Evaluation Method for Green Construction Demonstration Projects in China Based on G-TOPSIS

Abstract

Although the construction industry has played an important role in promoting national economic growth, over the past decades, construction activities have caused serious negative impacts on the ecological environment. Faced with this challenge, many countries have made promoting the greening of the construction industry one of their development goals. As a high-level demonstration project for green construction, Green Construction Demonstration Projects (GC-DPs) play a significant role in improving the level of green construction and promoting the green development of the construction industry. This study aims to establish a reasonable GC-DP evaluation method to promote the development of green construction in China. An evaluation index system is constructed, including five criterion layers, 20 main factors, and 60 sub-factors. Sixty sub-factors correspond to 60 specific indicators. The combination optimization of subjective and objective weights of indicators is conducted using game theory, and the comprehensive weights of indicators are calculated. Furthermore, a GC-DP evaluation model based on the gray TOPSIS method with game theory combination weights (hereinafter referred to as G-TOPSIS) is established. Finally, a case study is carried out to verify the feasibility of the proposed method. Expert scoring and actual engineering data are used to calculate indicator weights, and game theory is utilized to balance the influence of subjective and objective factors. Results show that the evaluation results obtained using the proposed G-TOPSIS method are consistent with the actual situation of the project.

1. Introduction

The construction industry plays a significant role in meeting social needs, improving quality of life [1], and promoting national economic growth [2,3]. However, at the same time, construction activities have had serious negative impacts on the ecological environment, resources, and energy. According to the World Business Council for Sustainable Development (WBCSD), about 50% of global energy is used in the construction industry. Various types of pollution caused by the construction industry account for about 1/3 of the global total pollution [4]. It has been reported that nearly 30% of greenhouse gas emissions in developing countries are related to construction [5]. Compared with industries such as industry and transportation, the construction industry is considered to be the largest primary energy consumption field [6]. If no effective measures are taken, the International Energy Agency predicts that energy consumption in the construction industry will increase by 50% by 2050 [7].

Energy consumption during the construction phase accounts for 23% of the building’s life cycle, making it the most critical stage in terms of energy [8]. Furthermore, the construction process plays an important role in both direct and indirect environmental impact [9]. As an organic combination of building and sustainability, green construction is a key link in reducing ecological damage and realizing green development in the construction industry. Many countries regard construction management as an important part of the green building evaluation system [10]. The implementation of green construction is a concrete practice of carrying out the scientific development concept in the construction industry, as well as an important measure to realize national sustainable development. Hence, Green Construction Demonstration Project (GC-DP) came into being. Carrying out GC-DP activities is not only conducive to strengthening the green construction awareness of industry personnel, but also conducive to promoting the development and growth of the industry. Additionally, cooperative learning in demonstration projects contributes to cultivating green innovation awareness. However, in recent years, the important basis of GC-DP evaluation has not been updated. This problem has led to slow development of green construction in China, with only a few benchmark GC-DPs. Thus, there is an urgent need for constructing a GC-DP evaluation model to promote the development of green construction in China.

This study aims to construct a GC-DP evaluation method to provide guidance for the development of green construction. The contributions of this study are as follows: (1) A set of GC-DP evaluation index systems is constructed. (2) The game theory combination weight method is applied to calculate comprehensive weights of indicators, which balances the influence of subjective and objective factors. (3) A G-TOPSIS comprehensive evaluation model is constructed. Based on the calculation results of the G-TOPSIS comprehensive evaluation, deficiencies of the declared projects in the same period can be identified, which is helpful for formulating effective measures to improve the green construction level of the project. The remaining sections are organized as follows. Section 2 introduces existing studies on green building and green construction evaluation. In Section 3, problems existing in current standards in China are summarized. The focus of the GC-DP evaluation index system is identified. In Section 4, the research methodology and calculation results of indicator weights are described. In Section 5, MJ-1 new residential project is used to verify the feasibility of the proposed method. The rationality of the GC-DP evaluation index system and the practicality of the G-TOPSIS comprehensive evaluation model are explored in Section 6. Finally, the main findings, limitations of the study and future research directions are summarized in Section 7.

2. Literature Review

The concept of green construction is developed in the context of green buildings. Green buildings focus on improving building energy efficiency and reducing construction’s negative impacts on the environment and resources [7]. Green construction refers to adhering to the principles of reduction, reuse, and recycling in construction, achieving the unity of social, economic, and ecological benefits. However, relying solely on green construction makes it difficult to meet the requirements of green buildings. High-quality design, scientific operation, and maintenance management are indispensable elements for constructing green buildings [11]. As a new construction concept, green construction is not only used to build green buildings but can also realize the construction of traditional buildings, prefabricated buildings, healthy buildings, and other forms of construction. Furthermore, it can reduce the impact of the construction process on the environment and maximize the realization of a sustainable construction process.

2.1. Green Buildings Evaluation

In the 1970s, the global energy crisis broke out, and countries successively carried out exploration and research on energy conservation [12]. In 1990, Britain established BREEAM, the world’s first green buildings evaluation system [13]. Since then, there have been a quantitative criterion for green buildings. As a leader in green buildings evaluation, BREEAM has become a reference model for many countries to develop green buildings evaluation systems. In 1994, the United States Green Building Council (USGBC) formulated LEED (V1) according to its development characteristics of green buildings. LEED has undergone adjustments and updated in its development process. In 2019, LEED (V4.1) was launched globally [14]. LEED is currently the most widely used framework [15]. Ashuri and Durmus-Pedini [16] found that after more than 20 years of development, green buildings can save 50% of energy consumption, reduce operating costs by 8%, and increase investment return by 6.6%, compared with traditional buildings in the United States. In 2001, Sustainable Building Association of Japan developed CASBEE with environmental efficiency as its core. They then released a pilot version for global use in 2015. Subsequently, Singapore’s Green Mark, Australia’s Green Star, Germany’s DGNB, and Korea’s G-SEED were launched successively. The evaluation system for green buildings worldwide is gradually improving.

Some scholars have also carried out research on green buildings. Bai and Liu [17] proposed a green buildings performance evaluation index system to evaluate the sustainability of buildings. BIM was employed to facilitate the process of green building assessment. Nguyen and Macchion [18] used mean scoring and fuzzy synthetic evaluation method to assess the risk of green buildings. In more specific terms, assessed risks of human health and safety in green buildings.

In the 1980s, Standard for Energy Efficiency Design of Civil Buildings (Heating Residential Buildings) (JGJ26-86) [19] was published. With technological innovation, terms for building energy efficiency in the standard have been increased from 30% to 75% [20]. In 1999, Taiwan established the EEWH with the theme of “Ecology, Energy, Waste and Health”. By 2015, the standard had been revised seven times. In 2001, “China Ecological Housing Technology Assessment Manual” was issued, indicating the formal implementation of green buildings in China. LEED standard was introduced to China in 2003. However, due to the significant differences in architectural design standards and the high cost of certification, the implementation of LEED faces numerous obstacles [10]. In 2006, the Ministry of Construction and the General Administration of Quality Supervision, Inspection and Quarantine of the People’s Republic of China jointly issued the national standard “Green Building Evaluation Standard (GB/T 50378-2006)” [21] (hereinafter referred to as “06 version Green Standard”). In 2014, the “Green Building Evaluation Standard (GB/T 50378-2014)” [22] (hereinafter referred to as “14 version Green Standard”) officially replaced the “06 version Green Standard”. It added a new chapter on “construction management”, which made clear provisions for green construction. The development of green buildings has entered a new era in China. In 2019, the Ministry of Housing and Urban-Rural Development approved the “Green Building Evaluation Standard (GB/T 50378-2019)” [23] (hereinafter referred to as “19 version Green Standard”) as the national standard. The “19 version Green Standard” reconstructed the technical index system of green building evaluation, and raised the performance requirements of green buildings.

Shan and Hwang [24] compared the core evaluation contents of the latest version of nine green building standards that are influential worldwide. The analysis results indicated that although the rating system of green buildings varies from country to country, factors such as “site”, “water”, “energy”, “indoor environmental quality”, “materials”, “waste and pollution”, “management” and “innovation” are always important assessment factors. Even though the above factors are not directly listed as chapter headings in some standards, they are still reflected in these standards. In general, the eight core factors can be summarized into three aspects. First, resource conservation and utilization, including land, water, energy and materials, accounts for half of the eight core elements. Thus, they are the top priority in the evaluation of green buildings. Second, environmental protection and comfort, including waste, indoor environmental quality, focuses on controlling the emission of various hazardous substances during construction and recycling construction waste. The third aspect is about management and innovation. From design to operation and maintenance, scientific management can make the implementation of green building more effective. The rating of green buildings is beneficial to promote the green transformation of the construction industry. The differences in the setting of evaluation levels in existing standards across countries are provided in

Table 1. Certification levels of green building standards.

Name of Standard	Country	Year	Four Levels	Five Levels
BREEAM	UK	2018		✓
LEED	USA	2019	✓
BEAM	China	2020	✓
CASBEE	Japan	2018		✓
Green Star	Australia	2019	✓
Green Mark	Singapore	2018	✓
Evaluation Standards for Green Buildings	China	2019	✓
DGNB	Germany	2018	✓
GBI	Malaysia	2019	✓

. The setting of the certification structure level is crucial for the implementation and review of green buildings. It not only serves as a reference standard for green building proponents to carry out design and construction, but also determines the certification workload and certification cost of experts. Most of the nine standards adopt a 4-level certification structure.

2.2. Green Construction Evaluation

The implementation of green buildings around the world contributes to the advancement of green construction. At present, some countries have incorporated the inspection of the green level during the construction stage into the green building evaluation standards. In 2007, the Green Construction Guidelines provided guidance for the development of green construction in China. In 2010, the Evaluation Standard for Green Construction of Building (GB/T 50640-2010) (hereinafter referred to as “10 version Standard”) [25], was published and implemented. It filled the gap in the evaluation system of green construction in China. In the same year, the China Construction Industry Association carried out GC-DP evaluation activities with the 10 version Standard as the core.

Some scholars have explored the concept, feasibility, and implementation measures of green construction. Tam et al. [26] divided the evaluation indicators of green construction into two categories: management performance indicators (MPI) and operational performance indicators (OPI). A “Green Construction Assessment (GCA)” system applicable to Hong Kong was proposed. Skouloudis et al. [27] divided the evaluation of green construction sustainability into three dimensions: environment, society, and economy. Qin and X.W. Zou [28] compared the characteristics of lean construction with green construction and proposed that green construction is the deepening of lean construction. Liu and Lin [29] proposed ecological indicators for green building construction by applying a slack-based data envelop analysis approach, in which resource conservation and environmental protection were both incorporated. Zhang, Wu and Liu [7] analyzed the economic feasibility of green construction from two dimensions: the whole life cycle of buildings and project participants. They concluded that green buildings are economically feasible. However, factors such as overestimation of incremental costs by participants, asymmetric market information, diversification of contract structures, and instability of energy prices hinder the development of green construction. In addition, they concluded that specific green technologies could help save the cost of green certification and life cycle of industrial manufacturing buildings.

In existing studies, management, energy saving, environmental protection, and economy are often utilized as the core of green construction evaluation. The quantification degree of indicators has been relatively high. However, some scholars have only studied one aspect of the environmental protection, saving materials and material resources, water saving and water resource utilization, energy conservation and energy utilization, and land conservation and resource protection, so their applicability to actual green building evaluation is weak. Management and comprehensive benefits are not fully considered when constructing secondary indexes. Additionally, the strong ambiguity of indicators, difficulty in data collection, and poor operability are also reasons that hinder the promotion and application of green construction evaluation. Furthermore, the model is difficult to fully consider the characteristics of qualitative and quantitative indicators, so the effective data cannot be fully utilized.

3. Evaluation Index System for GC-DP in China

3.1. Existing Problems in Green Construction Evaluation Standards

“10 version Standard” is the core document for evaluating traditional GC-DPs, as shown in

Table 2. Summary of “10 version Standard”.

Stages of Evaluation	Contents of Evaluation	Setting of Weight	Evaluation Methods	Evaluation Level
Foundation and foundation engineering Structural engineering Decoration and mechanical Electrical installation engineering	Environmental protection Saving materials and material resources Water saving and water resource utilization Energy conservation and energy utilization Land conservation and resource protection	One level	Linear weighting method	Not up to standard [0, 60) Qualified: [60, 80] Good: [80, 100]

.

The problems existing in the current green construction evaluation standard are analyzed, and the main improvement direction of GC-DP evaluation is determined.

• Poor operability of evaluation process. From a macro perspective, the evaluation system has strong significance for guiding construction enterprises to carry out green construction and standardizing specific practices. However, the evaluation of the completion level of green construction is relatively complex.
• Deviation from engineering practice. The current green construction evaluation index system has more qualitative indicators and less quantitative indicators, and does not consider socio-economic factors.
• Subjectivity of the evaluation method. The evaluation of the green construction level of the project is mainly based on expert scoring, definitions, and other forms, which affects the accuracy of evaluation results.

3.2. Determination of Evaluation Index System of GC-DP

According to the top-down decomposition strategy of the factor analysis method, the criterion layer of the evaluation system is determined first. The evaluation index system of GC-DP in this paper includes five criteria layers, namely environmental protection and control, conservation and use of resources and energy, comprehensive construction management, comprehensive benefit analysis, and green construction innovation and effectiveness analysis. Based on the literature review and standard comparison, 66 sub-factor indicators (i.e., the initial selection of indicators) were identified. Then, questionnaires were distributed to 23 experts, as detailed in

Table 3. Interviewed experts’ background information.

Occupation	Employer	Education	Proportion
Engineering practitioner	Construction organizations	Bachelor, Master	30.43%
Projects advisory	Designing institutes	Master, Ph.D.	30.43%
Scientific researcher	Universities	Ph.D.	39.14%

. A 5-level Likert scale was used, and experts rated the importance of 66 indicators based on personal theoretical and practical knowledge. Test of Kendall’s coefficient of concordance W was conducted on the questionnaire results. According to statistical analysis results, 59 indicators were identified as intermediate indicators. The concentration of expert opinions on the other seven initial indicators is poor. Therefore, expert interviews were conducted in the form of questionnaires to determine the trade-offs between these seven indicators. Finally, 60 sub-factor indicators were determined, as shown in Figure 1. Then, by categorizing the 60 sub-factor indicators, the main factor layer is obtained.

4. Methodology

The overall study framework is illustrated in Figure 2.

4.1. Subjective Weight Calculation Model Based on Fuzzy Analytic Hierarchy Process (FAHP)

FAHP has been used in many different fields. Fuzzy numbers are used to represent ratios in pair-to-pair comparisons, which solves the imprecision of traditional AHP with the advantage of fuzzy logic [30]. FAHP is more adaptable to the complexity of GC-DP evaluation indicators and improves the accuracy of decision-making. In the calculation process, compared with traditional methods, the consistency test is omitted, and the calculation is more convenient. The specific calculation steps are as follows.

Step 1: Construct fuzzy complementary matrix $A$ of GC-DP index.

$$A={(a_i{}_j)}_{m\times m}$$

(1)

where m represents the number of evaluation indicators; a_ij represents the fuzzy relationship affiliation obtained by two-by-two comparison of GC-DP evaluation indicators M_i and M_j.

Step 2: Construct fuzzy consistent judgment matrix $B={(b_{ij})}_{m\times m}$.

$$b_{ij}=\frac{b_i-a_{ij}}{2m}+0.5$$

(2)

$$b_i={\displaystyle \sum _{j=1}^m{a}_{ij}}$$

(3)

Step 3: Calculate the subjective weight of GC-DP index.

$$w_i=\frac{2{\displaystyle \sum _{j=1}^m{b}_{ij}-1}}{m(m-1)}$$

(4)

where $b_{ij}$ represents the value of row $i$ and column $j$ in matrix B.

4.2. Objective Weight Calculation Model Based on Entropy Weight Method

The entropy weighting method determines the target weights by determining the degree of dispersion of the dataset. The smaller the degree of discrete change in data, the smaller the entropy value, and the more effective the information. Using actual data to calculate indicator weights is more realistic and objective [31]. This paper collects information of four projects declared as GC-DPs in the same period to calculate the objective weights of indicators, as shown in

Table 4. Information of four declared GC-DPs.

Project Name	Duration of Construction	Declared Level	Building
MJ-1	Two years and three months	Provincial	Residential buildings
Project 2	Three years	Municipal	Residential buildings
Project 3	Two years and five months	Municipal	Residential buildings
Project 4	Two years and seven months	Provincial	Residential buildings

. The source of raw data for the entropy weighting method is shown in Figure 3. The background information of three professionals involved in the entropy weighting method is provided in

Table 5. Interviewed experts’ background information.

Expert	Occupation	Employer	Level of Education	Experience
1	Lecturer	University	Ph.D	5 years
2	Manager	Design company	Master	10 years
3	Engineer	Construction company	Master	7 years

. The procedures are as follows [32,33].

Step 1: Construct the original matrix G.

$$G={(g_{ij})}_{n\times m}=\left(\begin{array}{cccc}{g}_{11}& g_{12}& \cdots & g_{1m}\\ g_{21}& g_{22}& \cdots & g_{2m}\\ ⋮& ⋮& ⋮& ⋮\\ g_{n1}& g_{n2}& \cdots & g_{nm}\end{array}\right)$$

(5)

where g_ij represents the jth indicator of the ith declared GC-DP project; $n$ represents the number of GC-DP projects declared.

Step 2: Construct the standardized matrix S.

$$S={(s_{ij})}_{n\times m}=\left(\begin{array}{cccc}{s}_{11}& s_{12}& \cdots & s_{1m}\\ s_{21}& s_{22}& \cdots & s_{2m}\\ ⋮& ⋮& ⋮& ⋮\\ s_{n1}& s_{n2}& \cdots & s_{nm}\end{array}\right)$$

(6)

$$s_{ij}=\frac{g_{ij}}{{\displaystyle \sum _{i=1}^n{g}_{ij}}}$$

(7)

where $s_{ij}$ represents the percentage of indicator values for the ith participating item under the jth indicator.

Step 3: Calculate the entropy value of the evaluation index.

$$e_j=-\alpha {\displaystyle \sum _{i=1}^n{s}_{ij}\cdot \ln }{s}_{ij}$$

(8)

$$\alpha =1/\ln n$$

(9)

When $s_{ij}=0$, define $\underset{s_{ij}\to 0}{\lim }(s_{ij}\cdot \ln s_{ij})=0$.

Step 4: Calculate the entropy weight of the evaluation index.

$${\omega }_j=\frac{(1-e_j)}{{\displaystyle \sum _{j=1}^m(1-e_j)}}$$

(10)

4.3. Comprehensive Weight Calculation Model Based on Game Theory

The combination weight in the game theory is different from the traditional simple linear combination weight. The core idea is to coordinate conflicts and maximize benefits. In fact, the relationship between indicators is comprehensively considered, and the subjective and objective weights are balanced [34]. The game theory not only considers the subjective judgment of experts, but also takes into account the impact of project construction data on rating results, making the determination of weights more reasonable. The calculation steps are as follows [35,36].

Step 1: Construct a set of possible weight vectors ${\omega }_k$.

$${\omega }_k=\left\{{\omega }_{k1}, {\omega }_{k2}, \dots , {\omega }_{km}\right\}$$

(11)

where $k$ represents the number of methods used for weighting calculations. FAHP and entropy weight method are employed to calculate the index weights of the GC-DP evaluation system. Therefore, $k=1, 2$.

Set the linear combination weight coefficient $\partial =\left\{{\partial }_1, {\partial }_2, \dots {\partial }_k, \dots , {\partial }_K\right\}$. K = 2. Then, any linear combination of these vectors can be obtained.

$$\omega ={\displaystyle \sum _{k=1}^K{\partial }_k{\omega }_k^T}$$

(12)

Step 2: Optimal combination.

The balance point between different weights is sought to minimize the deviation of $\omega$ and ${\omega }_k$, and optimization of ${\partial }_k$ is performed in Equation (12). The optimal weight of $\omega$ is calculated, and the objective function is obtained.

$$\min {‖{\displaystyle \sum _{k=1}^K{\partial }_k^{}{\omega }_k^T-{\omega }_k}‖}_2$$

(13)

Step 3: Normalize the optimal combination coefficients ${\partial }_k$.

$${\partial }_k{}^{*}=\frac{{\partial }_k}{{\displaystyle \sum _{k=1}^K{\partial }_k}}$$

(14)

Step 4: Calculate the comprehensive weight based on the game theory.

$${\omega }^{\ast }={\displaystyle \sum _{k=1}^K{\partial }_k^{\ast }{\omega }_k^T}$$

(15)

4.4. G-TOPSIS Comprehensive Evaluation Model

The basic principle of the TOPSIS method is that the optimal solution must satisfy certain conditions, i.e., the shortest distance from the positive ideal solution and the longest distance from the negative ideal solution [37]. The limitations of TOPSIS can be compensated by defining the gray relational coefficients through the gray relational analysis (GRA) [38]. Therefore, the GRA is integrated with TOPSIS for evaluation of declared GC-DPs. The detailed procedure is summarized as follows [39].

Step 1: Construct the weighted standardized decision matrix D.

The original matrix G constructed in Equation (5) is processed as follows to obtain the normalized decision matrix.

$$g_{ij}^{*}=\left\{\begin{array}{c}\frac{g_{ij} - g_{m\min }}{g_{m\max } - g_{m\min }} \mathrm{Benefit}\text{-}\mathrm{based} \mathrm{indicators}\\ \frac{g_{m\max } - g_{ij}}{g_{m\max } - g_{m\min }} Cost\text{-}based indicators\end{array}\right.$$

(16)

$$D={(d_{ij})}_{n\times m}={(g_{ij}^{*}\cdot w_j^{*})}_{n\times m}(i=1,2,\dots ,n; j=1,2,\dots ,m)$$

(17)

Step 2: Calculate the positive ideal solution and the negative ideal solution

$$D^{\pm }=\left\{d_1^{\pm },d_2^{\pm },\cdots ,d_j^{\pm }\right\}$$

(18)

$$\begin{array}{l}{d}_j^{+}=\left\{\begin{array}{c}\max (d_{ij}) (j\in J^{+})\\ \min (d_{ij}) (j\in J^{-})\end{array}\right.\\ d_j^{-}=\left\{\begin{array}{c}\min (d_{ij}) (j\in J^{+})\\ \max (d_{ij}) (j\in J^{-})\end{array}\right.\end{array}$$

(19)

where $J^{+}$ represents benefit-based indicators and $J^{-}$ represents cost-based indicators.

Step 3: Calculate the distance to the optimal/inferior GC-DP level of the declaration.

$$l_i^{\pm }=\sqrt{{{\displaystyle \sum _{j=1}^m\left(d_{ij}-d_j^{\pm }\right)}}^2}$$

(20)

Step 4: Calculate the gray correlation coefficient between each declared project and the optimal/inferior green construction level.

$${\delta }_{ij}^{\pm }=\frac{\underset{i}{\min } \underset{j}{\min }\left|d_j^{\pm }-d_{ij}\right|+\rho \underset{i}{\max } \underset{j}{\max }\left|d_j^{\pm }-d_{ij}\right|}{\left|d_j^{\pm }-d_{ij}\right|+\rho \underset{i}{\max } \underset{j}{\max }\left|d_j^{\pm }-d_{ij}\right|}$$

(21)

where ρ represents the resolution coefficient. Considering there is a strong resolution of the construction sample data at this time, ρ = 0.5.

The gray correlation coefficient matrix for each declared item is calculated.

$${\delta }^{\pm }=\left[\begin{array}{cccc}{\delta }_{11}^{\pm }& {\delta }_{12}^{\pm }& \cdots & {\delta }_{1m}^{\pm }\\ {\delta }_{21}^{\pm }& {\delta }_{22}^{\pm }& \cdots & {\delta }_{2m}^{\pm }\\ ⋮& ⋮& ⋮& ⋮\\ {\delta }_{n1}^{\pm }& {\delta }_{n2}^{\pm }& \cdots & {\delta }_{nm}^{\pm }\end{array}\right]$$

(22)

Step 5: Calculate the degree of gray correlation between each declared project and the optimal/inferior green construction level.

$$o_i^{\pm }=\frac{1}{m}{\displaystyle \sum _{j=1}^m{\delta }_{ij}^{\pm }}$$

(23)

Step 6: Dimensionless treatment of the calculations in steps 3 and 5.

$$L_i^{\pm }=\frac{l_i^{\pm }}{\underset{i}{\max } l_i^{\pm }}$$

(24)

$$O_i^{\pm }=\frac{o_i^{\pm }}{\underset{i}{\max } o_i^{\pm }}$$

(25)

Step 7: Linear combination of the results of the dimensionless processing.

$$Z_i^{\pm }={\mu }_1{L}_i^{\pm }+{\mu }_2{O}_i^{\mp }$$

(26)

where ${\mu }_1$ represents the preference of the response decision maker and ${\mu }_2$ represents the shape preference of the sample data.

Step 8: Calculate the degree of fit of each declared project to the optimal/inferior green construction level.

$$C_i=\frac{Z_i^{-}}{Z_i^{+}+Z_i^{-}}$$

(27)

Step 9: Calculate the composite score $Q_i$ for each GC-DP filing program [40].

$$Q_i=100-{\displaystyle \sum _{i=1}^m{C}_i{f}_i/4}$$

(28)

4.5. Index Weight Calculation Results

In order to minimize the impact of experts’ professional and educational backgrounds on the results, this study conducted a rigorous screening of experts in the field of green construction. Twenty-three experts with high professional level and rich practical experience who have been engaged in green construction-related work or research for more than 3 years have been selected, as shown in Table 3. The judgment matrix constructed by 23 experts is calculated using MATLAB programming. The weights calculated by experts are then weighted and averaged. The subjective weight ${\omega }_i$ of each indicator is thus obtained and presented in

Table 6. Weight calculation results of evaluation indexes of GC-DP.

Criterion Layer		Main Factor Layer		Sub-Factor Layer
Index	Weight ${\mathit{\omega }}^{\ast }$	Index	$\mathbf{Weight} {\mathit{\omega }}^{\ast }$	Index	$\mathbf{Subjective} \mathbf{Weight} {\mathit{\omega }}_{\mathit{i}}$	$\mathbf{Objective} \mathbf{Weight} {\mathit{\omega }}_{\mathit{j}}$	$\mathbf{Comprehensive} \mathbf{Weight} {\mathit{\omega }}^{\ast }$
A	0.2661	A1	0.0425	A11	0.0185	0.0030	0.0146
				A12	0.0133	0.0088	0.0122
				A13	0.0199	0.0034	0.0157
		A2	0.0282	A21	0.0141	0.0049	0.0118
				A22	0.0083	0.0019	0.0067
				A23	0.012	0.0028	0.0097
		A3	0.0226	A31	0.0182	0.0001	0.0136
				A32	0.0113	0.0023	0.0090
		A4	0.0821	A41	0.0155	0.0011	0.0118
				A42	0.0145	0.1981	0.0611
				A43	0.0108	0.0044	0.0092
		A5	0.0582	A51	0.0119	0.0002	0.0089
				A52	0.0133	0.0011	0.0102
				A53	0.0146	0.1112	0.0391
		A6	0.0325	A61	0.0112	0.003	0.0091
				A62	0.0082	0.0001	0.0061
				A63	0.0094	0.0095	0.0094
				A64	0.0098	0.0023	0.0079
B	0.2935	B1	0.0652	B11	0.0152	0.003	0.0121
				B12	0.0157	0.0331	0.0201
				B13	0.0144	0.0049	0.0120
				B14	0.0128	0.0021	0.0101
				B15	0.0138	0.0025	0.0109
		B2	0.0792	B21	0.0128	0.0061	0.0111
				B22	0.0165	0.0336	0.0208
				B23	0.0139	0.0075	0.0123
				B24	0.0154	0.0928	0.0350
		B3	0.1094	B31	0.0168	0.0519	0.0257
				B32	0.0127	0.0029	0.0102
				B33	0.0135	0.1283	0.0426
				B34	0.0163	0.0214	0.0176
				B35	0.0167	0.0033	0.0133
		B4	0.0397	B41	0.0139	0.0016	0.0108
				B42	0.0129	0.003	0.0104
				B43	0.0115	0.0008	0.0088
				B44	0.0123	0.0021	0.0097
C	0.1452	C1	0.0426	C11	0.0163	0.0004	0.0123
				C12	0.0223	0.0037	0.0176
				C13	0.0169	0.0005	0.0127
		C2	0.0389	C21	0.0111	0.0034	0.0091
				C22	0.0196	0.002	0.0151
				C23	0.0189	0.0025	0.0147
		C3	0.0412	C31	0.0177	0.0022	0.0138
				C32	0.017	0.0022	0.0132
				C33	0.0174	0.0046	0.0142
		C4	0.0225	C41	0.0097	0.002	0.0077
				C42	0.0099	0.0023	0.0080
				C43	0.0087	0.0014	0.0068
D	0.1631	D1	0.0719	D11	0.0253	0.0484	0.0312
				D12	0.0356	0.0556	0.0407
		D2	0.0437	D21	0.0355	0.0019	0.0270
				D22	0.0222	0.0007	0.0167
		D3	0.0475	D31	0.019	0.0285	0.0214
				D32	0.0346	0.0011	0.0261
E	0.1318	E1	0.0435	E11	0.0331	0.0018	0.0252
				E12	0.0241	0.0014	0.0183
		E2	0.0555	E21	0.0298	0.0035	0.0231
				E22	0.0238	0.0577	0.0324
		E3	0.0328	E31	0.0233	0.0101	0.0199
				E32	0.0161	0.0034	0.0129

. The objective weight ${\omega }_j$ of each indicator calculated using MATLAB programming is displayed in Table 6. The subjective weight ${\omega }_i$ and objective weight ${\omega }_j$ are substituted into the game theory calculation model. The optimal combination of ${\alpha }_1=0.7462$ and ${\alpha }_2=0.2538$ is obtained after normalization. The comprehensive weight ω* of each indicator is shown in Table 6. By balancing the subjective weights obtained from FAHP through game theory and the objective weights obtained from the entropy weight method, the comprehensive weights of sub-factor indicators are obtained. The weight of the main factor is the sum of the comprehensive weights of the sub-factor indicators under it.

The importance ranking of indicators at the main factor layer is displayed in Figure 4. It can provide guidance for decision makers to make green construction decisions and manage the construction process.

5. Case Study

Four GC-DP projects in Table 3 are used for case study to testify the proposed G-TOPSIS method.

Step 1: Construct the weighted standardized decision matrix D.

$$D=\left[\begin{array}{ccccccccc}0.0146& 0.0122& 0.0157& 0.0100& \cdots & 0.0231& 0.0324& 0.0100& 0.0129\\ 0.0068& 0& 0.0105& 0.0087& \cdots & 0.0128& 0.0054& 0& 0.0063\\ 0& 0.0037& 0& 0& \cdots & 0& 0& 0& 0\\ 0.0107& 0.0079& 0.0031& 0.0118& \cdots & 0.0189& 0.0216& 0.0199& 0.0105\end{array}\right]$$

Step 2: The optimal/inferior level of green construction is calculated for the current four participating projects.

$$\begin{array}{l}{D}^{+}=\left[\begin{array}{ccccccccc}0.0146& 0.0122& 0.0157& 0.0118& \cdots & 0.0231& 0.0324& 0.0199& 0.0129\end{array}\right]\\ D^{-}=\left[\begin{array}{ccccccccc}0.0000& 0.0000& 0.0000& 0.0000& \cdots & 0.0000& 0.0000& 0.0000& 0.0000\end{array}\right]\end{array}$$

Step 3: Based on the Euclidean distance, the distance to the optimal/inferior green construction level is derived as $l^{\pm }$ for each of the four projects.

$$\begin{array}{l}{l}_i^{+}={\left[\begin{array}{cccc}0.0392& 0.0899& 0.1497& 0.0522\end{array}\right]}^T\\ l_i^{-}={\left[\begin{array}{cccc}0.1372& 0.0890& 0.0037& 0.1149\end{array}\right]}^T\end{array}$$

Step 4: Calculate the gray correlation coefficient matrix ${\delta }^{\pm }$ for the four projects and the optimal/inferior green construction level with respect to the 60 evaluation indicators.

$$\begin{array}{l}{\delta }^{+}=\left[\begin{array}{ccccccccc}1.000& 1.000& 1.000& 0.9465& \cdots & 1.000& 1.000& 0.7539& 1.000\\ 0.7976& 0.7153& 0.8536& 0.9084& \cdots & 0.7470& 0.5308& 0.6049& 0.8233\\ 0.6771& 0.7832& 0.6604& 0.7220& \cdots & 0.5692& 0.4853& 0.6049& 0.7035\\ 0.8882& 0.8784& 0.7071& 1.000& \cdots & 0.8776& 0.7388& 1.000& 0.9288\end{array}\right]\\ {\delta }^{-}=\left[\begin{array}{ccccccccc}0.6771& 0.7153& 0.6604& 0.7526& \cdots & 0.5693& 0.4853& 0.7539& 0.7035\\ 0.8176& 1.000& 0.7447& 0.7786& \cdots & 0.7051& 0.8498& 1.000& 0.8286\\ 1.000& 0.8920& 1.000& 1.0000& \cdots & 1.000& 1.000& 1.000& 1.000\\ 0.7402& 0.7939& 0.9091& 0.7220& \cdots & 0.6182& 0.5858& 0.6049& 0.7436\end{array}\right]\end{array}$$

Step 5: Calculate the degree of gray correlation $o^{\pm }$ between the four projects and the optimal/inferior level of green construction.

$$\begin{array}{l}{o}_i^{+}={\left[\begin{array}{cccc}0.9587& 0.8059& 0.6863& 0.8990\end{array}\right]}^T\\ o_i^{-}={\left[\begin{array}{cccc}0.7104& 0.8270& 0.9982& 0.7410\end{array}\right]}^T\end{array}$$

Step 6: Calculation results of the Euclidean distance $l^{\pm }$ and the degree of gray correlation $o^{\pm }$ are made dimensionless.

$$\begin{array}{l}{L}^{+}={\left[\begin{array}{cccc}0.2618& 0.6006& 1.000& 0.3486\end{array}\right]}^T\\ L^{-}={\left[\begin{array}{cccc}1.000& 0.6484& 0.0270& 0.8370\end{array}\right]}^T\\ O^{+}={\left[\begin{array}{cccc}1.000& 0.8407& 0.7131& 0.9378\end{array}\right]}^T\\ O^{-}={\left[\begin{array}{cccc}0.7116& 0.8285& 1.000& 0.7423\end{array}\right]}^T\end{array}$$

Step 7: Linear combinations of the results of dimensionless processing.

$$\begin{array}{l}{Z}_i^{+}={\left[\begin{array}{cccc}0.4867& 0.7145& 1.000& 0.5455\end{array}\right]}^T\\ Z_i^{-}={\left[\begin{array}{cccc}1.000& 0.7445& 0.3700& 0.8874\end{array}\right]}^T\end{array}$$

Step 8: Calculate the degree of fit $C_i$ with the optimal level of green construction for the current four declared projects.

$$C_i={\left[\begin{array}{cccc}0.6726& 0.5103& 0.2701& 0.6193\end{array}\right]}^T$$

Step 9: Calculate GC-DP composite score $Q_i$ for the four declared projects.

$$Q={\left[\begin{array}{cccc}90.1049& 77.3628& 62.5983& 88.1136\end{array}\right]}^T$$

From the calculation results, it can be seen that the order of the four projects affixing progress is: MJ-1 > Project 4 > Project 2 > Project 3. It reveals that MJ-1 has the highest affixing degree with the optimal green construction level.

6. Discussion

The total construction area of the MJ-1 new residential project is 142,500 square meters. The plot ratio of this project is 2.0. Additionally, this project adopts a reinforced concrete shear wall structure. The overall construction concept is “green energy saving, low carbon, low energy consumption”. The construction unit conducted a thorough green organization design in the early stages of construction. Furthermore, a number of advanced green technologies have been adopted in the construction process. All loose particles in this project were covered by dense screens. Real-time monitoring of air and noise pollution was carried out. Wastewater has been recycled, and electricity consumption has been regularly accounted for and corrected. Occupational health management measures were organized and implemented by the project manager during the construction phase. The material clerk strictly controlled the inspection of materials entering and leaving the site. BIM technology was used for pipeline delivery during the construction process, which improved the construction efficiency of complex nodes. Thus, MJ-1 has the highest affixing degree with the optimal green construction level. The evaluation result of MJ-1 is consistent with its actual performance and close to the score based on the “10 version Standard” (i.e., 97.25). This shows the rationality of the GC-DP evaluation index system and the practicality of the comprehensive evaluation method based on the game-theoretic combination of G-TOPSIS. The reason why the comprehensive score is lower than the score based on the “10 version Standard” is that the GC-DP evaluation index system considers more indicators. In addition, the proposed method restricts the influence of subjective factors on weight and better utilizes actual construction data, thereby ensuring the rationality of the results to a certain extent.

By comparing the quantitative data of a key indicator of a solution with that of a positive ideal solution, the direction for improvement of the solution can be determined. According to the evaluation results, the green construction effect of Project 3 is the worst. The problem of project 3 can be found in the gray correlation coefficient matrix of 60 evaluation indicators (i.e., Appendix A). From the matrix, it can be seen that for Project 3, among all indicators, four of them have the lowest values, namely sewage discharge (A42, 0.3333), monitoring and recording of energy consumption of transportation building materials and waste (B33, 0.4174), economic value added of green construction (D12, 0.4289), and emission of construction waste (A53, 0.4385). The results indicate that there is a significant gap between the four indicators of Project 3 and the optimal level of green construction. Based on the issues reflected by these four indicators, project 3 can take the following measures for improvement: (1) The rainwater from the construction area and the wastewater generated by vehicle flushing and dust suppression spraying should be directed to the collection pipeline network. The construction water purified by sedimentation should meet quality standards. (2) Materials should be obtained as close as possible to the project site. Except for materials that require special processing and customization, most construction materials should be produced 100 km or even closer to the site. The selection of pre-mixed concrete suppliers should be as close to the construction site as possible to shorten the delivery distance. The construction site should utilize as much waste as possible. (3) Efforts should be made to save resources and construction costs as much as possible through reasonable technological and management innovation. When selecting technical solutions, consideration should be given to the actual situation of the construction site. Incremental costs should be controlled, and quality management should be strengthened. (4) The slope should be backfilled with construction waste at the construction site. Demolished and abandoned construction components or wood and metal products should be utilized as much as possible.

The core content of the 10 version Standard mainly includes environmental protection, saving materials and material resources, water saving and water resource utilization, energy conservation and energy utilization, and land conservation and resource protection. The evaluation index system proposed in this paper takes into account social and economic factors, making it more comprehensive. In terms of weight setting, only the first-level indicators are considered in 10 version Standard, while this study refines the weight to third-level indicators. In addition, 10 version Standard adopts a linear weighting method, which directly multiplies the weight by the sum of the scores of the second-level indicators under the first-level indicator, which is relatively subjective. The G-TOPSIS comprehensive evaluation model in this study balances the influence of subjective and objective factors, making the evaluation results more reasonable. Other countries do not have separate green construction standards, and most consider them to be certain aspects of green building. Moreover, the sustainable development strategies formed by the environmental characteristics of different countries are different. The GC-DP evaluation focuses more on energy conservation and utilization in this study. However, due to the high population density in Europe, BREEAM has significantly higher requirements for land use and ecology than other countries. In addition, Australia is an arid country, and the utilization of water resources plays a prominent role in the Green Star. Therefore, when adopted by other countries, indicators and weights can be revised and adjusted according to their policies and regulatory requirements.

The framework system of the GC-DP evaluation model proposed in this study is applicable to all building types. However, when the model is specifically applied to a certain other building type, it is necessary to modify the evaluation indicators and their weights. For example, the green construction of an existing building renovation includes the demolition phase and the construction phase. Therefore, the construction unit should pay attention to the utilization of existing building demolition materials and the recycling and cleaning of construction waste in the construction management design. The construction plan for the main renovation process should also be optimized based on advanced technical level and environmental effects. Large public buildings have the characteristics of large volume, multiple design specialties, and long construction cycles, which have a great demand for water resources and material use. Therefore, attention should be paid to conserving water resources and materials when constructing large public buildings.

7. Conclusions

In this study, a comprehensive GC-DP evaluation index system was constructed by comparing and analyzing the relevant literature and standards. The system consists of five criteria layers, 20 main factors, and 60 sub-factors. 60 sub-factors correspond to 60 indicators. The subjective weights of the indicators were calculated using FAHP, and the objective weights of the indicators were calculated employing the entropy weighting method. Then, the game theory was utilized to optimize the combination of subjective and objective weights to determine comprehensive weights of indicators. The G-TOPSIS comprehensive evaluation model was constructed for GC-DP evaluation. Finally, a case study including four GC-DP projects was conducted to verify the reasonableness of the index system for GC-DP and the practicality of the G-TOPSIS comprehensive evaluation model. The deficiencies of the worst project were determined, and corresponding improvement suggestions were provided.

There are some limitations to this study. The comprehensive evaluation model of GC-DP is not applicable to all building types. Different types of buildings vary greatly in their construction methods. This paper only focuses on the green construction of new residential buildings. Public buildings, renovations of existing buildings, and other cases are not considered. Furthermore, the GC-DP evaluation indicator system in this study involves multiple indicators and is difficult to manually operate. Therefore, a GC-DP evaluation system will be developed to achieve paperless evaluation and improve the efficiency and feasibility of green building evaluation. In future work, more cases will be used for validation to explore the GC-DP evaluation model with wider applicability. In addition, differences in the selection of evaluation indicators and weights among different types of buildings will be studied, and other algorithms (such as machine learning algorithms, neural networks, and artificial intelligence techniques) will be combined to improve the accuracy of the model.