Cause Analysis of Hindering On-Site Lean Construction for Prefabricated Buildings and Corresponding Organizational Capability Evaluation

School of Mechanics and Civil Engineering, China University of Mining and Technology, Xuzhou 221116, China Research Center for Digitalized Construction and Knowledge Engineering, China University of Mining and Technology, Xuzhou 221116, China Nantong Prefabricated Buildings and Intelligent Structures Institute, Nantong 226000, China School of Management Science and Real Estate, Chongqing University, Chongqing 400044, China


Introduction
e sudden outbreak and rapid spread of the new coronavirus "COVID-19" have caused huge losses in lives and economies around the world. Since prefabricated buildings have the advantage of rapid construction, they earn lots of time for humans in the global fight against the new coronavirus "COVID-19," such as Leishenshan Hospital and Huoshenshan Hospital. is emergency-driven mode makes prefabricated buildings win more global attention than other types of buildings. However, the market-driven prefabricated buildings are encountering various challenges, which hinder the sustainable development of prefabricated buildings [1]. In the context of saturated construction market, more and more construction companies are entering the era of low profits. Lots of prefabricated construction companies expect to change the current situation of low profits and enhance their overall competitiveness via lean construction (LC). Solaimani and Sedighi [2] believed that LC was an effective way to increase project benefits and explored "how Lean helps achieve and maintain sustainability in construction sector." Francis and omas [3] found that LC and sustainable construction held certain common objectives in improving resource efficiency and minimizing waste. Hence, if prefabricated buildings want to achieve the sustainable development, they need to implement LC.
Many countries and regions are encountering different challenges during the LC implementation, such as United States, United Kingdom, Norway, Brazil, and Chile [4]. Innella et al. thought that the latest challenge of LC is its implementation in prefabricated buildings [5]. A report from the "Annual Meeting of the Architectural Society of China-Construction Management Research Section (2019)" shows that the on-site LC for prefabricated buildings has developed rapidly in Japan, which discovers lean production from the automotive industry. Scholars, such as Innella et al. [5], Heravi and Firoozi [6], Goh and Goh [7], Li et al. [8], Yu et al. [9], Nascimento et al. [10], Bamana et al. [11], and Shewchuk and Guo [12], pay more attention to the on-site LC methods and techniques for prefabricated buildings. However, few scholars explore the causes hindering the on-site LC for prefabricated buildings and the corresponding organizational capability evaluation. Related research regarding other types of buildings may provide value references for the implementation of onsite LC of prefabricated buildings, but prefabricated buildings have their particularities, especially their special construction processes. e problem is that prefabricated buildings have fallen into an eddy in terms of on-site LC: one aspect is that people believe in the good performance of on-site LC [13], but the other aspect is that actual prefabricated projects are difficult to successfully implement on-site LC [14]. Hence, there are some unknown causes hindering the implementation of on-site LC in prefabricated buildings. Alternatively, known causes have not been effectively valued and resolved by scholars and construction organizations. ese causes are hereinafter referred to as barriers.
is study aims to identify and analyse the critical barriers to on-site LC for prefabricated buildings and further evaluate the construction organization's capability to solve these critical barriers.
is study will have the following contributions. (1) It assists managers make more reasonable and feasible countermeasures according to the identified critical barriers and corresponding weights so as to improve the performance of on-site construction for prefabricated buildings. (2) It makes managers more aware of the organizational capability to deal with these critical barriers so as to identify and improve the vulnerable spot. (3) It expands the theoretical boundary of LC from the perspective of the specific prefabricated construction industry rather than the general construction industry. Other sections of this study are structured as follows. In Section 2, the existing relevant literature is reviewed and analysed. In Section 3, a research methodology framework is proposed. In Section 4, a critical barrier system is established, and then the exploratory factor analysis is implemented to calculate the weight of each barrier and extract a few common components. In Section 5, a typical large-scale project is selected to evaluate its construction organizational capability to deal with the barriers, and then the obtained results are analysed and discussed in depth. In Section 6, the conclusions and future work are summarized.

Barriers to On-Site Lean Construction for Prefabricated
Buildings. Lean construction (LC) aims to strive for perfection during project construction, including zero waste, zero accidents, lower pollution, higher quality, and many more. Its core is to eliminate all unnecessary waste rather than accept unsatisfactory events reluctantly. e waste in LC is considered as all nonvalue-added activities, which generate direct or indirect costs but do not add any value to the product [15]. So far, many lean methods (or techniques) have been considered suitable for the construction industry, including just-in-time (JIT), last planner system (LPS), prefabrication, value stream mapping (VSM), total quality management (TQM), concurrent engineering, and 5S and 6S on-site management [16]. ey are believed to help improve quality, increase efficiency, reduce waste, enhance the construction management practices, and promote sustainable development [17,18]. From traditional construction to prefabricated construction is a transformation of construction mode, which needs the reform of supporting management ideas and methods. LC and prefabricated buildings have similarities and collaborative relationships in terms of origin, construction processes, and other aspects, which are compatible with each other. With the adoption of lean concepts and tools in traditional construction, many studies are also exploring their application in prefabricated construction.
ese studies cover different topics: on-site scheduling [19], lean learning [20], quality improving [21], resources [22], environment [23], and many more. All these studies have shown that LC is suitable for being integrated into the on-site construction of prefabricated buildings. However, the combination of LC and prefabricated on-site construction is still in its infancy, and there are still lots of barriers during its promotion and application. So far, only a few studies discuss the barriers to implementing the on-site LC for prefabricated buildings. In the research of Oral et al., the main barriers to implementing JIT in the Turkish prefabrication sector are the financial difficulties and the demand uncertainties while those in other developing countries are poor demand conditions, poor supply conditions, and unstable economic environments [24]. eir research does not focus on the on-site construction for prefabricated buildings, but is oriented to the Turkish entire prefabrication sector. In addition, their research also shows that it is necessary to explore the LC for prefabricated buildings in developing countries, in which the barriers encountered by these countries are different each other. In the research of Said, the reason for low lean utilization in the United States is that most contractors are still trying and learning the best practices of prefabrication [25]. In the research of Dave et al. and Xu et al., the low adoption of information technology will affect the leanness of prefabricated construction [26,27]. China is the largest developing country, where prefabricated buildings are in its infant stage. Obviously, given the high consistency and compatibility between LC and prefabricated buildings, it is essential to investigate the feasibility of integrating LC with prefabricated buildings in the context of China, which offers an opportunity for examining effectiveness and barriers of LC in the prefabricated domain.

Barriers to Lean Construction Being Not Specifically for Prefabricated
Buildings. Prefabricated buildings have many different characteristics in on-site construction compared with other types of buildings. On the one hand, the construction methods similar to assembling building blocks can shorten construction time, improve environmental performance, ease labour requirement, and reduce workload and have other advantages beyond traditional construction methods [28,29]. On the other hand, compared with the traditional cast-in-place processes, prefabricated construction projects request higher requirements for mechanical equipment (e.g., cranes) [30], work environment [31], higher standardization [32], and information technology [33]. However, prefabricated buildings are also a member of the construction industry and have common attributes. Hence, studies on LC implementation barrier being not specifically for prefabricated buildings may provide valuable reference for this paper, as shown in Table 1.
From the above studies, due to the cultural and economic differences among countries and the special construction process of prefabricated buildings, it is of necessity to make an in-depth analysis of integrating LC with prefabricated buildings in the context of China.

Organizational Capability Evaluation of Lean Construction for Prefabricated Buildings.
Organizational capability is generally positively correlated with the effect of barrier solving. Organizational capability includes employee governance, employee mentality, and employee abilities. An organizational capability evaluation is the premise and basis of overcoming the barriers to implementing on-site LC for prefabricated buildings. Hence, after these barriers have been identified and analysed, it is necessary to assess whether a construction organization has sufficient capabilities to deal with these barriers. However, few directly related studies are found via Web of Science, Engineering Index, and China National Knowledge Infrastructure. Indirectly related studies may provide valuable references. For organizational capabilities in the AEC (Architecture, Engineering, and Construction) industry, previous studies have explored a wide range of topics, including organizational resilience [52], occupational safety and health abilities [53], project management competencies [54,55], resource allocation capabilities [56], risk management capabilities [57,58], and contract management capabilities [59]. ese research results show that when organizations are in different contexts, the attributes constituting their organizational capabilities are often different and require corresponding evaluation indicators. As the key to ensuring the required performance of construction projects, on-site construction management has higher requirements for organizational capabilities [60]. However, these previous studies on organizational capabilities lacked attention to on-site construction. With the transformation and upgrading of prefabrication technology, prefabricated buildings need more on-site LC management, which puts forward higher requirements for organizational capabilities. In view of the research progress of AEC's organizational capabilities, there is an urgent need for methods that can accurately evaluate organizational capabilities to deal with barriers to LC implementation in prefabricated construction sites.

Framework Design of Research Methodology.
Prefabricated buildings have a unique construction system in the construction industry and are encountering lots of unknown critical barriers in implementing on-site LC. In addition, China also owns its special economic, cultural, and political environment. In terms of applicable conditions, exploratory factor analysis (EFA) is not limited by priori information while confirmatory factor analysis (CFA) requires sufficient priori information. Hence, EFA instead of CFA is selected by this research. EFA is a technique for dimension reduction and classification, which can synthesize many original observable variables with dependent relationships into a few common latent variables with independent relationships [61,62]. Due to the complexity of on-site construction for prefabricated buildings, a single method cannot fully achieve the identification and analysis of critical barriers, as well as the evaluation of organizational capabilities at the same time. Multimethod collaboration is regarded as an effective way. Hence, a research methodology based on multimethod collaboration is designed, as shown in Figure 1. It consists of three modules, namely, the exploratory factor analysis model regarding critical barriers Advances in Civil Engineering and the exploratory factor evaluation model regarding critical barriers, as well as the findings and suggestions of this study. Figure 1, the multimethod collaboration involves data collection and analysis methods. Data collection methods include literature analysis, field survey, semistructured interview, online consultation, questionnaire survey, and construction company survey. Data analysis methods contain pre-exploratory factor analysis, descriptive statistics analysis, formal exploratory factor analysis, and fuzzy comprehensive evaluation. ere is close interactions among them. In the exploratory factor analysis model regarding critical barriers, the combination of literature analysis and    [34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51].

Principle Description of Research Methodology. In
Common barriers Countries/regions previous (or subsequent) field survey is used to preliminarily identify the critical barriers to on-site LC. e previous field survey is the previous project survey experiences rather than the survey conducted just for the critical barriers. e subsequent field survey is that researchers take some initial critical barriers identified from the literature analysis to investigate the project specifically. e combination of the semistructured interview and online consultation is used to further improve the critical barriers. Related documents are sent to the selected experts in advance for better preparation during the interview and consultation. e questionnaire survey with large sample is adopted to gain the scores for all critical barriers. e factor analysis method, including preexploratory factor analysis and formal exploratory factor analysis, is used for preprocessing data and extracting several common components (namely, common latent variables) via dimension reduction and classification. e descriptive statistics analysis is adopted to calculate the weight of each critical barrier. Factor analysis method corresponds to a classic mathematical expression [63], as shown in Formula (1). x i represents an original observable variable; f j represents a common latent variable; w ij represents the loading of x i on f j ; ε i represents a special factor; m and n represent the number of original observable variables and common latent variables, respectively; and m is greater than n. If the data regarding original observable variables are collected reasonably or are standardized, the value of ε i is zero: x � wf + ε, In the exploratory analysis evaluation model regarding organizational capabilities, the construction company survey is adopted to collect the detailed data of a prefabricated project. e questionnaire survey with small sample is used to gain the scores for extracted common components. e sample size should not be less than five. Fuzzy comprehensive evaluation (FCM) [64], an evaluation method based on fuzzy mathematics, include the establishment of evaluation index set, comment set, evaluation matrix, index weight vector, and comprehensive evaluation model. Since FCM can better solve some vague and difficult-to-quantify problems, it is selected to measure the capability level of a construction organization in dealing with these common components, as shown in Formulas (2)(3)(4). e evaluation index set consists of all common components, including C 1 , C 2 ,. . ., C i ,. . ., C m . e comment set is quantified by Likert scale, such as "1 � powerless," "2 � weaker capability," "3 � medium capability," "4 � higher capability," and "5 � fully capable." e S represents the evaluation matrix.
e W represents the index weight vector. e E represents the comprehensive evaluation model. e w i represents the weight of C i . Suppose N j represents any element in the evaluation index set. e s ij is the membership, representing the degree to which the evaluation index C i belongs to the comment N j . e result of fuzzy comprehensive evaluation is the comment corresponding to the largest element in the E:

Identification of Critical Barriers.
On the basis of extensive literature analysis and previous field survey experiences (e.g., Shanghai, Shenzhen, Harbin, Nantong, and Taizhou), some critical barriers are initially identified. e critical barriers identified in previous field surveys have not exceeded those identified in existing literature, which may be attributed to the fact that the authors do not have sufficient project experiences such as domain experts. erefore, it is necessary to invite experienced domain experts to further improve the identified and unidentified critical barriers.
Since prefabricated buildings are still in their infancy, the number of experts with more experience is limited. erefore, the selection criteria of experts in this study tend to be quality rather than quantity. Four experienced experts are finally selected. e initially identified critical barriers are sent to each expert in advance, which allows the experts enough time to think about the raised questions and decide when it is more appropriate to accept our semistructured interview. ree of them accept online interviews for 30∼60 minutes, while one only accepts brief consultation on account of his company's regulations. e basic information related to these experts is shown in Table 2.
e semistructured interview mainly involves three aspects. (1) e nonlean phenomena or behaviours of on-site construction for prefabricated buildings. (2) e critical barriers to on-site LC for prefabricated buildings. (3) Follow-up feedback and auxiliary data collection. e latter expert is allowed to see the opinions of the previous expert during the interview and consultation. In addition to improve the barrier system of on-site LC, a preliminary scoring using Likert scale is performed simultaneously for a pilot test. e comprehensive treatment results of all experts are shown in Table 3.

Characteristics for Questionnaires.
A questionnaire is designed to collect more professionals' views on the importance of critical barriers to on-site LC. In the questionnaire, these professionals are set as the respondents who have the experience or knowledge of on-site construction for prefabricated buildings. e impact level is evaluated by the Advances in Civil Engineering 5-point Likert Scale: 1 for "negligible," 2 for "lower," 3 for "medium," 4 for "higher," and 5 for "very high." All questionnaires are distributed via the Internet rather than on site. A total of 185 questionnaires are returned from March 6, 2020, to April 27, 2020. In order to further ensure the quality of data, this research excludes some questionnaires from the respondents who have no experience or knowledge of on-site construction for prefabricated buildings. Finally, the number of valid questionnaires is 154, which is more than the 150 needed for exploratory factor analysis [66]. e sample size of questionnaires used in this study is not large enough although it meets the basic requirements of exploratory factor analysis. is is mainly due to the fact that prefabricated buildings account for a relatively small proportion in the construction market. Figure 2 shows the distribution information of respondents' work units, while Figure 3 indicates the distribution information of the years that these respondents study or engage in on-site construction for prefabricated buildings. For a prefabricated project, owners, supervision units, design units, construction units, and precast-component factories stay at the construction site for a long time, and their proportion is 77.273%. e parameter x represents the number of experience years. e respondents with more than two years of experience account for 61.688%. Hence, the statistics show that the majority of respondents have sufficient qualifications, which ensure the quality of the data collected.

Reliability and Validity Test for Questionnaires.
Reliability refers to the stability of the collected data, which means the results of multiple measurements for the same object are similar. Reliability is divided into internal reliability and external reliability. Internal reliability is used to determine whether multiple items in the questionnaire measure the same concept or content. Cronbach's alpha coefficient (see Formula (5)), a common approach, is adopted to verify the internal reliability. α represents the Cronbach's alpha coefficient, and its value greater than 0.7 is acceptable [67]. k is the number of items on a scale. x i is the i th item, and x is the set of all items. σ 2 x is the variance of the total sample, which represents the variance among the total score of each item scored by each respondent on a scale. σ 2 x i is the variance of the observation sample, which represents the variance among each respondent's score on a certain item: SPSS (Statistical Product and Service Solutions), a professional statistical software, is used to calculate the reliability of questionnaires. Table 4 shows the results of reliability analysis of questionnaires using SPSS. Cronbach's alpha is 0.946. erefore, these collected questionnaires are very good in terms of reliability.
Validity refers to the degree to which the measured result reflects what people want to examine. e more the measurement result matches what people want to examine, the higher the validity is. Validity is further divided into content validity, structure validity, and standard validity. e Kaiser-Meyer-Olkin (KMO) and Bartlett's ball test, an integrated method for measuring structural validity, is adopted to verify the validity of questionnaires. e KMO test is to examine the correlation and partial correlation among variables, as shown in Formula (6). A(x i , x j ) represents the sum of squares of correlation coefficients between all variables. B(x i , x j ) represents the sum of squares of partial correlation coefficients between all variables. KMO owns the following levels: excellent (KMO >0.9), good (KMO >0.8), acceptable (KMO >0.7), questionable (KMO >0.6), and unacceptable (KMO <0.5) [68]. Bartlett's ball test is used to examine whether each variable is independent. If the significance (Sig.) in Bartlett's ball test is less than 0.05, then there is a correlation among the variables. is means that common latent variables can be extracted from the original variables: e SPSS is adopted again for calculating the validity of questionnaires. Table 5 indicates the results of validity analysis of questionnaires using SPSS. e result shows that KMO is 0.918 with the significance (Sig.) 0.000. is means that there is a strong correlation among the variables. Hence, these collected data are appropriate for exploratory factor analysis.

Result Analysis of Screening and Ranking for Critical
Barriers. Data preprocessing aims to remove some unimportant barriers from the thirty-one original barriers. It will be achieved through pre-exploration factor analysis. In preexploration factor analysis, the principal component analysis (PCA) method is used for component extraction. Since the number of all components is equal to that of all original observable variables (namely, all original barriers), the initial commonality of each variable is one. A variable with  extraction communality below 0.5 should be ignored [69]. Hence, the variable B12 with extraction communality 0.451 should be deleted. After data preprocessing, the number of original observable variables is reduced from 31 to 30. Table 6 shows the weights and rankings of thirty original observable variables. According to Tables 3 and 6, the following analysis and discussion are carried out on the barriers to implementing on-site LC for prefabricated buildings. irty critical barriers to on-site LC are identified and analysed from the perspective of the specific prefabricated buildings rather than all types of buildings, and the top five barriers are "inadequate professional management capability of managers" (weight � 3.657%), "insufficient standardization" (3.628%), "absence of lean awareness/thinking" (3.611%), "inadequate professional skills of workers" (3.589%), and "lack of appropriate lean technology/tools" (3.543%). Hence, the professional management capability of managers is the primary factor affecting the on-site LC.
is finding is consistent with the culture of the construction industry in China, where managers are the key to the success of a task. Existing managers are mostly transferred from nonfabricated buildings (e.g., cast-in-place concrete buildings) to prefabricated buildings and lack the professional management capability to match the on-site LC of prefabricated buildings. According to the previous field surveys and semistructured interviews, internal training or external recruitment is suggested to solve the inadequate professional management capability of managers. Compared with nonfabricated buildings, the standardization is a prominent feature of prefabricated buildings, such as process standardization, equipment standardization, and operation standardization. However, standardization is difficult to achieve in real life so that prefabricated buildings do not show its lean advantages in terms of on-site construction. e suggestion from consulted experts is that the standardization of on-site construction should be based on a sufficient understanding of construction process. e absence of lean awareness/thinking is roughly consistent with the previous study which thinks the lack of a long-term philosophy is a very important factor [35]. Hence, how to make employees have the awareness/thinking of LC is the focus of the future. In addition, the training of workers' professional skills and the adoption of lean technology should be taken seriously. Lean construction education/ training is considered as an effective solution to overcome the inadequate professional skills of workers, such as the installation training, grouting training, and caulking training of precast components. In order to solve the lack of appropriate lean technology/tools, the introduction of related advanced technologies is a good choice. e barriers that rank 6th to 10th are "lack of effective supervision and control" (weight � 3.543%), "absence of support from senior     leaders (e.g., project manager and government)" (3.532%), "lack of training and education regarding LC" (3.492%), "insufficient funds during construction" (3.475%), and "cooperation problems outside the construction department" (3.470%). e barriers that rank 6th to 8th still require improvements in terms of management and training. Besides, we recommend maintaining a certain amount of reserves and establishing a good collaboration mechanism among all units. Other barriers after ranking 10th also need to be valued according to the time and costs to be paid.

Result Analysis of Common Component Extracting and
Naming. After data preprocessing, a formal exploratory factor analysis begins to be implemented. Table 7 indicates the commonality before and after data preprocessing. In the formal exploratory factor analysis, the extraction commonality of each variable is greater than 0.5. Cronbach's alpha and KMO change to 0.945 and 0.919, separately. e significance (Sig.) is still 0.000. All components and the common components from them are extracted via the principal component analysis (PCA) method, as shown in Table 8. e eigenvalue of a component refers to the sum of squared loadings of all original observable variables on this component, namely, the variance of this component. e percentage of variance for a component is equal to its eigenvalue divided by the sum of eigenvalues of all components. For the extracted thirty components, their total variance is 30. If the eigenvalue of a component is not greater than 1, this means that the component can only explain no more than one original observable variable. Hence, the six components with the eigenvalue greater 1 should be selected as the common components, including Component 1, Component 2, Component 3, Component 4, Component 5, and Component 6. e common components are also known as the common factors or the common latent variables. In the column "Rotation Sums of Squared Loadings" of Table 8, the six common components account for 64.668% of the total variance explained, which is greater than the 60% required by sufficient construct validity [70,71]. Scree plot (see Figure 4) is adopted to test the rationality of common component extraction. Since the curve in Figure 4 begins to converge from the sixth component, it is appropriate to retain the first six components as common components.
Compared with the initial thirty barriers, the extracted six common components have weaker correlations. Moreover, these common components will save lots of time and effort in exploratory factor evaluation. However, only when the six common components have specific names and contents can they be used for subsequent exploratory factor evaluation. Hence, the original barriers are correspondingly classified into the six common components via the varimax rotation method, which changes the axis position to redistribute the percentage of variance explained by each component. e varimax rotation method will make the component structure simpler and does not change the cumulative percentage of the six common components. e rotated component matrix is shown in Table 9.

Basic Information of Project
Case. Different construction organizations have different abilities to deal with the thirty critical barriers encountered by on-site LC for prefabricated buildings. A construction organization is formed temporarily based on an engineering project. Once a project is over, the employees will return to their original companies or departments. Hence, the capability evaluation of a construction organization in dealing with the barriers should be based on a specific project case. e relevant findings related to this specific project case will provide a reference for similar projects in the future. A large-scale project case with detailed information is provided by a construction company in China. is project with a total construction area of 250,000 square meters adopts a general contracting mode called EPC (Engineering Procurement Construction). e "cast-in-place + prefabricated" frame structure is one of its three structural systems. e main structure of 1#, 2#, 4#, and 6# in this project is a prefabricated structure, and the prefabrication rate is 50%. Precast concrete components include beams, stairs, and semiprecast slabs. Figure 5 shows some scenario information for the project case. Figure 5(b) is one of the many buildings in this project and adopts "castin-place + prefabricated" frame structure. e construction organization attaches great importance to the on-site construction, which establishes a smart construction site cloud platform, a remote video monitoring system, a security risk self-checking system, a dust monitoring and reduction system, and many more. ey aim to strive for perfection rather than accept unsatisfactory events reluctantly during the construction.

Result Analysis of Capability Evaluation for Construction
Organization. In project practice, simple and effective methods are often favoured by construction organizations. If project members evaluate the capability of the construction organization to deal with the thirty barriers one by one, it is very time consuming and laborious. Moreover, there is a very strong correlation between these thirty barriers, which is not conducive to obtaining a scientific evaluation value. In order to solve these problems, the six common components with specific names instead of the thirty barriers are scored by six project members. e positions of the six project members are the project chief engineer, the project technical manager, the manager of company's engineering department, the construction quality controller, the project chief engineer of subcontractors, and the project executive manager of subcontractors. ese project members have sufficient qualifications in the whole construction organization. e evaluation and calculation results are shown in Formulas (7)(8)(9). e evaluation matrix S represents the scoring of the six project members. e weight vector W is obtained by normalizing the rotated variance percentage corresponding to each common component. e largest element in the comprehensive evaluation model E is 0.670, which corresponding to the level "fully capable." According to the maximum membership principle, the capability level of the construction organization is determined as "fully capable": W � 0.348 0.235 0.162 0.090 0.088 0.077 , E � W × S � 0 0 0.110 0.220 0 .670 .
According to Tables 8 and 9 and (7)-(9), the following analysis and discussion are carried out on the capability evaluation of construction organization. A total of six common components are extracted and nominated, namely, "management-related barriers" (% of variance � 22.526%), "skills and knowledge-related barriers" (15.175%), "barriers regarding the construction industry itself" (10.472%), "barriers regarding objective construction conditions" (5.851%), "barriers regarding materials and leaders" (5.661%), and "barriers related to prefabrication rate" (4.983%). eir cumulative percentage is 64.668%, which is greater than the 60% required by sufficient construct validity. en, the fuzzy comprehensive evaluation method is used to evaluate the capability of the construction organization to deal with all thirty critical barriers. e six common components extracted from the thirty barriers are scored by six employees from the project case, and the comprehensive evaluation result indicates that this construction organization is fully capable of dealing with the barriers to on-site LC. e worst level given by the six project employees owns a membership of 0.110, which is not lower than "medium ability." Based on the original data provided by the construction company, the reasons for the higher overall evaluation are as follows: (1) use of EPC mode; (2) establishment of standardized management system; and (3) attention and investment in advanced technology. e EPC mode gives the general contractor greater powers, which only needs to be responsible to the owner and can manage all subcontractors in a unified manner. is makes the construction organization more capable to solve some challenges in construction (e.g., the barriers to on-site LC). However, not every construction company has such a high capability. In addition to the contracting mode, the implementation effect and organizational capability of onsite LC are also related to the property (e.g., state-owned or privately owned) and scale (e.g., large or small) of construction companies [3,42].

Conclusion and Future Work
In order to solve the barriers of on-site LC for prefabricated buildings, this study develops a methodology framework based on multimethod collaboration. e methodological framework include three models, namely, the exploratory factor analysis model regarding critical barriers, the exploratory factor evaluation model regarding organizational capabilities, and the important findings and suggestions. Various data collection and analysis methods in these models cooperate with each other to achieve the effect of "1 + 1 > 2." Some complex mathematical formulas are also integrated into these models to increase the rigor and scientificity of the research. A large amount of data is calculated and analysed via the first two models to obtain many original results, which lay the foundation for making corresponding suggestions.
In the exploratory factor analysis model regarding critical barriers, the thirty-one critical barriers of on-site LC for prefabricated buildings are identified through literature analysis, field survey, and semistructured interview. After conducting a pre-exploratory factor analysis, the "resistance to changes in lean construction" is deleted. After conducting a formal exploratory factor analysis, the retained thirty barriers are prioritized, and the top five barriers are "inadequate professional management capability of managers" (weight � 3.657%), "insufficient standardization" (3.628%), "absence of lean awareness/thinking" (3.611%), "inadequate professional skills of workers" (3.589%), and "lack of appropriate lean technology/tools" (3.543%). Besides, the six common components accounting for 64.668% of the total variance are extracted and named, including "managementrelated barriers" (% of variance � 22.526%), "skills and knowledge-related barriers" (15.175%), "barriers regarding the construction industry itself" (10.472%), "barriers regarding objective construction conditions" (5.851%), "barriers regarding materials and leaders" (5.661%), and "barriers related to prefabrication rate" (4.983%).
In the exploratory factor evaluation model regarding organizational capabilities, it uses a few common components instead of a large number of initial observable variables.
is evaluation model not only eliminates the interference of correlation but also saves a lot of time and effort of evaluators. In order to demonstrate this evaluation model, a large-scale prefabricated project case is selected to measure its construction organization's capability to solve on-site LC barriers represented by the six common components.
e capability level of this construction organization is determined as "fully capable." For the barriers within the top five and the comprehensive evaluation result of case, this study conducts an indepth analysis. e corresponding substantive suggestions are provided by domain experts as follows (listed in the order of priorities): (1) An internal training or external recruitment is suggested to solve the inadequate professional management capability of managers. e internal training should take the form of seminars and training courses that invite senior prefabricated project management experts to participate. e external recruitment needs to focus on the management experience, lean skills, and leadership of managers in prefabricated projects.
(2) e standardization of on-site construction needs to be based on a sufficient understanding of the construction process: Firstly, the overall construction process needs to be scientifically and reasonably divided. Secondly, the main tasks of each functional department within a project are clearly defined. irdly, the standardized subprocesses need to be explained further. Finally, the process executors are trained to deepen their understanding of all standardized subprocesses. (3) How to make employees have the awareness/ thinking of LC is the focus of the future. If necessary, LC should be enforced by inserting LC clauses in a contract. (4) Lean construction education/training is considered as an effective solution to overcome the inadequate professional skills of workers. After the education/ training, it is necessary to assess the professional skills of workers and incorporate the assessment results into the performance related to wages and promotions. (5) It is also feasible to introduce LC technologies by visiting and learning some advanced foreign companies to continuously enrich the on-site LC of domestic prefabricated buildings. (6) In view of the excellent performance of EPC mode in promoting on-site LC of the project case, the EPC mode is suggested to be encouraged and promoted in future prefabricated projects.
e established methodology framework integrating various methods and mathematical formulas proposes a new idea for the barrier analysis and corresponding organizational capability evaluation of on-site LC from the perspective of the specific prefabricated construction industry. Since prefabricated buildings have particularity in terms of the construction mode compared with other buildings, the methodology framework further expands the current boundary of LC methodology. e findings and suggestions will provide a valuable reference and guidance for the prefabricated construction industry to solve the barriers regarding on-site LC. For prefabricated construction projects, this study will improve the utilization rate of on-site lean construction and the capability level of construction organization. However, this study only selects the construction organization adopting the EPC mode as a case, which may be slightly insufficient in terms of convincing. Hence, the capability evaluation regarding different types of construction organizations may be a direction for future research.

Data Availability
All data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.