Research on Influencing Factors of Knowledge Transfer among Prefabricated Construction Workers

Abstract

To identify the influencing factors and internal mechanism of knowledge transfer among prefabricated construction workers (PCWs), this study constructs a knowledge transfer behavior model for PCWs from various perspectives, including behavioral attitude, subjective norms, physiological perceived control, behavioral habits, and behavioral intention based on a modified Theory of Planned Behavior (TPB). It also employs a Structural Equation Model (SEM) for hypothesis validation and utilizes System Dynamics (SD) to simulate the knowledge transfer process of PCWs. Through empirical and simulation research, three conclusions are drawn: (1) Knowledge transfer willingness has a promoting effect on knowledge transfer behaviors. (2) Trusted relationships, organizational culture, physiological perceived control, and behavioral habits all have a promoting effect on workers’ knowledge transfer willingness and subsequently promote their knowledge transfer behaviors. (3) Among these factors, the strength of knowledge transfer willingness has the most significant impact on workers’ behavioral habits, followed by trusted relationships, physiological perceived control, and organizational culture. Additionally, when comparing the results of this study with knowledge transfer influencing factors of traditional construction workers (TCWs), it is found that trusted relationships and self-efficacy have a promoting effect on the knowledge transfer willingness of PCWs as well as TCWs. However, the impact of organizational culture, environmental perception, and behavioral habits on the knowledge transfer willingness of PCWs is more pronounced compared to TCWs. Based on TPB, this study constructs a suitable theoretical model to analyze the knowledge transfer process of PCWs by mining their group characteristics, and the research results establish a systematic analysis framework for the study of knowledge transfer behavior of PCWs. It also provides scientific suggestions for the formulation of targeted measures to enhance the willingness and efficiency of knowledge transfer of PCWs.

1. Introduction

The construction industry, as a traditional labor-intensive sector, is currently facing severe challenges such as labor shortages, a significant aging workforce, and low levels of knowledge and skills among construction workers. The industrial transformation of construction workers is urgently needed. Prefabricated construction, due to its production of several prefabricated components in a controlled environment with standardized design and mechanized manufacturing processes [1,2], is considered an effective means to achieve the industrialization of construction workers [3]. However, the widespread adoption and application of prefabricated construction also pose higher requirements for the knowledge and skills of construction workers. Currently, the majority of prefabricated construction workers (PCWs) come from traditional construction backgrounds and lack specialized knowledge in prefabricated construction. Additionally, factors such as entrenched professional experience, age, and professional conduct contribute to a slow uptake of new knowledge systems among these workers. This creates a contradiction with the urgent need for skilled PCWs due to the widespread promotion of prefabricated construction at present. How to scientifically and effectively enhance the skills and knowledge level of PCWs is crucial for realizing the advantages of prefabricated construction, including boosting construction efficiency, shortening construction periods, and reducing construction costs. So, there is an urgent need for practical and effective methods to enhance their knowledge and skills, and to provide a knowledgeable and innovative workforce for the development of prefabricated construction.

Knowledge transfer among construction workers helps narrow the knowledge gap and improve their overall knowledge and skill levels, leading to enhanced job performance [4]. Additionally, knowledge transfer enables construction workers to summarize lessons learned, innovate in the utilization of knowledge, reduce work errors, and improve the quality of job completion [4,5]. For prefabricated construction enterprises, the higher the effectiveness of knowledge transfer, the greater the level of job synergy among workers, ultimately enhancing the overall competitiveness of the enterprise [6]. However, the lack of effective knowledge transfer methods in the construction industry, particularly for PCWs, is a significant issue [7,8]. Compared to traditional construction workers (TCWs), PCWs need to learn specific new techniques through training and practice [9]. They must understand and become familiar with the performance, uses, and installation methods of prefabricated components. Their unique characteristics as a workforce, lower educational levels, and significant differences between prefabricated and traditional construction methods result in lower efficiency of knowledge transfer and limited improvement in skill levels, thereby impacting the development of prefabricated construction. To explore effective knowledge transfer methods, it is necessary to clarify the influencing factors and mechanisms of knowledge transfer among PCWs, to improve the efficiency of knowledge transfer. Based on the Theory of Planned Behavior, this study summarized the influencing factors of knowledge transfer of PCWs, combined with the SEM and SD to reveal the mechanism of action between the influencing factors, and put forward specific guidance suggestions for managers to effectively carry out knowledge management.

2. Related Works

2.1. Knowledge Transfer

Teece [10] proposed that knowledge transfer refers to the process through which enterprises transfer resources such as information and technology to help accumulate new knowledge, thus narrowing the knowledge and technological gap between different regions. Gilbert and Cordey-Hayes [11] defined knowledge transfer as the process of absorbing knowledge, integrating it, and effectively reutilizing it. Argote and Ingram [12] defined knowledge transfer as the process by which an organization transfers its expertise to another organization. Although different scholars have slightly different definitions of knowledge transfer, overall, it involves the transmission of knowledge from the knowledge sender to the knowledge receiver, who integrates, absorbs, and applies the acquired knowledge.

2.2. Theory of Planned Behavior

The TPB proposed by Ajzen in 1991 [13] suggests that behavioral intention directly influences individual behavior, and behavioral intention is influenced by three factors: attitude toward the behavior, subjective norms, and perceived behavioral control. Ibrahim and Heng [14] utilized TPB to explore the potential factors affecting knowledge sharing in small- and medium-sized enterprises. Tung et al. [15] examined the influencing factors on transfer willingness in the areas of occupational health and safety training among construction workers. They conducted data collection through a questionnaire survey with Australian construction workers and tested the transfer willingness model based on the TPB using Partial Least Squares Structural Equation Modeling (PLS-SEM). They performed a multi-group analysis to identify similarities and differences in factors influencing transfer willingness between managerial/professional and non-managerial/blue-collar construction workers. Wang et al. [16] proposed a framework encompassing measurement, absorptive capacity, communication, technology, transfer intent, and trust, based on theories of knowledge creation, the TPB, and organizational trust models. They explored determinants of knowledge transfer behavior from academia to industry collaboration. Man et al. [17] quantitatively studied factors influencing construction workers’ acceptance of personal protective equipment. They integrated theories including the Technology Acceptance Model, TPB, risk perception, and safety climate to explain the degree of acceptance of personal protective equipment among construction workers. Zheng et al. [18] employed TPB to validate the effectiveness of organizational culture on knowledge management within organizations. Zhao et al. [19] applied the TPB to investigate the mechanism by which job satisfaction among new-generation construction workers influences their professional behavior. They established a theoretical model of influence mechanisms based on TPB and SEM, aiming to identify key pathways that promote the development of professional behavior among construction workers. Furthermore, TPB has been used to explain knowledge sharing behaviors among construction team members [4]. TPB has been demonstrated to effectively explore the factors and mechanisms influencing the behavior and intentions of construction workers, thus providing a method for studying knowledge transfer behavior among construction workers. In this study, TPB is applied to investigate the knowledge transfer behaviors and constraining factors among PCWs, exploring the influence and mechanisms of attitude toward the behavior, subjective norms, and behavioral intention on knowledge transfer. This research aims to provide theoretical guidance for improving knowledge transfer behaviors and intentions among PCWs.

2.3. Factors Affecting Knowledge Transfer

Trusted relationships are considered crucial factors influencing knowledge transfer among organizational members. Trust among members can effectively facilitate the understanding of knowledge and promote the process of knowledge transfer [20,21,22]. Yazan [23] found that organizational culture plays a significant role in successful knowledge transfer within the construction industry. Owusu-Manu et al. [24] pointed out a significant positive correlation between organizational culture and knowledge transfer in the construction industry. Nesheim and Smith [25] discovered that an individual’s self-perception significantly influences knowledge sharing in projects. Moreover, the willingness of individuals to engage in knowledge transfer plays a decisive role in its success [20]. Zhou et al. [26] indicated that the stronger the knowledge provider’s willingness, the more likely their knowledge will be transmitted and received by the knowledge receiver, resulting in better knowledge transfer outcomes. From the above research, it can be inferred that trusted relationships, organizational culture, self-efficacy, and individual transfer willingness are crucial factors affecting knowledge transfer.

Rahat et al. [27] analyzed the gap between knowledge transfer and training transfer, using this as a basis to assess the disparities in individual capability for knowledge transfer, on-site hazard recognition, and workers’ safety performance related to training transfer. Huang et al. [28] employed the SEM to investigate the process of on-site safety knowledge transfer and its relationship with safety in the work environment, empirically exploring the relationships among various constructs in the process of construction site safety knowledge transfer. To date, research on knowledge transfer among construction workers has predominantly focused on factors influencing safety knowledge and awareness, as well as willingness for health training, with a lack of studies examining the impact of trusted relationships, organizational culture, physiological perceived control, and knowledge transfer willingness derived from the Theory of Planned Behavior on knowledge transfer. Maslow [29] pointed out that when individuals are in a state of physical decline or an unhealthy work environment, their enthusiasm and responsiveness in performing tasks and learning will significantly decrease. Compared to traditional construction methods, prefabricated construction provides workers with a more stable and conducive work environment [30]. Therefore, to highlight the work characteristics of construction workers and effectively differentiate the knowledge transfer behaviors between PCWs and TCWs, this study also includes the worker’s physical condition and perception of the external environment as influencing factors in knowledge transfer.

Zhou et al. [31] took social cognitive theory as the theoretical framework to study the factors affecting the knowledge transfer effect of TCWs and the mediating role of knowledge transfer intention, designed a questionnaire on knowledge transfer, and analyzed the knowledge transfer situation of 288 construction workers by using structural equation model. It is concluded that knowledge self-efficacy, blood relationship, trusted relationship, organizational culture, geographical relationship and other factors affect the effectiveness of knowledge transfer among construction workers. Cao et al. [1] explored knowledge transfer paths and transfer characteristics within worker groups and provided a theoretical basis for formulating new measures to improve knowledge and skills in worker groups. However, compared to TCWs, PCWs involved in assembly-line production in prefabrication factories tend to reduce cognitive effort during task execution by establishing work habits [32]. Their work often involves repetitive operations with fixed and standardized features [33]. Aarts and Dijksterhuis [34] suggested that when a specific behavior is repeated multiple times, future behavior tends to occur unconsciously and automatically. Christensen et al. [35] proposed that the more familiar workers are with their job, the more they rely on their work habits during actual work. However, excessive reliance on work habits can lead to reduced reflection on job content and weaken individual cognitive abilities, thereby affecting the process of knowledge transfer. Therefore, this study includes behavioral habits as a factor influencing knowledge transfer among PCWs.

This study, based on the existing literature and the characteristics of PCWs, has modified the TPB model to propose a theoretical model for knowledge transfer behaviors among PCWs, as illustrated in Figure 1 below.

3. Research Model and Hypotheses

3.1. Knowledge Transfer Willingness

According to TPB, behavioral intention refers to an individual’s subjective willingness to perform a specific behavior, which significantly influences the occurrence of actual behavior [13]. The stronger the individual’s intention to engage in a particular behavior, the more likely that behavior is to occur [33]. Szulanski et al. [36] defined knowledge transfer intention as the inclination of knowledge senders to share knowledge with other members of the organization. Research by Ren et al. [21] found that higher knowledge transfer intention leads to better quality of knowledge transfer. Ibrahim and Heng [14] proposed that a stronger intention for knowledge transfer leads to higher efficiency in knowledge transfer. Based on the above analysis, the following hypothesis is proposed:

H1. 

The knowledge transfer intention of PCWs is positively correlated with their knowledge transfer behaviors.

3.2. Trusted Relationships

According to TPB, behavioral attitude refers to an individual’s degree of endorsement towards a particular behavior, indicating a positive or negative evaluation of the behavior. Research has shown that behavioral attitude has a positive influence on behavioral intention. Trust is defined as one party’s willingness to be influenced by another party’s actions without considering the other party’s capabilities and taking positive actions [37]. Feng and Zhang [38] pointed out that trusted relationships can improve interpersonal relationships between knowledge senders and recipients and enhance the efficiency of knowledge transfer. Several scholars confirmed the importance of trusted relationships in knowledge transfer among project teams [5,20,21,26]. Based on the above analysis, this study combines the factors influencing knowledge transfer among PCWs, selects trusted relationships as an equivalent variable for behavioral attitude in the modified TPB model, and proposes the following hypothesis:

H2. 

Trusted relationships among PCWs have a positive effect on their knowledge transfer intention.

3.3. Organizational Culture

According to TPB, subjective norms reflect the influence of the individual’s organization on their behavioral execution. In the construction industry, subjective norms manifest as the behaviors and collective beliefs of project managers and colleagues [39]. Organizational culture is a shared belief system that implicitly influences members’ attitudes and behaviors [40]. Gunasekera and Chong [8] points out that a positive organizational culture contributes to active knowledge transfer activities among organizational members. Mainga [41] emphasizes the creation of a “blameless culture” within an organization, which can motivate members to contribute knowledge and enhance their desire to learn, avoiding ridicule for exposing their ignorance. Based on the above analysis, this study combines the factors influencing knowledge transfer among PCWs, selects organizational culture as an equivalent variable for subjective norms in the modified TPB model, and proposes the following hypothesis:

H3. 

Organizational culture has a positive effect on the knowledge transfer intention of PCWs.

3.4. Physiological Perceived Control

According to TPB, perceived behavioral control can influence behavioral intention [13]. Yu et al. [42] introduced the concept of Physiological Perceived Control (PPC) in studying construction workers’ safety knowledge sharing behavior and demonstrated its positive impact on such behavior. Physiological perceived control comprises two aspects: perceived behavioral control, which refers to an individual’s perception of the difficulty of performing a behavior [13], a concept similar to self-efficacy [42]; and physiological perception, which refers to an individual’s perception of the physiological environment while performing a behavior. Phung et al. [43] and other studies found that knowledge self-efficacy can promote organizational members’ knowledge sharing behavior. Additionally, workers can achieve optimal behavioral performance only in a suitable environment [42]. Uncomfortable work environments can lead to insomnia and decreased attention, reducing the desire to learn knowledge [44]. Based on the above analysis, the following hypothesis is proposed:

H4. 

Physiological perceived control (including self-efficacy and physiological environmental perception) positively affects the knowledge transfer intention of PCWs.

3.5. Behavioral Habits

Behavioral habit is defined as the frequency of past behavior [45]. Liu et al. [33] incorporated behavioral habits into the TPB to study the safety behavior of PCWs. Wong and Lee [46], using a modified TPB model, verified that behavioral habits can influence the behavioral intention of steel structure precast factory workers. Behavioral habits can also lead to higher cognitive phenomena, such as thinking habits. Lowe [47] used neural networks to explain how habit-based and goal-directed learning systems facilitate knowledge transfer. Prefabricated factories adopt an assembly-line production mode, which breaks down work tasks into smaller components, improving the learning curve of tasks and enabling workers to become more familiar with their work and develop habits [48]. However, excessive reliance on work habits can reduce workers’ desire to think about and seek knowledge, affecting their psychological processes and leading them to believe that simple work knowledge does not need to be shared with colleagues. Feng and Zhang [38] found that behavioral habits had no significant effect on the intention of construction workers to engage in unsafe work behaviors. Therefore, this study proposes the following hypothesis:

H5. 

Behavioral habits have positive effects on the knowledge transfer intention of PCWs.

The assumptions of this study are integrated into a modified TPB model, as shown in Figure 2.

4. Methodology

To accurately identify the factors influencing knowledge transfer behaviors among PCWs and understand the interaction mechanisms among these factors, this study applies the TPB and integrates existing research to summarize the influencing factors of knowledge transfer among PCWs. It utilizes the SEM to reveal the impact mechanisms of trust, organizational culture, behavioral habits, and knowledge transfer intention on knowledge transfer among PCWs. Furthermore, employing the SD simulation, it depicts the process of each influencing factor and provides targeted recommendations for improving knowledge transfer efficiency and knowledge and skill levels among PCWs.

The research process of this study consists of three stages.

✓

In the first stage, a literature review is conducted, and a questionnaire survey is distributed to explore the types and influence levels of the main factors affecting knowledge transfer among PCWs.

✓

In the second stage, the SEM is employed to examine the reliability, validity, and goodness of fit of the data obtained in the first stage. The theoretical hypotheses proposed in the third part are tested, and the relationships between different influencing factors and PCWs are determined.

✓

In the third stage, SD is utilized, and a simulation model is developed to simulate and validate the relationships identified in the second stage. This stage aims to uncover the impact pathways and underlying mechanisms of different influencing factors on knowledge transfer among PCWs. A flow chart of the research process is shown in Figure 3.

4.1. Questionnaire Investigation

This study, based on a comprehensive understanding of the group characteristics and overall knowledge level of PCWs, adheres to several principles for questionnaire design to ensure the objectivity and scientific validity of data collection:

(1)

Clarity and Simplicity: The language used in the questionnaire is simple and clear, ensuring that it is easily understood by the workers. This approach prevents misunderstandings or ambiguities among the participants, promoting accurate responses.

(2)

Controlling Question Quantity: While ensuring that all relevant indicators are covered, the questionnaire strictly controls the number of questions. Given that PCWs have diverse job tasks and long working hours, it is important to keep the questionnaire concise to avoid participants feeling overwhelmed or providing rushed, insincere responses, which could compromise the data’s accuracy.

(3)

Using Positive Statements: All questions in the questionnaire are presented in a positive manner. During the pre-survey process, it was found that using negatively framed questions, such as “I don’t always believe the knowledge my workmates share with me, and it may not be helpful to me” or “I rarely encounter new problems at work, so I don’t often discuss work-related knowledge with workmates”, could lead to misunderstandings among PCWs, potentially affecting the validity of the data.

(4)

Using a Five-Point Likert Scale: A five-point Likert scale is employed for PCWs to rate their attitudes towards various questions. Scores of 1 to 5 correspond to different levels of agreement, with 1 indicating “Strongly Disagree”, 5 indicating “Strongly Agree”, and 2, 3, and 4 representing varying degrees of agreement in between.

(5)

Pilot Testing: To ensure the questionnaire’s effectiveness, a pre-survey was conducted with a randomly selected group of 30 workers before formally distributing the questionnaire. Based on the data and feedback from the pre-survey, any questions with unclear or inappropriate wording in the questionnaire were modified to improve clarity and understanding.

The questionnaire was composed according to the five types of latent variables: trusted relationships, organizational culture, physiological perceived control, behavioral habits, knowledge transfer willingness, and knowledge transfer setting measurement items. No less than 3 measurement items correspond to each type of latent variable, and the abbreviation of each potential variable is added with a number to indicate the number of each measurement item. For example, the first measurement item of trusted relationships, “I believe that the knowledge taught by my workmates is correct and helpful to me”, is numbered TR1. The second measurement item of organizational culture, “Managers like to arrange for experienced employees to help new employees learn knowledge”, is numbered OC2. The corresponding items and sources of the five types of latent variables are shown in

Table 1. Measurement items and sources.

Latent Variables	Measurement Item Numbers	Measurement Items	Sources
Trusted Relationships	TR1	I believe that the knowledge taught by my workmates is correct and helpful to me.	Sun et al. (2019) [20]
	TR2	When I teach my workmates knowledge; they are willing to learn and trust me.
	TR3	At work, employees are very sincere in sharing their knowledge.
Organizational Culture	OC1	Managers encourage knowledge sharing.	Ni et al. (2018) [5]; Cao (2020) [1]
	OC2	Managers like to arrange for experienced employees to help new employees learn knowledge.
	OC3	Workers often share knowledge in their daily work.
Physiological Perceived Control	PPC1	I can pass on my knowledge to my workmates and let them get it.	Zhang and Ng (2013) [4]; Phung (2019) [43]; Zhou et al. (2022) [31]; Yu et al. (2021) [42]
	PPC2	When workmates teach me knowledge, I can understand and get it.
	PPC3	I am in good health at work and enjoy learning and sharing knowledge.
	PPC4	Compared to a construction site, I find the environment of a prefabricated component factory more conducive to learning and sharing knowledge
Behavioral Habits	BH1	I have been doing the same kind of work for the past month. My work is very simple, and the workload is very heavy.	Liu (2018) [2]
	BH2	When the workload is heavy, I am also willing to find time to exchange knowledge with my workmates.
	BH3	I take the initiative to think about new problems in my work, and I am willing to share them with my workmates.
	BH4	I am familiar with my work, but I still need to exchange knowledge with my workmates.
	BH5	Sometimes I will change different jobs, and I am willing to share different working knowledge with my workmates.
Knowledge Transfer Willingness	TKW1	I am willing to share all my knowledge with my workmates.	Sun et al. (2019) [20]; Zhou et al. (2020) [26]
	TKW2	The workmates are willing to share their knowledge with me.
	TKW3	When my workmates share knowledge with me, I am willing to study hard.
Knowledge Transfer	TK1	The knowledge shared by my workmates helps me solve problems in my work.	Sun et al. (2019) [20]
	TK2	At work, I have gained a lot by learning, and I am very satisfied with it.
	TK3	I learn new knowledge at work and use it to improve my work ability.

.

The questionnaire data collection was conducted from 20 December 2022 to 1 March 2023. The questionnaires were distributed in multiple locations, including Hainan, Beijing, Tianjin, Shaanxi, and Shanxi, among other regions. A total of 772 questionnaires were collected. The research team identified invalid questionnaires that had missing responses, exhibited obvious consistency in answers, or selected multiple options for the same question. These invalid questionnaires were excluded, resulting in a final sample size of 329 valid questionnaires. The number of valid questionnaires meets the requirement of the SEM, which typically requires a sample size of at least 200 [49]. The effective response rate for the questionnaires is 42.62%, exceeding the recommended minimum response rate of 20% suggested by Rowley [50].

The surveyed PCWs were predominantly male, with the age range primarily between 30 and 40 years old. Their educational attainment was mainly at or below the level of junior high school, which aligns with the overall characteristics of construction workers in China. Among the respondents, 76.6% were ordinary workers, while the rest included team leaders, management staff, and technical backbones within the workforce. Specific information about the surveyed workers is provided in

Table 2. Sample description.

Category	Classification	Number of Cases	Proportion (%)
Gender	Male	236	71.7
	Female	93	28.3
Age	≤20	4	1.2
	21–30	71	21.6
	31–40	127	38.6
	41–50	92	28.0
	≥51	35	10.6
Education level	Bachelor’s degree and above	19	5.8
	Junior high School	161	48.9
	Below junior high school	99	30.1
	Below high school	50	15.2
Length of service	3–5	68	20.7
	6–8	16	4.9
	≥8	16	4.9
	≤3	229	69.6
Position	Team leader	16	4.9
	Management staff	23	7
	Technical backbone	8	2
	General workers	252	76.6

.

4.2. Structural Equation Model

The SEM allows for the study of the relationships among variables and takes into consideration the characteristics of measurement error [51]. Therefore, the SEM is employed to test the measurement model established in this study. In this study, covariance-based SEM (CB-SEM) was adopted, based on the hypothesis of multivariate normal distribution and maximum likelihood estimation, and causality in the model was tested by analyzing the covariance matrix between variables. Initially, an evaluation of the model’s reliability and validity was conducted. Reliability is typically assessed using the Cronbach’s alpha coefficient. If 0.35 < Cronbach’s alpha < 0.7, it indicates moderate reliability; if 0.7 < Cronbach’s alpha < 0.9, it demonstrates high reliability; and if Cronbach’s alpha > 0.9, it suggests excellent reliability [52].

Validity comprises both structural validity and discriminant validity. Structural validity can be evaluated based on three criteria: (1) factor loadings should exceed 0.6; (2) Composite Reliability (CR) should exceed 0.7; and (3) Average Variance Extracted (AVE) should be greater than 0.5 [53]. Gefen et al. [54] proposed that when the square root of AVE for each latent variable is greater than the correlations between that variable and other variables in the same row and column, it indicates good discriminant validity.

Once the model’s reliability and validity have been confirmed through testing, an evaluation of the path coefficients in the model is conducted based on data analysis results.

4.3. System Dynamics Model

SD is a theoretical approach to analysis and research based on feedback loops, originally introduced by J.W. Forrester in 1990 [55]. Its system structure includes causal loop diagrams, stock-flow diagrams, differential equations, and simulation platforms. In this study, an SD model for the knowledge transfer among PCWs is constructed using causal loop diagrams and stock-flow diagrams. Differential equations are used to represent the relationships between variables, and the model is placed within a simulation platform for analytical simulations.

When using SD for modeling and simulation, it is necessary to define the boundary of the system to ensure the accuracy of the simulation model and that the simulation process is more realistic. Combined with the empirical analysis results in Section 4.1 and Section 4.2, this study proposes the following basic hypotheses for the knowledge transfer process of PCWs:

H6. 

There is a knowledge gap in the process of knowledge transfer. At the beginning of the simulation, the knowledge stock of the knowledge receiver is smaller than that of the knowledge sender.

H7. 

Trusted relationships, organizational culture, physiological perceived control, behavioral habits, and knowledge transfer willingness of knowledge sender and knowledge receiver all have influence.

H8. 

There is a knowledge threshold in the process of knowledge transfer, that is, the ratio of the knowledge stock of the knowledge sender to the knowledge stock of the knowledge receiver. When the knowledge threshold is too large, that is, the difference between the knowledge stock of the sender and the receiver is small, the sender will worry about the decline of his status and choose not to transfer knowledge.

H9. 

The workers in the prefabricated component factory are fixed and do not move during the simulation.

H10. 

The knowledge transfer process of PCWs is continuous and complex, and a simulation period of 36 months is assumed in this study.

5. Data Collection and Analysis

5.1. Reliability, Validity, and Goodness-of-Fit Tests

The reliability and validity test results of the questionnaire are presented in

Table 3. Reliability and validity.

Latent Variables	Measurement Items	Cronbach’s Alpha	Factor Loadings	CR	AVE
Trusted Relationships	TR1	0.786	0.714	0.789	0.556
	TR2		0.781
	TR3		0.740
Organizational Culture	OC1	0.754	0.665	0.756	0.509
	OC2		0.699
	OC3		0.772
Physiological Perceived Control	PPC1	0.819	0.718	0.823	0.528
	PPC2		0.652
	PPC3		0.795
	PPC4		0.761
Behavioral Habits	BH1	0.863	0.747	0.864	0.560
	BH2		0.765
	BH3		0.786
	BH4		0.716
	BH5		0.727
Knowledge Transfer Willingness	TKW1	0.840	0.777	0.829	0.618
	TKW2		0.793
	TKW3		0.788
Knowledge Transfer	TK1	0.866	0.823	0.867	0.685
	TK2		0.813
	TK3		0.847

and

Table 4. Correlations among latent variables.

	Trusted Relationships	Organizational Culture	Physiological Perceived Control	Behavioral Habits	$\sqrt{\mathit{A}\mathit{V}\mathit{E}}$
Trusted Relationships	1	-	-	-	0.746
Organizational Culture	0.329	1	-	-	0.713
Physiological Perceived Control	0.11	0.004	1	-	0.733
Behavioral Habits	−0.017	0.011	0.033	1	0.749

, respectively. After calculation, the Cronbach’s alpha values of all variables were greater than 0.7, indicating that the scale had passed the reliability test. Additionally, all factor loadings were greater than 0.6, CR was greater than 0.7, and AVE was greater than 0.5, indicating that the scale had good convergent validity. Furthermore, the square root of AVE for each latent variable was greater than the correlation between the same row and column of latent variables, demonstrating that all variables had good discriminant validity. The goodness-of-fit analysis results for the scale data are presented in

Table 5. Fitting results of the model.

Evaluation Indicators	Test Results	Evaluation Indicators	Test Results
χ2	237.614	RMSEA < 0.08	0.032
χ2/df < 3	1.335	NFI > 0.9	0.923
p < 0.05	0.05	IFI > 0.9	0.980
GFI > 0.9	0.937	TLI > 0.9	0.976
AGFI > 0.9	0.918	CFI > 0.9	0.979

, and all indicators meet the standard criteria.

5.2. Structural Path Coefficient

This research draws on the results of the SEM to create System Dynamics causal loop diagrams and simulation flowcharts. The knowledge transfer process of PCWs is simulated using the simulation platform provided by Vemsim ple. The hypothesis test results of the model are shown in Figure 4.

The findings are as follows:

Hypothesis H1 is supported: Knowledge transfer intention positively influences knowledge transfer behaviors (β = 0.82, p < 0.001).

Hypothesis H2 is supported: Trusted relationships positively influence knowledge transfer intention (β = 0.33, p < 0.001).

Hypothesis H3 is supported: Organizational culture positively influences knowledge transfer intention (β = 0.19, p < 0.001).

Hypothesis H4 is supported: Physiological perceived control positively influences knowledge transfer intention (β = 0.37, p < 0.001).

Hypothesis H5 is supported: Behavioral habits have a significant promoting effect on knowledge transfer intention (β = 0.30, p < 0.001).

5.3. Causal Relational Model Based on SD

Based on the existing literature and the SEM analysis results in this study, the knowledge transfer process among PCWs is influenced by the knowledge transfer willingness of both knowledge receivers and knowledge senders, which is affected by trusted relationships, organizational culture, physiological perceived control, and behavioral habits. Additionally, the knowledge stock of the knowledge sender depends on the innovative knowledge generated and the knowledge eliminated by the sender, while the knowledge stock of the knowledge receiver depends on the knowledge transfer amount and the knowledge eliminated by the receiver. Based on the above analysis, this study proposes six causal loop diagrams for knowledge transfer among PCWs, as shown in Figure 5.

(Loop 1) Knowledge sender’s knowledge stock (+) → Behavioral habits (+) → Cognitive level (+) → Knowledge sending willingness (+) → Knowledge transfer amount (+) → Knowledge sender’s knowledge stock (+)

(Loop 2) Knowledge sender’s knowledge stock (+) → Environmental perception level (+) → Physiological perceived control (+) → Knowledge sending willingness (+) → Knowledge transfer amount (+) → Knowledge sender’s knowledge stock (+)

(Loop 3) Knowledge sender’s knowledge stock (+) → Self-efficacy (+) → Physiological perceived control (+) → Knowledge sending willingness (+) → Knowledge transfer amount (+) → Knowledge sender’s knowledge stock (+)

(Loop 4) Knowledge receiver’s knowledge stock (+) → Knowledge transfer satisfaction (+) → Trusted relationships (+) → Knowledge receiving willingness (+) → Knowledge transfer amount (+) → Knowledge receiver’s knowledge stock (+)

(Loop 5) Knowledge receiver’s knowledge stock (+) → Knowledge transfer satisfaction (+) → Self-efficacy (+) → Physiological perceived control (+) → Knowledge receiving willingness (+) → Knowledge transfer amount (+) → Knowledge receiver’s knowledge stock (+)

(Loop 6) Knowledge transfer amount (+) → Organizational communication (+) → Organizational culture (+) → Knowledge receiving willingness (+) → Knowledge transfer amount (+)

Please note that the “+” symbol in the loops represents a positive influence or relationship between the variables.

Based on the simulation requirements of SD and considering the characteristics of knowledge transfer among PCWs, the model proposes the following five hypotheses:

✓

There exists a knowledge gap during the knowledge transfer process. At the beginning of the simulation, the knowledge stock of the knowledge receiver is less than that of the knowledge sender.

✓

Trusted relationships, organizational culture, physiological perceived control, and behavioral habits all have an impact on the knowledge transfer willingness of both knowledge senders and knowledge receivers.

✓

There is a knowledge threshold in the knowledge transfer process, i.e., the ratio of knowledge stock between knowledge senders and knowledge receivers. When the knowledge threshold is too large, meaning the difference in knowledge stock between the sender and the receiver is small, the sender may be concerned about a decline in their advantage in knowledge reserves and choose not to transfer knowledge.

✓

There are no personnel changes among workers in the prefabrication factories during the simulation.

✓

The knowledge transfer process among PCWs is continuous and complex. The simulation period for this study is 36 months.

5.4. Simulation Based on SD

Based on the causal loop diagram in Figure 5 and the aforementioned hypotheses, a System Dynamics model for knowledge transfer among PCWs has been developed, as shown in Figure 6. The model consists of 2 stock variables (the state variables), 5 rate variables (the rates of change), 14 auxiliary variables (intermediate variables used to calculate other variables), 2 exogenous variables (external inputs to the model), and 8 constants (fixed values in the model). The variable definitions, initial parameter settings, and equation designs are presented in

Table 6. Equation design.

Variable Name	Equation Setting
Knowledge Stock of Knowledge Senders	INTEG (Sender Innovation volume + 0.2 × Knowledge Transfer Volume-Sender elimination volume, 50)
Knowledge Stock of Knowledge Recipients	INTEG (Receiver Innovation Volume + Knowledge Transfer Volume-Receiver elimination volume, 10)
Sender Innovation Volume	Sender Innovation Rate × Knowledge Stock of Knowledge Senders
Sender Elimination Volume	STEP (Sender elimination rate × Knowledge Stock of Knowledge Senders, 6)
Knowledge Transfer Volume	DELAY1I (IF THEN ELSE (Knowledge Transfer Volume < 0.9, 0.34 × LN (Knowledge Transfer Satisfaction) + (1 − 0.22 × LN (Knowledge Gap)) + 0.44 × LN (Knowledge acceptance willingness), 0), 4, 0)
Receiver Innovation Volume	Receiver Innovation Rate × Knowledge Stock of Knowledge Recipients
Receiver Elimination Volume	STEP (Receiver elimination rate × Knowledge Stock of Knowledge Recipients, 6)
Knowledge Threshold	IF THEN ELSE (Knowledge Stock of Knowledge Recipients/Knowledge Stock of Knowledge Senders < 0.9, Knowledge Stock of Knowledge Recipients/Knowledge Stock of Knowledge Senders, 0.9)
Knowledge Gap	Knowledge Stock of Knowledge Senders-Knowledge Stock of Knowledge Recipients
Organizational Communication	0.66 × Knowledge Transfer Volume + 0.34 × Knowledge Transfer Satisfaction
Organizational Culture	0.3 × Sense of Identity + 0.3 × Sense of Belonging + 0.4 × Organizational Communication
Knowledge Transfer Satisfaction	0.56 × Knowledge Stock of Knowledge Recipients
Knowledge Acceptance Willingness	0.34 × Trusted Relationships + 0.14 × Organizational Culture + 0.24 × Physiological Perceived Control + 0.36 × Cognitive Ability
Knowledge Sending Willingness	0.27 × Physiological Perceived Control + 0.44 × Organizational Culture + 0.15 × Cognitive Ability + 0.14 × Trusted Relationships
Environmental Perception Level	0.48 × Environmental Perception Level + 0.52 × LN (“Self-efficacy”)
Self-efficacy	0.61 × Knowledge Stock of Knowledge Senders + 0.39 × Knowledge Transfer Satisfaction
Physiological Perception Control	0.48 × Environmental Perception Level + 0.52 × LN (Self-efficacy)
Behavioral Habits	0.45 × Knowledge Stock of Knowledge Senders × RAMP (0.42, 1, 36)
Thinking Level	0.6 × Behavioral Habits × RAMP (0.45, 1, 36)
Cognitive Ability	0.22 × Learning Efficiency + 0.33 × Learning Ability + 0.44 × Thinking Level
Trusted Relationships	0.24 × Emotional Trust + 0.52 × Competence Trust + 0.24 × Knowledge Transfer Satisfaction
Sender Innovation Rate	WITH LOOKUP (TIME, ([(0, 0), (36, 0.1)], (0, 0.06), (36, 0.08)))
Receiver Innovation Rate	WITH LOOKUP (TIME, ([(0, 0), (36, 0.1)], (0, 0.05), (36, 0.07)))

.

Based on the System Dynamics simulation results shown in Figure 7, the following conclusions can be drawn:

✓

As the knowledge transfer process progresses, the knowledge stock of both knowledge senders and knowledge receivers continuously increases, and the growth rate of knowledge stock becomes faster within the 36 months.

✓

Along with the knowledge transfer process, the knowledge transfer willingness of both knowledge senders and knowledge receivers continuously increases, and the growth rate of knowledge transfer willingness becomes faster within the 36 months.

✓

The values of the four auxiliary variables, trusted relationships, physiological perceived control, behavioral habits, and organizational culture, also continuously increase over the simulation period.

This indicates that during the knowledge transfer process, PCWs increase the frequency of communication among themselves, leading to the establishment of deep trusted relationships. Over time, the degree of trust deepens, and the influence of organizational culture strengthens through the frequent communication and collaboration of workers. Additionally, with increasing work experience, workers accumulate more skills and knowledge, forming a thinking framework for knowledge related to prefabricated construction under the influence of behavioral habits. PCWs become more familiar with their work environment and the knowledge they possess, leading to an enhanced perception of environmental control and self-efficacy, and an increasing ability of physiological perceived control.

6. Discussion

6.1. Sensitivity Analysis

To further explore the impact of trusted relationships, organizational culture, physiological perceived control, and behavioral habits on the knowledge transfer willingness of PCWs, this study adjusted the values of these variables to observe their effects on the simulation results.

(1)

Trusted relationships

The impact of changes in trusted relationships on knowledge transfer willingness is shown in Figure 8. Increasing trusted relationships within the organization has a significant enhancing effect on both the knowledge sending willingness and knowledge receiving willingness of PCWs. Trusted relationships effectively reduce the defensive psychology of knowledge senders and receivers, alleviating concerns of knowledge theft or misuse by senders and ensuring the authenticity and reliability of knowledge received by the recipients.

(2)

Organizational culture

The sensitivity analysis results demonstrate that strengthening organizational culture can enhance PCWs’ knowledge transfer willingness, as shown in Figure 9. In the early stages of knowledge transfer, when the organizational culture is in its initial establishment phase, the impact of improving organizational culture on knowledge transfer willingness may not be significant. However, in the later stages of knowledge transfer, as the atmosphere of knowledge sharing within the organization becomes more pronounced, experienced knowledge senders are more inclined to actively share knowledge with others under the influence of the overall culture. Knowledge receivers also become more receptive to learning the knowledge transferred by the knowledge senders.

(3)

Physiological perceived control

The sensitivity analysis results demonstrate that increasing physiological perceived control can enhance the willingness of knowledge receivers and knowledge senders, as shown in Figure 10. PCWs with higher physiological perceived control have greater self-efficacy and environmental awareness, enabling them to adapt more quickly to the working environment of the prefabrication factories. This increased confidence in their knowledge transfer abilities leads to a higher willingness to learn new knowledge and share knowledge with others.

(4)

Behavioral habits

The sensitivity analysis results demonstrate that strengthening behavioral habits significantly influences the willingness of knowledge senders and knowledge receivers, as shown in Figure 11. PCWs’ work habits contribute to their improved understanding of knowledge related to prefabricated construction, reducing barriers to comprehension during knowledge transfer. Additionally, they develop specific thinking habits, enhancing the efficiency of knowledge integration and internalization, thus increasing their knowledge transfer willingness. The stronger their behavioral habits, the deeper their familiarity with the knowledge, leading to increased confidence in knowledge sharing and learning, creating a positive feedback loop.

6.2. Comparison with TCWs’ Knowledge Transfer

Through empirical research and simulation studies on the factors influencing knowledge transfer among PCWs, this study found that trusted relationships, organizational culture, physiological perceived control (including self-efficacy and environmental awareness), and behavioral habits have a positive impact on PCWs’ knowledge transfer willingness. Knowledge transfer willingness, in turn, plays a facilitating role in PCWs’ knowledge transfer behaviors. Comparing the results of this study with the research on factors influencing knowledge transfer among TCWs conducted by our research team previously [31], it was observed that trusted relationships and self-efficacy have promoting effects on knowledge transfer willingness for both PCWs and TCWs. However, there were significant differences in the impact mechanisms of organizational culture, environmental awareness, and behavioral habits between the two groups.

(1)

Factors that affect the knowledge transfer of both PCWs and TCWs

This study found that trusted relationships and self-efficacy have promoting effects on knowledge transfer willingness for both PCWs and TCWs. Trusted relationships serve as a bridge for communication among team members. Whether for PCWs or TCWs, effective knowledge transfer requires communication with others within the project team. Trusted relationships among project members help establish positive interpersonal connections, reducing defensive attitudes during knowledge transfer. This enables knowledge senders to share their knowledge without fear of theft and knowledge receivers to accept knowledge without doubting its accuracy. Moreover, trusted relationships enhance the sense of belonging and identification within the team, providing a sense of achievement for knowledge senders through positive feedback from recipients and a sense of belonging for knowledge receivers through sincere assistance from senders. Trusted relationships also improve communication efficiency and strengthen the knowledge transfer willingness for both PCWs and TCWs. Thus, trusted relationships are essential elements in organizational communication and have a positive impact on knowledge transfer for both PCWs and TCWs.

Knowledge self-efficacy directly influences the behavioral patterns of construction workers in complex environments, and individuals with stronger self-efficacy are more proactive in knowledge transfer. Both PCWs and TCWs engage in knowledge transfer, which involves the processes of knowledge summarization and internalization by knowledge senders and knowledge receivers, respectively. In these processes, self-efficacy reflects the confidence of both senders and receivers in accomplishing the knowledge transfer task. Knowledge senders with high self-efficacy actively participate in internal organizational training and learning exchange activities, continuously improving their knowledge system. Likewise, knowledge receivers with high self-efficacy actively address difficulties during the knowledge transfer process, making efforts to resolve knowledge learning challenges. Self-efficacy embodies an individual’s self-awareness capacity, and it influences both PCWs and TCWs in their knowledge transfer endeavors.

(2)

Factors that have different impacts on knowledge transfer between PCWs and TCWs

This study found that organizational culture, environmental perception, and behavioral habits significantly influence the knowledge transfer willingness of PCWs, subsequently impacting their knowledge transfer behaviors. The mechanisms through which these factors influence knowledge transfer in PCWs differ significantly from TCWs.

Organizational culture significantly influences the knowledge transfer willingness of PCWs, while its impact on TCWs is not significant. This difference is attributed to the temporary and transient nature of traditional construction teams, where construction teams are formed and disbanded as projects progress. As a result, it becomes challenging for TCWs to develop a strong sense of identification and belonging within the team, leading to a reduced willingness for knowledge transfer. On the other hand, kinship and geographical relationships are essential influencing factors in knowledge transfer among TCWs. Due to the specialization of work roles and the unique labor patterns of traditional construction projects, these kinship and geographical relationships act as significant ties that maintain the work and life relations among workers. However, this can also foster a sense of “closed circles” within the organization, hindering the establishment of a strong organizational culture. In contrast, within the prefabrication factories, companies have more standardized management systems for workers, and the recruitment of workers follows more formal procedures compared to the traditional kinship-based subcontracting team model. The prevalence of casual labor is also relatively lower. PCWs have fixed job positions and workplaces, leading to greater job stability and a higher degree of professionalization. Thus, the organizational environment for PCWs is more conducive to cultivating a positive organizational culture, fostering a knowledge-sharing atmosphere within the team, and promoting workers’ proactive engagement in knowledge transfer.

The impact of environmental perception on the knowledge transfer willingness of PCWs is more significant compared to TCWs. This is because the production mode in the prefabrication factories significantly reduces the number of safety risks and the severity of accidents compared to the traditional on-site construction mode. Working in a relatively safer prefabrication factory enhances PCWs’ awareness of environmental safety, resulting in a more significant promotion effect on their knowledge transfer willingness.

The impact of behavioral habits on PCWs’ knowledge transfer willingness is more pronounced compared to TCWs. This is because TCWs face a complex and dynamic construction environment, and the nature of work tasks varies across different construction projects, making it challenging for them to form significant behavioral habits that directly influence knowledge transfer. In contrast, prefabricated factories adopt a streamlined production mode, breaking down work tasks into smaller modules, which improves the learning curve of tasks. This enables PCWs to become more familiar with and proficient in their work, leading to the formation of habits. Such habits alleviate the understanding barriers during knowledge transfer and subsequently enhance their willingness to engage in knowledge transfer. Experienced PCWs often possess a higher level of cognitive control over knowledge within their work domain, making them more conducive to facilitating knowledge transfer.

7. Conclusions and Suggestions

7.1. Conclusions

This study adopts the TPB as the theoretical framework to explore the influencing factors and mechanisms of knowledge transfer among PCWs. Empirical research results demonstrate that trusted relationships, organizational culture, physiological perceived control, and behavioral habits have a positive impact on PCWs’ knowledge transfer willingness, and knowledge transfer willingness also positively influences knowledge transfer behaviors.

✓

Trusted relationships have a positive impact on PCWs’ knowledge transfer willingness. It enhances the willingness of knowledge senders to actively share knowledge with recipients. Trusted relationships also ensure the authenticity of knowledge during the transfer process.

✓

Organizational culture has a positive impact on PCWs’ knowledge transfer willingness. Under the active encouragement of managers and the prevailing atmosphere of knowledge sharing, knowledge senders are motivated to proactively share knowledge to gain recognition and trust from superiors and colleagues.

✓

Physiological perceived control has a positive impact on PCWs’ knowledge transfer willingness. It includes two aspects: self-efficacy and perception of the working environment. Individuals with stronger physiological perceived control in the prefabrication factories have higher self-identification with their knowledge level and are more adept at adjusting their knowledge acquisition approach in response to environmental changes.

✓

Behavioral habits have a positive impact on PCWs’ knowledge transfer willingness. Experienced PCWs often have a higher cognitive grasp of knowledge in the field of work, and their willingness to engage in knowledge transfer is enhanced.

✓

Knowledge transfer willingness has a positive impact on PCWs’ knowledge transfer behaviors. In prefabrication factories, the stronger the willingness for knowledge transfer between knowledge receivers and knowledge senders, the higher the likelihood of knowledge transfer behaviors occurring.

The simulation study further verified that trusted relationships, organizational culture, physiological perceived control, and behavioral habits have varying degrees of positive effects on knowledge transfer intention. Among them, the willingness of knowledge senders and receivers is more sensitive to changes in behavioral habits.

7.2. Suggestions

Based on the impact of various factors on knowledge transfer, this study proposes relevant recommendations for the knowledge management of PCWs, guiding managers to implement targeted measures to improve PCWs’ knowledge enhancement methods and efficiency.

✓

The establishment of trusted relationships and organizational culture can enhance the trust among organizational members and stimulate the proactivity and enthusiasm of workers to share knowledge. Managers can foster a conducive knowledge-sharing environment among PCWs by organizing skill assistance activities, conducting technical discussions, or arranging regular recreational events. Furthermore, companies can implement knowledge-sharing incentives and promotion mechanisms to increase PCWs’ sense of respect and recognition derived from knowledge transfer activities, thereby enhancing their willingness to engage in knowledge transfer.

✓

A safe and comfortable working environment facilitates PCWs’ quick adaptation to their work environment. Standardized work procedures help cultivate positive behavioral habits and improve learning efficiency among PCWs. Therefore, it is essential to improve internal rules and regulations, such as the admission and training system for PCWs, strict on-site operation procedures, and effective supervision and management measures, ensuring that workers can perform their tasks safely and efficiently on the assembly line while avoiding the development of undesirable work habits.

This study proposes a theoretical model of the influencing factors of knowledge transfer among prefabricated construction workers based on the Theory of Planned Behavior. Additionally, it employs the System Dynamics methodology to dynamically simulate the knowledge transfer process. This provides a theoretical basis and feasible recommendations to promote knowledge transfer behavior among prefabricated construction workers, enhancing productivity and fostering standardized construction practices and high-quality workforce transformation in the construction industry. It lays a foundation for further exploration of methods for knowledge transfer among prefabricated construction workers. Future research will explore effective methods of knowledge transfer among prefabricated construction workers based on trusted relationships, organizational culture, physiological perceived control, behavioral habits, and knowledge transfer willingness, examining their positive impacts on knowledge transfer. However, this study also has certain limitations. It only considers knowledge transfer among prefabricated construction workers working within prefabrication plants and does not include on-site installation workers in the discussion. Future research could utilize big data, algorithmic analysis, and other methods to incorporate more information related to knowledge transfer among prefabricated construction workers into the research sample. By identifying the influencing factors of knowledge transfer among prefabricated construction workers and considering both prefabrication plant workers and on-site installation workers, future studies could provide a more comprehensive analysis of knowledge transfer in prefabricated construction.