Evaluating Stakeholders’ Decisions in a Blockchain-Based Recycling Construction Waste Project: A Hybrid Evolutionary Game and System Dynamics Approach

Abstract

To promote efficient construction waste recycling and reuse, a novel waste management approach based on blockchain technology was introduced to the industry. However, adopting blockchain platforms in construction waste recycling and reuse may impact the behavioral strategies of stakeholders and impede the prediction of the specific impacts of stakeholders’ decisions. Accordingly, this study addresses two primary questions: (1) What are the collaborative framework and the behavioral evolution trends of multiple stakeholders within the context of blockchain? (2) How can the behavioral strategies of multiple stakeholders be systematically coordinated to achieve efficient construction waste recycling and reuse driven by blockchain? To answer these questions, a tripartite game model combined with system dynamics was constructed. In this model, we aimed to elucidate the internal organizational framework, analyze the dynamic evolution process, and assess the influence of decisions made by multiple stakeholders at the individual level. It also offers corresponding policy recommendations for efficient construction waste recycling and reuse driven by blockchain at the system level. This study offers three innovations. First, it considers the decision-making of multiple stakeholders as an interdependent and coevolutionary process to overcome the defects of analyzing only one type of participant. Second, in contrast to the static analysis method, it employs a dynamic system approach to deeply analyze the evolving structures of blockchain-based projects. Third, it provides a theoretical framework for the practical implementation of blockchain-driven platforms in managing construction waste recycling and reuse, thus fostering effective policy development and management practices. This framework aims to promote sustainable development in construction waste recycling and reuse projects in China as well as globally.

1. Introduction

The construction industry has experienced rapid growth due to accelerated urbanization. However, this expansion has come at a cost, with the construction industry becoming one of the largest energy consumers and carbon emitters [1]. The processes involved in constructing new buildings and renovating, expanding, and demolishing structures generate substantial construction waste. This exacerbates issues related to resource depletion, environmental pollution, and urban degradation [2]. Recycling and reusing construction waste are considered the most effective strategies to address these challenges, offering benefits such as resource conservation, pollution reduction, and economic stimuli [3]. In fact, researchers argue that 100% of construction waste generated from new construction projects can be reused or recycled [4].

Despite the continuous refinement of policies promoting construction waste recycling and reuse (CWRR), the actual recycling rate remains relatively low in practical implementation. This is fundamentally attributed to information asymmetry, inadequate supervision, and imperfect incentive mechanisms within the CWRR process [5]. The current management methods primarily include administrative supervision and centralized information platforms for managing CWRR. Administrative supervision is predominantly government-led; however, other factors often influence its effectiveness, including regional differences, legislation, law enforcement, construction and demolition practices, and recycling infrastructure. These factors inevitably lead to challenges related to inadequate supervision and information asymmetry [6]. Centralized information platforms designed to manage CWRR projects heavily rely on unilateral data uploads, posing difficulties in ensuring data authenticity, reliability, and real-time accuracy [7]. Typically, in situations where third-party platforms are involved, government oversight and intervention become difficult, potentially resulting in a lack of control and vulnerability to fraud and manipulation [8]. To address these issues, innovative management mechanisms and strategies for CWRR projects are of utmost importance.

Scholars have introduced a novel waste management approach based on blockchain technology, which has already found application in various waste management domains, including industrial [9], urban [10], electronic [8], and medical wastes [7]. Blockchain technology ensures data security and reliability while enabling information sharing and providing robust support for government oversight [11,12]. Furthermore, blockchain-based incentive mechanisms have been proven as effective in facilitating waste transfer and recycling [13]. Indeed, promoting efficient CWRR project management through blockchain technology is of significant importance, as it reduces environmental pollution and yields substantial economic and societal benefits [14,15].

CWRR projects face different challenges, including a low resource utilization rate, slow growth of the recycling industry chain, and limited coordination among stakeholders [16]. These challenges underscore the essential need for government policy support and enhanced collaboration among stakeholders in CWRR projects, particularly in adopting blockchain technology [5]. The adoption of blockchain platforms impacts the behavioral strategies of stakeholders and collaborative effects within CWRR projects. Moreover, it complicates the prediction of specific impacts of related factors and consequently hinders the determination of the value offered by implementing blockchain as a solution [7]. The complexity of CWRR projects involves a system comprising multiple stakeholders and processes [17,18], and within this system, decision dynamics influence each other [16]. For example, construction enterprises (CEs) in CWRR projects may adjust their recycling strategies based on the influence of recycling enterprises (REs). Meanwhile, government policies may influence REs to adjust their strategies, encouraging participation in blockchain platforms to enhance profitability.

However, past research has primarily focused on waste processing efficiency and the recycling behavior of individual stakeholders in CWRR. Meanwhile, the collaboration between the other two key stakeholders in CWRR, namely CEs and REs, as well as the impact of government supervision actions on their collaborative efforts, has often been overlooked. The decision-making processes of the multiple stakeholders in the blockchain-based CWRR projects are interdependent and intertwined. In this sense, collaboration among multiple stakeholders is crucial for improving the efficiency of CWRR projects [18]. In this study, we aimed to address the following research questions: (1) What are the collaborative framework and the behavioral evolution trends of multiple stakeholders in CWRR projects within the context of blockchain? (2) How can the behavioral strategies of multiple stakeholders be systematically coordinated to achieve efficient CWRR driven by blockchain?

Existing studies have indicated that an evolutionary game analysis, as a solution for addressing dynamic and complex system problems involving multiple participants, can be employed to explore changes in participants’ behavioral strategies over time and their sensitivity to specific influencing factors [12]. It can provide insights into how these strategies evolve. Game theory is good at analyzing the interactions amongst multiple players, using a mathematical framework to foresee the strategic behaviors of players in terms of their expected benefits [19]. In contrast to classical game theory, the evolutionary game theory assumes that players have bounded rationality and incomplete information on the environment [20,21]. In fact, the government, CEs, and REs are neither entirely rational nor completely informed about their collaborative intentions and environmental conditions [18,22]. Therefore, evolutionary game theory offers a mechanism to analyze the behavior of such a system. Moreover, it is easy to compute and can delineate optimal and stable behavioral strategies [23]. Unlike the static equilibrium state in the multi-objective optimization, which aims to find a Pareto front or a similar solution and maintains this equilibrium without change [24], the equilibrium state of evolutionary game theory requires a dynamic adjustment process that evolves through multiple game interactions among players [18]. Therefore, evolutionary game theory offers advantages over other control modeling methods in analyzing the dynamics of strategy change. In actual situations, as the dynamic system evolves over time, the government, CEs, and REs compare payoffs and adjust their strategies in response to their strategic choices. Evolutionary game theory provides a basis for exploring the dynamic iterative situations and fits well with the dynamic nature of a CWRR complex system, facilitating collaborate behavior strategies for blockchain technology.

To provide a clearer and more intuitive understanding of the dynamic evolution process inherent in the CWRR projects within the context of a blockchain-based complex system, this study combined a multi-agent evolutionary game analysis and system dynamics (SD) to simulate the behavior and coevolution process of various stakeholders.

Therefore, the main objectives of this study were to (1) figure out the inside organizational framework in CWRR projects driven by blockchain platforms and analyze the dynamic evolution process of decisions made by governments, CEs, and REs using a tripartite evolutionary game model; (2) deeply analyze the influence of strategic behavior of multiple stakeholders at the individual level and offer corresponding policy recommendations for efficient CWRR driven by blockchain through a tripartite evolutionary game SD model.

The novelty of this study could be summarized as (1) proposing a tripartite game model for governments, CEs, and REs to analyze the effects of their collaborative strategy choices on the evolution of CWRR projects driven by blockchain. This study integrates the tripartite game model into an SD analysis system, which is conducive to the decision-making processes of the three parties in the blockchain-based CWRR projects as an interdependent and coevolutionary process. This model can overcome the defects of analyzing only one type of participant in the CWRR projects.

(2) From the individual- and system/group-level perspectives, there is clarifying the influence of stakeholders’ strategy choices at the individual level on the evolution of the whole blockchain-based CWRR projects at the group level. In contrast to static analysis methods adopted in previous studies, the tripartite evolutionary game SD simulation system focuses more on the internal dynamic and evolutionary structures of the blockchain-based CWRR projects.

(3) There is providing a theoretical framework to implement practical actions for adopting blockchain-driven platforms in CWRR and achieving efficient CWRR project management. Through exploratory scenario simulation research of the dynamic effects of various policies and measures on the evolution of blockchain-based CWRR projects, this study offers relevant policy recommendations for effective waste management policy design. This is conducive to exploring effective policies and management methods to promote the overall development of CWRR projects in China and other countries and regions.

The structure of this study is as follows: Section 2 is the literature review. Section 3 constructs a tripartite evolutionary game model. Section 4 builds an SD simulation model. Section 5 discusses the findings based on simulation results and provides policy recommendations. Lastly, Section 6 summarizes the content of this study and highlights its contributions and limitations.

2. Literature Review

2.1. Resource Management of Construction Waste

Construction waste refers to the solid waste generated from construction, reconstruction, and demolition activities; this term is often interchanged with construction and demolition waste [25]. Across the globe, countries and regions have increasingly shifted their focus toward the sustainable management of construction waste and the promotion of recycling [26]. In their study on tools for CWRR, Wang et al. (2020) [27] observed that computer vision performs effectively in CWRR but may lack precision in certain situations. Furthermore, they developed robots based on field construction, demolition waste classification, and recycling using vision-based technology. Lai et al. (2016) [28] conducted a study on the online system of the Taiwan Environmental Protection Agency and determined that it could track waste flow and monitor the quantity generated. Additionally, previous research has emphasized that obtaining information on construction waste effectively addresses the challenges encountered in the CWRR process.

However, while these studies have contributed to the theoretical understanding of construction waste, they have not proposed actionable measures to effectively promote CWRR. Therefore, there is a pressing need to introduce an innovative CWRR management model into the CWRR projects [5,29].

2.2. Waste Management Based on Blockchain Technology

Blockchain technology, also called distributed ledger technology, was initially applied in the context of the Bitcoin cryptocurrency [30]. As a decentralized infrastructure and distributed ledger agreement, blockchain is well suited for ensuring data security and establishing trust in automation and smart development [31]. IBM has identified five significant advantages of blockchain technology in transforming industries: enhanced transparency, heightened security, upgraded traceability, improved efficiency and speed, and lower costs [32]. These advantages make blockchain application in the construction environment particularly promising, potentially disrupting traditional construction industry models. They are pivotal in the digital and intelligent transformation of the construction sector [33].

Blockchain technology can improve the management of electronic, medical, transboundary, agricultural, industrial, and municipal solid wastes. This application has facilitated the secure storage and sharing of waste information, as well as the transparent tracing and supervision of waste streams [8,9,10]. In addition, numerous studies have devised incentives to encourage stakeholders’ participation. Refs. [8,34,35] utilized token incentives, while [7,36] recommended the use of reputation or credibility as incentives. The proposed system adopts Ethereum’s blockchain public network architecture, providing greater system security as it is protected from local risk situations, such as data loss, power outage, attempts to gain access, and other typical issues [8,35].

Considering that the digital transformation of construction waste, as suggested in previous studies, is an effective measure to promote recycling [5,29], this study proposes that the adoption of blockchain technology can effectively address the issues of information asymmetry, regulatory gaps, and imperfect incentive mechanisms in CWRR.

2.3. Behavioral Strategies of Stakeholders in Construction Waste

The recycling of construction waste is a multifaceted and intricate endeavor, involving various stakeholders and connections [17,37]. It necessitates collaborative efforts from relevant governmental departments, environmental protection entities, construction material enterprises, CEs, other affiliated sectors, and the general public. Additionally, it requires the formulation of pertinent government policies [18,38].

Scholars have extensively examined the cooperative mechanisms among stakeholders in the CWRR projects, focusing on the micro-level decision-making of these stakeholders. Guo et al. (2022) [39] built a four-party model through an evolutionary game analysis, involving the government, CEs, REs, and consumers. They analyzed behavioral strategies in the recycling and sales phases, providing valuable insights into recycling practices. Su (2020) [40] crafted a three-party evolutionary game model using Shanghai as a case study, concluding that the combinations of stable strategy from the government, CEs, and recycling stakeholders involve the supervision of the market, procurement of recycled products, and production of high-quality recycled products. The strategies of CEs are notably influenced by prices of products, reputational interests, and resource fees and taxes.

In summary, the intricate web of relationships among stakeholders in the CWRR projects poses challenges for the government to obtain real-time insights into the industry. Consequently, this hampers its ability to forecast and devise effective strategies for action [18]. The integration of blockchain technology into the CWRR process will inevitably introduce further complexity to the behaviors and decisions of stakeholders, making it even more challenging for governments to determine appropriate courses of action [41]. An evolutionary analysis of game theory, as a tool for addressing dynamic systems involving multiple participants, can facilitate the understanding of how participants’ strategies evolve over time [42,43]. Therefore, within the context of implementing blockchain technology for a CWRR platform, simulating stakeholder behaviors in the projects through an evolutionary game analysis can effectively facilitate cooperation among stakeholders and guide them in making effective decisions in the context of blockchain adoption. This approach offers a promising avenue for unraveling the intricacies of decision-making games and stability strategies among the various stakeholders during the CWRR process.

3. Research Methodology

This study adopted the evolutionary game method. An evolutionary game analysis is typically applied to tackle issues in dynamic systems comprising multiple actors whose strategies change dynamically over time [39,44]. CWRR is a complex ecological project; hence, introducing blockchain technology for management will inevitably have a corresponding impact on the construction industrial chain. To determine the internal organizational framework of the construction industrial chain, along with the competitive and cooperative relationships between various stakeholders, and to understand the coevolutionary pattern of the construction industrial chain, a scientific CWRR evolution game model needs to be developed.

Based on the practical requirements of blockchain-based CWRR, this section analyzes the evolutionary game process and evolutionary stabilization strategies for the choice of multiple stakeholders’ behaviors in the blockchain-based CWRR complex system. In addition, from a dynamic and system perspective, we constructed a tripartite evolutionary game model and an SD simulation system. The evolutionary game SD model for the blockchain-based CWRR projects was numerically simulated using Vensim PLE and Matlab software to explore the influence of strategy choices made by multiple stakeholders on the evolution of the blockchain-based CWRR projects. Subsequently, relevant policy recommendations from the evolutionary strategies of stakeholders at the individual level were offered based on the simulation results. Finally, based on the results, the specific impact of adopting blockchain technology on stakeholders’ behavior and strategy decisions in the CWRR projects is clarified. The research flow in Figure 1 illustrates the whole modeling and simulation process in this study.

3.1. Problem Description

3.1.1. Behavioral Strategies of Multiple Stakeholders

A blockchain-driven framework for reusing and recycling construction waste has been established in our previous studies [45]. This process involves three key stakeholders and six smart contracts, as shown in Figure 2. The blockchain-based framework for CWRR considers the government, CEs, and REs as nodes. These stakeholders participate in the corresponding transaction processes to maintain the operation of the blockchain-based CWRR projects. CEs and REs are the waste producers and recyclers, respectively, and to use this blockchain-driven platform, they must log into the system to register relevant information. The government, acting as the platform manager, will review and verify this information.

In an evolutionary game, players are assumed to be bound rationally, and then the dynamic evolution process of their strategies is subsequently explored [18,46]. Three main stakeholders are considered in the blockchain-based CWRR projects: the government, CEs, and CWRR enterprises.

(1)

Behavioral strategies of the government

The government is the master planner and policymaker in CWRR, promoting the active participation of enterprises in CWRR, facilitating the smooth progress of CWRR, and encouraging positive cycle management of resources [18]. The government participates in the management of CWRR; however, it cannot manage all of the relevant information on CWRR. Therefore, it can only improve the awareness of relevant enterprises on recycling and reuse of products through policy initiatives and carry out relatively light supervision through feedback mechanisms. Alternatively, the government promotes CWRR by firmly introducing blockchain technology and formulating strict supervising policies.

(2)

Behavioral strategies of CEs

CEs produce construction waste and play a significant role in its recycling. Their participation contributes to the CWRR rate and facilitates sustained development in the construction industry [5]. CEs choose whether to actively recycle by comprehensively considering costs, economic benefits, and recycling policy advantages. In some cases, CEs do not adopt CWRR, but passive recycling strategies, and are unwilling to recycle. This will lead to speculative behavior, including illegal dumping and mixed landfilling. However, in other cases, CEs may be willing to adopt scientific methods to address construction waste, take the initiative to recycle, and adopt active recycling strategies that can promote the recycling of resources and maximally protect the environment.

(3)

Behavioral strategies for CWRR enterprises

REs are the main stakeholders involved in the recycling of construction waste and the production of recycled products; thus, they have a vital role in CWRR [18]. REs choose their strategies based on government policies and profit balance analyses. They may decide to transfer CWRR information to the blockchain-based CWRR projects; that is, adopt participation strategies for the blockchain-based CWRR projects and release real-time, authentic, transparent, and shared recycling information. Alternatively, if REs lack sufficient revenue or production profits to update information technology and improve recycling levels, they may be forced to select non-participation strategies for blockchain-based CWRR projects.

From the above analysis, it can be concluded that the government has two strategic choices as a regulator: strict supervision of G₁ and light supervision of G₂. CEs’ behavioral strategies are divided into active recycling of CE₁ and passive recycling of CE₂. Meanwhile, REs have two strategies for the blockchain-based CWRR projects: the participation of RE₁ and non-participation of RE₂.

3.1.2. Tripartite Game Model Assumptions

Based on the above game relationship, the following assumptions are proposed according to possible actions taken by the three stakeholders in the blockchain-based CWRR projects.

Assumption 1. 

Actors are all rationally bounded; they can adapt to environmental changes and dynamically adjust strategies for recycling.

Assumption 2. 

The government supervises the CWRR process. When CWRR enterprises participate in CWRR, the government builds an information platform through strict supervision. Assuming that the probability of the government selecting a strict supervision strategy G1 is x, the probability of the government selecting a light supervision strategy is 1 − x, where 0 ≤ x ≤ 1.

Assumption 3. 

If the probability of CEs selecting an active recycling strategy CE₁ is y, the probability of choosing a passive recycling strategy CE₂ is 1 − y, where 0 ≤ y ≤ 1.

Assumption 4. 

When the REs select the strategy of participation in the recycling platform (i.e., strategy RE₁) and the probability of recycling construction waste is z, the probability of choosing the non-participate strategy (i.e., strategy RE₂) is 1 − z, where 0 ≤ z ≤ 1.

Assumption 5. 

If the REs opt for the RE₂ recycling platform strategy (not participation), the cost of REs is C₂. If the REs choose to participate in the RE₁ recycling platform strategy, the recycling cost is C₃ > C₂ > 0, which is because REs’ participation will incur additional technical costs, compared with non-participation [47]. The revenue from recycling services provided by REs is I₁. When REs provide real-time recycling information and services by participating in the recycling platform, CEs obtain positive utility [48], set as S₁. Assuming the government supervises an illegal treatment of construction waste by REs owing to the conflict of interest between CEs and REs [16], the REs will be sanctioned, set as L, after investigation; when REs choose to publish false recycling information to trick CEs into recycling for profit, CEs will have a negative effect [22,49], set as D₁; if CEs choose a passive recycling strategy, CE₂, their treatment cost for construction waste will be C₁.

Assumption 6. 

When the government selects the strict supervision strategy G1, it incurs supervision costs and may also choose to take punitive measures based on the different strategies of REs [16]. If the government selects the strict supervision strategy G1, the decision-making and publicity cost is set as D0. If REs publish false recycling information or carry out other illegal waste disposal behaviors, they will be fined [18,50], which is set to D2. Once this issue has been resolved, the CEs will receive extra compensation, R1; if the CE recycles through the blockchain-driven platform, it will receive the corresponding compensation [51,52], set as L.

Assumption 7. 

The cost incurred by the government in encouraging REs to build a blockchain-driven platform for CWRR is C₀. If the government implements the light supervision strategy G₂ and the illegal activities of REs are not stopped or sanctioned, the CE will gradually lose trust in the government [53,54]. This will propagate negative impressions, generating additional damaging effects to the government [54], which is set as S₀. In contrast, when the government regulatory authorities choose strict supervision strategy G₁ and take punitive measures against the illegal acts of REs, the government will gain positive benefits_, including improved trust [18,55], set as R₀.

Assumption 8. 

When REs engage in illegal activities, the actions will be recorded and rapidly disseminated, resulting in losses, set as F; these losses include a damaged reputation, decreased order volume, and reduced industry influence [18,56]. If the behaviors of REs are legal when they adopt the recycling platform strategy RE₁, they will receive positive benefits R₂, such as improved industry recognition and increased order volume [16,55].

Based on the method of an evolutionary game analysis and the above assumptions,

Table 1. Description of related parameters of the game model.

Parameter	Implication(s)
${\mathrm{I}}_0$	Benchmark government revenue from recycling
$D_0$	Decision-making and publicity costs of the government’s choice of strict supervision
$C_0$	Cost of building a blockchain-driven platform under the government’s strict supervision
$R_0$	Positive benefits obtained by the government when adopting strict supervision measures
$S_0$	Negative effects of government under light supervision
$L$	Penalties for illegal acts taken by REs or compensation obtained by CEs when they choose to actively recycle
${\mathrm{S}}_1$	Under the active recycling strategy, CEs earn extra income by selecting blockchain-driven platform recycling
$C_1$	Treatment cost of passive recycling chosen by CEs
$R_1$	When the government strictly supervises, the CEs will receive extra compensation for the non-participation of the recycling enterprise in illegal activities
$D_1$	Negative effects of CEs selecting passive recycling
$C_2$	Additional costs incurred by REs not participating
$C_3$	Recycling costs associated with RE participation
${\mathrm{I}}_1$	Revenue from recycling services provided by REs
$R_2$	REs participate in recycling, resulting in positive benefits
$D_2$	RE fines for publishing false information or illegally disposing of construction waste
$F$	Recycling the extra losses caused by illegal acts of enterprises
x	Probability of government selecting strict supervision strategy G1
y	Probability of CEs selecting active recycling strategy CE1
z	Probability of REs selecting to participate in recycling platform RE1

describes the relevant parameters and implications.

3.2. Model Construction

3.2.1. Income Payoff Matrix

Given the above model assumptions and the related model parameters, a tripartite game tree for the government, CEs, and REs was constructed (Figure 3).

We conclude that there are eight game strategy combinations, involving the government, CEs, and REs: strict supervision, active recycling, and participation (G₁, CE₁, RE₁); strict supervision, passive recycling, and participation (G₁, CE₂, RE₁); strict supervision, active recycling, and non-participation (G₁, CE₁, RE₂); strict supervision, passive recycling, and non-participation (G₁, CE₂, RE₂); light supervision, active recycling, and participation (G₂, CE₁, RE₁); light supervision, passive recycling, and participation (G₂, CE₂, RE₁); light supervision, active recycling, and non-participation (G₂, CE₁, RE₂); light supervision, passive recycling, and non-participation (G₂, CE₂, RE₂).

3.2.2. Replicated Dynamic Analysis

Based on the research by Liu et al. (2021) [13] and Du et al. (2020) [16], this study adopted the replication dynamic analysis method in evolutionary games. In this approach, participants observe the outcomes of various strategies. When they find that a particular strategy yields higher benefits—that is, the income of the selected strategy in the game process is higher than the average income of other strategies—they replicate this action and choose the more successful strategy. Replication dynamics uses a dynamic differential equation that depicts the probability of all strategy groups by selecting a specific strategy [15]. The general expression is as follows:

$${\displaystyle \frac{d_{x_i}}{d_t}}=x_i\left[\left(u_{x_i},x\right)-u(x,x)\right]$$

(1)

where $x_i$ is the probability of selecting strategy i, $\left(u_{x_i},x\right)$ is the expected benefit of selecting strategy i, and $u(x,x)$ is the average benefit of other strategies. According to Equation (1), the replication dynamic equations of the government, CEs, and REs in the blockchain-based CWRR projects under their respective strategies can be obtained as follows:

① Dynamic equation for government replication

The expected benefit of the government under strict supervision strategy G₁ is assumed to be $U_{G1}$ and the expected benefit under light supervision strategy G₂ is considered $U_{G2}$. The average benefit choosing strategy is taken as $\overline{U_G}$ and the replication dynamic equation for the government in the blockchain-based CWRR projects can be expressed as $F\left(x\right)={\displaystyle \frac{d_x}{d_t}}$.

$$\begin{array}{ll}{U}_{G1}& =yz\left(I_0+R_0-C_0-D_0\right)+\left(1-y\right)z\left({I_0+R_0-C_0-D}_0\right)+y\left(1-z\right)\left(I_0+D_2+R_0-L-C_0-D_0\right)\\ & \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+\left(1-y\right)\left(1-z\right)\left(I_0+D_2+R_0-C_0-D_0\right)\\ & =I_0+z\left(R_0-C_0-D_0\right)+\left(1-z\right)\left(D_2+R_0-C_0-D_0-yL\right)\end{array}$$

(2)

$$\begin{array}{ll}{U}_{G2}& =yz\left(I_0+R_0-C_0\right)+\left(1-y\right)z\left({I_0+R_0-C}_0\right)+y\left(1-z\right)\left(I_0-S_0-C_0\right)+\left(1-\mathrm{y}\right)\left(1-\mathrm{z}\right)\left(I_0-S_0-C_0\right)\\ & =I_0+z\left(R_0+S_0\right)-S_0-C_0\end{array}$$

(3)

$$\overline{U_G}=x{U}_{G1}+(1-x)U_{G2}$$

(4)

$$\begin{array}{ll}F\left(x\right)& ={\displaystyle \frac{d_x}{d_t}}=x\left(U_{G1}-\overline{U_G}\right)=x\left(1-x\right)\left(U_{G1}-U_{G2}\right)\\ & =x(1-x)\left[z\left(yL-S_0-D_2-R_0\right)+D_2+R_0-D_0-yL+S_0\right]\end{array}$$

(5)

② Dynamic equation for CE replication

The expected benefit of CEs under the active recycling strategy CE₁ in blockchain-based CWRR projects is assumed to be $U_{CE1}$, and the expected benefit of CEs under the passive recycling strategy is $U_{CE2}$. The strategy average benefit is assumed as $\overline{U_{CE}}$ and the replication dynamic equation for CEs in the system can be expressed as $F\left(y\right)={\displaystyle \frac{d_y}{d_t}}$.

$$\begin{gathered} U_{CE1}=xz{S}_1+x\left(1-z\right)(R_1-D_1)+\left(1-x\right)z{S}_1+\left(1-x\right)\left(1-z\right)\left(-D_1\right) \\ =z{S}_1+\left(1-z\right)(x{R}_1-D_1) \end{gathered}$$

(6)

$$\begin{array}{ll}{U}_{CE2}& =xz\left(-C_1-L\right)+x\left(1-z\right)\left(R_1+L-C_1-D_1\right)+\left(1-x\right)z\left(-C_1\right)+\left(1-x\right)\left(1-z\right)\left(-D_1-C_1\right)\\ & =-xz\left(R_1+2L\right)+x\left(R_1+L\right)-\left(1-z\right)D_1-C_1\end{array}$$

(7)

$$\overline{U_{CE}}=y{U}_{CE1}+(1-y)U_{CE2}$$

(8)

$$\begin{array}{ll}F\left(y\right)& ={\displaystyle \frac{d_y}{d_t}}=y\left(U_{CE1}-\overline{U_{CE}}\right)=y\left(1-y\right)\left(U_{CE1}-U_{CE2}\right)\\ & =y(1-y)\left[2xzL-x\left(R_1+L-D_1\right)+z{S}_1+C_1\right]\end{array}$$

(9)

③ Dynamic equation for RE replication

The expected benefit of REs participating in the blockchain-based CWRR projects is assumed to be $U_{RE1}$, and the expected benefit of not participating in the blockchain-based CWRR projects is taken as $U_{RE2}$. The strategy average benefit for REs is assumed to be $\overline{U_{RE}}$ and the replication dynamic equation for REs in the system can be expressed as $F\left(z\right)={\displaystyle \frac{d_x}{d_t}}$.

$$\begin{gathered} U_{RE1}=xy\left(I_1+R_2-C_3\right)+x\left(1-y\right)\left(I_1+R_2-C_3\right)+\left(1-x\right)y\left(I_1+R_2-C_3\right)+\left(1-x\right)\left(1-y\right)\left(I_1+R_2-C_3\right) \\ =I_1+R_2-C_3 \end{gathered}$$

(10)

$$\begin{gathered} U_{RE2}=xy\left(-C_2-D_2-F\right)+x\left(1-y\right)\left(-C_2-D_2-F\right) \\ +\left(1-x\right)y\left(-C_2-F\right)+\left(1-x\right)\left(1-y\right)\left(I_1-C_2-F\right) \\ =-C_2-F-x{D}_2+\left(1-x\right)\left(1-y\right)I_1 \end{gathered}$$

(11)

$$\overline{U_{RE}}=z{U}_{RE1}+(1-z)U_{RE2}$$

(12)

$$\begin{gathered} F\left(z\right)={\displaystyle \frac{d_x}{d_t}}=\mathrm{z}\left(U_{RE1}-\overline{U_{RE}}\right)=\mathrm{z}\left(1-\mathrm{z}\right)\left(U_{RE1}-U_{RE2}\right) \\ =z(1-z)\left[I_1+R_2-C_3+C_2+F+x{D}_2-\left(1-x\right)\left(1-y\right)I_1\right] \end{gathered}$$

(13)

3.2.3. Stability Analysis

To obtain the three-party stable strategy for the evolutionary game dynamic system, the replication dynamic equations of the government, CEs, and REs in the blockchain-based CWRR projects under their respective strategies were set to 0:

$$F\left(x\right)=0,F\left(y\right)=0,F\left(z\right)=0$$

(14)

Based on the stability theorem of differential equations and the evolutionarily stable strategy (ESS) properties, the stable state must show good robustness against small disturbances using an evolutionary stability strategy [57,58]. Specifically, when $x$ < $x^{*},$ it must be >0; when $x$ is > $x^{*},$ it must be <0 [58,59]. Therefore, to achieve the evolutionarily stable strategy ESS, the following parameters must be met: $F\left(x\right)=0$ and $F^{\prime }\left(x\right)<0$. The same is true for $F\left(y\right)$ and $F\left(z\right)$.

The derivatives of the replication dynamic equations for the government, CEs, and REs are as follows:

$$F^{\prime }\left(x\right)=(1-2x)\left[z\left(yL-S_0-D_2-R_0\right)+D_2+R_0-D_0-yL+S_0\right]$$

(15)

$$F^{\prime }\left(y\right)=(1-2y)\left[2xzL-x\left(R_1+L-D_1\right)+z{S}_1+C_1\right]$$

(16)

$$F^{\prime }\left(z\right)=(1-2z)\left[I_1+R_2-C_3+C_2+F+x{D}_2-\left(1-x\right)\left(1-y\right)I_1\right]$$

(17)

① Stability analysis of the government’s behavior strategy

Based on the above theory, an incremental stability analysis of the government’s behavior strategy was performed, and the following scenario hypotheses were proposed:

Proposition 1. 

(1) According to Equation (5), when $y=y^{*}={\displaystyle \frac{D_0+(z-1)\left(S_0+D_2+R_0\right)}{(1-z)L}}, F\left(x\right)=0$, indicating a stable boundary and that the government’s strategy will not change. (2) According to Equation (5), when $y\ne y^{*}$t, $F\left(x\right)=$0, and two evolutionary stable points of x = 0 and x = 1 can be obtained, respectively [18].

Proof 1. 

According to Equations (5) and (15), two cases are discussed as follows:

When $y>y^{*}$, $F^{\prime }\left(0\right)>0$ and $F^{\prime }\left(1\right)<0$, i.e., x = 1 is the evolutionary stability strategy point ESS, indicating that the government will choose the strict supervision strategy G₂ [18].

When $y<y^{*}$, $F^{\prime }\left(0\right)<0$ and $F^{\prime }\left(1\right)>0$, i.e., x = 0 is the evolutionary stability strategy point ESS, indicating that the government will select the light supervision strategy G₂ [18]. □

② Stability analysis of the CEs’ behavior strategy [18].

An incremental stability analysis of the CEs’ behavior strategy was conducted, and the following scenario assumptions were proposed:

Proposition 2. 

(1) According to Equation (9), when$z=z^{*}={\displaystyle \frac{x\left(R_1+L-D_1\right)-C_1}{2xL+S_1}}, F\left(y\right)=0$ represents a stable boundary, indicating that the CE strategy was unchanged. (2) According to Equation (9), when $z\ne z^{*}$, $F\left(y\right)=0$, and y = 0 and y = 1 are the two evolutionary stable points.

Proof 2. 

According to Equations (9) and (16), two cases are discussed as follows:

When $\mathrm{z}>z^{*}, { F}^{\prime }\left(0\right)>0$ and $F^{\prime }\left(1\right)<0$; that is, y = 1 is the evolutionary stability strategy point ESS, indicating that the CEs will choose the active recycling strategy CE₁ in the blockchain-based CWRR projects [18].

When $z<z^{*},{ F}^{\prime }\left(0\right)<0$ and $F^{\prime }\left(1\right)>0$; y = 0 is the evolutionary stability strategy point ESS, indicating that CEs will choose the passive recycling strategy CE₂ in the blockchain-based CWRR projects [18]. □

③ Stability analysis of REs’ behavior strategy

Finally, an incremental stability analysis of the behavior strategy of REs was conducted, and the following scenarios were proposed:

Proposition 3. 

(1) According to Equation (13), when $x=x^{*}={\displaystyle \frac{-y{I}_1-R_2+C_3-C_2-F}{\left(1-y\right)I_1+D_2}}$,$F\left(z\right)=0$ represents a stable boundary, indicating that the strategy of the REs will remain unchanged. (2) According to Equation (13), when  $x\ne x^{*}$, $F\left(z\right)=0$; z = 0 and z = 1 represent the two evolutionary stable points.

Proof 3. 

According to Equations (13) and (16), two cases are discussed as follows:

When $x>x^{*}, F^{\prime }\left(0\right)>0$ and $F^{\prime }\left(1\right)<0$; that is, z = 1 is the evolutionarily stable strategy point ESS, indicating that REs will select the participation strategy in the blockchain-based CWRR projects (i.e., strategy RE₁) [18].

When $x<x^{*},{ F}^{\prime }\left(0\right)<$0 and $F^{\prime }\left(1\right)>0$; that is, z = 0 is the evolutionarily stable strategy point ESS, indicating that REs will select the non-participation strategy in the blockchain-based CWRR projects (i.e., strategy RE₂) [18]. □

4. Numerical Simulation Analysis

To fully present the significant parameters and all possible future developments in the evolution of blockchain-based CWRR projects, this study presents the influence of tripartite behavioral strategies on evolutionary pathways. This provides a clearer and more intuitive understanding of the dynamic evolution process inherent in the CWRR projects in blockchain technology [60].

This study used Vensim PLE 10.1 and Matlab 2018b to simulate the evolution of behavioral strategies of government, CEs, and REs in the blockchain CWRR projects. While conducting a numerical simulation analysis on the tripartite evolutionary game SD model, it was necessary to assign initial value to the parameters based on specific blockchain-driven CWRR projects. This study aimed to use the tripartite evolutionary game model to reveal the evolution trends of the tripartite game, the stability of equilibrium points, and influence of the behavioral strategy choices of the three parties on the evolution results of the system. Therefore, an accurate model parameter value may not be strictly necessary. In addition, owing to the complexity of the model and the imperfect data in the actual project operations, it is often difficult to obtain the accurate parameter data required for the numerical simulation of the game model [18,61]. Based on model assumptions and considerations of relevant suggestions from experts who have participated in actual CWRR projects in China (e.g., the construction waste dump in Guangdong Province of China) as well as related studies on construction waste recycling [16,18,50,56,62] and blockchain technology [48,51,52], the initial parameters were set for the numerical simulation analysis (

Table 2. Initial parameter settings for simulation.

Parameter	$I_0$	$D_0$	$C_0$	$R_0$	$S_0$	$L$	${\mathrm{S}}_1$	$C_1$	$R_1$	$D_1$	$C_2$	$C_3$	${\mathrm{I}}_1$	$R_2$	$D_2$	$F$
Value	4	5	3	3	5	2	4	3	2	4	3	5	6	5	3	5

). Simultaneously, the initial setting conditions for $C_3{>C}_2$ were satisfied.

4.1. Simulation of the Tripartite Evolutionary Game Model Based on SD

Based on the above assumptions and analysis of the evolutionary game, Vensim PLE 10.1 was used to construct a tripartite evolutionary game SD model for the blockchain-based CWRR system comprising three subsystems, i.e., government, CEs, and REs (Figure 4).

As indicated in Figure 4, the tripartite evolutionary SD model for the blockchain-based CWRR projects included three horizontal variables (x, y, z), three rate variables (F(x), F(y), F(z)), six intermediate variables ($U_{G1}$, $U_{G2}$, $U_{CE1}$, $U_{CE2}$, $U_{RE1}$, $U_{RE2}$), and sixteen other external variables. The external variables were the values taken for the variables in the tripartite game tree between government, CEs, and REs (see Table 1 and Figure 3). Moreover, in this tripartite evolutionary game SD model, the flow equations and the functional relationships between the three horizontal variables, three rate variables, and the intermediate variables were primarily based on the replication dynamic equations presented in Equations (5), (9), and (13).

As indicated in Figure 4, the tripartite evolutionary SD model for the blockchain-based CWRR projects included three horizontal variables (x, y, z), three rate variables (F(x), F(y), F(z)), six intermediate variables ($U_{G1}$, $U_{G2}$, $U_{CE1}$, $U_{CE2}$, $U_{RE1}$, $U_{RE2}$), and sixteen other external variables. The external variables were the values taken for the variables in the tripartite game tree between government, CEs, and REs (Figure 3). Moreover, in this tripartite evolutionary game SD model, the flow equations and the functional relationships between the three horizontal variables, three rate variables, and the intermediate variables were primarily based on the replication dynamic equations presented in Equations (5), (9), and (13).

Therefore, as indicated in Figure 5, with the evolutionary game proceeding, regardless of the initial strategy or the mutation degree of the three parties in the blockchain-based CWRR projects, the final evolutionary outcome of the system is at a stable strategy point of $(0, 1, 1)$. This indicates that the government ultimately chooses the light supervision strategy, the CEs select the active recycling strategy, and the REs choose the participation strategy.

4.2. Influence of Strategy Choices from an Individual Level on the Evolution Results of the System

In this study, to reflect the complexity, dynamics, and uncertainty of a blockchain-based CWRR social system, SD and replication dynamic analysis methods in evolutionary games are adopted and a tripartite game evolutionary SD model is constructed (see Figure 4). In the game, all three parties are boundedly rational. Upon identifying strategies yielding higher profits, they adjust their initial choices. Through the continuous mutual learning and imitation of strategy choices among the parties, the system reaches a new equilibrium state. Therefore, to explore how individual-level strategy choices of the three parties influence the system’s evolution, we conduct numerical simulations of the tripartite game evolutionary SD model using Matlab 2018b (See source code of algorithms of the model in Appendix A).

4.2.1. Influence of the Strategies of Two Parties on the Evolution Result

Based on the analysis of the tripartite evolutionary game SD model in Figure 4, the strategies of two parties (i.e., two subsystems) can be influenced by the same intermediate variables. For example, the intermediate variable L (penalties for illegal acts taken by REs or compensation obtained by CEs) can have an impact on both the subsystem of government and the CEs. Moreover, the intermediate variable D₂ (fines of REs for publishing false information or illegally disposing) can have an impact on both the subsystem of government and the REs.

The simulation results of the impact of the intermediate variable L on the system evolution are depicted in Figure 6 with the initial policies set to (x = 0.1, y = 0.1, z = 0.1) and (x = 0.5, y = 0.5, z = 0.5) and L = 0 and 100, respectively. The comparison of Figure 6a,b indicates that the increase in the penalties for illegal acts taken by REs or compensation obtained by CEs can cause the government to choose the light supervision strategy. Moreover, the CEs in the blockchain-based CWRR projects will choose an active recycling strategy and stabilize at 1 (y = 1), regardless of the changes in the compensation obtained by CEs (L), or even when it becomes 0 (no compensation). Similarly, in Figure 6c,d, when the initial decision values were adjusted to (x = 0.5, y = 0.5, z = 0.5), the strategy probabilities of the government and the CEs rapidly stabilized at 0 and 1, respectively, but the steady state and the equilibrium point $(0, 1, 1)$ were not affected. The above result indicates that the penalties for illegal acts taken by REs or compensation obtained by CEs can reduce the government supervision costs and control the fluctuations of the tripartite game in the system by bringing the system to a stable state, $(0, 1, 1)$.

The simulation results of the impact of the intermediate variable D₂ on the system evolution are depicted in Figure 6e–h with the initial policies set to (x = 0.1, y = 0.1, z = 0.1) and (x = 0.5, y = 0.5, z = 0.5) and D₂ = 0 and 100, respectively. The comparison and analysis of Figure 6e,f indicate that the increase in fines of REs for publishing false information or disposing of waste illegally (D₂) can significantly accelerate the participation behavioral strategy of REs in the blockchain-based CWRR projects. However, the increase in D₂ (0→100) led to big fluctuations in the evolutionary process of governmental decisions. As confirmed by the research of Su et al. (2020) [63] and Du et al. (2020) [16], the main reason for this fluctuation is that excessively high punitive measures (e.g., D₂ = 100 in this model) will exacerbate illegal disposition under the information asymmetry, and in this case, the government will tend to choose the strict supervision strategy (x→1 in Figure 6f). However, once REs choose a participation strategy in the CWRR projects with blockchain technology (z = 1 in Figure 6f), the government will eventually choose a light supervision strategy (x = 0 in Figure 6f) due to increased information transparency [64]. Similarly, Figure 6g,h compare the results.

In addition, by comparing Figure 6e,g or Figure 6f,h, we observed that when the initial strategies of three parties are assumed as (x = 0.5, y = 0.5, z = 0.5), the evolutionary process of the system becomes stable and reaches equilibrium point $(0, 1, 1)$. This finding indicates that the adjustment of fines of REs for publishing false information or illegally disposing of waste (D₂) can impact the fluctuations in the evolutionary game process of blockchain-based CWRR projects.

4.2.2. Influence of the Strategies of One Party on the Evolution Result

As depicted by the tripartite evolutionary game SD model in Figure 4, changes in some intermediate variables can impact the evolution results of the whole blockchain-based CWRR industry chain by influencing the strategies of a single stakeholder. The influence of critical factors on the strategy selection of government, CEs, and REs was simulated (Figure 7).

Influence of Strategies of Government on the Evolution Result

In the simulation analysis of the factors (intermediate variables) influencing government strategy evolution (x), this study explored the impact of positive incentive effect $R_0$ and negative supervision effect $S_0$ and strict supervision decision-making and publicity costs $D_0$ (see Figure 7a,b). As shown in Figure 7a, the increase in positive incentive effect $R_0$ and negative supervision effect $S_0$ can promote a smoother trend of the strategies of government towards light supervision (x = 0). The simulation results indicate that the government needs to pay attention to the social effects caused by the adoption of blockchain technology. As demonstrated by the studies of Jiang et al. (2023) [48] and Liu et al. (2022) [52], corresponding measures must be adopted to ensure that REs can legally comply with the recycling of construction waste through blockchain technology. As shown in Figure 7b, reducing decision-making and publicity costs of strict supervision ($D_0$) can encourage the government to adopt strict supervision strategies (x = 1). When supervision cost $D_0$ is below a certain value (e.g., $D_0=0$), governments in the blockchain-based CWRR projects tend to maintain strict supervision strategies (x = 1). Based on the simulation results in Figure 5, the change trends of government strategies will ultimately affect the overall evolution result of the blockchain-based CWRR projects.

Influence of Strategies of CEs on the Evolution Result

In the simulation analysis of the factors (intermediate variables) influencing CEs’ strategy evolution (y), this study simulates the impact of treatment costs of passive recycling chosen by CEs $C_1$, negative effects of CEs selecting passive recycling $D_1$, and extra compensation to CEs for the non-participation of the REs in illegal activities $R_1$ (see Figure 7c–e).

As shown in Figure 7c–e, the values of the three influencing factors affect the strategy evolution of CEs in blockchain-based CWRR projects. The larger the value of influencing factors, the faster the CEs will prefer to adopt the active recycling strategy (y = 1), and moreover, the sensitivity of the evolution of strategies to different influencing factors is different. In contrast, CEs are most sensitive to changes in treatment costs during passive recycling ($C_1$) (see Figure 7c), followed by negative effects of CEs ($D_1$) (see Figure 7d) and then extra compensation obtained ($R_1$) (see Figure 7e). Notably, when the treatment cost of passive recycling ($C_1$) is zero, CEs do not tend toward the passive recycling (y = 0) strategy but toward the stability strategy (y→1) at a very slow speed. The reason is that relevant CWRR information is constantly publicized and recognized by the industry through the application of blockchain technology [51,52]. CEs will eventually realize that the benefits of active recycling are far higher than those of passive recycling; hence, they will finally choose active recycling. Based on the simulation results in Figure 5, the change trends of CEs’ strategies will ultimately affect the overall evolution result of the blockchain-based CWRR projects.

Influence of the Strategies of REs on the Evolution Result

In the simulation analysis of the factors (intermediate variables) influencing REs’ strategy evolution (z), this study assessed the impact of the cost of participation ($C_3)$ and non-participation in blockchain-based CWRR projects ($C_2)$, extra losses for illegal acts ($F$), and revenue, and the positive effect of participating in recycling platform income (${\mathrm{I}}_1+R_2$). The simulation results are shown in Figure 7f–h.

Given the simulation results in Figure 7f–h, it can be concluded that the aforementioned three groups of influencing factors markedly influence the strategic stability of REs in blockchain-based CWRR projects. Of these factors, the cost of participating in the recycling platform $C_3$ and the cost of non-participation $C_2$ exert the greatest impact on the strategic evolution and stability of REs (see Figure 7f). As confirmed by the studies of Gopalakrishnan et al. (2021) [47] and Jiang et al. (2023) [48], the high cost of this strategy (e.g., high technical cost threshold) will exceed the affordability of REs. In this case, when the cost reaches a certain value (e.g., $C_2$ = 0 and $C_3$ = 20 in this model), REs can even ignore the other effects of the blockchain-based CWRR projects (e.g., ${\mathrm{I}}_1+R_2$ as shown in Figure 7g) and give up the participation strategy (z = 1). Under this circumstance, REs in the blockchain-based system will then prefer to choose a non-participation strategy, potentially leading to negative strategic outcomes for overall blockchain-based CWRR projects [48].

4.3. Numerical Simulation Result Analysis

Through numerical simulation assessing the impact of individual-level strategy choices on system evolution, the influence pathways of various significant parameters (refer to Table 2) on the behavioral strategies of governments, CEs, REs, and the overall evolution of blockchain-based CWRR projects can be observed, as shown in Figure 6 and Figure 7. To deeply analyze the behavioral strategy decisions of the government, CEs, and REs in the blockchain-based CWRR project system, the final simulation results are summarized in

Table 3. Simulation results of the influence of significant parameters.

Behavioral Strategy Choices of Tripartite Stakeholders	Parameters	Influence Path: Parameters → Behavioral Strategy Choices of Stakeholders → the Overall Evolution Result of the System	Significance of the Parameters
Probability of government selecting strategy x ($\mathbf{0}\mathbf{\le }\mathit{x}\mathbf{\le }\mathbf{1}$)	L: Penalties for illegal acts taken by REs or compensation obtained by CEs	① The increase in L can encourage the government to choose light supervision strategy; ② the larger the value of L, the faster the government will choose light supervision strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 6)	Main parameter L has significant influence on the strategy choice of the government
	D2: Fines of REs for publishing false information or illegally disposing	① The increase in D2 can encourage the government to choose light supervision strategy; ② the larger the value of D2, the more hesitant the government will be when choosing light supervision strategy; ③ control the fluctuations of the tripartite game in the system by bringing the system to a stable state, $(0, 1, 1)$ (see Figure 6)	Main parameter D2 has significant influence on the strategy choice of the government
	${\mathit{R}}_{\mathbf{0}}:$ Positive benefits obtained by the government when adopting strict supervision measures	① The increase in ${\mathit{R}}_{\mathbf{0}}$ can encourage the government to choose light supervision strategy; ② the smaller the value of ${\mathit{R}}_{\mathbf{0}}$, the faster the government will choose light supervision strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7a)	Main parameter ${\mathit{R}}_{\mathbf{0}}$ has significant influence on the strategy choice of the government
	${\mathit{S}}_{\mathbf{0}}$: Negative effects of government under light supervision	① The increase in ${\mathit{S}}_{\mathbf{0}}$ can encourage the government to choose light supervision strategy; ② the smaller the value of ${\mathit{S}}_{\mathbf{0}}$, the faster the government will choose light supervision strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7a)	Main parameter ${\mathit{S}}_{\mathbf{0}}$ has significant influence on the strategy choice of the government
	${\mathit{D}}_{\mathbf{0}}$: Decision-making and publicity costs of the government’s choice of strict supervision	① The decrease in ${\mathit{D}}_{\mathbf{0}}$ can encourage the government to adopt strict supervision strategies (x = 1); ② when supervision cost ${\mathit{D}}_{\mathbf{0}}$ is below a certain value (e.g., $D_0=0$), governments tend to maintain strict supervision strategies; ③ the equilibrium point of the overall system is $\left(0, 1, 1\right)$ or $\left(1, 1, 1\right)$ (see Figure 5 and Figure 7b)	Main parameter ${\mathit{D}}_{\mathbf{0}}$ has significant influence on the strategy choice of the government
Probability of CEs selecting strategy y ($\mathbf{0}\mathbf{\le }\mathit{y}\mathbf{\le }\mathbf{1}$)	L: Penalties for illegal acts taken by REs or compensation obtained by CEs	① The increase in L can encourage the CEs to choose active recycling strategy and stabilize at 1 (y = 1); ② no matter how compensation obtained by CEs (L) changes or even becomes 0 (no compensation), CEs choose active recycling strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 6)	Secondary parameter L has little influence on the strategy choice of the CEs
	${\mathit{C}}_{\mathbf{1}}$: Treatment costs of passive recycling chosen by CEs	① The increase in ${\mathit{C}}_{\mathbf{1}}$ can encourage the CEs to choose active recycling strategy; ② the larger the value of ${\mathit{C}}_{\mathbf{1}}$, the faster the CEs will prefer to adopt the active recycling strategy (y = 1); ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7c)	Main parameter ${\mathit{C}}_{\mathbf{1}}$ has significant influence on the strategy choice of the CEs
	${\mathit{D}}_{\mathbf{1}}$: Negative effects of CEs selecting passive recycling	① The increase in ${\mathit{D}}_{\mathbf{1}}$ can encourage the CEs to choose active recycling strategy; ② the larger the value of ${\mathit{D}}_{\mathbf{1}}$, the faster the CEs will prefer to adopt the active recycling strategy (y = 1); ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7d)	Main parameter ${\mathit{D}}_{\mathbf{1}}$ has significant influence on the strategy choice of the CEs
	${\mathit{R}}_{\mathbf{1}}$: Extra compensation to CEs for the non-participation of the REs in illegal activities	① The increase in ${\mathit{R}}_{\mathbf{1}}$ can encourage the CEs to choose active recycling strategy; ② the larger the value of ${\mathit{R}}_{\mathbf{1}}$, the faster the CEs will prefer to adopt the active recycling strategy (y = 1); ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7e)	Main parameter ${\mathit{R}}_{\mathbf{1}}$ has significant influence on the strategy choice of the CEs
Probability of REs selecting strategy z ($\mathbf{0}\mathbf{\le }\mathit{z}\mathbf{\le }\mathbf{1}$)	D2: Fines of REs for publishing false information or illegally disposing	① The increase in D2 can significantly accelerate the participation behavioral strategy of REs (z = 1); ② regardless of the changes in the fines of REs (D2), or even in cases where fines become 0 (no fines), REs choose the participation strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 6)	Secondary parameter D2 has little influence on the strategy choice of the REs
	${\mathit{C}}_{\mathbf{2}}$: Additional costs incurred by REs not participating	① The increase in ${\mathit{C}}_{\mathbf{2}}$ can encourage the REs to choose participation strategy (z = 1); ② when it reaches a certain value, ${\mathit{C}}_{\mathbf{2}}$ = 0, REs will then prefer to choose non-participation strategy; ③ the equilibrium point of the overall system is $\left(0, 1, 1\right)$ or $\left(0, 1, 0\right)$ (see Figure 5 and Figure 7f)	Main parameter ${\mathit{C}}_{\mathbf{2}}$ has significant influence on the strategy choice of the REs
	${\mathit{C}}_{\mathbf{3}}$: Recycling costs associated with RE participation	① The increase in ${\mathit{C}}_{\mathbf{3}}$ can cause the REs to choose participation strategy (z = 1); ② when it reaches a certain value, ${\mathit{C}}_{\mathbf{3}}$ = 20, REs will then prefer to choose non-participation strategy; ③ the equilibrium point of the overall system is $\left(0, 1, 1\right)$ or $\left(0, 1, 0\right)$ (see Figure 5 and Figure 7f)	Main parameter ${\mathit{C}}_{\mathbf{2}}$ has significant influence on the strategy choice of the REs
	$\mathbf{F}$: Recycling the extra losses caused by illegal acts of enterprises	① The increase in $\mathbf{F}$ can cause the REs to choose the participation behavioral strategy (z = 1); ② the larger the value of $\mathbf{F}$, the faster the REs will adopt the participation strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7g)	Main parameter $\mathbf{F}$ has significant influence on the strategy choice of the REs
	${\mathbf{I}}_{\mathbf{1}}\mathbf{+}{\mathit{R}}_{\mathbf{2}}:$ Revenue and positive effect of participating in the income of recycling platform	① The increase in $({\mathbf{I}}_{\mathbf{1}}+{\mathit{R}}_{\mathbf{2}})$ can cause the Res to choose the participation behavioral strategy (z = 1); ② the larger of the value of $({\mathbf{I}}_{\mathbf{1}}\mathbf{+}{\mathit{R}}_{\mathbf{2}})$, the faster the REs will adopt the participation strategy; ③ the equilibrium point of the overall system is $(0, 1, 1)$ (see Figure 5 and Figure 7h)	Main parameter ${\mathbf{I}}_{\mathbf{1}}$ and ${\mathit{R}}_{\mathbf{2}}$ have significant influence on the strategy choice of the REs

.

Based on the influence of the aforementioned parameters on the behavioral strategy choices and the overall system evolution, they can be classified as primary and secondary factors, as shown in Table 3. Recognizing the significance of these parameters, supervising authorities facilitate more effective collaboration among multiple stakeholders to enhance the efficiency of CWRR projects driven by blockchain technology.

5. Discussion and Implications

The results of this study confirm that blockchain-based CWRR projects form the common value chain in circumstances that benefit all stakeholders. This enables all stakeholders in the blockchain-based CWRR projects to collaborate and share information for coordinated development. The simulation results obtained in this study provide implications for the supervising authorities (governments) implementing the blockchain-based CWRR projects.

• From the perspective of the government: The government is decisive in promoting blockchain-based CWRR

As demonstrated by the simulation results in Figure 6 and Figure 7a,b, in blockchain-based CWRR projects, the government, as the master planner and policymaker, should conduct a multi-agent collaborative management mechanism. First, the government should establish a sound and strict supervision policy on CWRR. This includes setting up a reward and punishment system and a coherent “top–down” regulatory system. For example, the government should increase the subsidy rate based on the financial capacity of the local government and simultaneously establish an appropriate penalty mechanism to avoid exorbitant disposal fees [65]. Additionally, increasing the cost of construction waste treatment in the passive recycling of construction waste by CEs can decrease the possibility of illegal dumping. As presented in the model of this study, increasing $C_1$ from 0 to 10 can promote the transition of CEs’ behavior towards the active recycling strategy (see Figure 7c). The government should also implement consequences for the illegal treatment of construction waste by REs and control the quality of recycling products to keep the CEs’ purchase intentions. This would ensure that the construction waste generated by CEs flows to REs as raw materials, resulting in multi-collaborative CWRR projects and promoting the sustainable development of the CWRR supply chain [66].

Second, the simulation results in Figure 7a,b indicate that the government must actively guide relevant enterprises to improve the awareness and implementation of CWRR. This indicates that the government can comprehensively improve the efficiency of CWRR for economic and social sustainability by carrying out demonstration projects to popularize the benefits of CWRR, vigorously developing a circular economy, and promoting the flow of recycled products into the market.

Finally, the government should formulate relative standards for assessing the affordability of adopting blockchain technology by REs, as indicated by the increase in the cost of REs’ participation in the recycling platform shown in Figure 7f. This includes focusing on the status of REs in the process of adopting blockchain technology, issuing appropriate subsidies according to the level of support for REs, reducing their technical cost threshold (i.e., the cost of REs participating in blockchain-driven recycling platform), and providing support to adopt new technologies [47].

Therefore, the government must regard a large CWRR company (or listed company) as a host enterprise of the chain and maintain close collaboration with other relevant stakeholders. To achieve efficient CWRR, the government must oversee construction management and environmental protection, while CEs must manage waste transportation throughout the life cycle [18].

• From the perspective of CEs: Attention should be paid to the negative effects of construction waste treatment costs and the influence of compensation

As shown in the simulation results in Figure 7c–e, if the CEs select a passive recycling strategy, the government should apply strict supervision policies to increase the cost of construction waste treatment by the CEs. This can be achieved by increasing the cost of construction waste treatment by themselves, as indicated by the increase in the treatment costs of passive recycling chosen by CEs in Figure 7c. In addition, the government should strengthen publicity and education to enhance the recycling awareness of the CEs through financial incentives (e.g., tax incentives), demonstration projects, media platform promotion, research cooperation, and community interaction, thus promoting active recycling [53].

Moreover, the government must carefully consider the losses incurred by CEs owing to the illegal acts of REs, as indicated by the increase in extra compensation to CEs for REs’ illegal activities. This can be achieved by providing legal and policy guarantees to ensure that CEs with recycling intentions obtain legal compensation when their interests are infringed upon. For example, regulations on construction waste management in various regions should emphasize corresponding safeguard measures for CEs [67] to further motivate the recycling efforts of CEs across the industry.

In addition, the government should promote the market entry of construction waste recycled resources (CWRRs), revise perceptions of recycled materials, and establish a collaborative network for producing and recycling construction waste products. These initiatives will significantly enhance the willingness of CEs to recycle and enhance the quality of construction waste. Recycling in China is still in its infancy, necessitating the establishment of quality standards for CWRR products and their promotion through government procurement and demonstration projects to highlight the advantages of materials recycled from CWRR [68].

• From the perspective of REs: Participation costs, illegal punishment, and participation benefits should be considered

The simulation results in Figure 7f–h indicate that reasonable control of the ratio of illegal punishment and the cost and benefit of REs’ participation in the CWRR projects makes REs’ recycling behavior passively correlated with government supervision. In the tripartite game analysis of the blockchain-based CWRR projects, the preferable strategy combination is light supervision, active recycling, and participation; hence, government supervision is the dominant factor in adopting the blockchain-based CWRR projects.

First, the strength, particularly the technical capabilities of REs, should be promoted because progressive waste recycling technology can produce high-quality recycled products that meet public standards [18]. In this scenario, government incentives and guidance, as well as technical support from research institutions, are crucial in advancing recycling technologies. Recycling technology in China is still immature, as only low- and medium-strength concrete can be produced for road surfaces and shock-absorbing cushion layers. Nevertheless, a large number of valuable wastes, such as light fixtures, precious metals, plastic, glass, and wooden products, can be recycled. Incentives and guidance from the government and research cooperation among research institutions can improve the reuse and recycling of construction waste and facilitate the production of high-quality recycled materials [68,69].

Meanwhile, REs can also introduce innovative technologies and develop their own waste recycling technology to meet the standard performance indicators set by the government and receive incentives. When the cost of participating in the blockchain-based CWRR projects constructed by the government is high and the benefits of utilizing blockchain technology are significant, REs can construct their own private blockchain-driven platform for recycling. The government will be unable to directly supervise and deter the REs’ illegal activities.

Therefore, governments should establish CWRR projects based on blockchain technology. Meanwhile, reducing the participation costs for REs and guiding them to participate in the recycling platform are conducive to law enforcement and government supervision. This blockchain-based CWRR project is also convenient for CEs to safeguard their rights and interests and enhance the credibility of the government. Moreover, due to the flow of recycled construction waste products to the market, good-quality products can decrease the amount of required natural materials and gain social recognition. This will increase the benefits of participating in the recycling platform and prompt REs to adopt the participating strategy.

The use of blockchain technology can facilitate information sharing among CEs, REs, and governments, providing strong support for government supervision and incentives to address practical barriers. Meanwhile, in this study, the organizational framework of the construction industrial chain system was clarified, revealing that collaboration among these three parties can be strengthened. Therefore, with the policy recommendations proposed for the three parties’ evolutionary game strategies at the individual level, more efficient CWRR projects based on blockchain technology at the system (group) level can be achieved. Figure 8 presents a framework for efficient CWRR projects driven by a blockchain platform based on the aforementioned implications. Although the proposed theoretical framework for efficient blockchain-based CWRR projects in this study is transferable and suitable for the same type of CWRR projects in other countries, the specific degree of implementation of tripartite stakeholders’ strategies needs to be adjusted based on the different practical situations and barriers in these countries or regions. For example, the appropriate rates for the government’s reward/punishment may vary across different regions and countries. Another situation is that some countries or regions already have well-established incentive mechanisms and good awareness of CWRR (e.g., countries such as Japan and South Korea where the construction waste recycling rate has reached over 90%) [70]. Therefore, the behavioral strategy choices for tripartite stakeholders at the individual level will inevitably be adjusted based on this background. This is because when the tripartite game model established in this study is applied to these countries or regions, the model assumptions, model parameters, and the initial value of parameters will be adjusted to better align with local practices.

6. Conclusions

Given that the CWRR project remains in its infancy and continues to face several challenges, collaborative growth cannot be achieved solely through market-driven mechanisms [18]. At this stage, macro-level government supervision is required to improve the coordinated development of CWRR projects and introduce innovative management models for recycling and reuse [18,40]. Therefore, this study introduces the evolution and development of the CWRR projects’ ecosystem based on innovative blockchain technology from a government perspective. First, using a practical analysis, a tripartite evolutionary game model involving the government, CEs, and REs was established. Subsequently, through a simulation analysis based on SD, the stable strategies of the three parties and the factors influencing their evolution of stable strategies were examined. Finally, based on the simulation results, relevant policy recommendations from the evolutionary strategies of stakeholders at the individual level were offered for the blockchain-based CWRR projects at the group level.

This study analyzes the impact of adopting blockchain technology on the behavioral strategies of stakeholders within the CWRR industry chain and provides strategic recommendations. It integrates three parties, namely the government, CEs, and REs, into a coevolutionary framework, expanding the existing studies on CWRR in the context of blockchain and promoting the relevant modeling practices of the CWRR projects. The tripartite game modeling framework can also help to connect previous dispersed research on government supervision and enterprise recycling decisions. It offers a scientific and rational basis for the adoption of blockchain-driven platforms and, at the same time, fosters collaboration among stakeholders in the CWRR industry chain to drive high-quality development. Moreover, based on the tripartite evolutionary game SD model, this study adopted exploratory scenario simulations to analyze the influence of the strategy choices made by the three parties on the evolution of blockchain-based CWRR projects. This provides practical insights and the capacity to quantify differences in various policies and measure interventions on the CWRR projects in the context of blockchain. Particularly, through simulation and a theoretical framework of efficient blockchain-based CWRR projects, this study offers scientifically sound management recommendations to foster collaboration among stakeholders in the CWRR projects and implement blockchain-based CWRR practices in China and other countries and regions: (1) as the master planner and policymaker, the government should conduct a multi-agent collaborative management mechanism; (2) more attention should be paid to the negative effects of construction waste treatment costs and the influence of compensation of CEs to promote active recycling; (3) participation costs, illegal punishment, and participation benefits of REs’ participation should be considered.

However, it should be noted that the model in this study mainly focuses on the Chinese context for CWRR projects driven by blockchain. Owing to the differences in legislation, law enforcement, construction and demolition practices, recycling infrastructure, administration supervision, CWRR market environment, and the popularization and promotion of blockchain technology, the adaptation of the model and research results requires adjustments to fit local conditions. To adapt this model to different geographical or regulatory environments, assumptions and initial parameter values for numerical simulations should align with specific local CWRR and blockchain technology application contexts.

This study has certain limitations. It uses simulations and the reports of other studies for the model verification and analysis. It would be more convincing to verify the model and research results with actual case data. In addition, the stakeholder model only considers a three-party game for CWRR, overlooking other key entities within the projects, such as transportation enterprises, research institutions, and sales enterprises. Future research should develop a comprehensive multi-stakeholder game model for blockchain-based CWRR projects supported by real-case data.