. Some scholars [21–23] also combined the BIM with the traditional management system to improve maintenance efficiency.

In China, however, the lifecycle bridge management system based on the BIM technology is still under development. This paper introduces the development of a BIM-based BMS, which integrates the geographic information system (GIS), develops the necessary Industry Foundation Classes (IFC) and International Framework for Dictionaries (IFD) standards, which are still in blank in China, implements the Web-BIM oriented management, and realizes the collaborative work among all related users. This BIM-based bridge management system is supposed to improve the efficiency of maintenance management in order to satisfy the keen demand of maintenance management from substantive bridges in China.

## **2. System Goal and Requirement**

With the huge inventory of Chinese bridges, maintenance of these bridges becomes especially important and challenging. An informationalized management system is helpful and necessary. Many BMS have been developed, and fairly good practical results have been achieved, such as GIS-based BMS [24], visual inspection data-based BMS [25], and consecutive condition assessments-based BMS [26]. In this paper, we introduce a BIM-based BMS. The main target is to increase the management efficiency, make it visualized and easy to use, and provide a collaborative management platform for all people involved in the projects. To realize it, the bridge management system should:

(1) Integrate the BIM with the GIS. As a node of the transportation network, geographic information is crucial to the bridge. Moreover, our management is not only focused on a single bridge but the bridges at a certain transportation line or a certain area. Information of the group of the bridges helps to grasp the overall condition of the transportation and gives us guidance in maintenance.

(2) Realize the Web operable BIM system. Confining the users to computers is no longer welcome. The system should be operable at the Web, reducing the dependency on heavy BIM software, and, most importantly, making the portable device based operation possible.

(3) Have a unified database. The database of the system should support different kinds of data. Information should have the same coding rule so that all data can be searched, retrieved, and analyzed. Interaction between systems can also be implemented.

(4) Realize the function of structural inspection and monitoring. Condition of the bridges reveals the safety of the bridges. Structural inspection, health monitoring, as well as the performance evaluation are therefore the most important in the management system.

(5) Realize the collaborative management. By collaborating, all people with different jobs work together; collaborative management is a key issue to increase the efficiency of the management.

(6) Achieve the multi-scale visualization during the management. Visualization is a significant advantage of the BIM-based management system. It makes the system easy to understand and easy to grasp for people.

As a popular informationalized method, the term BIM itself has several definitions, such as a product, a collaborative process, and a facility life cycle management requirement [27]. It can also be regarded as a combination of the concepts of building information modeling and building information management. Three-dimensional representation, information integration, and digital models are the three significant features of BIM. The collaborative management system utilizes the advantages of BIM to provide better service for bridge maintenance, which makes up for deficiencies of the traditional management system. Considering the merit of the BIM technology, a BIM-based bridge management system is developed in this paper to improve the maintenance efficiency and release the management pressure of large numbers of existing bridges.

## **3. Key Points of the Proposed Bridge Management System**

## *3.1. BIM Standard Extension*

As mentioned earlier, some relevant standards are still incomplete or blank in the bridge industry in China. The lack of uniform BIM standards has become one of the main obstacles to the application and the development of BIM [28]. Compared with other industries, civil engineering projects usually include many stages—planning, design, and construction—and involve many participants. Different types of BIM software are adopted in the process. Without a unified BIM standard, it is hard to achieve information sharing among different stages and software, which leads to a large number of repetitive works. It is necessary to establish information storage and exchange standard. Otherwise, the values of BIM are not realized fully.

## 3.1.1. IFC Standard Extension

The IFC standard is an open international standard for the expression and the exchange of building product data, supporting data exchange, and sharing throughout the life of a building project. The architecture of the IFC standard consists of four levels; from the bottom to the top, these are resource layer, core layer, interoperability layer, and domain layer.

At present, the IFC standard is incomplete, and it is hard to cover the whole life-cycle information of a bridge [29]. Thus, it is necessary to make an extension by using EXPRESS language based on the existing IFC framework. The bridge member, such as beam and abutment, can directly refer to IfcBuildingElement in the existing IFC framework. However, structural defect, inspection, and evaluation information should be defined by extending the attribute set. Table 1 defines the attribute set of bridge defect information and reinforcement information. Tables 2 and 3 list the specific attribute name and the type of the attribute set. Table 4 lists the attribute definition of the concrete crack. Finally, the bridge maintenance information is attached to the bridge members through IFC relationship entity. In this way, the IFC framework targeting at bridge maintenance is built completely, which provides a uniform standard for information exchange and delivery. Bridge information models can be imported and exported by mainstream BIM software such as Autodesk, Bently, Dassault, etc. Since the unstructured data cannot be directly expressed in the existing IFC standard, it is necessary to extend to encoding the attribute information. Parameter and

attribute information sets can be added to unstructured data, including disease information, assessment information, files, pictures, and videos. As shown in Figure 1, the attribute information set combined with the structured information through data sources is compressed together and integrated with IFC parsing. Thus, the entity model of the IFC extension for bridge disease and evaluation is completed, and the structural defect extension rules are formed.



**Table 2.** Attribute definition of structural defect information.


**Table 3.** Attribute definition of reinforcement information.



**Figure 1.** Expression of attribute information using Industry Foundation Classes (IFC) extension.

## 3.1.2. IFD Standard Extension

The descriptions of information are usually distinguished in different languages and under different situations, which makes information uncertain and inaccurate. The IFD defines a globally unique identifier for each conception. In China, IFD standards have been established in the field of architecture and railway industry [30]. However, the IFD standards in the highway field are still blank in China. This paper proposes the IFD standard for Chinese high way industry and makes a supplement for the highway bridge maintenance.

International Standard Organization (ISO) provides the information classification framework (ISO 12006-2). This framework [31] divides the building information into three categories: construction resources, construction processes, and construction results. Construction results are produced by applying construction resources to the construction process, as Figure 2 shows. Among them, the primary construction resources include products, tools of production, roles of bridge engineers, roles of bridge engineering organizations, information, materials, attributes, etc. Meanwhile, the construction process includes the behavior of bridge engineering, the project phase of bridge engineering, the professional field, etc. The construction results consist of the structures classified by function, morphology, and so on. The classification and the encoding of the bridge maintenance information can refer to or extend from the existing IFD standards. For the cases where the current standards cannot satisfy the needs, information needs to be classified and encoded based on the ISO 12006-2 framework.

**Figure 2.** Classification framework.

The numerical method is applied to encode the information. The coding consists of two parts: table name and coding of different levels (each two numbers represent one level), which are separated by a dividing line. A part of coding is to directly quote IFD standards in the field of construction, while another part is to extend the building standards. Coding should independently compile the information. Among them, codes with initial numbers 71, 72, 41, 74, and 78 indicate bridge type, bridge component, attribute, bridge behavior, and bridge characteristics, respectively. The coding principles of the bridge type and the bridge component are shown in Figure 3. Because of the limitation of space, Table 5 only lists a part of the engineering behavior information coding in the bridge maintenance phase. Table 6 gives a part of the structural defect, evaluation, and reinforcement information coding of the bridge. Defect levels, associated maintenance, and repair priorities are defined as well.

**Figure 3.** Coding principles using International Framework for Dictionaries (IFD) standard extension: (**a**) bridge type, (**b**) bridge component.


**Table 5.** Engineering behavior information.



## *3.2. Structural Modeling Method*

In the bridge life-cycle planning, the information model is expected to be established during the design and the construction phases and then delivered to the maintenance phase. However, at present, most existing bridges did not apply the BIM technology in design and construction. Besides, some as-built information models make it hard to satisfy the needs of bridge maintenance due to the restriction of software. As a result, a bridge information model for maintenance needs to be rebuilt in most cases. This part introduces a structural modeling method by using Autodesk Inventor, taking the Grand Canal Bridge modeling process as an example. The background of Grand Canal Bridge is introduced later in Section 5.

Autodesk Inventor is modeling software that has strong curve design capability, powerful assembly, and information integration functions, which gives it the ability to build the model of the Grand Canal Bridge. Since the workload of modeling is large, it is necessary to unify the basic information such as coordinate system, unit, and storage path before building the model. The specific modeling steps are as follows:

(1) Establish component models. The Grand Canal Bridge is divided by structural components such as bridge decks, abutments, pile foundations, etc., and each component is modeled separately by creating a "part" in the Autodesk Inventor. However, to build all the components one by one is inefficient. The software provides a parametric method, which can be used for building different types of components effectively.

(2) Assembly of components. The separate components need to be assembled to form the bridge model. For components with geometric constraints, users can assemble them directly. For component types without direct constraint relationships, the reference surfaces and the reference points need to be created to help determine the location of the components. The assembly details can be seen in Figure 4.

(3) Level adjustment. The requirements of the model for maintenance are somewhat different from design and construction. The model should have a hierarchical characteristic in the maintenance phase because the structural safety assessment is usually carried out upwards from the bottom layer to the top layer. Engineers could adopt the degrading function of the Autodesk Inventor to adjust the hierarchical relationship among the components. The final model of the Beijing-Hangzhou Grand Canal Bridge is shown in Figure 5.

(4) Information integration. After completing the model building, the information of design and construction need to be attached to the bridge model. Autodesk Inventor provides common information addition. If the types of information are not provided, users also can make a customized expansion. The Bill of Material (BOM) table can be used for adding information in bulk.

**Figure 4.** Modeling process.

**Figure 5.** The Grand Canal Bridge model.

## *3.3. Web-BIM Oriented Management*

Traditional bridge management systems are usually based on computers. Users need to install all the software at computers and finish all the work in front of the computers. However, from another point of view, users are confined to those computers. Nowadays, portable devices, such as pads and cell phones, give significant convenience and flexibility to their management works. However, portable devices do not have powerful enough central processing units (CPU) or sufficient memory and cannot support heavy BIM software such as the Revit. Also, expensive BIM software needs to be installed at each user's computer. Data such as inspection data and monitoring data usually need to be copied into some memory devices and then inputted into the system. All of these aspects make the management system expensive and inefficient. Web-BIM techniques have been developed and applied in the building management system but seldom in the bridge management system. Here, we propose the Web-BIM based management system with the lightweight techniques, thus the system can be operated at Web and could adapt to different users.

To develop the Web-BIM system, the structural model is required to be built using BIM software such as Revit, and then the model is exported and transferred to a certain file with OBJ format. Such OBJ files can then be uploaded to the server and used as a calling file to the webpage. By running the JavaScript software and calling the WebGL program, the structural model can then be transferred to a 3D web file and can be displayed on internet browsers.

For a bridge structure, especially the large-scale bridges, the converted structural model is often very huge and can hardly be displayed and operated smoothly. The lightweight model technique is necessary to reduce the data volume of the model. To realize that, a three-step lightweight approach is implemented as follows:

(1) Establish a hybrid space index. Separate the model into different scales and build the index. Trim the model to show only the information within the visual field.

(2) Establish the Levels of Detail (LOD) to the model data. Erase the information related to the unnoticeable details.

(3) Reduce the rendering batch numbers. Incorporate the objects with the same texture and render them together so that rendering efficiency can be significantly increased.

## **4. Design of BIM-based maintenance management system**

The development of the bridge management system requires the cooperation of different specialties. The system takes management as the basic requirement, information technology as the means, and bridge professional knowledge as the basis. This part introduces the detailed process of the system design.

## *4.1. Computer Network Design*

As Figure 6 shows, C/S (client/server) models and B/S (browser/server) models are currently mainstream network structures. The C/S model belongs to a distributed application structure that can partition tasks or workloads between the providers of a resource or service, called servers, and service requesters are called clients. Often, there is communication between clients and servers over a computer network on separate hardware, but both the client and the server may reside in the same system. The device utilization is highly efficient, and the security can be ensured. However, this structure is less scalable and is generally limited to local networks. The cost for the system's maintenance and upgrade is high. The B/S model developed along with the rise of the Internet. This model centralizes the core part of the system's functional implementation to the server, which greatly simplifies the client computer load and reduces the cost of system maintenance and up-gradation. Users only need a web browser to get access to the system without installing a client. B/S model's security is relatively poor, which is why some measures such as permission control are needed.

**Figure 6.** Computer network.

In the bridge maintenance project, a large number of stakeholders are involved. The management system usually has many users, including investors, managers, and engineers. They are likely to get access to the system from different locations and with different access methods [such as a local area network (LAN), a wide area network (WAN), and Internet/Intranet]. Taking the convenience and the cost into consideration, the B/S structure is in accordance with the Web-BIM oriented management and therefore more applicable to the BIM-based management system.

### *4.2. System Framework*

The BIM-based system mainly includes four parts: data collection system, data center, model layer, and evaluation system. The framework of the system is shown in Figure 7, which displays the connections and the relationship between the different modules of the system.

The data center plays a significant role in the system, which consists of four layers: cloud resource layer, access layer, storage layer, and application layer. The cloud resource layer provides computing services by renting existing mature cloud computing resources for implementing various functions of the data center. The access layer provides a unified interface for the import of different types of data. The storage layer provides centralized storage of all bridge management-related data. The application layer includes the implementation of various user-oriented functions such as data query, data management, and database management tools.

The data center obtains static and dynamic data of bridges from scattered data resources through the data acquisition system of bridge maintenance. The dynamic data collection system includes two parts: the front-end collection program and the background management program. The front-end collection program is mainly the mobile APP software, which is responsible for real-time collection, input of maintenance data, and uploading information to the background management program. During the process, the background management program is responsible for managing the collected data and generating reports.

The bridge automatic analysis and evaluation system relies on the data center to obtain various types of data required for bridge condition assessment, bridge data statistics, and in-depth analysis. Through the calculation and the analysis of relevant models and algorithms, output evaluation, statistics, and analysis results are obtained and submitted to the data center. Also, the evaluation results would be presented in the bridge model.

The model layer is one of the important distinguishing points of the BIM-based bridge maintenance management system compared with the traditional systems. Because the system is based on the three-dimensional model for operations, it can make many maintenance services more intuitive and vivid. Considering that the size of the model is huge and takes up too much memory, the model is then transformed into a lightweight model to better display in the portable terminals. The model layer correlates the information such as inspection, evaluation, and maintenance reinforcement in the database with the component of the bridge, thus different information can be displayed through the model. Therefore, the model layer part effectively connects the data layer part and the functional layer part together and plays a central role in the whole system framework.

**Figure 7.** System framework.

## **5. The application on the Grand Canal Bridge**

## *5.1. Project Overview*

The Beijing–Hangzhou Grand Canal is the longest artificial river in the world, starting at Beijing, passing through Tianjin and provinces of Hebei, Shandong, Jiangsu, and Zhejiang, and finally arriving in the city of Hangzhou. The total length of the Grand Canal reaches 1776 kilometers. The Beijing-Hangzhou Grand Canal Bridge is located at the south of Yangzhou, Ning yang Expressway. As is shown in Figure 8, the main bridge across the Grand Canal is a double-tower semi-floating system concrete cable-stayed bridge, the length of which is 464 meters (108 m + 248 m + 108 m). On each side of the main bridge is 11-span (6 × 30 m + 5 × 30 m) prefabricated, prestressed concrete approach bridge.

**Figure 8.** The Grand Canal Bridge: (**a**) bridge deck; (**b**) whole bridge.

## *5.2. Implementation in Grand Canal Bridge*

Beijing-Hangzhou Grand Canal Bridge is classified in the second level according to its present structural condition. Many shrinkage cracks have appeared on the cable-stayed bridge tower and the main beams of the bridge. Some protective covers of the cable anchorages have deformed severely. Moreover, there are a lot of longitudinal cracks on the box girder of the approach bridge and transverse cracks on the wet joints. Some of the laminated rubber bearings have been damaged.

In order to realize the effective management and maintenance of the Beijing-Hangzhou Grand Canal Bridge, the BIM-based management system was built. Firstly, the bridge model was established using Inventor, and a special lightweight processing that could significantly reduce the data volume was implemented to form the model layer. Then, the lightweight model was connected to the GIS system, which showed the geographic information of the bridge, as Figure 9 shows. Also, it is associated with the bridge maintenance information database. The database is an important foundation for further evaluation and decision-making, which are the core parts of the function layer.

**Figure 9.** System homepage.

With the Web-BIM oriented approach, the stakeholders of the bridge can get access to the system through the Internet. The homepage of the system includes three parts: function modules, bridge model, and static information, as Figure 9 shows.

## 5.2.1. Static information module

Static information management in this BMS platform was established and is shown in Figure 10. This system accumulates the scattered information in the design and the construction phases to formulate the "bridge life card". The results indicate that users could choose the members in the bridge model to query corresponding static information such as design drawings, design parameters, material usage, etc. Moreover, the model tree is one type of dendrogram based on the hierarchy characteristic of the model, which is provided for accurate localization of members in the model. As Figure 11 shows, users can unfold the dendrogram to click the pile foundation of thirteenth span, which was highlighted in the model automatically.

**Figure 10.** Visualization management of static information.

**Figure 11.** Pile foundation of thirteenth span in the model tree.

## 5.2.2. Inspection and evaluation module

Inspection and evaluation have great significance for BMS. Based on this, we firstly designed an inspection workflow to show how the bridge managers create and manage the inspection projects on the website. As Figure 12 shows, the inspectors could log into the system through a mobile application at the spots, collect all the structural defect information of the bridge, and then submit to the system. Large bridges have high numbers of members; inspectors can localize the member quickly by choosing it on the lightweight bridge model. In order to improve the accuracy of defect information description, the system provides a standardized template according to the statistics and the classification of common defects. After inspectors submit the defect information, managers can check it in the system, as Figure 13 shows.

**Figure 12.** Inspection workflow.

**Figure 13.** Structural defect information in inspection module.

Based on the defect information collected by the mobile terminal, the evaluation system can make automatic analysis and calculate the scores of the members according to the bridge maintenance standard. Then, the bridge information model displays different colors on the members to intuitively describe the technical condition of different levels of the whole bridge. Figure 14 shows the evaluation results of the Grand Canal Bridge. The members in green color stayed in a safe state, and the members

in yellow suffered light damage. The red color means the members have been damaged severely, and corresponding measurement should be considered.

**Figure 14.** Evaluation module.

Repair and reinforcement information of the bridge is stored in the system and can be searched by users. The defects of the bridge, such as cracking, usually develop as the time grows. When the defect develops to a certain extent and threatens the structure safety, the system automatically sends text messages to bridge managers to remind them to create a repair and reinforcement project. Besides, the system regularly generates statistics, graphs, and tables according to inspection and monitoring information. Based on these functions, the BIM-based system helps to make auxiliary maintenance decisions. As shown in Figure 15, users can view the relevant information on the repair plan in this module. After this reinforcement and repair, one can determine whether the maintenance is qualified or not according to the actual situation. The maintenance can be displayed through the model, and the repair results can be check by the managers, as shown in Figure 16. Based on this, the detailed information of the location and the effects of this reinforcement can be obtained.


**Figure 15.** The repair plan of bridge components.

**Figure 16.** The information of component maintenance in repair and reinforcement module.

## *5.3. Results and system benefits*

After this maintenance system was built, it was applied to the company in charge of the maintenance of the Grand Canal Bridge. Feedback was quite positive. Web-BIM oriented management mode allows all users easy access to the system. Portable devices have also been applied in the bridge inspection procedure. By scanning the tag at a component, the component can be identified, and related structural defects can be directly reported by text description, picture, or report via the portable device on site. The established BMS also provides a good platform upon which all users can collaboratively work together. Owners, engineers, economists, managers, and other related staff can share their data, release/receive tasks, track the progress, and even communicate with each other. Multi-scale visualization of the bridge model helps to understand the structural condition easily. Structural analysis and diagnosis algorithms can be embedded in the system and make the system smart. All of these make the maintenance activity very convenient and efficient. It will be helpful for the maintenance of many bridges in China, and its application is promising.

## **6. Conclusions**

The number of bridges has increased significantly within the last three decades in China. The large number of the bridges, however, brings great pressure to the owners, the industry, and even the government for their maintenance. An effective and efficient bridge management system has become indispensable to meet the massive demand of bridge maintenance.

In this paper, a bridge management system based on the BIM technology was proposed, developed, and introduced. Necessary IFC and IFD standards were proposed as supplements according to the actual needs of bridge maintenance, filling the blank in BIM standards of the bridge industry of China. Web-BIM oriented management was proposed, and portable devices were introduced to the system as well as the maintenance activity. The bridge modeling method was developed to build the proper information model for bridges. Lightweight methods were also introduced to reduce the volume of the model and make the system run smoothly. The system framework and functions were designed. A BMS for a real long-span cable-stayed bridge was also built and established, which incorporates the GIS and satisfies the main maintenance functions, such as information management, bridge inspection, condition evaluation, repair and reinforcement, multi-scale visualization, and collaborative management.

The proposed BIM-based bridge management system could improve management efficiency and satisfy the need for maintenance of substantive bridges in China. The future of such a management system is promising. However, at present, the BIM model cannot be connected to the finite element model (FEM), which strongly limits the structural analysis ability. Also, data in the system are not mined deep enough. Integration of the BIM model and the FEM model as well as the deep mining of data using big data technology will make the system much more powerful and useful and could provide a better platform for efficient bridge management.

**Author Contributions:** Conceptualization, C.W. and Y.D.; methodology, C.W. and Y.D.; software, Z.X., Z.Y. and Y.X.; formal analysis, Z.Z. and S.L.; investigation, Z.Z. and S.L.; data collection, Z.X. and F.Y.; writing—original draft preparation, Z.Z. and S.L.; writing—review and editing, C.W. and Z.Z.; funding acquisition, C.W. and Y.D.; supervision, C.W.

**Funding:** The research was supported by the National Key Research and Development Program of China (2017YFC0806001), the National Natural Science Foundation of China (No. 51578140), the Postgraduate Research & Practice Innovation Program of Jiangsu Province (No. SJCX18\_0027).

**Conflicts of Interest:** There is no conflict of interest.

## **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Promoting Owners' BIM Adoption Behaviors to Achieve Sustainable Project Management**

## **Hongping Yuan 1, Yu Yang 2,\* and Xiaolong Xue <sup>1</sup>**


Received: 20 June 2019; Accepted: 12 July 2019; Published: 18 July 2019

**Abstract:** Although building information modeling (BIM) has a promising future in the architecture, engineering and construction industry, its wider adoption and implementation is desired. Grounded with a technology-organization-environment (TOE) framework and the theory of technology acceptance model (TAM), this study extracted "social influence", "organizational support", "BIM technical features", and "government BIM policies" as four key external antecedents—in reference to the particular BIM practices in China—and proposed a model to predict project owners' BIM adoption behaviors. To test the proposed model, structural equation modeling (SEM) analysis was applied for configuration analyses on a sample of 188 project owners from the Chinese construction industry. Results show that BIM technical features, and government BIM policies have positive effects on perceived usefulness, but social influence and organizational support have no significant influence on perceived usefulness. Furthermore, both social influence and BIM technical features have positive effects on perceived ease of use, while organizational support and government BIM policies have no significant influence on perceived ease of use. Attitude plays a significant intermediary role among perceived usefulness, perceived ease of use and behavior intention. Additionally, attitude significantly affects behavior intention, and behavior intention can also affect BIM adoption behavior. This study is the first attempt to investigate project owners' behaviors toward BIM adoption and the findings are expected to provide a better understanding of the essential elements of project owners' BIM adoption behaviors and guide industry practitioners in developing proper strategies to achieve more effective BIM implementation.

**Keywords:** building information modeling; project owner; attitude; behavior; technology acceptance model

## **1. Introduction**

In the last decade, with the rise of information technologies (ITs), a paradigm shift of industrial informatization has translated into a critical national strategy [1–3]. As a pillar of the domestic economy, the architecture, construction and engineering (AEC) industry in China is on the cusp of transition from an extensive and high-consumption pattern to a new one driven by high efficiency, sustainability and informatization. According to Eastman et al. [4], BIM is "a new approach to design, construction, and facilities management, in which a digital representation of the building process (is used) to facilitate the exchange and interoperability of information in digital format". Therefore, BIM is an innovative paradigm of building information digitalization resorting on certain specific technologies or software which integrates cash flows, information flows, logistics throughout the project lifecycle and reduces the information asymmetry, unforeseen changes and re-doings effectively, turning the utopia of construction visualization to reality. In the meantime, existing literature suggests the integration of environment and economic assessment for the promotion of sustainable construction

is considerably important [5,6], and it happens that BIM is an ideal tool which can integrate the assessment of sustainable construction as well as resource management efficiently, such as benefit-cost analysis of economically sustainable design, energy-consumption analysis for a sustainable built environment assessment, architectural information sharing for sustainable facilities management and stakeholder relationship management. Therefore, a wide adoption and application of BIM is bound to strike and even overturn the traditional development patterns of the Chinese AEC industry, embedding these sustainable assessments throughout project lifecycle and thus contributing to sustainable project management.

As such, BIM adoption has become one of the central topics among AEC studies over the past decade. Previous studies have attempted to determine the major factors motivating BIM adoption among project stakeholders, with the aid of various means including questionnaire surveys, interviews, and case studies. For instance, Cao et al. [7] found that the motivation of design units and general contractors in BIM adoption is closely linked to the characteristics of organizational nature and project scale. Meanwhile, evidence from the comparative case study in China and Australia indicated that BIM adoption strategies vary in building construction and infrastructure engineering industries [8]. There are also studies that have compiled and sorted a collection of factors that influence BIM adoption, such as effective leadership and organizational support [9–12], sufficient BIM human resources [13,14], and the availability of information and technology [11,15]. Similarly, the large amount of capital required for BIM adoption and application is also a Gordian knot to be unhitched by potential BIM participants [13,16,17]. Moreover, the lack of universal standards of BIM implementation [18] and the indistinct legal bounds of a series of work outcomes related to BIM (such as the BIM model) make the environment of BIM implementation immature, and this restricts the adoption and application of BIM [14].

The above studies are significant in promoting a wider adoption and application of BIM through addressing major barriers. However, unfortunately, the majority of current literature has ignored a vital stakeholder in AEC projects—the project owner, (e.g., government, real estate developers), who takes overall responsibility for project investment, initiation, construction or even the operation and management of facilities. Holding preponderant advantages in project planning and controlling the entire project lifecycle, the project owner could enhance project performance by requiring and driving other stakeholders (such as architects, general contractors and so on) to get involved in BIM adoption and implementation. Some implications from prior literature have proven the important role of the project owner in driving BIM adoption. For example, Ling et al. [19] pointed out that the superiority of project owners can effectively promote the application of innovative technologies. In a recent study, Cao et al. [7] also found that project owners' support for BIM application can facilitate better stakeholder cooperation and get more stakeholders engaged in BIM implementation. However, these studies are mainly conducted based on qualitative analysis or mostly focus on identifying factors hindering BIM acceptance and adoption and, thus, fail to reveal the mechanism that drives BIM adoption behavior. In addition, very limited studies have attempted to investigate project owners' BIM adoption behaviors, though some of them have suggested that project owners are critical in promoting BIM acceptance and adoption.

Given the research gap, we believe that understanding why the project owner adopts BIM is an important step in increasing the use of BIM within projects and potentially improving BIM adoption efficiency. The research questions we are attempting to answer are "What factors influence project owners' acceptance of BIM? And how do these factors result in project owners' final BIM adoption behaviors?" To address these research questions, we have developed and tested a model integrating the TOE framework and the theory of the technology acceptance model (TAM) to explain project owners' BIM adoption behaviors.

We also believe that the lack of a theoretical foundation for this stream of research has limited the contributions of previous research and prevented project stakeholders from understanding what makes effective BIM adoption and implementation. It is necessary to understand the effects of these factors hindering BIM adoption and determine the critical path impact on this behavior to develop effective BIM adoption measures and design practical strategies that can lead to wider BIM application. The present study helps BIM researchers describe how project owners' adoption behavior can be driven and pilot project owners make informed decisions as to what strategies they can use to promote BIM application in their projects and organizations.

The rest of this paper is organized as follows. The next section briefly reviews research on BIM adoption factors, the theory of the TAM and the technology–organization–environment framework. Then, we present research hypotheses and the research model, followed by an introduction of the research method including the instrument development and validation process. After that, we present data analyses and results, and suggest the implications for research and practice as well as limitations of the current research. Finally, we conclude this paper with a brief summary in the conclusions.

## **2. Literature Review**

## *2.1. Influencing Factors of BIM Adoption*

BIM has been recognized as a pivotal information technology in the AEC sectors due to its strength of integrating the continuous flows of funds, information and logistics throughout the project lifecycle [4]. Admittedly, the adoption and application of BIM would inevitably make a profound impact on driving the development of informatization in the AEC industry. Thus, BIM has attracted a lot more attention from AEC researchers in recent years. For instance, Gu et al. [20] posited that although BIM develops with promising prospects, both technical and non-technical factors hamper its diffusion. Based on BIM implementation practices in China, Cao et al. [7] found that the BIM competitiveness of construction firms is closely related to the social network structure in which they are located. An empirical study by Son et al. [21] showed that top management support, subjective norms, and technical compatibility are most important in affecting designers' BIM adoption. In addition, qualified employees, efficient leadership, the availability of information, and the complexity of the project itself are also fundamental factors for successful BIM implementation [11,13]. Based on investigations on potential BIM adopters in the UK, Howard et al. [22] suggested that performance expectations do not directly affect the adoption bias of BIM potential adopters, but improving the strategic policy and the incentive mechanism would be a great help for accelerating BIM diffusion. Liu et al. [23] pay more attention to factors of BIM cooperation from the individual, technological and organizational dimensions among design and construction firms. For BIM users, more emphasis was placed on the information quality needed when implementing BIM and related exoteric services, because these factors will directly affect their satisfaction with BIM [24]. Additionally, based on institutional theory, Cao et al. [25] found that the homomorphism from both mandatory and imitation systems would significantly affect the application of BIM at the project level, and the support from project owners would be conducive to the acceptance of BIM in certain circumstances. By conducting a case study based on the theory of innovation diffusion, Gledson et al. [26] identified the inter-organizational factors driving BIM diffusion at the project level and provided exhaustive schemes to address the individual, managerial, environmental, and technological challenges experienced by construction firms in the process of BIM diffusion.

## *2.2. Technology Acceptance Model*

Currently, the technology acceptance model proposed by Davis [27] (see Figure 1) has become a classical and parsimonious model, which has been widely used to explain the behavior of information technology adoption or acceptance. In line with the TAM, perceived usefulness and perceived ease of use are regarded as two essentials to explain the use of a technology [27]. According to the theoretical framework of the TAM, an individual's information technology adoption behavior is determined by his/her behavioral intention. Moreover, attitude and perceived usefulness influence his/her behavioral intention of using a technology, which would necessarily in turn affect the actual system usage. As a key construct, attitude is influenced by both perceived usefulness and perceived ease of use. Davis [28]

also found that, to some extent, an individual's perceived ease of use of a particular technology will affect his/her perceived usefulness of this technology. In addition, external variables (such as technical features, user intervention, etc.) can indirectly affect user behavior through perceived usefulness and perceived ease of use [28]. Because there is no specific and strict constraint to external variables, the TAM is powerful in explaining user behavior of an information system with high parsimony and conciseness. As such, the TAM has been widely applied to predict various types of technology acceptance behaviors, including Smart Grid [29], virtual reality [30], LNG [31], and transportation [32]. In the context of BIM adoption, the research of Lee et al. [10] was one of the few existing studies that applied the TAM to investigate the BIM adoption of designers, contractors and engineers. However, it still did not approach the crucial role that project owners play in BIM adoption processes.

**Figure 1.** Classical technology acceptance model.

## *2.3. Technology–Organization–Environment Framework (TOE)*

The TAM, however, has some limitations when extended beyond the workplace because its fundamental constructs do not fully reflect a variety of the user task environments and constraints. Furthermore, Legris et al. [33] also suggested that the TAM is a useful model but needs to be integrated into a broader model that includes variables related to both human and social factors. To take these limitations of the TAM theory into account, in this present study, we incorporated the technology–organization–environment framework (TOE) (see Figure 2) as the theoretical foundation to clustering various impacts on project owners' BIM adoption. This framework describes factors that influence technology adoption and its likelihood, and the process by which a firm adopts and implements technological innovations is jointly influenced by the technological, organizational, and environmental contexts.

**Figure 2.** Technology–organization–environment framework.

Since its origination, the TOE framework has aroused increasing attention and been applied in the elements and factors studies of technology innovation among research fields such as tourism, manufacturing (3D printing, RFID, etc.), business (electronic data interchange, customer relationship management, etc.), and project management [34–39]. As indicated by previous studies, the TOE framework is supported by an abundance of empirical results, so it offers a solid foundation to unravel the conundrum behind project owners' BIM adoption decisions.

## **3. Theory and Hypotheses**

In this study, we extend the classical TAM by identifying "social influence", "organizational support", "BIM features", and "government BIM policies" based on the TOE framework as external variables and propose a conceptual model to predict project owners' BIM adoption behavior (see Figure 3). The model is explained with detailed research hypotheses in the following section.

**Figure 3.** A technology–organization–environment (TOE)- and technology acceptance model (TAM)-based model of project owners' building information modeling (BIM) adoption behavior.

## *3.1. BIM Behavioral Intention and BIM Adoption Behavior*

Behavioral intention is defined as "an indication of an individual's readiness to perform a given behavior". Therefore, behavioral intention is assumed to be an immediate antecedent of behavior [40]. Based on findings from case studies, Arayici et al. [41] found that the rapid promotion of BIM in the UK is benefitted precisely from the user's adoption decision, which leads to wide BIM adoption. Accordingly, we propose the following hypothesis:

**H1.** *Behavioral intention will have a positive influence on BIM adoption behavior.*

## *3.2. Attitude toward BIM and BIM Behavioral Intention*

Attitude refers to one's subjective positive or negative judgment of a technology. Previous studies have shown that attitude has a certain influence on behavioral intention. Through investigating American consumers' behavior of car buying, Etter [42] found that attitudes can significantly affect purchase intentions. In addition, in a Taiwanese study focusing on online shopping, Wu [43] also found that attitudes directly influence purchasing decisions. Hereby, we propose the following hypothesis:

**H2.** *Attitude toward BIM adoption will positively a*ff*ect the BIM behavioral intention.*

## *3.3. Perceived Usefulness and BIM Behavioral Intention*

Perceived usefulness refers to the degree to which a person believes that using a particular technology would enhance his/her job performance [27]. Previous studies have proved that perceived usefulness has a direct effect on users' behavioral intention to use a technology [44,45]. Studies on BIM adoption in South Korea indicate that the perceived usefulness of BIM will significantly affect the behavioral intention of all parties involved in the construction industry [21,46]. Based on the comparative study of the acceptance of BIM in South Korea and the United States, Lee et al. [47] found that perceived usefulness had a significant impact on BIM behavioral intention at both the individual level and organizational level. Accordingly, we propose the following hypothesis:

**H3.** *Perceived usefulness will positively a*ff*ect BIM behavioral intention.*

## *3.4. Perceived Usefulness, Perceived Ease of Use and Attitude toward BIM Adoption*

In line with the TAM, perceived usefulness and perceived ease of use are essential variables framing an individual's technology acceptance behavior. Studies on the adoption of information technology have shown that perceived usefulness and perceived ease of use can have a significant impact on users' attitudes toward BIM Adoption [27,48]. In addition, in the context of BIM adoption, perceived ease of use also shows a positive effect on perceived usefulness [24]. We, thus, propose the following hypotheses:

**H4.** *Perceived usefulness of BIM will positively a*ff*ect attitude toward BIM adoption.*

**H5.** *Perceived ease of use will positively a*ff*ect attitude toward BIM adoption.*

**H6.** *Perceived ease of use has a positive e*ff*ect on perceived usefulness of BIM.*

## *3.5. Social Influence, Perceived Usefulness and Perceived Ease of Use*

Social influence refers to the degree of an individual's perception that most people who are important to him think he should or should not perform the behavior in question [49]. The rationale for a direct effect of social influence on perceived usefulness and perceived ease of use is that people may choose to accept the same perspective, even if they are not so favorable toward the opinions. If they believe one or more important referents (such as superiors, peers, partners, etc.), they are sufficiently motivated to comply with the referents (especially superiors) based on affiliation and/or respect [50]. In an early study on bandwagon innovation diffusion, Rosenkopf and Abrahamson [51] pointed out that bandwagons have a positive feedback loop in which information generated by more adoptions creates a stronger bandwagon pressure, and a stronger bandwagon pressure prompts more adoptions. Therefore, a successful BIM application by competitors and partners will, to some extent, affect project owners' perception of the usefulness and usability of this innovative technology, which, in turn, affects BIM adoption as a whole [11,25]. Accordingly, this study proposes the following hypotheses:

**H7a.** *Social influence has a positive influence on project owners' perceived usefulness of BIM.*

**H7b.** *Social influence has a positive influence on project owners' perceived ease of use of BIM.*

## *3.6. Organizational Support, Perceived Usefulness and Perceived Ease of Use*

Organizational support refers to an individual's perception on the degree of policy, resources and other kinds of support provided by the organization for the use of technology. Herein, it refers to the support provided by the project organization to the project owner to adopt BIM. Organizational support carries great weight in motivating employees' potential, allocating resources and enhancing work performance [52–54]. It is easily understandable that sufficient organizational support will exert an incentive effect on employees and improve their sense of organizational backup. Gaining strong support from their organizations, employees will have a sense of being trusted which fulfills their expectation, making them more dedicated to their job, and more likely to demonstrate that they can achieve the organizational goals. However, if employees lose the necessary support (such as information, resources, equipment or training, etc.), their work procedures and work quality will be adversely affected, leading to employees' becoming upset and eventually frustrated [55,56].

In the information technology field, Lin et al. [57] found that organizational high-level support can improve employees' perceived usefulness and perceived ease of use of information technology. Recently, Song et al. [12] implied that, as a new project management technology, successful BIM adoption cannot be realized without the superincumbent financial and policy support in the early stage of software and hardware procurement and personnel training. Furthermore, there is some evidence showing that due to the fact that BIM is often not launched or advocated by the leader or

decision-makers in organizations, it often fails to allocate sufficient human, material and financial resources to support BIM adoption [58]. Based on the above, we propose two hypotheses:

**H8a.** *Organizational support to adopt BIM will positively impact project owners' perceived usefulness of BIM.*

**H8b.** *Organizational support to adopt BIM will positively impact project owners' perceived ease of use of BIM.*

## *3.7. BIM Technical Features, Perceived Usefulness and Perceived Ease of Use*

BIM technical features normally reflect the fitness, ease, compatibility and interoperability of BIM application. In line with a theory of innovation diffusion, Rogers [59] pointed out that the application of an innovation technology needs to be consistent with the existing value, demand and the experience of potential adopters. In particular, when it comes to introducing or adopting a new technology, firms will compare it with the existing technology, and consider the relevant advantages and characteristics of the two technologies in various aspects. Despite the huge potential value, if BIM is ineffective at interoperating or fitting current work procedures, it will not likely be accepted and adopted by project owners within a short period, as these project owners would be greatly concerned about the risk of abundant inputs (such as financial investments, human resources, etc.). Kim et al. [46] identified major obstacles to BIM adoption, including the actual software operation, the complexity of BIM workflows and the gap between the actual expectations of organizations. Evidence from previous studies clearly indicates that the lack of compatibility between different BIM software hinders the successful application of BIM in the construction industry [13,60]. For example, the compatibility of BIM will significantly affect users' perceived ease of use [21]. Accordingly, this study proposes the following hypotheses:

## **H9a.** *BIM technical features will have a positive e*ff*ect on project owners' perceived usefulness of BIM.*

**H9b.** *BIM technical features will have a positive e*ff*ect on project owners' perceived ease of use of BIM.*

## *3.8. Government BIM Policies, Perceived Usefulness and Perceived Ease of Use*

Government BIM policies generally refer to related policies issued by the government to promote BIM adoption. As shown in the Report of Business Value of BIM in China [58], the respondents who are project owners asserted that lacking first-hand experience deters them from joining BIM adoption and application practices. Even after the project owners adopted BIM, they usually need to emulate the existing projects which successfully applied BIM to guide their actual BIM applications and implementation. Thus, if the government could launch BIM pilot projects in batches and develop guidance for BIM application, this would significantly reduce the difficulty of BIM application, which would be likely to attract more project owners to adopt BIM in the first place. Some countries' experience has proven that appropriate financial subsidies would merit BIM adoption and application (such as in Singapore). For example, Succar [61] and Eadie et al. [62] indicated that government policies are among the primary factors influencing BIM adoption. Therefore, we propose the following hypotheses:

**H10a.** *Government BIM policies will have a positive e*ff*ect on project owners' perceived usefulness of BIM.*

**H10b.** *Government BIM policies will have a positive e*ff*ect on project owners' perceived ease of use of BIM.*

## **4. Research Method**

## *4.1. Measurements and Pilot Survey*

A structured questionnaire with two sections was designed and used for data collection. The first section covered demographic information of respondents including gender, age, education background, position, work experience and BIM experience. The second section included 24 measurement items (see Table 1 for details) which were designed to elicit project owners' assessments of BIM adoption

on a five-point Likert-type scale, with 1–5 indicating "strongly disagree", "disagree", "generally", "agree" and "strongly agree", respectively. All of these measurements were adopted from existing studies and reworded to render the items relevant to BIM adoption for project owners in China. Specifically, perceived usefulness (PU), perceived ease of use (PEOU), attitude (AT), and behavioral intention (BI) were developed based on the measures previously validated by Davis [27,28,46,63,64] and Xu et al. [18], and were reworded in accordance with the context of BIM adoption among Chinese project owners. Social influence (SI) was adopted from Kim et al. [46], Venkatesh and Davis [63], and contains the two dimensions of authoritative influences by intra-firm and inter-firm individuals and associations. Organizational support (OS) was operationalized to reflect the different impacts that the organization's resources exert on BIM adoption. Similar items had previously been validated by Xu et al. [18] and Cao et al. [25]. The construct of BIM technical features (TF) is derived from the research of Xu et al. [18], Kim et al. [46] and Song et al. [12] on the basis of BIM adoption practice in mainland China, with three items indicating the degree of interoperability and compatibility and fitness of BIM in project owners' daily tasks. Referring to the study of Song et al. [12], we replenished and extended the connotations of the vital construct, "Government BIM Policies"(GP), with distinctive Chinese characteristics. Therewith, three BIM experts who have eminent experience in BIM research and application were invited to participate in the pilot survey. Shortly after the pilot survey was conducted, the experts were asked to provide advice regarding the refinement of items and their personal understanding of BIM adoption. According to the experts' advice, items with ambiguity were refined, and the items with tautology were eliminated. After that, we sent the modified measurement items to these experts again and asked them to review whether the amendments strictly complied with their intentions to ensure the applicability of measurement items. The ultimate measurement items of these constructs are provided in Table 1.


## *4.2. Sampling and Data Collection*

In this study, we targeted project owners in mainland China involved in BIM adoption as qualified respondents for data collection. A total of 300 questionnaires were distributed by means of face-to-face interviews (number: 200) and an online survey platform (number: 100). In the face-to-face part, before the formal survey, we conducted an interview with each interviewee to ensure that he/she had first-hand experience of BIM practice. As for the online channel, targeted delivery was the only step taken to send the questionnaire to the preselected project owners involved in BIM adoption. In addition, to obtain sufficient samples, a snowball sampling method was utilized to increase the sample size as we invited the surveyed respondents to share more information regarding knowledgeable participants in other BIM projects or organizations. Any questionnaire with incomplete information or missing values was excluded. Finally, 188 valid questionnaires were received (156 (83%) from the face-to-face interviews and 32 (17%) from the online platform). The valid response rate of face-to-face interviews was 78%, and that of the online survey was 32%. Among these respondents, 64.9% were male, and the remaining respondents were female. All of them were practitioners undertaking tasks directly related to BIM practice in the client departments. The demographics of the respondents under investigation are presented in Table 2.


**Table 2.** Demographics of the respondents (N = 188).

## **5. Data Analyses and Results**

In this study, confirmatory and discriminant factor analyses of the measurement model were first conducted in order to assess the reliability and validity of the proposed constructs. Afterwards, the maximum likelihood estimate (MLE) method of the structural equation model (SEM) was employed to validate the hypotheses and the fitness of the proposed model.

## *5.1. Measurement Validation*

In general, reliability and validity were the two most common indicators used to evaluate the measurement model. The reliability of the measurement for each construct can be assessed on the basis of Cronbach's α coefficient. Previous studies suggested that a Cronbach's α greater than 0.7 indicates acceptable reliability [66,67]. All of the Cronbach's α coefficient values in the present study are more than the threshold of 0.7 (see Table 3 for detailed values), which indicates good reliability.

With regard to validity, convergent and discriminant validity should both be taken into account. On one hand, convergent validity is usually assessed by three indices: composite reliability (CR), average variance extracted (AVE), and standardized factor loadings. For the composite reliability (CR), values of 0.7 or higher are recommended, according to Nunnally et al. [66] and Nunnally and Bemstein [68]. The CR values in this study range from 0.811 to 0.933, which satisfy the recommended value of 0.7 (see Table 3). In addition, as one of the indices to access the convergent validity, the AVE is often used by examining the construct relative to the amount of variance attributed to the measurement error [69]. With regard to Segars [70], the AVE value for each construct which exceeds the threshold of 0.5 is acceptable. In our study, the AVE values all meet the acceptable requirement (which all range from 0.633 to 0.846). Moreover, values of all standardized factor loadings in this study are above the threshold of 0.7. Therefore, all of the indices are satisfied at acceptable levels, demonstrating the convergent validity of the measure model.


**Table 3.** Convergent validity of the measurement model.

Note: \*\*\* *p* < 0.001.

On the other hand, discriminant validity is mainly used to demonstrate the non-correlation between one given construct and the others which ought not be correlated with the given one [69]. Normally, the discriminant validity of one item is judged based on whether it can be easily determined as good or bad by comparing the square root of the AVE for the given construct with the correlations between that construct and all others. If the square roots of the AVE of one given construct are greater than all correlation coefficients of other constructs, it implies that the given construct is more likely to be strongly correlated with its own indicators than the other constructs in the model. In this study, the square roots of all the average variances extracted (the diagonal elements) are greater than the values of the off-diagonal correlation coefficients in the corresponding columns in Table 4, which confirms good discriminant validity as a whole.


**Table 4.** Correlation matrix and the square of average variance extracted.

Note: The diagonal numbers underlined represent the square of average variance extracted.

## *5.2. Hypotheses Testing*

With the aid of AMOS 21.0, the maximum likelihood estimate (MLE) method in the structural equation model (SEM) was employed to validate the hypotheses and the fitness between the proposed model and the collected data.

The fitness of the proposed model is revealed by the indices of the ratio of the Chi-square model and degrees of freedom (χ2/df), goodness-of-fit (GIF), root mean square error approximation (RMSEA), normed fit index (NFI), comparative fit index (CFI), incremental fit index (IFI), and Tacker–Lewis index (TLI). The recommend criteria of a goodness-of-fit and the values of these indices derived from this study are shown in Table 5. Despite the GFI and NFI being slightly lower than the recommended acceptable value of 0.90, they are close enough to suggest that the model fits the data reasonably well.

**Table 5.** Evaluation of overall fitness of the conceptual model.


Then, a path analysis is carried out to test the hypotheses. As the results in Table 6 show, nine of fourteen hypotheses are supported. Similar to findings in some previous studies [22,40,63,71], behavioral intention (BI) has a significant positive impact on behavior (β = 0.698, t = 10.581, *p* < 0.001), supporting H1. Furthermore, attitude has a significant positive impact on behavioral intention (β = 0.886, t = 5.854, *p* < 0.001), which means that H2 is supported. However, an unexpected outcome is that PU has no significant impact on BI; thus, H3 is not supported. Furthermore, both PU and PEOU have significantly positive impacts on attitude (β = 0.476, t = 6.060, *p* < 0.001; β = 0.404, t = 4.391, *p* < 0.001), and therefore H4 and H5 are supported. Meanwhile, H6 is also supported given that PEOU has a significantly positive influence on PU (β = 0.282, t = 3.388, *p* < 0.001). Social influence (SI), on the one hand, is found to have a significantly positive impact on PEOU (β = 0.134, t = 2.901, *p* < 0.01), supporting H7b, but on the other hand, SI has no significant influence on PU (β = 0.134, t = 1.735). OS has no significant impact on either PU or PEOU (β = −0.123, t = −1.779; β = 0.134, t = 1.765). In addition, the results show that TF has significant impacts on both PU and PEOU (β = 0.489, t = 4.586, *p* < 0.001; β = 0.286, t = 2.565, *p* < 0.01). Therefore, both H9a and H9b are supported by the empirical results. GP has a significant impact on PU (β = 0.291, t = 4.309, *p* < 0.001), while it has no significant impact on PEOU (β = −0.012, t = −0.155). Thus, H10a is supported but H10b is not.


**Table 6.** Results of the tested hypotheses.

Note: \*\*\* *p* < 0.001; \*\* *p* < 0.01; \* *p* < 0.05.

#### **6. Discussion and Implications**

#### *6.1. Discussion*

The above results reveal that most of the proposed research hypotheses are well supported. As revealed by the results, project owners' behavior intention has a significantly positive impact on their BIM adoption behaviors, and their attitude will also positively influence their behavioral intention toward BIM adoption, which is consistent with the findings of Davis et al. [27], Ajzen [40], Yuan et al. [72] and Liu et al. [73]. However, we also found that the impact of perceived usefulness (PU) on behavioral intention is insignificant, which is contrary to the prediction of the classical TAM. Differing from previous TAM-based studies confirming that perceived usefulness (PU) has a significantly positive influence on attitude (AT), this study shows that attitude mediates between perceived usefulness and perceived ease of use. In BIM adoption cases for project owners, the impact of perceived usefulness (PU) on behavioral intention (BI) can be formed only across the "bridge" of attitude (AT). In addition, attitude is composed of three elements: inner feelings, emotions, and intentions, and these three elements are intersected with each other [74]. Therefore, perceived usefulness alone, without strong subjective inner feelings or desires, cannot transfer this perceived usefulness into a powerful driven force influencing behavioral intention, which echoes the dilemma that project owners are unwilling to step into actual BIM adoption although they have perceived the usefulness and great potential of BIM [58,75]. Both perceived usefulness (PU) and perceived ease of use (PEOU) positively influence the attitude. Specifically, the stronger the project owners' perceived usefulness and perceived ease-of-use of BIM, the more positive their attitude toward BIM adoption will be.

Additionally, social influence has no significant effect on perceived usefulness, while its impact on the perceived ease of use is significant. The explanation lies in that the impact of social influence on perceived usefulness will diminish as users' personal experience and cognition deepen over time [63]. Therefore, before the actual adoption, project owners' knowledge and beliefs about BIM are "vague and bandwagen", and they must therefore rely more on the opinions of others (such as top management and partners) as a basis for their intentions [76]. After implementation, when more information and details about BIM's strengths and weaknesses become clearer through direct experience, the social influence weakens [77]. On the other hand, social influence has a significant impact on perceived ease of use, indicating that when users' perceived ease of use of information technology is consistent with

the external world, unlike perceived usefulness, their belief that the information technology (i.e., BIM) is easy to use will be further strengthened.

This study also presents an unexpected but interesting finding that organizational support has no significant influence on either perceived usefulness or perceived ease of use, which is in contrast to many previous research findings claiming that organizational support is a critical successful factor for BIM adoption and implementation [11,13,18]. These results could largely be explained by the differences in the demand of resources input and the actual effort exerted by top management to promote BIM adoption and implementation. Furthermore, the statements of the majority of survey respondents provide easily understandable reasons for why, in their organizations, management's support for BIM is inadequate to support comprehensive BIM application, which leads to a lack of necessity and incentive to use BIM. This parallels previous findings that employees' perception of organizational support influences whether employees will improve their organizational commitment and support organizational goals [55,78]. In turn, this finding indicates that organizational support will directly influence staff's perception and feelings regarding whether BIM is useful and easy to use, which consequently affects their ultimate adoption behavior.

Furthermore, it can be seen from Table 4 that BIM technical features (such as interoperability, compatibility, etc.) are the most important factors for determining project owners' decision on whether to adopt BIM. The result is consistent with the findings of previous studies [13], which state that the fitness and interoperability of BIM to the current tasks are critical factors influencing the owners' perceptions of the usefulness and ease of use of BIM and will ultimately constitute a BIM adoption behavior. Therefore, the matching degree of BIM's own task technology and how to improve users' perceived usefulness must be considered.

Besides, government BIM policies have a significantly positive effect on perceived usefulness. This is consistent with previous studies which revealed that the existence of government-led initiatives to promote BIM implementation within the industry is one of the critical success factors for extensive BIM adoption and diffusion [62,79]. On the one hand, government policies, such as subsidies, can directly reduce the BIM application costs, which in turn improve project benefits and attract more project owners to get involved in BIM adoption. On the other hand, a universal BIM standard supported by government will reduce the difficulty to interoperate among different special platforms [11,18,58]. Meanwhile, almost all project owners in our survey believe that government support for policies is very helpful for their BIM adoption.

## *6.2. Theoretical Implications*

This study enriches the theoretical literature in three main areas.

Firstly, based on the theory of the TAM, this study categorizes different dimensions of factors and elements—which impact individuals' perceptions of the usefulness and ease of use of BIM—affecting project owners' BIM adoption behavior in the Chinese AEC sector, with an ultimate aim to explain why project owners might adopt BIM. The factors capable of explaining project owners' BIM adoption behaviors are tested and validated by using a structural equation model, and the intrinsic motivation and action mechanism of project owners' BIM adoption behavior are revealed. The findings provide a deeper understanding for explaining the causal factors leading to BIM adoption behaviors.

Secondly, this study's findings also provide valuable insights into the TOE framework and the theory of the TAM in a specific context. The results show that organizational support has no significant impact on either perceived usefulness or perceived ease of use, which can be explained by Eisenberger's findings "perceived organizational support is the premise of behavior, organizational support without perception cannot work even if the support is already provided [52]", indicating that there seems to be a certain precondition that organizational support could positively affect the behavior. Beyond this, another attracting result is that attitude mediates between perceived usefulness and perceived ease of use, which contracts with the findings of Venkatesh et al. [64], but agrees with the findings of Howard et al. [22].

Thirdly, this study extends the theory of the TAM by integrating the TOE framework, revealing that technical, organizational and environmental variables are significantly related to behavioral intention. These variables are intermediated by two distinct constructs (PU and PEOU) and attitude (AT) in a BIM adoption context. Furthermore, results also demonstrate that most of the proposed hypotheses are well supported and the causal relationships among the postulated constructs in the model are analyzed. As such, the model in our study provides an elaborated explanation of the key factors forming the behavioral intentions of project owners toward BIM adoption. In other words, the model offers important insights into the reasons behind project owners' willingness to adopt BIM. By investigating BIM adoption from project owners' perspectives, this study also responds to and reinforces the concern of Ling et al. [19], i.e., focusing on the adoption behavior of other project participants would add more dimensions and shed more light on construction innovation.

## *6.3. Practical Implications*

BIM is often recognized as a promising platform for project stakeholders (including the project owner) to capture complete information throughout the project lifecycle, and to utilize the available data for sustainable design, sustainability rating analysis and sustainable facilities management. The findings of this study will help project owners to understand the impact and interaction of the external constraints and their own subjective perceptions of BIM adoption, based on which successful BIM adoptions and construction sustainability will be increased by some effective incentives and strategies.

Responding to many previous studies on technology acceptance, attitude, perceived usefulness and perceived ease of use are key determinants of behavioral intention [28,40,43,48], which will lead to the ultimate BIM adoption behavior. Thus, project owners should break the traditional mindset, production-organization mode and work procedures to form a positive attitude to embrace a brand-new or even subversive paradigm based on BIM, leading to long-term sustainable growth not only for the organizations but also for the construction industry.

Among these proposed external antecedents, the technical feature is found to be of the utmost importance to perceived usefulness and perceived ease of use. This finding provides insights revealing that project owners must pay attention to the technical features (such as interoperability and compatibility) of the introduced BIM platform or tools, which would greatly enhance the likelihood of successful BIM adoption and sustainability rating analysis from the point of view of technical feasibility.

Furthermore, social influence has a significant impact on the perceived ease of use of project owners, affecting project owners' BIM adoption through attitudes and behavioral intentions. It is suggested that intensifying the dissemination of BIM's benefits and peer experience exchanges could enhance project owners' acknowledgement of BIM's benefits, thus effectively helping project owners' BIM adoption. Furthermore, as mentioned above, the effect of social influence diminishes as experience is gained. As such, project owners should attach importance to establishing a good corporate environment to embrace BIM adoption.

Lastly, government policies also have a positive impact on the perceived usefulness of BIM to project owners, which in turn indirectly affects their BIM adoption. This finding indicates that external incentives from government will help project owners' BIM adoption. In this regard, launching BIM pilot programs and tax exemption could be effective ways to create a favorable environment for promoting project owners' BIM adoption activities.

## **7. Conclusions**

Based on the theory of the TAM, this study attempts to explain project owners' BIM adoption behaviors by investigating how different dimensions of factors and elements—which impact individuals' perceptions of usefulness and ease of use of BIM—influence project owners' BIM adoption behavior in the Chinese construction sector. The factors affecting project owners' BIM adoption are tested and validated by using a structural equation model, and the intrinsic motivation and action mechanism of project owners' BIM adoption behavior are revealed.

The results indicated that most of the proposed hypotheses are well supported and the causal relationships among the postulated constructs in the model are analyzed. The model in our study provides an elaborated explanation of the key factors influencing the behavioral intentions of project owners toward BIM adoption. Particularly, the results reveal that BIM technical features and government BIM policies have positive effects on perceived usefulness, but social influence and organizational support do not significantly influence perceived usefulness. Furthermore, both social influence and BIM technical features have positive effects on perceived ease of use, while organizational support and government BIM policies do not significantly influence perceived ease of use. Attitude plays a significant intermediary role among perceived usefulness, perceived ease of use and behavior intention. Additionally, attitude significantly affects behavior intention, and behavior intention can also affect BIM adoption behavior. The findings of this study are expected to provide a better understanding of the essential elements of project owners' BIM adoption behaviors and guide industry practitioners in developing proper strategies to achieve more effective BIM implementation.

There are also limitations. Although this study deepens the understanding of project owners' BIM adoption intentions and behavior, a wider range of variables can be considered to enhance the model's robustness to more accurately predict project owners' BIM adoption behaviors. Also, despite some previous studies indicated that project features (such as project size, nature, delivery types, etc.) need to be taken into account when it comes to BIM adopting strategies, the limitation in sample data blocks us to conduct further examinations. Hence, research focusing on the influence of project features (such as project size, nature, delivery types, etc.) on BIM adoption behaviors should be further developed. Besides, extensive studies should be conducted to examine the generality of the proposed model in different countries' practice and background to expand the situations to which it applies.

**Author Contributions:** Conceptualization, H.Y. and X.X.; methodology, H.Y. and Y.Y.; software, Y.Y.; validation, Y.Y.; formal analysis, Y.Y.; investigation, Y.Y. and H.Y.; resources, H.Y. and X.X.; data collection, Y.Y.; writing—original draft preparation, Y.Y.; writing—review and editing, H.Y.; funding acquisition, H.Y. and X.X.

**Funding:** This research was funded by the Major Program of the National Social Science Fund of China (Grant number: 18ZDA043), and the National Natural Science Foundation of China (Grant number: 71573216; 71671053).

**Conflicts of Interest:** There is no conflict of interest.

## **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Sustainability* Editorial Office E-mail: sustainability@mdpi.com www.mdpi.com/journal/sustainability

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com

ISBN 978-3-0365-2629-4