Identification of Key Drivers and Path Transmission of Carbon Emissions from Prefabricated Buildings: Based on System Dynamics

Abstract

In order to achieve the ‘dual carbon’ goal, based on the DEMATEL-ISM model, 19 main factors affecting the carbon emissions of prefabricated buildings were preliminarily identified from five dimensions, including government decision-making, technical environment, social economy, energy consumption, and market supply and demand. The logical relationship, hierarchical structure, and importance between the factors were clarified, and finally, the four influencing factors were determined. According to the causal feedback relationship between the above four factors in the system flow from 2010 to 2030, eight different control scenarios were proposed, and the impact and change trend of each control scenario on the reduction of carbon emissions of prefabricated buildings were analyzed. The research results show that the key factors for carbon emissions from prefabricated buildings include 14 outcome factors and 5 cause factors, and that the causal factors are key drivers. They are the standard specification system, the incremental cost of prefabricated buildings, investment in scientific and technological innovation, and the level of prefabricated integrated technology. The key factors were structurally stratified from the essential level to the superficial level in four tiers. The first tier of the standard specification system is the surface causal factor affecting carbon emissions from prefabricated buildings. Investment in scientific and technological innovation in the second and third tiers, and the level of prefabricated integrated technology are the causes of the transition. The incremental cost of prefabricated buildings at the fourth level is the essential causal factor. Finally, based on the data related to carbon emissions of prefabricated buildings in Yunnan, China, and verified in eight regulatory scenarios, the results of the study can effectively reveal the carbon emission reduction transmission path of prefabricated buildings, which can provide a reference for the development of prefabricated buildings and carbon emission reduction strategies.

1. Introduction

The construction industry is now facing a growing carbon emissions problem. The Research Report on China’s Building Energy Consumption and Carbon Emissions (2023) states that the total amount of carbon emissions from the whole process of buildings in China is 5.01 billion tCO₂, accounting for 47.1% of the national energy consumption carbon emissions [1]. As the industry with the highest proportion of carbon emissions, the construction industry has become an important area for energy conservation and emission reduction. Accelerating the carbon emission reduction process in the construction industry plays a significant role in achieving the dual carbon goal. At the 75th Session of the United Nations General Assembly, China officially proposed that it strives to peak carbon dioxide emissions before 2030 and achieve carbon neutrality before 2060, which is the ‘dual carbon’ goal [2]. With the advantages of green and low carbon, prefabricated buildings in the construction industry have gradually become an important direction in the transition to sustainable development. The prefabricated building was in an exploratory stage at the beginning of the founding of the country, entered a trough stage after the reform and opening up, and finally stepped into a rapid development stage in 2016. At present, the proportion of new construction areas of prefabricated buildings in China has exceeded 25%, and the contribution of controlling carbon emissions from prefabricated buildings to reducing carbon emissions from the construction industry has gradually increased. As a result, identifying the key drivers of carbon emissions from prefabricated buildings and clarifying the relationship between carbon emission factors and carbon emissions within the system can effectively reveal the critical path of carbon emission reduction in prefabricated buildings and provide a reference for prefabricated buildings’ development and carbon emission reduction strategies.

2. Literature Review

Most countries and regions have taken several measures to reduce carbon emissions from the construction sector. Firstly, the state reduces carbon emissions by issuing low-carbon policies. Jezzini et al. [3] reduced carbon emissions through low-carbon procurement policies. Lin et al. [4] believed that a carbon tax could be effective in achieving CO₂ emission reductions. Cao et al. [5] believed that the widespread adoption of low-carbon subsidy policies could curb carbon emissions. Secondly, green and low-carbon materials could be used in construction. Henrion et al. [6] combined low-implied carbon technology with concrete, thereby reducing maintenance emissions from concrete. Miller et al. [7] applied two clinker substitutions for cement, an approach that has the potential to mitigate CO₂ emissions from the global cement industry. Athira et al. [8] used straw ash as a building material. Finally, green infrastructure could be scaled up. Liu et al. [9] concluded that green roofs can be a solution to reduce carbon emissions. Manso et al. [10] used green walls to improve and restore urban environments. Coma et al. [11] concluded that a double-skin green façade has a high potential for energy savings.

In recent years, experts and scholars have conducted research on the factors affecting carbon emissions from prefabricated buildings. Zhan et al. [12] determined the main factors affecting carbon emissions in the production stage of building materials through quantitative analysis. Sun et al. [13] analyzed the factors affecting carbon emissions in the construction stage of prefabricated buildings. Wei et al. [14] studied the driving factors of carbon emissions in the construction and operation stages of buildings. The above studies mainly focus on the micro level of prefabricated buildings and analyze the factors affecting carbon emissions from different stages of the entire life cycle. However, research on the macro level of carbon emissions from prefabricated buildings is relatively scarce.

Experts and scholars have conducted a large number of empirical studies on quantitative carbon emission techniques for prefabricated buildings. Du et al. [15] evaluated the key factors affecting carbon emissions and their influence relationships by applying the structural equation model (SEM) from the perspective of the prefabricated building supply chain. Chen et al. [16] used the improved STIRPAT model to compare the factors affecting the energy consumption and carbon emission intensity of large public buildings and obtained the degree of influence of each indicator on carbon emissions. Wang et al. [17] determined the influencing factors of carbon emission reduction by enterprises by constructing a dynamic evolutionary game model between prefabricated building component manufacturers and the government. The evaluation of the influencing factors of carbon emissions of prefabricated buildings mainly adopts qualitative analysis methods, while the SEM model, STIRPAT model, game model, and other methods are complex in calculation and poor in intuitiveness, and cannot perform causal relationship analysis, sensitivity analysis and scenario regulation scheme analysis of the influencing factors.

Wang et al. [18,19,20] mainly calculated the carbon emissions of buildings at different stages based on the carbon emission factor or life cycle assessment (LCA) method. These studies took construction projects as the background that focused on comparing the differences in energy consumption and carbon emissions between prefabricated buildings and traditional cast-in-place buildings with different structural types. However, this not only needs to consider the characteristics of prefabricated buildings themselves, but also the impact of related aspects such as government decision-making, technical environment, and market supply and demand.

In summary, the factors affecting carbon emissions of prefabricated buildings are numerous and highly correlated, and it is a complex system. The DEMATEL-ISM and SD models can be used to study the carbon emission reduction path of prefabricated buildings. This study constructed an index system of factors affecting carbon emissions of prefabricated buildings through the literature survey method, explored the relationship and mechanism of action between various influencing factors, and established a hierarchical structure model on this basis. Then, according to the results of the hierarchical analysis, sensitivity indicators were set, and scenario regulation schemes were designed, respectively, and then simulation experiments were carried out on various regulation schemes in Yunnan Province from 2010 to 2030. By simulating the optimal path of carbon emission reduction of prefabricated buildings, the change range and trend of building carbon emissions could be predicted.

3. Identify Factors Affecting Carbon Emissions of Prefabricated Buildings

Questionnaires were designed to investigate the factors affecting carbon emissions of prefabricated buildings, and factors with weak influence were eliminated, corrected, and simplified. Nineteen influencing factors were selected from five dimensions, including government decision-making, technical environment, social economy, energy consumption, and market supply and demand (

Table 1. Factors affecting carbon emissions of prefabricated buildings.

Dimension	Influencing Factors	Factor Explanations
Government decision-making	Policy guidance efforts X1	Government policy guidance and publicity on the development of assembled buildings.
	Government incentives X2	Government economic subsidies, such as tax incentives and financial subsidies, for projects using assembly construction.
	Standard specification system X3	Measurement of assembly building design specifications, quality inspection of parts and components, construction quality acceptance and other processes.
Technological environment	The investment in scientific and technological innovation X4	Amount of inputs to support the development of innovative activities in science and technology in the field of assembled buildings.
	The level of prefabricated integrated technology X5	Technical level of construction activities using assembly methods, including assembly construction capacity, component production capacity, modularization design capacity, etc.
Social economy	Populations X6	Population actually and regularly residing in an area for more than six months.
	GDP X7	GDP, which expresses the level of economic development of a country or region.
	Per capita disposable income X8	Sum of income that residents have at their disposal for final consumption expenditures and savings.
	Willingness of enterprises to develop X9	Reflecting the degree of willingness of enterprises to invest in the construction of assembled buildings.
	Consumer acceptance X10	Mass acceptance of assembled buildings.
	Consumer purchasing power X11	Ability of the masses to purchase assembled buildings.
Building energy consumption	Building carbon emissions X12	Sum of greenhouse gas emissions from the production and transportation of building materials, construction and demolition, and operation phases of buildings.
	Building energy consumption X13	Quantity of building materials, such as steel and cement, and energy consumption, such as coal, oil, and natural gas, generated during the stage of assembled building construction.
	Energy intensity of assembled buildings X14	Energy consumption per unit of floor area or per unit of function over the entire life cycle of the assembled building.
	Energy carbon emission factor X15	Consumption, structural composition and share of each energy source in the production and construction process.
Market supply and demand	Demand and supply of assembled buildings X16	Consumer demand for assembled buildings, sum of supply from all producers of assembled buildings.
	The incremental cost of prefabricated buildings X17	Increased cost of assembled buildings compared to traditional cast-in-place concrete buildings.
	Assembly building quality X18	Quality and quality safety in assembled buildings.
	Scale of assembled buildings X19	Size and scale of assembled buildings.

).

Government decision-making mainly refers to the policy guidance and support provided by the government to promote the development of the prefabricated building industry. The technical environment mainly describes the technical support required for prefabricated buildings. Social economy refers to the level of social and economic development that can determine the foundation and development space of the prefabricated building industry. Building energy consumption refers to the efficiency of energy structure and building energy consumption. Market supply and demand refer to the description of the prefabricated building market environment and the scale of industrial development.

4. Construction and Analysis of DEMATEL-ISM Model

Decision-making Trial and Evaluation Laboratory (DEMATEL) is a model for visualizing complex causal relationships [21]. In this paper, DEMATEL’s research goal is to simplify the structure of carbon emission causality for prefabricated buildings using matrices and diagrams, calculating the causal degree and centrality of each influencing factor, and a multi-level hierarchical structure model was constructed to clarify the logical relationship and hierarchical structure between the factors [22], ultimately revealing the key motivating factors. The Interpretative Structural Modeling Method (ISM) is a structure for analyzing the effect of one variable on other variables [23], which consists of three modeling languages: words, series, and discrete mathematics [24]. The research goal of ISM is to decompose the carbon emission system of prefabricated buildings into four subsystems, transitioning from superficial to essential expression of their hierarchical relationships, thus facilitating the objective analysis of complex problems. As a result, the DEMATEL-ISM model can determine the visualization of causality and the hierarchical structure among the factors of carbon emissions from prefabricated buildings with a relatively small computational burden.

4.1. Determination of the Direct Impact Matrix

The expert scoring method was used for data collection, and experts with an average of 10 years of relevant experience participated in the questionnaire, including three experts from academia, two government experts from the URA, and five experts from the construction industry. The score sheet sets the factors affecting carbon emissions of prefabricated buildings were set from X₁ to X₁₉. Then, experts and relevant construction industry staff were invited to participate in the questionnaire and score the degree of influence between two factors. A 0–4 score standard was adopted (0, 1, 2, 3, and 4 represent no influence, small influence, medium influence, large influence, and strong influence, respectively). Finally, the initial direct influence matrix A was established (

Table 2. Direct impact matrix A.

	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14	X15	X16	X17	X18	X19
X1	0	2	3	2	2	1	2	2	2	3	2	3	3	3	2	3	2	3	3
X2	2	0	3	3	2	1	3	3	3	4	3	3	4	3	3	3	4	2	3
X3	3	3	0	2	4	2	2	2	3	3	3	4	3	4	3	4	3	4	4
X4	2	3	4	0	4	2	3	2	3	3	3	4	4	4	4	4	3	4	4
X5	2	2	4	3	0	3	4	2	4	4	3	4	4	4	4	4	4	4	4
X6	1	1	2	1	1	0	3	4	3	3	3	3	3	3	2	4	2	3	3
X7	2	3	2	3	2	3	0	4	3	4	4	4	4	4	3	4	2	4	4
X8	3	3	2	2	2	4	4	0	4	4	4	3	4	4	3	4	3	4	4
X9	2	3	2	2	2	2	3	4	0	3	3	4	4	4	3	4	3	4	4
X10	3	3	2	2	2	3	4	4	4	0	4	4	4	4	4	4	2	4	4
X11	2	3	2	2	2	3	4	4	3	4	0	3	3	3	3	3	3	3	4
X12	3	3	4	4	4	2	3	3	4	4	3	0	4	4	4	4	3	4	4
X13	3	3	3	4	4	3	4	4	4	4	3	4	0	4	4	4	4	4	4
X14	3	3	3	4	4	3	4	4	4	4	3	4	4	0	4	4	4	4	4
X15	2	2	2	4	4	2	3	3	4	4	3	4	4	4	0	3	3	3	3
X16	3	3	2	3	3	3	4	4	4	4	3	4	4	4	4	0	3	4	4
X17	2	4	3	4	4	3	4	4	4	4	4	4	4	4	3	4	0	4	4
X18	3	2	3	2	3	3	3	3	3	4	4	4	4	4	4	4	3	0	4
X19	3	3	3	2	3	3	3	4	3	4	4	4	4	4	4	4	2	4	0

). The $x_{ij}$ elements of matrix A indicated the degree of direct influence of the factor $x_i$ on the factor $x_j$. When $i$ = $j$, $x_{ij}$ = 0.

4.2. Normalization Directly Affects the Matrix

The row maximum method was used to normalize the direct influence matrix (Formula (1)), and the calculation results are shown in

Table 3. Standard direct impact matrix B.

	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14	X15	X16	X17	X18	X19
X1	0.000	0.029	0.044	0.029	0.029	0.015	0.029	0.029	0.029	0.044	0.029	0.044	0.044	0.044	0.029	0.044	0.029	0.044	0.044
X2	0.029	0.000	0.044	0.044	0.029	0.015	0.044	0.044	0.044	0.059	0.044	0.044	0.059	0.044	0.044	0.044	0.059	0.029	0.044
X3	0.044	0.044	0.000	0.029	0.059	0.029	0.029	0.029	0.044	0.044	0.044	0.059	0.044	0.059	0.044	0.059	0.044	0.059	0.059
X4	0.029	0.044	0.059	0.000	0.059	0.029	0.044	0.029	0.044	0.044	0.044	0.059	0.059	0.059	0.059	0.059	0.044	0.059	0.059
X5	0.029	0.029	0.059	0.044	0.000	0.044	0.059	0.029	0.059	0.059	0.044	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.059
X6	0.015	0.015	0.029	0.015	0.015	0.000	0.044	0.059	0.044	0.044	0.044	0.044	0.044	0.044	0.029	0.059	0.029	0.044	0.044
X7	0.029	0.044	0.029	0.044	0.029	0.044	0.000	0.059	0.044	0.059	0.059	0.059	0.059	0.059	0.044	0.059	0.029	0.059	0.059
X8	0.044	0.044	0.029	0.029	0.029	0.059	0.059	0.000	0.059	0.059	0.059	0.044	0.059	0.059	0.044	0.059	0.044	0.059	0.059
X9	0.029	0.044	0.029	0.029	0.029	0.029	0.044	0.059	0.000	0.044	0.044	0.059	0.059	0.059	0.044	0.059	0.044	0.059	0.059
X10	0.044	0.044	0.029	0.029	0.029	0.044	0.059	0.059	0.059	0.000	0.059	0.059	0.059	0.059	0.059	0.059	0.029	0.059	0.059
X11	0.029	0.044	0.029	0.029	0.029	0.044	0.059	0.059	0.044	0.059	0.000	0.044	0.044	0.044	0.044	0.044	0.044	0.044	0.059
X12	0.044	0.044	0.059	0.059	0.059	0.029	0.044	0.044	0.059	0.059	0.044	0.000	0.059	0.059	0.059	0.059	0.044	0.059	0.059
X13	0.044	0.044	0.044	0.059	0.059	0.044	0.059	0.059	0.059	0.059	0.044	0.059	0.000	0.059	0.059	0.059	0.059	0.059	0.059
X14	0.044	0.044	0.044	0.059	0.059	0.044	0.059	0.059	0.059	0.059	0.044	0.059	0.059	0.000	0.059	0.059	0.059	0.059	0.059
X15	0.029	0.029	0.029	0.059	0.059	0.029	0.044	0.044	0.059	0.059	0.044	0.059	0.059	0.059	0.000	0.044	0.044	0.044	0.044
X16	0.044	0.044	0.029	0.044	0.044	0.044	0.059	0.059	0.059	0.059	0.044	0.059	0.059	0.059	0.059	0.000	0.044	0.059	0.059
X17	0.029	0.059	0.044	0.059	0.059	0.044	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.000	0.059	0.059
X18	0.044	0.029	0.044	0.029	0.044	0.044	0.044	0.044	0.044	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.044	0.000	0.059
X19	0.044	0.044	0.044	0.029	0.044	0.044	0.044	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.059	0.029	0.059	0.000

.

$$B={\displaystyle \frac{A}{\underset{1\le i\le n}{\max }({\displaystyle \sum _{j=1}^n{x}_{ij}})}}$$

(1)

4.3. Calculation of the Integrated Impact Matrix

When the standard direct impact matrix was multiplied by itself, all values of the matrix would approach zero ($\underset{k\to \infty }{\lim }{B}^k=0$). The integrated influence matrix T was calculated by Formula (2). The solution is shown in

Table 4. Integrated impact matrix.

	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14	X15	X16	X17	X18	X19
X1	0.181	0.227	0.237	0.227	0.239	0.203	0.269	0.270	0.281	0.309	0.265	0.308	0.312	0.312	0.278	0.311	0.241	0.305	0.312
X2	0.246	0.239	0.277	0.282	0.282	0.241	0.333	0.334	0.347	0.377	0.327	0.363	0.381	0.368	0.343	0.366	0.311	0.346	0.367
X3	0.274	0.296	0.251	0.284	0.326	0.270	0.338	0.339	0.367	0.385	0.345	0.398	0.389	0.402	0.363	0.401	0.315	0.393	0.402
X4	0.277	0.315	0.325	0.275	0.346	0.287	0.374	0.362	0.391	0.410	0.368	0.423	0.428	0.428	0.400	0.426	0.335	0.418	0.428
X5	0.288	0.314	0.337	0.329	0.303	0.313	0.403	0.378	0.421	0.440	0.383	0.440	0.445	0.445	0.416	0.444	0.361	0.435	0.445
X6	0.204	0.222	0.231	0.222	0.233	0.198	0.295	0.310	0.307	0.321	0.290	0.320	0.324	0.325	0.289	0.337	0.250	0.317	0.325
X7	0.272	0.309	0.291	0.310	0.311	0.296	0.325	0.383	0.383	0.415	0.375	0.414	0.420	0.420	0.379	0.418	0.315	0.410	0.420
X8	0.291	0.316	0.297	0.303	0.317	0.316	0.389	0.336	0.405	0.424	0.382	0.410	0.429	0.428	0.386	0.427	0.335	0.419	0.428
X9	0.262	0.298	0.280	0.286	0.300	0.272	0.354	0.369	0.327	0.387	0.348	0.399	0.405	0.404	0.365	0.403	0.316	0.395	0.404
X10	0.292	0.316	0.298	0.304	0.318	0.303	0.389	0.392	0.406	0.370	0.383	0.424	0.430	0.429	0.400	0.428	0.322	0.419	0.429
X11	0.251	0.287	0.269	0.274	0.287	0.275	0.354	0.356	0.355	0.385	0.293	0.371	0.376	0.376	0.350	0.375	0.303	0.367	0.389
X12	0.305	0.331	0.340	0.346	0.363	0.303	0.393	0.395	0.425	0.444	0.387	0.389	0.450	0.450	0.420	0.448	0.352	0.439	0.449
X13	0.316	0.344	0.339	0.359	0.375	0.329	0.423	0.424	0.441	0.462	0.402	0.461	0.412	0.467	0.436	0.466	0.378	0.456	0.467
X14	0.316	0.344	0.339	0.359	0.375	0.329	0.423	0.424	0.441	0.462	0.402	0.461	0.467	0.412	0.436	0.466	0.378	0.456	0.467
X15	0.265	0.289	0.286	0.318	0.332	0.276	0.360	0.361	0.389	0.406	0.353	0.406	0.411	0.411	0.329	0.397	0.322	0.389	0.398
X16	0.300	0.326	0.308	0.327	0.343	0.312	0.401	0.403	0.418	0.438	0.381	0.437	0.443	0.443	0.413	0.386	0.346	0.433	0.442
X17	0.307	0.362	0.343	0.363	0.380	0.333	0.428	0.430	0.447	0.468	0.421	0.467	0.473	0.473	0.441	0.472	0.328	0.462	0.473
X18	0.289	0.300	0.309	0.301	0.330	0.300	0.372	0.374	0.389	0.421	0.379	0.420	0.425	0.425	0.397	0.424	0.332	0.360	0.425
X19	0.296	0.320	0.316	0.308	0.337	0.307	0.381	0.397	0.411	0.431	0.388	0.430	0.435	0.435	0.406	0.434	0.327	0.425	0.379

.

$$T=(B+B^2+\cdots +B^k)={\displaystyle \sum _{k=1}^{\infty }{B}^k}=B{(I-B)}^{-1}$$

(2)

where I is the unit matrix.

4.4. Calculation the Degree of Influencing, Influenced, Centrality and Causal Contribution

The degree of influencing refers to the sum of the rows in the matrix T, which represents the value of the combined influence of the elements of each row on all other elements, denoted as $D_i$, then:

$$D_i={\displaystyle \sum _{j=1}^n{x}_{ij}},(i=1,2,\dots ,n)$$

(3)

The degree of being influenced refers to the sum of the columns in the matrix T and represents the value of the combined influence of the elements in each column on all other elements, denoted as $C_i$, then:

$$C_i={\displaystyle \sum _{j=1}^n{x}_{ij}},(i=1,2,\dots ,n)$$

(4)

The degree of centrality is the position of the factor in the evaluation system and the magnitude of the role it plays, and the degree of influence and the degree of influence of the factor are added together to obtain the degree of centrality, which is noted as $M_i$, then:

$$M_i=D_i+C_i$$

(5)

The degree of causal contribution is the result of subtracting the degree of influence and the degree of influence, denoted as $R_i$, then:

$$R_i=D_i-C_i$$

(6)

If the degree of causal contribution is greater than zero, it means that the factor has a great influence on other factors and is called a causal factor. Conversely, if the degree of causal contribution is less than zero, it means that the factor has a small influence on other factors and is called a result factor [25]. The calculation results are shown in

Table 5. Analysis of DEMATEL calculation results of factors affecting carbon emissions of prefabricated buildings.

Xi	Influencing Factors	Influence	Influenced	Centrality	Order of Centrality	Cause	Factor Attributes
X1	Policy guidance efforts	5.086	5.231	10.317	19	−0.145	Result factor
X2	Government incentives	6.130	5.755	11.885	17	0.375	Causal factor
X3	Standard specification system	6.540	5.671	12.211	16	0.868	Causal factor
X4	The investment in scientific and technological innovation	7.015	5.776	12.792	15	1.239	Causal factor
X5	The level of prefabricated integrated technology	7.342	6.098	13.441	13	1.244	Causal factor
X6	Populations	5.319	5.463	10.782	18	−0.144	Result factor
X7	GDP	6.866	7.002	13.869	12	−0.136	Result factor
X8	Per capita disposable income	7.039	7.040	14.078	8	−0.001	Result factor
X9	Willingness of enterprises to develop	6.573	7.349	13.922	11	−0.776	Result factor
X10	Consumer acceptance	7.052	7.754	14.806	6	−0.703	Result factor
X11	Consumer purchasing power	6.292	6.871	13.163	14	−0.579	Result factor
X12	Building carbon emissions	7.429	7.742	15.171	3	−0.313	Result factor
X13	Building energy consumption	7.756	7.855	15.611	1	−0.099	Result factor
X14	Energy intensity of assembled buildings	7.756	7.854	15.610	2	−0.098	Result factor
X15	Energy carbon emission factor	6.696	7.246	13.942	10	−0.550	Result factor
X16	Demand and supply of assembled buildings	7.301	7.829	15.130	4	−0.528	Result factor
X17	The incremental cost of f prefabricated buildings	7.872	6.167	14.039	9	1.705	Causal factor
X18	Assembly building quality	6.970	7.645	14.616	7	−0.675	Result factor
X19	Scale of assembled buildings	7.163	7.849	15.011	5	−0.686	Result factor

, and by sorting according to the centrality, the causal factors were (X₁₇, X₅, X₄, X₃, X₂) and the result factors were (X₁₃, X₁₄, X₁₂, X₁₆, X₁₉, X₁₀, X₁₈, X₈, X₁₅, X₉, X₇, X₁₁, X₆, X₁).

4.5. Cause-Effect Diagram

A Cartesian rectangular coordinate system was established with centrality $M_i$ as the horizontal coordinate and causality $R_i$ as the vertical coordinate to draw a cause-effect relationship diagram (Figure 1). Since centrality represented the importance of a factor, it could be seen from the figure that the first quadrant had high causality and centrality, the second quadrant had high causality and low centrality, the third quadrant had low causality and centrality, and the fourth quadrant had low causality and high centrality.

4.6. Computing the Reachability Matrix

Calculate the overall impact matrix $E={[e_{ij}]}_{n\times n}=I+T$ based on the synthesis matrix T.

Threshold $\lambda$ was introduced to exclude influence relationships with lesser degrees of influence, by calculating the mean $\alpha$ and standard deviation $\beta$ of the overall influence matrix E, which were summed to give the threshold $\lambda$.

$$\lambda =\alpha +\beta$$

(7)

where $\alpha$ = 0.361, $\beta$ = 0.063, $\lambda$ = 0.424.

Calculate the reachability matrix $F={[f_{ij}]}_{n\times n}$ (

Table 6. The reachability matrix F.

	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14	X15	X16	X17	X18	X19
X1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
X2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
X3	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
X4	0	0	0	1	0	0	0	0	0	0	0	0	1	1	0	1	0	0	1
X5	0	0	0	0	1	0	0	0	0	1	0	1	1	1	0	1	0	1	1
X6	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0
X7	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0
X8	0	0	0	0	0	0	0	1	0	1	0	0	1	1	0	1	0	0	1
X9	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0
X10	0	0	0	0	0	0	0	0	0	1	0	0	1	1	0	1	0	0	1
X11	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0
X12	0	0	0	0	0	0	0	0	1	1	0	1	1	1	0	1	0	1	1
X13	0	0	0	0	0	0	0	1	1	1	0	1	1	1	1	1	0	1	1
X14	0	0	0	0	0	0	0	1	1	1	0	1	1	1	1	1	0	1	1
X15	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0
X16	0	0	0	0	0	0	0	0	0	1	0	1	1	1	0	1	0	1	1
X17	0	0	0	0	0	0	1	1	1	1	0	1	1	1	1	1	1	1	1
X18	0	0	0	0	0	0	0	0	0	0	0	0	1	1	0	0	0	1	1
X19	0	0	0	0	0	0	0	0	0	1	0	1	1	1	0	1	0	1	1

), then:

$$f_{ij}=\{\begin{array}{c}1,e_{ij}\ge \lambda (i,j=1,2,\dots ,n)\\ 0,e_{ij}<\lambda (i,j=1,2,\dots ,n)\end{array}$$

(8)

where $f_{ij}$ is the factor in the reachability matrix F and $e_{ij}$ is the factor in the overall impact matrix E.

4.7. Building Hierarchical Model

According to the reachable matrix F, the regions and levels were divided, and the reachable set $R(x_i)$, the antecedent set $S(x_i)$, and the intersection of the two were determined, respectively, where $R(x_i)$ represents the set of factors that influence $S(x_i)$ on other factors. The calculation formula is:

$$R(x_i)=\left\{x_i\right|F_{ij}=1\}(i,j=1,2,\dots ,n)$$

(9)

$$S(x_i)=\left\{x_i\right|F_{ij}=1\}(i,j=1,2,\dots ,n)$$

(10)

When $R(x_i)\cap S(x_i)=R(x_i),(i=1,2,\dots ,n)$, it means that all the corresponding factors $x_i$ in $R(x_i)$ can find antecedents in $S(x_i)$. Then $R(x_i)$ is taken as the highest level factor, and then the corresponding rows and columns in $R(x_i)$ are deleted from the reachable matrix F. This operation should be repeated until all the rows and columns have been crossed out. The specific division is shown in

Table 7. Classification of factors affecting carbon emissions of prefabricated buildings.

Xi	Reachable Set	Antecedent Set	Intersection
X1	(1)	(1)	(1)
X2	(2)	(2)	(2)
X3	(3)	(3)	(3)
X4	(4, 13, 14, 16, 19)	(4)	(4)
X5	(5, 10, 12, 13, 14, 16, 18, 19)	(5)	(5)
X6	(6)	(6)	(6)
X7	(7)	(7, 17)	(7)
X8	(8, 10, 13, 14, 16, 19)	(8, 13, 14, 17)	(8, 13, 14)
X9	(9)	(9, 12, 13, 14, 17)	(9)
X10	(10, 13, 14, 16, 19)	(5, 8, 10, 12, 13, 14, 16, 17, 19)	(10, 13, 14, 16, 19)
X11	(11)	(11)	(11)
X12	(9, 10, 12, 13, 14, 16, 18, 19)	(5, 12, 13, 14, 16, 17, 19)	(12, 13, 14, 16, 19)
X13	(8, 9, 10, 12–16, 18, 19)	(4, 5, 8, 10, 12, 13, 14, 16, 17, 18, 19)	(8, 10, 12, 13, 14, 16, 18, 19)
X14	(8, 9, 10, 12–16, 18, 19)	(4, 5, 8, 10, 12, 13, 14, 16, 17, 18, 19)	(8, 10, 12, 13, 14, 16, 18, 19)
X15	(15)	(13, 14, 15, 17)	(15)
X16	(10, 12, 13, 14, 16, 18, 19)	(4, 5, 8, 10, 12, 13, 14, 16, 17, 19)	(10, 12, 13, 14, 16, 19)
X17	(7–10, 12–19)	(17)	(17)
X18	(13, 14, 18, 19)	(5, 12, 13, 14, 16, 17, 18, 19)	(13, 14, 18, 19)
X19	(10, 12, 13, 14, 16, 18, 19)	(4, 5, 8, 10, 12, 13, 14, 16, 17, 18, 19)	(10, 12, 13, 14, 16, 18, 19)

. As shown in the table, the final stratification result is L1 = {1, 2, 3, 6, 7, 9, 10, 11, 15, 18, 19}, L2 = {12, 13, 14, 16}, L3 = {4, 5, 8}, L4 = {17}.

According to the elements at each level, a hierarchical structure model of factors affecting carbon emissions of prefabricated buildings was constructed, as shown in Figure 2.

The causal factors obtained from Table 5 were sorted according to centrality. The top four factors were selected as the influencing factors of carbon emissions of prefabricated buildings, which were the incremental cost of prefabricated buildings, the integrated technology level of prefabricated buildings, the investment in scientific and technological innovation, and the standard and specification system.

As can be seen in Figure 2, there is a complex interrelationship between the carbon emission influencing factors of prefabricated buildings. The incremental cost of prefabricated buildings will affect the preparation of the standard specification system, the capital investment in scientific and technological innovation, and the improvement of the level of prefabricated integrated technology, thus affecting the scale of prefabricated buildings. Therefore, the incremental cost of prefabricated buildings is a tier 4 factor and is the essential causal factor affecting carbon emissions from prefabricated buildings. Both the investment in scientific and technological innovation and the level of prefabricated integrated technology affect the amount of carbon emissions from buildings, the amount of energy consumed in buildings, and the energy intensity of prefabricated buildings, further affecting the magnitude of the change in carbon emissions. Therefore, the investment in scientific and technological innovation and the level of prefabricated integrated technology are the tier 3 factors, which are the transitional causal factors affecting the carbon emissions of prefabricated buildings. The standard specification system is a level 1 factor that directly affects the willingness of companies to implement management and carbon emission reduction, and it is a surface causal factor that affects the carbon emission of prefabricated buildings.

5. Simulation and Analysis of Carbon Emission Paths Based on SD Modeling

System dynamics (SD) models are mathematical models and computer simulations designed to aid in the understanding of complex nonlinear dynamic systems [26], which consist of three components: feedback loops, stocks, and flows [27]. The SD models select the time series data from 2010–2030 in Yunnan, China, and consider the social, political, technological, energy consumption, and market aspects, verify the interconnections among the key drivers affecting carbon emissions from prefabricated buildings through simulation modeling, and scientifically propose the emission reduction path.

The following basic assumptions are made for modeling the carbon emission system dynamics of assembled buildings:

• It is assumed that the overall level of socio-economic development and the total population of Yunnan Province maintains a steady growth, and that the carbon emission reduction work of various industries is being continuously and steadily promoted in the process, with no unexpected situations;
• During the study timeframe, it is assumed that the construction industry in Yunnan Province remains sustainable;
• The energy consumption of prefabricated buildings only considers oil, coal, natural gas, and electricity resources, as well as carbon emissions, as reference indicators of energy consumption, mainly in the form of carbon dioxide;
• The comparative object of the study in different scenario models is set to be both prefabricated buildings and traditional cast-in-place buildings.

5.1. Constructing System Flow Diagrams

Data related to carbon emissions from prefabricated buildings in Yunnan, China from 2010 to 2030 were used, and the SD model step was set to 1 year. From the five dimensions of government decision-making, technological environment, social economy, building energy consumption, and market supply and demand, the overall boundaries of the carbon emission system of prefabricated buildings are constructed as shown in Table 1, and the regulatory variables are set as shown in

Table 8. Sensitivity analysis of regulatory parameters.

Systems	Variant	Initial Value	Magnitude of Control
Government decision-making	Standardized system	0.63	−10% 10%
Market supply and demand	Incremental cost of prefabricated buildings	439.4	−10% 10%
Technological environment	Investment in scientific and technological innovation	1.06	10% 20%
	Level of prefabricated integrated technology	0.37	10% 20%

.

The data in the Yunnan Statistical Yearbook, China Statistical Yearbook, China Environment Statistics, and China Energy Data Report (2024) were mainly selected for value assignment and parameter setting. The specific system flow diagram is shown in Figure 3.

5.2. Selection of Scenario Parameters

The DEMATEL-ISM model was used to screen out four key influencing factors, regulate the key factors, observe the response of the entire system to the regulation behavior, and judge whether the desired goal was achieved.

Four parameters on the following were regulated, including standard and specification system, incremental cost of prefabricated buildings, investment in scientific and technological innovation, and prefabricated integrated technology level. The initial values of each variable are shown in Table 8.

5.3. Univariate Sensitivity Analysis

Sensitivity analysis is the process of determining how sensitive the model is to changes in parameter values and identifying leverage points in the model where a particular value of that parameter can have a significant impact on the behavioral patterns of the system [28]. By adjusting the four key drivers, namely, standard specification system, incremental cost of prefabricated buildings, investment in scientific and technological innovation, and level of prefabricated integrated technology, we will feedback on the sensitivity of the change in the magnitude of prefabricated carbon emissions, observe the conduction paths within the prefabricated carbon emission system, and obtain more scientific and reasonable emission reduction measures and pathways.

5.3.1. Standardized System

By modulating the variables of the standard specification system, the degree of change in carbon emission reduction under different scenario plans was observed. The simulation results are shown in Figure 4.

In the high-consumption scenario, where the standard system was reduced by 10%, the carbon emission reduction in 2030 would reach 64,987,770 tons, which was 330 tons less than the baseline scenario with a decrease of about 0.001%. In the low-carbon green scenario, where the standard system was improved by 10%, the carbon emission reduction in 2030 would reach 64,988,600 tons, which was 500 tons more than the baseline scenario, with an increase of about 0.001%.

The results have shown that changes in the standard system would directly affect the reduction of carbon emissions, and had a more obvious role in promoting the promotion and market incentives of prefabricated buildings.

5.3.2. Incremental Costs of Prefabricated Buildings

By modulating the incremental cost variables of prefabricated buildings, the degree of change in carbon emission reduction under different scenario plans was observed. The simulation results are shown in Figure 5.

In the low-carbon green scenario, where the incremental cost of prefabricated buildings was reduced by 10%, the carbon emission reduction in 2030 would reach 104,748,000 tons, with an increase of 29.41 million tons compared with the baseline scenario with an increase of about 39.03%. In the high-consumption scenario, where the incremental cost of prefabricated buildings was increased by 10%, the carbon emission reduction in 2030 would reach 64,988,100 tons, with a decrease of 10.35 million tons compared with the baseline scenario with a decrease of about 13.74%.

The results have shown that the increase in incremental costs might lead to a reduction in the supply of prefabricated buildings by construction-related parties, thereby reducing consumer demand for prefabricated buildings. The investment willingness of prefabricated building component-related production factories for prefabricated building projects would also be low, and they would choose traditional construction projects with low interest rates and high profits, which could ultimately lead to the emission reduction potential of prefabricated buildings failing to meet the expected standards.

5.3.3. Scientific and Technological Innovation Investment

By modulating the variables of scientific and technological innovation investment, the degree of change in carbon emission reduction under different scenarios was observed, and the simulation results are shown in Figure 6.

In the low-carbon green scenario, with a 10% increase in investment in scientific and technological innovation, carbon emissions would be reduced by 64,989,500 tons in 2030 (with an increase of 1400 tons) over the baseline scenario, with an increase of about 0.001%. In the enhanced low-carbon green scenario, with a 20% increase in investment in scientific and technological innovation, carbon emissions would be reduced by 64,990,800 tons in 2030 (with an increase of 2700 tons) over the baseline scenario (with an increase of about 0.04%).

The results have shown that as one of the main factors affecting the level of prefabricated innovation technology, the current low investment in technological innovation has hindered the technical level and development of prefabricated buildings to a certain extent. The effect of reducing the amount was relatively small, and the promoting effect was not significant, but it still had a driving effect on the enthusiasm of relevant scientific researchers and the transformation of scientific research results.

5.3.4. Prefabricated Integrated Technology Level

By modulating prefabricated integrated technology level, the degree of change in carbon emission reduction under different scenarios was observed. The simulation results are shown in Figure 7.

In the low-carbon green scenario, with a 10% improvement in the level of prefabricated integrated technology, the carbon emissions reduction in 2030 would reach 64,990,200 tons (with an increase of 2100 tons) over the baseline scenario (with an increase of about 0.003%). In the enhanced low-carbon green scenario, with a 20% improvement in the level of prefabricated integrated technology, the carbon emissions reduction in 2030 would reach 64,992,200 tons (with an increase of 4100 tons) over the baseline scenario (with an increase of about 0.006%).

The results have shown that the improvement of the level of prefabricated technology could effectively reduce the consumption of building materials and energy required for prefabricated building projects, thereby reducing carbon emissions and ultimately expanding the scale of prefabricated building development.

6. Analysis of Emission Reduction Paths and Selection of Energy-Saving Measures

6.1. Accelerate the Construction of Standard and Specification System

• Promote the formulation of the full life cycle evaluation system of prefabricated buildings, including design, construction, acceptance, and other links, to ensure that prefabricated buildings have a clear standard basis throughout their life cycle.
• Establish a supervision and inspection mechanism for the implementation of prefabricated building standards to ensure that various standards are effectively implemented and severely punish violations of standards, safeguard fair competition in the market, and protect the legitimate rights and interests of consumers.
• Formulate policies and regulations that are conducive to the promotion of prefabricated buildings, provide tax incentives and financial subsidies for projects that meet prefabricated building standards, and encourage enterprises and developers to actively adopt prefabricated building technology, as well as giving priority to the approval of prefabricated building projects and providing policy guarantees for the development of prefabricated buildings.

6.2. Reduce Incremental Costs

• By promoting the standardization and modular production of prefabricated building components, the demand for customized components can be reduced, and production costs can be reduced. Standardized component design can be reused in multiple projects, improve economies of scale, and reduce energy waste caused by repeated design and production.
• By optimizing the construction process, reducing wet operations (such as concrete pouring), and using efficient construction machinery, the incremental cost in the construction process can be reduced, the equipment operation time can be reduced, and energy consumption (such as electricity, fuel, etc.) can be reduced.

6.3. Increase Investment in Scientific and Technological Innovation

• Develop low-carbon, renewable, and environmentally friendly building materials, such as recycled concrete, low-carbon cement, lightweight and high-strength materials, and recyclable modular materials, which can reduce dependence on traditional high-carbon materials.
• The government and enterprises should increase investment in scientific research projects in the field of prefabricated buildings, build a platform for industry–university–research cooperation, promote close cooperation between universities, research institutes, and enterprises, and establish a special fund for prefabricated building science and technology innovation to support key scientific research projects and key technology research.
• Promote the application of building information modeling (BIM) and digital twin technology in prefabricated buildings, and realize the digitalization of the entire process of building design, construction, and management.

6.4. Improving the Level of Prefabricated Technology

• Formulate relevant standards and specifications to provide clear guidance and a basis for the construction of prefabricated buildings, clarify the technical requirements and capabilities that construction personnel need to have, and ensure the construction quality and safety of prefabricated buildings.
• Train construction personnel to master the operation skills of BIM software (such as Revit 2023), which can comprehensively manage and optimize the entire prefabricated building construction process; optimize the supply chain management, logistics scheduling and on-site construction of prefabricated buildings, which can reduce transportation and on-site energy consumption; integrate renewable energy systems (such as solar photovoltaic, wind energy, ground source heat pumps, etc.) into prefabricated buildings to reduce the dependence on traditional fossil fuels during the operation stage of buildings.
• Establish honorary titles or awards for outstanding prefabricated building construction personnel to encourage them to actively cultivate innovative awareness and technical capabilities in their positions, play a leading role in demonstration, and stimulate the enthusiasm and creativity of more construction personnel.

7. Conclusions

Previous studies on carbon emissions from prefabricated buildings have focused on calculating carbon emissions from the building itself from a microscopic point of view, with complex and poorly visualized data, and a lack of hierarchical progression between factors and simulation of dynamic adjustment schemes. In this study, we construct the DEMATEL-ISM evaluation model, macroscopically analyze the logical relationship and hierarchical structure among the influencing factors of carbon emissions from prefabricated buildings, and conclude that the key drivers of carbon emissions from prefabricated buildings are the standard specification system, incremental cost of prefabricated buildings, investment in scientific and technological innovation, and the level of prefabricated integrated technology, and utilized the SD model to conduct a simulation of the carbon emission conduction scenario regulation scheme. The SD model is used to simulate and validate the carbon emission conduction scenarios, to investigate the magnitude and trend of the impact of the regulation scheme on the reduction of building carbon emissions under different parameters, and, finally, to clarify the emission reduction conduction path. This provides improvement measures and realistic references to promote the long-term, low-carbon, and healthy development of the prefabricated building market. However, there are still some challenges in the current study, such as the need to adjust the parameters according to the actual situation due to the differences in the development of prefabricated buildings in different countries and regions, and the subjectivity in the determination of the target weights for different decision models. Future research can be devoted to solving these problems, further improving the carbon emission decision-making and optimization system for prefabricated buildings, and promoting the development of the construction industry in a more efficient and sustainable direction.