What Kind of Policy Intensity Can Promote the Development of Intelligent Construction in Construction Enterprises? Study Based on Evolutionary Games and System Dynamics Analysis

Abstract

Previous studies have focused on the fact that government policies are the key factors in promoting the development of intelligent construction in construction enterprises. However, how to select different forms of policy support and quantify the intensity of policy support, as well as the impact on the behavioral strategies of construction enterprises and the government, still needs in-depth exploration. This paper constructs an evolutionary game model between construction companies and the government, using the system dynamics simulation software Vensim to analyze the model under three different government policy support scenarios. The study explores how varying levels of policy support and key factors influence the strategic choices of the game participants, providing valuable insights for promoting the development of intelligent construction. The key findings are as follows: (1) The willingness to adopt intelligent construction is heavily dependent on policy incentives. The incentive effect of the three single policies is much lower than that of the combined policies, and only high-intensity special fund support (more than 8 CNY/m²) significantly promotes widespread adoption. Among combinations of policies, tax incentives coupled with special funds prove most effective. (2) The government’s decision to actively promote intelligent construction hinges on a cost–benefit analysis. Under medium to high levels of special fund support, medium to low levels of service support are more beneficial for reaching a stable state of intelligent construction implementation. (3) Reducing the incremental costs of intelligent construction transformation is the primary key factor in promoting construction. The findings contribute to a deeper understanding of how both the government and construction companies can adjust their strategies in response to policy changes, ultimately leading to more effective policy implementation and strategic decision-making.

1. Introduction

The construction industry is a crucial pillar of China’s national economy. However, currently, problems such as the extensive production mode, imperfect regulatory mechanisms, and low level of informatization have become increasingly prominent [1]. At the same time, the development of information technologies such as the Internet of Things and big data has gradually changed the organizational structures and production modes of various industries. The construction industry has an increasingly urgent need for informatization, and the construction industry urgently needs to transform and upgrade [2]. Intelligent construction, centered on information technologies such as cloud computing, BIM, the Internet of Things (IoT), big data, and artificial intelligence (AI), represents a new construction model that achieves information integration and efficient collaboration across the entire project lifecycle [3,4,5,6]. Developing intelligent construction is one of the key tasks in advancing new urban infrastructure based on digitalization, networking, and intelligence, which is significantly important for the transformation and sustainable development of China’s construction industry [7,8,9]. However, the overall level of intelligent construction in the industry is still in its nascent stage, and there are numerous obstacles to its promotion.

Compared with traditional construction methods, the promotion and application of intelligent construction require more investment, more stakeholders, and a more complex organizational structure [10]. The promotion of intelligent construction involves the collective participation and interaction of multiple stakeholders, primarily including construction companies (such as contractors, developers, and intelligent production enterprises) and the government. The significant heterogeneity of interests among these parties leads to obstacles in the transformation and upgrading of construction methods [11,12]. From the perspective of construction companies, there is an inherent profit motive focused on short-term cost–benefit analysis. Currently, the substantial investment in technology and the weak foundational platforms result in insufficient internal motivation for transformation, especially in the absence of adequate government incentives. From the viewpoint of government regulation, the government has gradually developed policies such as pilot programs and subsidies to promote intelligent construction. However, the corresponding policy support system remains underdeveloped, and the implementation is inconsistent, making it difficult to play a guiding role. Therefore, as a major stakeholder in the transformation of the construction industry towards intelligent construction, how government policies can effectively play an incentive role is particularly crucial [13,14]. This paper intuitively reflects the dynamic evolutionary process of the strategic choices of various stakeholders in the transformation of construction enterprises towards intelligent construction by constructing an evolutionary game model between construction enterprises and the government. It also uses Vensim simulation to study the mechanism by which construction enterprises and the government continuously adjust their transformation strategies according to the intensity of policy support. Based on the simulation results, suggestions are put forward to promote the development of intelligent construction in China.

The rest of the study is structured as follows: Section 2 introduces the literature and points out the contributions of this paper. Section 3 introduces the main research methods of this paper, puts forward the assumptions of evolutionary game theory, and constructs the model. Section 4 uses numerical simulation to analyze and compare the results obtained using parameter variations. Section 5 discusses the model and numerical simulations. Finally, Section 6 draws conclusions and offers suggestions for future actions.

2. Literature Review

2.1. Current Development Situation of Intelligent Construction

As a new technology and management approach for the transformation and upgrading of traditional construction models, intelligent construction has attracted a great deal of research by scholars both at home and abroad regarding its current development situation, technological applications, and development obstacles. This concept originated from Smart City (SC). SC is the intersection of traditional cities and information technology to achieve urban government and improve living standards through the application of new technologies [15,16]. SC is the key to achieving sustainable social development and can effectively improve the quality of life through SC construction [17]. Based on the development of the smart city concept, the construction industry started to introduce the concept of smart construction. Building Information Modeling (BIM) is beginning to be widely used in construction project management, providing architectural, engineering, and construction professionals with the tools to more effectively plan, design, and manage construction activities [18]. Under the influence of the explosive growth of building volume, the increasingly complex on-site environment, and the large number of participants, the demand for informatization in construction sites has become more and more urgent, and smart construction sites (SCS) have emerged as a result [19]. Smart construction sites (SCS) use information technology to develop intelligent systems and supervision platforms, which can help traditional construction sites solve many problems and accomplish safe, efficient, and high-quality construction [20]. The Internet of Things (IoT), blockchain, and Geographic Information System (GIS) are integrated with BIM to achieve a digital application of engineering projects [11,21,22]. With the introduction of the Industry 4.0 concept into the construction industry, the prefabricated construction method can effectively improve construction efficiency and safety. At the same time, through the combination of VR, sensors, the Internet of Things, etc., it has applications in aspects such as building monitoring, construction safety management, early prediction of building disasters, and asset management. This can enhance the comprehensive management and control capabilities of construction sites, reduce sick buildings, and increase healthy buildings [23,24,25,26].

However, there are still many deficiencies in the process of promoting intelligent construction. Chen et al. [27] noted that China has yet to develop a comprehensive and systematic understanding of the theories and key technologies related to intelligent construction, nor has it established a clear development direction. They pointed out that there are still many shortcomings in the promotion of intelligent construction. Ahn [28] found that the adoption of intelligent construction technologies in construction projects is relatively limited. He used the Quality Function Deployment (QFD) method to extract and prioritize the required technologies from different stakeholders, enabling decision-makers to address urgent issues strategically. Osunsanmi [29] conducted a survey of construction industry professionals regarding their willingness to adopt intelligent construction technologies and concluded that the degree of integration between intelligent construction and traditional construction technologies is a significant factor influencing whether professionals opt for intelligent construction. It is also found that, despite the availability and popularity of many technologies and typical cases in China [30], SCS is not widely adopted worldwide [31].

Therefore, scholars have also conducted research on the influencing factors of the transformation to intelligent construction. Ejidike [32,33,34] identified macroeconomic barriers and financing channels as significant obstacles to the transition to intelligent construction in economically underdeveloped countries. Gao et al. [35] used principal component analysis and multiple regression analysis to conclude that insufficient policy and industrial support, an inadequate coordination system, high construction costs, and low market recognition are major factors affecting the development of industrialized construction. Wang [36] highlighted the shortcomings in the current development of new industrialized construction and intelligent construction, noting issues such as imperfect mechanisms, weak industrial support, and insufficient motivation among market participants. Government policy support is one of the important factors influencing the transformation to intelligent construction [37,38,39,40,41]. The government plays a dominant role in formulating policies related to the construction industry [42,43]. This paper will focus on discussing the strategic choices of enterprises under government support.

The government promotes intelligent construction through different forms of policy support. Wang et al. [44] systematically reviewed the national policy framework for intelligent construction and local support policies in five areas: finance, land, taxation, awards, and talent. Li [45] argued that the formulation of government incentive policies should guide and regulate market behavior through economic interests to encourage enterprises to research and apply relevant intelligent construction technologies. Specific measures include tax support, fund refunds, financial rewards, and excellence incentive policies. Additionally, research support is crucial, such as promoting universities and research institutions to actively engage in the development of intelligent construction technologies, including specialized funding for R&D and innovation grants. Chen and Ding [27] pointed out that establishing an innovation investment incentive mechanism for key intelligent construction technologies can help enterprises accelerate the incubation of new technologies. Yang et al. [46] suggested that the government should adjust and formulate incentive policies with a focus on the introduction of technical talent. Jamil et al. [47] emphasized the need for the government to formulate relevant laws and regulations to incentivize and maintain processes related to intelligent construction technologies. Wei et al. [48] noted that mandatory government policies are crucial for regulating market behaviors, specifically through strict regulatory mechanisms and penalties for non-compliance. Radziejowska [34] and Soares C A P [28] argued that only by improving legal, financial, and tax incentive measures, subsidies, and specific credit limits can the development of the intelligent construction market and the participation of relevant stakeholders be promoted. Xing and Cao [49] highlighted that government grants and financial support are considered key to promoting green building development. Currently, various provinces and cities in China are implementing pilot construction plans to promote the transformation to intelligent construction, with key policy support measures including special fund subsidies, tax incentives, and service support [44,50]. However, how to select an appropriate policy form and incentive intensity based on the current development status of the industry to promote the development of intelligent construction is the focus of this study.

2.2. Evolutionary Game Theory

Game theory can apply mathematical models to investigate and infer strategy choices among stakeholders [51]. Under certain assumptions, it can better address the strategy choices made by different individuals for their benefit under the conditions of cooperation and competition [52]. The study of game theory and dynamic evolutionary processes, which have their roots in behavioral ecology and biological evolution, is combined in the new theory known as evolutionary game theory [53]. Evolutionary games are devised to address the multiple-equilibrium dilemmas in classical games. They can prove that a restricted number of interactions among individuals can generate certain equilibrium strategies [54]. In recent years, these games have been widely utilized in economic and management research. Evolutionary game theory does not mandate that agents possess complete rationality, believing instead that agents will continuously alter their strategies through learning and modeling. This makes it more aligned with the behavioral characteristics of organizations and individuals in the fields of new technology promotion, such as environmental protection, green and low-carbon initiatives, prefabricated buildings, etc. [55]. Qian believes that evolutionary game models are capable of elucidating the formation and evolution of group behaviors in the environmental protection domain [56]. Examples include the elevation of public environmental awareness [57] and the selection of corporate environmental strategies [58]. Through the analysis of the dissemination and adaptation processes of diverse strategies within groups, the underlying causes for the formation of acceptance behaviors can be revealed. This will provide supportive evidence for government policy-making and the attainment of governance goals [59]. Therefore, owing to their advantages of high theoretical compatibility, high interpretability, and high guidance value, evolutionary game models are frequently utilized in behavioral research within the new technology promotion field.

2.3. Application of Evolutionary Game Theory in Intelligent Construction

The promotion of intelligent construction requires the joint participation of various stakeholders, including the government and enterprises [60]. Scholars have analyzed the impact of the game behaviors among owners, construction companies, and technology service providers on the transition to intelligent construction. Wang [61] constructed an evolutionary game model between “government and developers”, concluding that the steady-state equilibrium strategy involves the government incentivizing developers, who, in turn, pursue maximum profit through their business models. Tang et al. [62] developed a bilateral evolutionary game model between owners and facility management service providers to explore the main factors affecting the adoption rate of key technologies in intelligent construction, suggesting that owners can mobilize organizational members’ enthusiasm through appropriate incentive mechanisms. Ye et al. [63] analyzed the factors affecting the adoption and improvement of key intelligent construction technologies by construction enterprises and technology suppliers, constructing a technology diffusion model based on complex network theory, which indicated that relevant government incentive policies can significantly enhance technology demand and improvement. Scholars have also examined the degree of impact of the game relationship between construction companies and the government on transformation. Feng et al. [64] and Sang et al. [65] established a tripartite evolutionary game model involving government, developers, and consumers in green building, providing recommendations from perspectives such as government rewards, regulatory intensity, and low-carbon trading mechanisms. Zhao et al. [66] performed a simulation analysis using system dynamics models, proposing reasonable suggestions from market changes, government policy formulation, and corporate development willingness perspectives. Zhu et al. [67] constructed an evolutionary game model involving the government, construction units, and technology suppliers, demonstrating through simulation that government subsidies, taxes, and penalties are key factors affecting the decisions of all participants, advocating for a dual incentive mechanism combining macro government incentives with adoption and development incentives. Zhang [68] established an evolutionary game model among the government, construction units, and construction workers, emphasizing that the benefits to construction workers from using intelligent construction technologies are key to promoting these technologies. It can be seen that through evolutionary game theory, the strategic choices of relevant stakeholders during the process of changes in their mutual behaviors can be analyzed.

2.4. Research Problems and Main Contributions

Current research on intelligent construction primarily focuses on the status and application of technologies, as well as the multi-stakeholder interactions among enterprises and government and between enterprises. It emphasizes the government’s leading role in guiding and encouraging the transformation and development of intelligent construction in China. However, most scholars discuss the impact of policies on the transformation of intelligent construction in a generalized manner, without conducting quantitative analyses of the incentive effects of different categories of policies or exploring the impact of policy combinations. The dynamic and uncertain relationship between government policies and the promotion of intelligent construction requires further investigation. Given this context, this paper will employ evolutionary game theory to analyze the dynamic evolution of strategy choices between construction enterprises and the government. It will consider three types of government policy support and apply system dynamics to simulate the game model. This study aims to examine the influence of varying levels of policy support on the strategic choices of both construction enterprises and the government, providing valuable insights for advancing the development of intelligent construction.

The main contributions of this paper are as follows: (1) Studying the strategic evolution of construction enterprises under government intervention, constructing an evolutionary game model between the government and construction enterprises, conducting a quantitative analysis of the impacts of the government’s single policies and combined policies on the willingness of construction enterprises to implement intelligent construction, determining the recommended thresholds for policy incentives and combination strategies, and overcoming the limitations of qualitative analysis of policy impacts. (2) Determining the incentive sequence for promoting the transformation of construction enterprises under the support of government policies through sensitivity analysis, which is helpful for the government to determine the phased focus when formulating policies.

3. Methodology

The theoretical framework in this study is mainly composed of three parts, as shown in Figure 1. The first part is to identify the key stakeholders and analyze the main types of incentive policies. Based on the analysis of interest demands and interest conflicts, it is clarified that the government and construction enterprises are the key stakeholders in the development of intelligent construction. Through literature research, the key policy support measures are determined, including special fund subsidies, tax incentives, and service support. Secondly, based on the evolutionary game theory, the payoff matrix of the game is calculated, the replication dynamic equations are listed, and the equilibrium point analysis is carried out. Finally, numerical simulations are conducted using the system dynamics method and Vensim PLE 7.3.5 to explore the evolution process of construction enterprises and the government under different policy types and policy support intensities. And through sensitivity analysis, the impact of external variables on the evolution is explored, and the key development strategies are derived.

3.1. Model Assumptions and Construction

3.1.1. Problem Description

In China, construction enterprises generally face low profit margins, and issues related to an aging workforce have become increasingly prominent. To enhance core productivity, improve efficiency, and reduce labor demand, these enterprises urgently need to transition from traditional extensive development to refined, intelligent approaches. However, uncertainties regarding upfront costs and investment returns have left many construction companies in a state of hesitation [69].

To accelerate the development of new productive forces and facilitate the implementation of intelligent construction, government support policies are crucial. Appropriate and reasonable incentive policies can create a conducive environment for transformation, reinforce the government’s positive guiding role, and alleviate the pressure on enterprises to transition. This paper analyzes the evolutionary game between the two main stakeholders—government and construction enterprises—during the promotion of intelligent construction in the construction industry. By assigning values to the incentive coefficients of three primary support policies, the study explores the impact of policy support intensity on the implementation of intelligent construction by construction enterprises.

The returns that construction enterprises realize during the construction process often dictate their development strategies. After the first round of the game, construction enterprises will assess their own profit situations before deciding on their strategies for the second round of the game [70]. Key factors in this game include incremental returns and incremental costs. On one hand, intelligent construction is an innovative model that integrates new information technologies with engineering construction, requiring significant upfront investments in costs and technical capabilities, including technology learning, equipment support, and talent training. In the short term, this can lead to reduced profits or even losses, significantly dampening the willingness of construction enterprises to transition. On the other hand, the smartization of construction methods can transform production processes, enhance quality and efficiency, facilitate the rational allocation of resources, and improve management levels and the digital image of enterprises, potentially increasing social benefits and added value.

From the government’s perspective, it plays supervisory, managerial, and service roles in construction management. By promoting intelligent construction and innovating the industry development model, the government can drive the transformation and upgrading of the construction sector, leading to high-quality industry growth, enhancing its image, and building social credibility. However, to facilitate the transition to intelligent construction methods, the government must implement incentive measures such as financial subsidies and service guidance to encourage construction enterprises to invest in their transformation. Excessive fiscal burdens may also influence the government’s strategic choices.

3.1.2. Assumptions and Parameters

Assumption 1.

Both parties in the game are of limited rationality and seek to maximize their own interests. Due to capacity constraints, it is difficult for both parties to obtain all market information. Both are operating under conditions of incomplete information, and they engage in dynamic games based on their returns to find optimal strategies.

Assumption 2.

The government and enterprises have two strategies during the game, and they can freely choose whether to support or implement intelligent construction methods. Enterprises aim to maximize their own interests by selecting strategies that are most beneficial for their development.

Assumption 3.

The probability of a construction enterprise choosing to implement intelligent construction is denoted as x (where 0 ≤ x ≤ 1), while the probability of not implementing intelligent construction is 1 − x. The government’s probability of actively promoting intelligent construction is denoted as y (where 0 ≤ y ≤ 1), and the probability of passively promoting it is 1 − y.

Assumption 4.

When using traditional construction methods, the returns for construction enterprises are E. The additional returns from implementing intelligent construction are also ∆E, which includes improvements in production management efficiency, enhanced quality of on-site construction and management, increased potential for winning bids, and improved social image. The costs incurred by construction enterprises for transformation are denoted as C1, which includes expenses such as talent training, procurement of hardware and software, and technological investments. During the initial stages of transformation, due to constraints in resources, capabilities, and scale, the additional returns are less than the incremental costs, i.e., ∆E < C1. Furthermore, traditional construction methods may have management flaws in green construction and quality safety supervision, resulting in penalties denoted as L.

Assumption 5.

The basic benefits for the government when traditional construction methods are used are denoted as G. Supporting intelligent construction yields positive impacts and incremental returns from reasonable allocation of social resources, denoted as ∆G. The development of new productive forces associated with intelligent construction creates numerous employment opportunities, shapes the government’s green and intelligent governance image, enhances social credibility, and increases international influence, contributing social benefits to the government denoted as S. The government can reduce the transformation costs for enterprises through tax incentives, such as lowering tax rates and waiving fees, with the tax incentive coefficient denoted as a (where 0 ≤ a ≤ 1) and the tax reduction ratio as m (where 0 ≤ m ≤ 1). Additionally, the government provides special fund subsidies for intelligent construction enterprises and projects, with the special fund amount denoted as F and the subsidy coefficient as b (where 0 ≤ b ≤ 1). The government also offers service support through policy formulation, training seminars, technical guidance, land support, and platform construction, with the government service cost denoted as C2 and the service provision coefficient as r (where 0 ≤ r ≤ 1). Relevant parameter settings are summarized in

Table 1. Notation description.

Parameters	Definition
x	Probability of construction companies adopting intelligent construction
E	Operating profits of construction companies using traditional construction methods
∆E	Additional profits achieved by construction companies using intelligent construction
C1	Incremental costs for construction companies using intelligent construction
L	Possible penalties for construction companies using traditional construction methods
y	Probability of government support for the development of intelligent construction
G	Basic benefits of the government from using traditional construction methods
∆G	Incremental benefits to the government from using intelligent construction
S	Additional social benefits for the government from supporting intelligent construction
a	Tax incentive coefficient
m	Tax incentive rate
F	Special fund allocation by the government for the development of intelligent construction
b	Special fund subsidy coefficient
C2	Cost of government-provided services for the development of intelligent construction
r	Service supply coefficient

.

3.1.3. Payoff Matrix and Replication Dynamic Equation

Based on the above assumptions, we can derive the evolutionary game payoff matrix for construction enterprises and the government, as shown in

Table 2. Payoff matrix for evolutionary game between government and construction enterprises.

Construction Enterprises	Government
	Actively Promoting (y)	Passively Promoting (1 − y)
Implement intelligent construction (x)	E + ∆E + amC1 + bF − C1	E + ∆E − C1
	G + ∆G + S − rC2 − amC1 − bF	G + ∆G
Implement traditional construction (1 − x)	E-L	E
	G − rC2 + S + L	G

.

3.2. Stability Analysis of the Evolutionary Game

3.2.1. Single Population Evolutionary Stability Analysis

1.

Evolutionary Stability Analysis for Construction Enterprises

The expected returns for construction companies choosing intelligent construction methods can be denoted as U₁. Expected returns for construction companies choosing traditional construction methods can be denoted as U₂. Average expected returns can be denoted as U_x. They can be formulated as follows:

$$U_1=y(E+∆E+amC1+bF-C1)+(1-y)(E+∆E-C1),$$

(1)

$$U_2=y(E-L)+(1-y)E,$$

(2)

$$U_x=x{U}_1+(1-x)U_2.$$

(3)

The replication dynamics equation for construction enterprises, denoted as F(x). This can be formulated using the following general form:

$$\mathrm{F}(\mathrm{x})={\displaystyle \frac{dx}{dt}}=\mathrm{x}({\mathrm{U}}_1-{\mathrm{U}}_{\mathrm{x}})=\mathrm{x}(1-\mathrm{x})\left[∆\mathrm{E}-\mathrm{C}1+\mathrm{y}(\mathrm{a}\mathrm{m}\mathrm{C}1+\mathrm{b}\mathrm{F}+\mathrm{L})\right].$$

(4)

In the context of the replication dynamics equation F(x), we set $y^{*}={\displaystyle \frac{C1-∆E}{amC1+bF+L}}$, where two specific scenarios can be analyzed based on the value of y.

Scenario 1: y = y*

In this case, it can be known that F(x) ≡ 0, indicating that all levels of x (the proportion of construction enterprises adopting intelligent construction) are stable.

Scenario 2: y ≠ y*

Here, we set F(x) = 0; this results in two stable points (x = 0 and x = 1).

The first derivative, F′(x), is crucial for analyzing stability; it can be formulated as follows:

$${\mathrm{F}}^{\prime }(\mathrm{x})={\displaystyle \frac{\partial F(x)}{\partial x}}=(1-2\mathrm{x})\left[∆\mathrm{E}-\mathrm{C}1+\mathrm{y}(\mathrm{a}\mathrm{m}\mathrm{C}1+\mathrm{b}\mathrm{F}+\mathrm{L})\right].$$

(5)

According to the principles of stability in differential equations, when F(x) = 0 and F′(x) < 0, the strategy choice of construction enterprises is in a stable state. Figure 2 represents the evolutionary phase diagram for construction enterprises’ strategy choices.

When y > y*, we have F′(x)|x = 1 < 0, indicating that at x = 1, the strategy of implementing intelligent construction is stable, placing it in space S1. Conversely, when y < y*, we find F′(x)|x = 0 < 0, meaning that at x = 0, the strategy of using traditional construction methods is stable, placing it in space S2. The sizes of S1 and S2 are related to y*. If ∆E, L, amC1, and bF increase, then S1 expands while S2 shrinks. This indicates that higher incremental returns from intelligent construction, increased special subsidies, tax incentives, and stronger penalties for traditional construction failures will encourage enterprises to choose the transition to intelligent construction.

On the other hand, if C1 increases, then S1 decreases while S2 increases. This suggests that if the incremental costs of transitioning to intelligent construction rise, construction enterprises are more likely to opt for traditional construction methods.

2.

Evolutionary Stability Analysis for Government

Expected returns for construction companies choosing intelligent construction methods can be denoted as V₁. Expected returns from passive government promotion can be denoted as V₂. Average expected returns can be denoted as Vy. They can be formulated as follows:

$$V_1=x(G+∆G+S-rC2-amC1-bF)+(1-x)(G-rC2+S+L),$$

(6)

$$V_2=x(G+∆G)+(1-x)G,$$

(7)

$$V_y=y{V}_1+(1-y)V_2.$$

(8)

The replication dynamics equation for government is denoted as F(y). This can be formulated using the following general form:

$$\mathrm{F}(\mathrm{y})={\displaystyle \frac{dy}{dt}}=\mathrm{y}({\mathrm{V}}_1-{\mathrm{V}}_{\mathrm{y}})=\mathrm{y}(\mathrm{y}-1)\left[\mathrm{x}(\mathrm{a}\mathrm{m}\mathrm{C}1+\mathrm{b}\mathrm{F}+\mathrm{L})+\mathrm{r}\mathrm{C}2-\mathrm{S}-\mathrm{L}\right].$$

(9)

In the context of the replication dynamics equation F(y), we set $x^{*}={\displaystyle \frac{L+S-rC2}{amC1+bF+L}}$, where two specific scenarios are analyzed based on the value of x.

Scenario 1: x = x*

In this case, it can be known that F(y) ≡ 0, indicating that all levels of y (the probability of government support for the development of intelligent construction) are stable.

Scenario 2: x ≠ x*

Here, we set F(y) = 0; this results in two stable points (y = 0 and y = 1).

The first derivative, F′(y), is crucial for analyzing stability; it can be formulated as follows:

$${\mathrm{F}}^{\prime }(\mathrm{y})={\displaystyle \frac{\partial F(y)}{\partial y}}=(2\mathrm{y}-1)\left[\mathrm{x}(\mathrm{a}\mathrm{m}\mathrm{C}1+\mathrm{b}\mathrm{F}+\mathrm{L})+\mathrm{r}\mathrm{C}2-\mathrm{S}-\mathrm{L}\right].$$

(10)

According to the principles of stability in differential equations, when F(y) = 0 and F′(y) < 0, the strategy choice of government is in a stable state. Figure 3 represents the evolutionary phase diagram for the government’s strategy choices.

When x > x*, we have F′(y)|_y=0 < 0, indicating that at y = 0, the strategy of the government passively promoting intelligent construction is stable, placing it in space S3. Conversely, when x < x*, we find F′(y)|_y=1 < 0, meaning that at y = 1, the strategy of the government actively promoting intelligent construction is stable, placing it in space S4. The sizes of S3 and S4 are related to x*. If amC1, bF, and rC2 decrease, then S4 increases while S3 shrinks. This indicates that reductions in special fund subsidies, tax incentives, and service support costs ease the burden on the government, thereby increasing the probability of government support. Conversely, if these factors do not improve, it becomes less favorable for the government to adopt an active promotion strategy.

Additionally, if S increases, then S4 expands while S3 contracts, suggesting that a greater positive social impact from the government’s active promotion of the intelligent construction transition will encourage further government support for such initiatives.

3.2.2. Government–Construction Enterprise System Evolutionary Stability Analysis

Based on the equations established in Formulas (4) and (9), we can analyze the evolutionary process of construction enterprises and the government. By setting F(x) = F(y) = 0, we can identify five local equilibrium points in the evolutionary game dynamics, which are as follows: (0, 0), (0, 1), (1, 0), (1, 1), (x*, y*), where $x^{*}={\displaystyle \frac{L+S-rC2}{amC1+bF+L}},y^{*}={\displaystyle \frac{C1-∆E}{amC1+bF+L}}$.

To assess the stability of the equilibrium points, we can use the Jacobian matrix [71]. The specific expressions for the determinant and trace at each equilibrium point are provided in

Table 3. Determinant and trace of the Jacobian.

Equilibriums	Det(J)	Tr(J)
(0,0)	−(∆E − C1)(rC2 − S − L)	(∆E − C1) − (rC2 − S − L)
(0,1)	(∆E − C1 + amC1 + bF + L)(rC2 − S − L)	(∆E − C1) + (amC1 + bF + L) + (rC2 − S − L)
(1,0)	(∆E − C1)[(amC1 + bF + L) + (rC2 − S − L)]	−(∆E − C1) − [(amC1 + bF + L) + (rC2 − S − L)]
(1,1)	−[(∆E − C1) + (amC1 + bF + L)][(amC1 + bF + L) + (rC2 − S − L)]	−[(∆E − C1) + (amC1 + bF + L)] + [(amC1 + bF + L) + (rC2 − S − L)]
(x*,y*)	M	N

. These expressions help in determining the stability characteristics of each point, allowing for a comprehensive understanding of the dynamics between the government and construction enterprises in the context of promoting intelligent construction practices.

To analyze the equilibrium points for the evolutionary game concerning the implementation of intelligent construction, we can apply the Friedman matrix method. This involves examining the Jacobian matrix J at each equilibrium point and determining, as shown in

Table 4. Stability analysis results for the evolutionary game between construction enterprises and the government.

Equilibriums	Eigenvalues λ		Results	Conditions of Stable
	λ1	λ2
(0,0)	∆E − C1	L − S − rC2	ESS	rC2 − S − L > 0
(0,1)	∆E − C1 + amC1 + bF + L	rC2 − S − L	ESS	∆E + amC1 + bF + L < C1; rC2 − S − L < 0
(1,0)	C1 − ∆E	−(amC1 + bF + rC2 − S)	saddle point	-
(1,1)	−(∆E − C1 + amC1 + bF + L)	amC1 + bF + rC2 − S	ESS	∆E + amC1 + bF + L > C1; amC1 + bF + rC2 < S

, if the conditions for local stability are met, Det(J) > 0, Tr(J) < 0.

This analysis categorizes the stability of the government–construction enterprise system into three scenarios based on the equilibrium points (0,0), (0,1), and (1,1), considering several key variables.

Scenario 1: (0, 0), stable when rC2 − S − L > 0

The government incurs service support costs that exceed the social benefits obtained from addressing issues caused by traditional construction methods. As a result, construction enterprises lack the incentive to adopt intelligent construction methods, and the government has insufficient motivation to support the transition.

Scenario 2: (0, 1), stable when ∆E + amC1 + bF + L < C1 and rC2 − S − L < 0

Here, the returns available to construction enterprises are less than the incremental costs associated with adopting intelligent construction. Consequently, there is a reduced enthusiasm for transitioning to intelligent construction, leading enterprises to prefer traditional methods.

Meanwhile, if the government finds that the service support costs are lower than the additional social benefits and fines, it may still choose to actively promote intelligent construction despite the financial implications of other supportive policies.

Scenario 3: (1, 1), stable when ∆E + amC1 + bF + L > C1 and amC1 + bF + rC2 < S

In this scenario, construction enterprises realize that the returns from adopting intelligent construction methods exceed the incremental costs, and the government’s costs related to taxes, special funds, and service support are lower than the social benefits. Under these favorable conditions, construction enterprises are likely to transition to intelligent construction, supported by active government promotion.

These situations highlight the significant impact of government policy support on the strategic choices of both construction enterprises and the government. The willingness of construction firms to transition largely depends on policy incentives, while high government service costs can diminish their willingness to support such transitions. The government should carefully design its policy support mechanisms and funding levels to ensure that, even under fiscal constraints, it can effectively incentivize construction enterprises to adopt intelligent construction methods.

4. Results

4.1. Numerical Simulation Based on System Dynamics

4.1.1. Construction of the System Dynamics Model

The evolutionary game process is dynamically evolving. System dynamics can effectively simulate the strategic choices and behavioral changes in game participants at different time stages and illustrate how the game outcomes change dynamically as participants adjust their strategies and environmental factors vary. Moreover, it can incorporate the influence of multiple strategies and complex external factors into the game process. At different development stages of intelligent construction, by setting different initial conditions and parameters, various possible evolutionary scenarios can be simulated, and the outcomes of the evolutionary game under different environments and conditions can be predicted. This model can be utilized to assist decision-makers in evaluating the effects of different policies and formulating corresponding strategies based on the development stages. Vensim is widely applied in research related to system dynamics. The simulation model, shown in Figure 4, is built in Vensim using a stock-and-flow diagram. It includes two stocks (representing the probability of construction enterprises adopting intelligent construction, x, and the probability of government support, y) and two flows (representing the rate of change in adoption, F(x), and the rate of change in government support, F(y)). Twelve external variables, derived from the payoff matrix of the game, are also included. The flow equations and variables in the system dynamics model are set according to Equations (1)–(9) from the game theory model. The simulation will use these components to model the dynamic interactions and evolution of strategies over time under different policy scenarios.

In the evolutionary game, by examining the benefit functions of different players under diverse strategies, one can discern the relationship between auxiliary variables and external variables. The advantage of this simulation model does not lie in its high degree of realism. Instead, its strength lies in its capacity to illustrate the internal laws governing the changes. Consequently, it does not require extremely accurate results [72]. In fact, the correctness of the structural design holds more significance than the precision of the parameter settings [73]. The model involves numerous parameters; relevant data in the model will be calculated based on statistical analyses of industry development, recently published policies from various cities, and the research findings from Xiamen’s first batch of intelligent construction pilot projects in 2024.

The average construction investment in Xiamen is 3500 CNY/m². Considering the average profit margin of listed construction companies at 5%, the profit from traditional construction methods is 180 CNY/m². According to the research on 10 pilot intelligent construction projects, the incremental cost of the projects is approximately 30 CNY/m². Currently, the pilot projects mainly focus on high-difficulty and large-scale projects. Cities like Zhengzhou, Shenzhen, and Hefei have formulated incentive policies for intelligent construction, providing subsidies of up to 2 million CNY, equivalent to 10 CNY/m². According to government disclosures, the service expenditures for platform development and personnel training are approximately 20 CNY/m².

The initial values for the parameters of the game system are shown in

Table 5. Initial parameter values.

Parameter Name	E	∆E	C1	L	G	∆G	S	m	F	C2
Initial Value	180	15	30	10	300	15	10	0.3	10	20

. The simulation will start at time 0 and end at time 2, with a step size of 0.0078125 years. The initial probabilities for the strategic choices of construction enterprises and the government are set at x = 0.5 and y = 0.5, respectively.

4.1.2. Analysis of Government Policy Support on System Evolution

(A)

Impact of Single Policy Support on Government–Enterprise Strategy Evolution

By adjusting the coefficients for tax incentives, special fund subsidies, and service provision, we explore how different levels of policy support influence the strategy evolution of construction enterprises and the government, as illustrated in Figure 5 and Figure 6.

From Figure 5, it can be observed that under varying levels of policy support, the strategy choices of construction enterprises gradually converge towards x = 0. This trend is primarily due to the fact that intelligent construction is still in its early development stages; enterprises have limited understanding, and the incremental costs exceed the perceived benefits, meaning that the potential advantages have not been fully realized. Furthermore, the effects of different policies and their respective support levels on the evolution of strategy choices vary to different extents.

In Figure 6, the results show that when the government implements three different levels of policy support, all scenarios stabilize at y = 1. This indicates that when only a single policy is in effect, the government is willing to adopt an active promotion strategy.

In the scenario where only tax incentives and service provision are available, construction enterprises tend to lean towards traditional construction methods. Under the same level of policy support, tax incentives have a stronger effect on promoting the transition of construction enterprises compared to service provision. This is mainly because tax incentives are more universally applicable, directly alleviating the tax burden related to the costs of equipment and technology needed for intelligent construction. On the other hand, service support primarily focuses on areas such as talent training and resource sharing. While these can gradually improve the environment for transition, creating an ecosystem takes time and does not promptly address the funding challenges faced by enterprises. When the government implements special fund subsidy projects, an increase in support intensity leads to a gradual shift in construction enterprises’ strategies from traditional methods to intelligent construction. Low-intensity fund subsidies are insufficient to motivate enterprises to transition; only high-intensity special fund subsidies can stabilize enterprises at x = 1.

(B)

Impact of Combined Policies on Government–Enterprise Strategy Evolution

(1)

Policy Combination of Tax Incentives and Special Funds

When the government simultaneously implements tax incentives and special fund policies, the impact on the strategy evolution of construction enterprises and the government is illustrated in Figure 7 and Figure 8.

From Figure 7, it can be seen that when the tax incentive coefficient is low (e.g., a = 0.05), only high-intensity special fund support enables construction enterprises to choose to transition to intelligent construction. As the tax incentive coefficient increases to 0.15, medium- to high-intensity special fund support can facilitate the transition of construction enterprises.

From Figure 8, it is evident that under different levels of support, the evolution results of the combined policies of tax incentives and special fund subsidies stabilize at y = 1.

• (2)

  Policy Combination of Tax Incentives and Service Provision

When the government simultaneously implements tax incentives and service support policies, the impact on the strategy evolution of construction enterprises and the government is illustrated in Figure 9 and Figure 10.

From Figure 9, it can be observed that the evolution results of the combination of tax incentives and service support stabilize at x = 0, indicating that construction enterprises tend to choose traditional construction methods. This is primarily because tax incentives typically only reduce costs related to the procurement of equipment for intelligent construction, covering about 30% of incremental costs, which has a limited promotional effect. As the tax incentive coefficient increases, the convergence speed of the strategy evolution for construction enterprises does not change significantly, suggesting that the benefits of service provision do not increase with higher tax incentive levels.

Additionally, the government’s strategy evolution remains stable at y = 1 under any intensity of the policy combination, with a noticeably slower convergence rate in cases of high-intensity service provision. This indicates that the combination of tax incentives and service provision does not provide effective incentives, and varying intensities of these policies have minimal impact on strategy choices.

• (3)

  Policy Combination of Special Funds and Service Provision

When the government simultaneously implements special fund subsidies and service support policies, the impact on the strategy evolution of construction enterprises and the government is illustrated in Figure 11 and Figure 12.

From Figure 11, it can be observed that when the intensity of special fund subsidies is low (e.g., b = 0.3), construction enterprises’ strategy evolution gradually approaches x = 1, indicating a preference for traditional construction methods, regardless of the level of service support. However, as the intensity of special fund subsidies increases to 0.8, various levels of service support can effectively encourage construction enterprises to transition to intelligent construction.

From Figure 12, it is evident that when the government invests high-intensity special fund subsidies and service support costs, it tends to adopt a more passive promotion strategy. This shift can further influence the strategies chosen by enterprises.

4.2. Key Driving Strategies Based on Sensitivity Analysis

In the current market environment, under medium to low levels of policy support, construction enterprises tend to choose traditional construction methods, while the government leans towards active promotion. It is particularly important to explore the key factors influencing the strategy choices of construction enterprises.

Using the transition from the strategy (traditional construction, active promotion) to (intelligent construction, active promotion) as an example, we will discuss the sensitivity of the proportion x (representing the adoption of intelligent construction methods) to external variables. Given that some external variables are difficult to change in the short term under current market conditions, we will primarily analyze the impact of incremental costs of intelligent construction C1, additional benefits, and government penalties L under strong incentive conditions.

In the context of unchanged original model data, we will set the following values under strong incentive conditions: a = 0.1 (considered as a high-tech enterprise); b = 0.3 (30% project incentive); r = 0.5 (government contribution at half the budget).

We will examine the impact of a 20% change in each individual variable on the proportion x of construction enterprises transitioning to intelligent construction. The parameter values are summarized in

Table 6. Parameter values for sensitivity analysis.

Parameter Name	Value
	Control Group	Experimental Group 1	Experimental Group 2	Experimental Group 3
C1	30	24	30	30
$∆\mathrm{E}$	15	15	18	15
L	10	10	10	12

below.

As shown in Figure 13, simulations reveal that when all variables change by 20%, the effectiveness of promoting the adoption of intelligent construction methods by construction enterprises follows the order: Experiment Group 1 > Experiment Group 2 > Experiment Group 3. This indicates that reducing the incremental costs of transitioning to intelligent construction, enhancing transition benefits, and increasing penalties for traditional construction methods can all encourage construction enterprises to adopt intelligent construction practices.

Among these, reducing the incremental costs for enterprises is the most effective. Enhancing transition benefits—such as improving enterprise credibility, increasing the probability of winning bids, and obtaining quality project awards—has a secondary positive effect. The impact of penalties is relatively moderate. However, as the incremental costs decrease, the government may begin to view the transition to intelligent construction as the industry norm, which could lead to a reduction in the intensity of its active promotion efforts.

5. Discussion

5.1. The Influence of the Mode and Intensity of Policy Support on the Strategic Choices of Construction Enterprises and the Government

As can be seen from Section 4.1.2 (A), three types of single policy support can, to a certain extent, encourage enterprises to explore and attempt intelligent construction. Under the same support intensity, the promoting effects of the policies of service provision, tax and fee concessions, and special fund rewards and subsidies increase successively. However, in the current situation of high-cost investment and lagging returns, it is difficult to give full play to their roles. Only the support of a high-intensity special fund (subsidizing at least 8 CNY/m² for a single project) can promote the transformation and upgrading of construction enterprises. In practice, in order to achieve the effectiveness of strategies, few strategy implementers adopt a single strategic tool. The combined use of multiple strategic tools is considered a more effective implementation strategy. This view can also be verified from Section 4.1.2 (B). The incentive effect of combined policies on construction enterprises is stronger than that of a single policy. Among the three combined policies, the policy combination of tax and fee concessions and special funds can most effectively promote construction enterprises to implement intelligent construction, mainly by reducing the input cost through tax and fee concessions and injecting special funds to replenish the cash flow. On the other hand, when the support intensity of both policies is low, they cannot play an incentive role. By comparing the three combined policies, it can also be found that only when the government replenishes the cash flow of construction enterprises is there a possibility of increasing the enterprises’ willingness to transform. Establishing assistance mechanisms, providing training and guidance, etc., are likely to only become a way for construction enterprises to free-ride, but they will be deterred from actual application. As the government invests more, it will lead to the waste of government finances. There are some differences between the conclusions in this part and the analysis of the incentive policies to promote enterprise digital transformation in Wang’s paper [50]. The main reason is that the digital transformation of small and medium-sized enterprises mainly focuses on internal management, and the main challenge is the technological barrier. However, intelligent construction involves the integration of multiple technologies, the procurement of materials and equipment, and the collaborative management of multiple parties. Before achieving the scale effect, the incremental cost is the main obstacle.

It is recommended that the government give priority to applying intelligent construction-related technologies to government projects and demonstration enterprises to cultivate pilot projects and enterprises and provide hierarchical subsidies for key application projects. Guide demonstration enterprises to give priority to participating in the evaluation of high-tech enterprises and enjoy honorary titles and high-tech tax reduction and exemption policies. In addition, key cities must become the priority areas for the development of intelligent construction site technologies and determine the number of pilot projects. At the same time, the application of intelligent construction should be increased to improve the quality of project construction and the construction level so as to achieve the transformation, upgrading, and sustainable development of the construction industry.

5.2. Sensitivity Analysis of Key Strategies

As can be seen in Section 4.2, the promotion of intelligent construction requires the joint realization of technology, market, motivation, and guarantee, which is similar to the view put forward by Feng [19]. Intelligent construction technologies involve the integration of emerging technologies such as the Internet of Things, artificial intelligence, cloud computing, and 5G. In the initial stage of development, the high cost needs to be reduced through technological innovation, which is the primary factor affecting promotion. In the initial stage of technology promotion, there is no need to pursue all-round application. Instead, it should start from the application points with a high cost-performance ratio and gradually expand the application scope from point to area. From the perspective of the construction market, key cities must become the priority areas for the development of intelligent construction technologies, and the number of pilot projects should be determined. As the number of application projects increases, a scale effect can also be formed, gradually reducing the incremental cost. The sustainable development of intelligent construction also requires a source of motivation. The government provides external motivation in terms of policies, training, and guidance, while enterprises provide internal motivation in terms of technology, talent, management, and operation. In terms of guarantee, the government needs to provide fiscal and tax policy support. After that, laws and regulations should be improved, and corresponding technical standards and management standards should be issued. According to the results of numerical simulation, supervision and protection are indispensable, but they are not the main factors affecting promotion in the initial stage of development. As the transformation and upgrading progress advances, the supervision of the implementation of policies and regulations can be strengthened, a supervision platform can be established, and appropriate penalties can be imposed.

5.3. Limitation and Future Work

This paper only considers the game between construction enterprises and the government. In fact, there are numerous stakeholders in the process of promoting intelligent construction, such as technology service providers, supervision units, etc. The hierarchical relationship among the government, enterprises, and projects also needs to be taken into account, and a three-party game or four-party game evolutionary model should be constructed to make the research results more instructive. In this paper, only single policies and combinations of two policies are considered for policy combinations. In the future, it is advisable to consider the evolutionary impacts of more policy combinations and different policy intensities on the strategic choices of multiple parties according to the changes in the development stages. The focus of this paper is on the impact of different policy combinations on the transformation and upgrading of intelligent construction. The sensitivity analysis only considers three external factors, but there are actually many influencing factors, all of which may have an impact on the strategic choices of stakeholders and can be further improved in future research.

6. Conclusions

This paper employs evolutionary game theory to analyze the dynamic evolution of strategy choices between construction enterprises and the government, using system dynamics to simulate the game model and explore the impact of three types of policy support and key factors on the strategic choices of the players. The study reveals the following findings:

(1)

The willingness of construction enterprises to transition relies heavily on policy incentives. Three types of single policy supports can moderately encourage enterprises to explore intelligent construction, but only high-intensity special fund support can promote the transformation of construction enterprises, at least more than 8 CNY/m² special subsidy. Among policies of equal support intensity, the promoting effects of service provision, tax incentives, and special fund subsidies increase sequentially. The combination of tax incentives and special funds proves to be the most effective, followed by the combination of special funds and service provision. The tax incentive policy can be implemented according to no more than the preferential treatment for high-tech enterprises. However, when the support intensity of these combinations is low, they do not exert significant incentivizing effects, and the combination of tax incentives and service provision fails to provide effective motivation regardless of intensity.

(2)

The government’s choice to actively promote depends on the balance between benefits and costs. Only when the government derives additional social benefits and the costs of penalties exceed the support costs, or when reputational gains from policies surpass subsidy costs, will it lean towards active support strategies. The government expresses a willingness to promote actively under a single policy support or combinations of tax incentives and special funds or tax incentives and service provision. Under medium to high levels of special fund support, medium to low levels of service support are more beneficial for reaching a stable state of intelligent construction implementation and active promotion, avoiding excessive financial burdens on the government. It is recommended that the special fund be no less than 5 CNY/m², and the cost of service provision be reduced to below 10 CNY/m².

(3)

In the context of government support, lowering incremental costs of transitioning to intelligent construction, enhancing transition benefits, and increasing penalties for traditional construction methods can all promote the adoption of intelligent construction practices. Among these, reducing incremental costs is the most effective, followed by enhancing transition benefits, while the impact of penalties is moderate.

To more effectively stimulate construction enterprises’ transition to intelligent construction and support the robust development of the construction market in China, the following measures and recommendations are proposed:

(1)

Strengthen supporting policies for intelligent construction: The government should expedite institutional development and improve the policy support system by reasonably setting combinations of policy support to address the issue of high incremental costs associated with intelligent construction. Compared to single policy support, the combination of tax incentives and special funds shows more pronounced effects. It is essential to adhere to the principle of moderation in policy support to avoid excessive financial burdens, which could reduce the government’s willingness to provide support. The government could adjust subsidy levels based on transformation progress to ensure effective investment and output ratios. Additionally, maximizing the role of policy guidance and government services is crucial by strengthening resource coordination, optimizing the technological incubation environment, and promoting the construction of new infrastructures such as 5G, IoT, AI, and big data, laying a foundation for the information technology-driven transition to intelligent construction.

(2)

Empower enterprises to reduce incremental costs: The promotion of intelligent construction is fundamentally a market economic activity and cannot rely solely on policy support. In the early stages, when the development level is low and market capabilities are limited, policy stimulation can be used for regulation. In the later stages, it is necessary to strengthen the primary role of enterprises. Enterprises should be guided to enhance their technological innovation capabilities and strengthen industry-university-research collaboration to shorten the research and application cycles of new materials and technologies. Actively promoting training for workforce skills will help accelerate the adoption of new technologies and minimize incremental costs associated with intelligent construction.