Prediction of Construction Waste Generation in China Based on Grey Model and Management Recommendations

Abstract

As urbanization and construction activities in China continue to accelerate, the management of construction waste has become crucial. This study comprehensively investigated the current status and challenges in construction waste management in China. Through the application of building area estimation methodology combined with the Grey Prediction GM (1,1) model, we analyzed historical waste generation patterns from 2000 to 2022 and projected future trends for the next 10 years. The results revealed significant regional disparities in waste generation, with the East China region contributing over 50% of the national total, while maintaining continuous growth. National construction waste generation is projected to reach 3.084 billion tons in 2032, highlighting escalating management challenges. This study identified several critical issues in China’s current management system, including incomplete statistical data, weak implementation of source reduction measures, underdeveloped classification systems, and a notably low resource utilization rate (below 10% as of 2022). Drawing on successful international practices and domestic pilot experiences, we proposed a comprehensive management framework emphasizing full-process supervision, enhanced data collection systems, improved classification management, advanced resource utilization technologies, and strengthened policy mechanisms. These proposals will foster the development of sustainable construction waste management in China’s transition, in parallel with the realization of circular economy principles within the construction sector.

1. Introduction

Construction waste management and disposal have become important issues in China’s urbanization process [1]. Construction waste generally refers to the solid waste that consists of discarded soil, materials, and other solid wastes generated from the construction, reconstruction, expansion, and demolition of any structure, building, or infrastructure plus residential renovation activities [2]. With the ever-building economy and urbanization in China, the creation of construction waste has been rapidly increasing and thus constitutes 30–40% of total urban waste [3,4]. As the demand for renewing dilapidated urban areas, especially by refurbishing old housing, increases, the generation rate of construction waste will continue to rise in years to come [5]. The volume of building waste in China is already high and varies greatly due to its breadth of sources and diverse composition. Mistreatment and mismanagement result in adverse environmental, economic, and social impacts [6,7]. Therefore, addressing the challenges posed by construction waste has become urgent.

China has attached much importance to the exploration of the construction waste management system and policy. In the 14th Five-Year Plan for Construction of Waste-Free Cities by China’s Ministry of Ecology and Environment in 2021, better statistical methods for building waste and the promotion of comprehensive usage were stressed [8,9]. In the 14th Five-Year Plan for Circular Economy Development, the National Development and Reform Commission stated that the recycling rate of construction waste must reach 60% by the year 2025 [10]. Rapidly increasing importance is attached to cutting back on construction waste as well. In the year 2020, China’s Ministry of Housing and Urban–Rural Development initiated the “Guidelines on Promotion of Construction Waste Reduction”. In the aftermath, cities like Beijing, Shanghai, and Guangzhou have come forward to introduce specific or updated policies on waste reduction [11]. This has further enriched the Chinese construction waste management system, which in turn stands to uplift the efficiency of using resources, foster a circular economy, and hasten the formation of cities that do not generate waste. This is in line with the U.N. Sustainable Development Goals developed in 2015, specifically Goal 9 (Sustainable Infrastructure), Goal 11 (Sustainable Cities and Communities), and Goal 12 (Responsible Consumption and Production) [12].

However, technical and managerial gaps prevail in the area of construction waste management. Most of the areas are still devoid of an all-inclusive collection and monitoring system. Hence, there is not enough accurate and timely generation of data about quantities of construction waste. Presently, this kind of situation severely hampers the feasibility of making a correct estimation of future waste volumes and, therefore, the ultimate management and utilization of resources. Additionally, the rate of resource utilization remains low in China. While developed countries in Europe and the US have achieved construction waste recycling rates exceeding 90%, China still predominantly relies on landfill or stockpiling, with the resource utilization rate below 10% [13,14,15]. As a result, urgent optimization of the management system is required.

In this study, we systematically reviewed the current situation and deficiencies in construction waste management. Using the building area estimation method and Grey Prediction Model, the annual generation of construction waste in China since 2000 was estimated, along with projections for seven administrative regions (excluding Hong Kong, Macau, and Taiwan). The future generation of construction waste over the next decade was also forecasted. Based on these findings, recommendations were put forward to optimize the construction waste management system, aiming to provide data support and a theoretical foundation for the improvement of construction waste management in China.

2. Methods

2.1. Estimation Methods and Data Sources for Construction Waste Generation

The analysis of construction waste generation in China presented unique challenges related to data availability and quality. This study primarily drew from the China Statistical Yearbook (2000–2023), which provides authoritative records of construction activities nationwide. While this source maintains rigorous data validation processes, the evolution of China’s construction waste statistical system had created temporal variations in data quality. The implementation of the Technical Standard for Construction Waste Treatment (CJJ 134-2009 [16]) in 2019 marked a significant improvement in data collection consistency, particularly enhancing the reliability of recent data.

Regional variations in reporting capabilities presented another consideration in data quality assessment. More economically developed regions typically maintained more comprehensive data collection systems, potentially influencing the observed patterns in waste generation across regions. To address these variations and ensure reliable analysis, this study employed multiple validation approaches and consistent methodology across all regions. The lack of a standardized statistical system for construction waste in China has resulted in data being predominantly sourced from provincial and municipal reports. This data limitation has led to discrepancies in estimates of national construction waste production, though there is general agreement that annual production falls within the billion-ton range. To address this statistical challenge, the Ministry of Housing and Urban–Rural Development introduced the Technical Standard for Construction Waste Treatment (CJJ 134-2009) in 2019, which proposed the building area estimation method as a standardized approach for regions without direct statistical data. This estimation method provided a systematic framework for calculating construction waste production by incorporating three key components: construction waste from new buildings (P₁), demolition waste (P₂), and renovation waste (P₃). The method utilized readily available parameters such as newly constructed floor area, demolition area, and number of residential households, applying specific conversion coefficients to estimate waste generation across different categories. The total construction waste production (P) was calculated as the sum of these three components, as follows [17,18]:

$$P=p_1+p_2+p_3$$

(1)

$$p_1=m_1\times i_1$$

(2)

$$p_2=m_2\times i_2$$

(3)

$$p_3=m_3\times i_3$$

(4)

where

$p_1$—annual generation of building construction waste, 10⁴ t;

$m_1$—annual construction area of the building, taken from the completed area of the house in the data of the National Bureau of Statistics [18], 10⁴ m²;

$i_1$—annual generation of construction waste per unit construction area [19], 0.055 t/m²;

$p_2$—annual generation of building demolition waste, 10⁴ t;

$m_2$—annual area of building demolition, taken as 10 per cent of the area of housing construction in the data of the National Statistical Office [18], 10⁴ m²;

$i_2$—annual generation of construction waste per unit of demolition area [20], 1.3 t/m²;

$p_3$—annual generation of building renovation waste, 10⁴ t;

$m_3$—annual renovation area of the building, taken as 10 per cent of the completed area of the house in the data of the National Bureau of Statistics [18], 10⁴ m²;

$i_3$—annual generation of construction waste per unit renovation area [21], 0.1 t/m².

This study analyzed data from the China Statistical Yearbook (2000–2023), focusing on floor area metrics of construction and completed housing. Floor area under construction included new housing starts, carried-over construction, resumed projects, completed housing, and suspended construction during the reporting period. Completed floor area encompassed buildings that met design requirements, passed inspections, and were ready for occupancy. To investigate the potential correlation between construction waste generation and economic development, as suggested by the “Prevention and Control Annual Report”, which indicated higher waste generation in larger, faster-growing cities, the analysis was conducted both nationally and regionally [22]. We employed China’s well-established seven-region classification system, a framework widely recognized in national policy-making and academic research for its effectiveness in capturing economic and administrative distinctions across the country. This classification system, encompassing 32 provinces, municipalities, and autonomous regions (excluding Hong Kong, Macao, and Taiwan), provided a comprehensive foundation for analyzing spatial variations in construction waste generation. The East China region (Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, and Shandong) represented the most economically developed coastal area with intensive urbanization and substantial construction activities, while the South China region (Guangdong, Guangxi, and Hainan) featured rapid urban expansion, particularly in the Greater Bay Area. The North China region (Beijing, Tianjin, Hebei, Shanxi, and Inner Mongolia) combined highly urbanized municipalities with resource-rich provinces, and the Central China region (Henan, Hubei, and Hunan) served as an emerging economic center experiencing accelerated infrastructure development. The Southwest region (Chongqing, Sichuan, Guizhou, Yunnan, and Tibet) presented unique challenges due to its mountainous terrain, while the Northeast region (Liaoning, Jilin, and Heilongjiang) represented traditional industrial bases undergoing transformation. The Northwest region (Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang) featured vast territories where construction activities were heavily influenced by infrastructure development projects. This regional framework, aligned with national development strategies and policy implementation mechanisms, enabled targeted analysis of waste generation patterns by accounting for variations in economic development levels, urbanization rates, construction practices, and local policies that collectively influence waste management approaches.

2.2. Prediction of Construction Waste Generation Based on the Grey Model

The selection of an appropriate forecasting model is crucial for accurate prediction of construction waste generation. While various prediction methods exist, including artificial intelligence-driven models, hybrid models, and traditional time series approaches like Autoregressive Integrated Moving Average (ARIMA), the Grey Prediction Model GM (1,1) was chosen for this study based on several key considerations. AI-driven predictive modeling, while powerful in many fields, requires substantial historical data to capture complex non-linear relationships and provide accurate predictions [23]. For instance, neural network models and random forest algorithms, despite their advanced capabilities, are less suitable for scenarios with limited data and insufficient explanatory variables [24,25]. While effective for time series forecasting, ARIMA assumes linear relationships and requires stationary data series with consistent statistical properties, making it less suitable for the complex patterns in construction waste generation data.

The Grey Prediction Model, established by Chinese scholar Ju-long Deng in 1982, provides distinct advantages for construction waste forecasting. This model excels in processing small sample sizes and incomplete information systems, making it particularly suitable for China’s construction waste data characteristics. Additionally, the model’s accuracy can be significantly enhanced through parameter optimization techniques [26,27] while maintaining computational efficiency compared to more complex models like ARIMA, which require extensive parameter selection.

The model operates by uncovering inherent patterns in data variation through correlation analysis of original sequences, generating regularized data sequences. Through the establishment of corresponding differential equations, the Grey Model enables the prediction of future development trends. This forecasting methodology is particularly suitable for analyzing systems with incomplete information or uncertain factors, making it an appropriate tool for construction waste generation prediction where comprehensive historical data may be limited.

2.2.1. Quasi-Smoothness Test and Quasi-Exponential Law Test

To determine the applicability of the Grey Prediction Model GM (1,1) for forecasting future construction waste generation, the model validation process requires compilation of historical waste production data to establish a discrete sequence. This sequence must then undergo quasi-smoothness and quasi-exponential regularity testing to verify its suitability for grey prediction modeling. The discrete original sequence of annual construction waste production was established as $X^{(0)}$ = {$X^{(0)}$(1), $X^{(0)}$(2),…, $X^{(0)}$(n)}. Through accumulation, a new sequence $X^{(1)}$ = {$X^{(1)}$(1), $X^{(1)}$(2),…, $X^{(1)}$(n)} was generated. The quasi-smoothness test was then conducted on the original data sequence $X^{(0)}$ using the ratio $p(t)={\displaystyle \frac{X^{(0)}(t)}{X^{(1)}(t-1)}}$, where t > 3 with the criterion that $p(t)$ should be less than 0.5 to satisfy the quasi-smoothness condition. Subsequently, a quasi-exponential regularity test was performed on $X^{(1)}$ using the ratio ${\sigma }^{(1)}(t)={\displaystyle \frac{X^{(1)}(t)}{X^{(1)}(t-1)}}$, where t > 3, and the ratio should fall between 1 and 1.5 to meet the quasi-exponential regularity condition. When both test criteria were satisfied, the Grey Prediction Model was validly established for forecasting construction waste production.

2.2.2. Establishing the Time Response Equation

After validation through consistency testing, the GM (1,1) model’s differential equation was established as

$${\displaystyle \frac{d{X}^{(1)}}{dt}}+a{X}^{(1)}=u$$

(5)

where $a$ represents the development coefficient and $u$ represents the internal control coefficient.

Equation (5) satisfied the initial condition, when $t=t_{(0)}$, and the solution for $X^{(1)}=X_{\left(t_0\right)}^{(1)}$ was

$$X_t^{(1)}=\left[X_{\left(t_0\right)}^{(1)}-{\displaystyle \frac{\mu }{a}}\right]e^{-a\left(t-t_0\right)}+{\displaystyle \frac{\mu }{a}}$$

(6)

For the discrete values sampled at equal intervals (noting that $t_0=1$), this was given by

$$X_{(k+1)}^{(1)}=\left[X_{(1)}^{(1)}-{\displaystyle \frac{\mu }{a}}\right]e^{-ak}+{\displaystyle \frac{\mu }{a}}$$

(7)

The values of $a$ and $u$ were calculated using the least squares method. Substituting $X_{(2)}^{(1)} , X_{(3)}^{(1)} ,\cdots X_{(N)}^{(1)}$ sequentially into formula (1.2), we obtained

$$\begin{array}{l}{X}_{(2)}^{(0)}+a{X}_{(2)}^{(1)}=u\\ X_{(3)}^{(0)}+a{X}_{(3)}^{(1)}=u\\ \dots \\ X_{(N)}^{(0)}+a{X}_{(N)}^{(1)}=u\end{array}$$

of which,

$$\begin{array}{l}{X}_{(2)}^{(0)}=\left[-X_{(2)}^{(1)},1\right]\left[\begin{array}{l}a\hfill \\ u\hfill \end{array}\right]\\ X_{(3)}^{(0)}=\left[-X_{(3)}^{(1)},1\right]\left[\begin{array}{l}a\hfill \\ u\hfill \end{array}\right]\\ \dots \\ X_{(\mathrm{N})}^{(0)}=\left[-X_{(\mathrm{N})}^{(1)},1\right]\left[\begin{array}{l}a\hfill \\ u\hfill \end{array}\right]\end{array}$$

Since ${\displaystyle \frac{\Delta X^{(1)}}{\Delta t}}$ involved the values of the cumulative column at two moments, it was more reasonable to replace it with the average of the two moments before and after, that is, to replace $X_{(i)}^{(1)}$ with $-{\displaystyle \frac{1}{2}}\left[X_{(i)}^{(1)}+X_{(i-1)}^{(1)}\right]$, $i$ = 1, 2, 3…N. Thus, we obtained,

$$\left[\begin{array}{l}{X}_{(2)}^{(0)}\hfill \\ X_{(3)}^{(0)}\hfill \\ \cdots \hfill \\ X_{(\mathrm{N})}^{(0)}\hfill \end{array}\right]=\left[\begin{array}{cc}-{\displaystyle \frac{1}{2}}\left[X_{(2)}^{(1)}+X_{(1)}^{(1)}\right]& 1\\ -{\displaystyle \frac{1}{2}}\left[X_{(3)}^{(1)}+X_{(2)}^{(1)}\right]& 1\\ -{\displaystyle \frac{1}{2}}\left[X_{(\mathrm{N})}^{(1)}+X_{(\mathrm{N})}^{(1)}\right]& 1\end{array}\right]\left[\begin{array}{l}a\hfill \\ u\hfill \end{array}\right]$$

such that $y=\left[\begin{array}{l}{X}_{(2)}^{(0)}\hfill \\ X_{(3)}^{(0)}\hfill \\ \cdots \hfill \\ X_{(\mathrm{N})}^{(0)}\hfill \end{array}\right]$

Then $B=\left[\begin{array}{cc}-{\displaystyle \frac{1}{2}}\left[X_{(2)}^{(1)}+X_{(1)}^{(1)}\right]& 1\\ -{\displaystyle \frac{1}{2}}\left[X_{(3)}^{(1)}+X_{(2)}^{(1)}\right]& 1\\ -{\displaystyle \frac{1}{2}}\left[X_{(\mathrm{N})}^{(1)}+X_{(\mathrm{N})}^{(1)}\right]& 1\end{array}\right],U=\left[\begin{array}{l}a\hfill \\ u\hfill \end{array}\right],y=BU.$

The least squares method was used to estimate $\widehat{U}=\left[\begin{array}{c}\widehat{a}\\ \widehat{u}\end{array}\right]={\left(B^{\mathrm{T}}B\right)}^{-1}{B}^{\mathrm{T}}y$. Substituting $\widehat{a}$ and $\widehat{u}$ into the equations, the time response equation was obtained:

$${\displaystyle \frac{d{X}^{(1)}}{dt}}-a{X}^{(1)}=u$$

(8)

2.2.3. Residual Test

After obtaining the time response equation, it was necessary to predict the original data sequence with the model and perform a residual test against the original results to examine the goodness of fit. The residual test included two key metrics: the absolute error (the difference between the predicted value and the actual value) and the mean relative error (the percentage of the absolute error relative to the actual value), as shown in Equations (9) and (10). A mean relative error within 10% indicated satisfactory model fitting accuracy.

$$\begin{array}{l}{\Delta }^{(0)}(i)=\left|X^{(0)}(i)-\widehat{X(i)}\right|\hspace{1em}i = 1, 2, 3,\cdots , N\\ \end{array}$$

(9)

$$\phi (i)={\displaystyle \frac{{\Delta }^{(0)}(i)}{X^{(0)}(i)}}\times 100\%$$

(10)

$$\overline{\phi (i)}={\displaystyle \sum _{i=1}^n\phi (i)}$$

(11)

The model’s accuracy was further validated by calculating the variance ratio (C) and the probability of small errors (P), where S₁ represents the standard deviation of the original series, S₂ represents the standard deviation of the residual errors, and P denotes the probability of small error occurrences.

$$\begin{array}{l}{S}_1=\sqrt{{\displaystyle \frac{{\displaystyle \sum {\left[x^{(0)}(i)-{\overline{x}}^{(0)}\right]}^2}}{n-1}}}\\ \end{array}$$

(12)

$$S_2=\sqrt{{\displaystyle \frac{{\displaystyle \sum {\left[{\Delta }^{(0)}(i)-{\overline{\Delta }}^{(0)}\right]}^2}}{n-1}}}$$

(13)

$$C={\displaystyle \frac{S_2}{S_1}}$$

(14)

$$P=P\left\{\left|{\Delta }^{(0)}(i)-{\overline{\Delta }}^{(0)}\hspace{1em}\right|<0.6745S_1\right\}$$

(15)

2.2.4. Predictive Model Accuracy Analysis

Following the residual test, the forecasting model’s accuracy was evaluated against established standards to validate its predictive reliability. The model’s predictive capability was assessed using a four-tier accuracy classification system, as detailed in

Table 1. Standard scale of calculation accuracy of Grey Forecast Model.

Model Accuracy Class	P	C
First grade, excellent	P ≥ 0.95	C ≤ 0.35
Second grade, good	0.80 ≤ P < 0.95	0.35 < C ≤ 0.5
Third grade, qualified	0.70 ≤ P < 0.80	0.50 < C ≤ 0.65
Fourth grade, unqualified	P < 0.70	C > 0.65

. According to these standards, higher accuracy levels correspond to smaller prediction errors, thereby indicating greater reliability and validity of the forecasted results. Only models meeting these specified accuracy criteria are considered suitable for forecasting applications.

2.2.5. Model Limitations and Constraints

The GM (1,1) model, while effective for our analysis, operates under specific constraints that warrant careful consideration. As established by Chen and Huang (2013), the model requires non-negative incremental data measured in years, with the dataset size carefully balanced to maintain prediction accuracy [28]. Two primary limitations emerged during our implementation. Firstly, the model was specifically designed for short-term prediction of monotonically increasing data and could not effectively handle strongly oscillating or non-linearly varying patterns. This limitation manifested in our regional analysis, where data from the Northwest and Northeast regions failed to pass the quasi-smoothness test and quasi-exponential test, indicating their unsuitability for GM (1,1) model prediction due to unstable growth patterns. In addition, through parameter sensitivity analysis involving perturbation of parameters a and u, we observed that the model’s relative error increased in later prediction stages, affecting its robustness. This characteristic resulted in decreased accuracy for long-term forecasting, as prediction errors accumulated over time. This limitation influenced our decision to restrict predictions to a 10-year timeframe, balancing prediction needed with accuracy considerations. These limitations underscored the importance of careful model application and result interpretation within the established constraints of the GM (1,1) methodology. While the model proved effective for regions showing stable growth patterns, its application required careful consideration of data characteristics and prediction timeframes.

3. Results and Discussion

3.1. Construction Waste Management: Current Situation and Deficiencies in China and Other Regions

3.1.1. Current Status of Construction Waste Management in the European Union

The European Union has developed a wide framework regarding construction waste management for this sector that amounts to the substantial use of resources (50% energy, 30% water resources, 50% of material consumption) and generates around 35% of total waste [29]. The legislative arsenal includes the Waste Framework Directive (2008/98/EC), EU Circular Economy Action Plan (2015), and EU Construction and Demolition Waste Protocol and Guidelines (2016) for the establishment of procedures of systematic management [30]. The European Union is to apply a systematic process of “waste pre-treatment—waste transportation—waste treatment and disposal” diversified treatment modes based on waste composition and recycling potential (Figure 1). With systematic processing procedures and diversified treatment means, the management of construction waste can be effectively, safely, and environmentally soundly achieved. According to the protocol, EU countries shall follow a closely linked process at each stage, with each step including different actions to manage waste effectively, securely, and in a manner that facilitates best practice. This holistic methodology will promote not only the improved recycling rate of construction waste, with due consideration to the environment, but also ensure the long-term stability of this problem. On-site recycling is slowly becoming the choice of most in the European Union because of its ease of handling and its efficiency [31].

When construction waste cannot be reused or recycled, incineration can be used for high-calorific-value compounds such as wood and plastics for energy recovery, which is applied for power generation or heating. Low on the priority list, backfilling is mainly applied to inert materials that are not easily degraded and have stable physical and chemical properties. Examples of these types of materials are waste concrete and waste asphalt. The landfill site is generally last to be used for waste disposal with strict categorization of the waste into hazardous, non-hazardous, and inert waste. This is because the EU requires that non-hazardous construction waste landfilling shall not exceed 10% to promote recycling and reuse. According to the statistics recorded and reported by the EU, since the implementation of such management measures in 2016, recycling of construction waste has continually gained an increasing recycling rate that stood at 90% by 2018. This administrative region has formed a highly referenced place from which very many other countries can study and learn [32].

However, recent research has indicated significant variations in construction waste management maturity across EU member states [33]. Countries in Northwestern Europe have demonstrated more advanced construction waste management practices, achieving higher recycling rates and better overall waste management performance. However, many EU member states still face challenges in waste prevention and show limited progress in reducing waste generation [33]. This regional disparity in management effectiveness highlights the complexity of implementing comprehensive waste management systems across areas with varying economic development levels and institutional capabilities.

3.1.2. Current Status of Construction Waste Management in the United States

Among the developed countries, the United States was the first to initiate measures in construction waste management. It has done so in a unique way by considering construction waste management as part of broader legislation on solid waste management rather than having independent legislation. Since 1965, the legislative installation has traveled a long way from the “Solid Waste Disposal Act” through the “Resource Recovery Act” of 1970 and to the present version of the “Resource Conservation and Recovery Act” of 1976—with no less than four major revisions since then. The industrial hazardous waste guidelines have further been strengthened by the implementation of the “Superfund Act” in 1980, which also encompasses construction wastes [34].

The United States implements a comprehensive “four-pronged management” approach encompassing “reduction”, “recycling”, “harmless”, and “industrialization” principles [35]. Reduction is emphasized as a critical strategy throughout all stages, from government regulation formulation to industry self-discipline, construction design, and on-site execution, aiming to achieve “zero discharge”. This source-control approach has proven more efficient than end-of-pipe treatment by reducing resource extraction, environmental damage, and associated costs. Notably, the United States addressed the significant proportion of concrete aggregate waste by permitting crushed cement concrete as coarse aggregate through the “Standard Specification for Concrete Aggregates” (ASTM C-33-82 [36]) in 1982 [37].

The successful implementation of “four-pronged management” and widespread adoption of recycled aggregate and concrete recycling technology have yielded significant achievements in construction waste management. These measures have effectively reduced environmental impacts and resource consumption while improving recycling efficiency, supporting the construction industry’s sustainable development. The United States has achieved a construction waste recycling rate of approximately 75%, with the remaining 25% disposed of through landfilling, establishing a valuable reference for global construction waste management practices.

3.1.3. Current Status of Construction Waste Management in Japan

Japan, as a densely populated and resource-scarce developed country, faces significant challenges in construction waste management, with construction waste accounting for 60% of the total illegal dumping waste. To address these challenges, the Japanese government established its regulatory framework through the “Basic Act on Establishing a Sound Material-Cycle Society” (1970, 2000), which mandates that construction waste must be reduced, made harmless, and be reused. The regulatory system was further strengthened in 2002 with the formal implementation of the “Construction Waste Recycling Law”, specifically dedicated to construction waste recycling. This law clearly defines waste classification, recycling processes, and stakeholder responsibilities while emphasizing enhanced departmental supervision [38].

Unlike the EU’s comprehensive process, Japan’s management strategy integrates circular economy principles through four main technological approaches: source reduction design, separation and treatment, recycling technologies, and the promotion of recycled production technologies. Source reduction and recycling technologies form the core components of this strategy. Japan’s approach is particularly distinctive due to its resource scarcity and higher construction material costs compared to Western countries, leading to more detailed waste classification and recycling processes. The government emphasizes high recycling efficiency from project design to material selection, ensuring buildings can be efficiently dismantled and recycled after 50 or 100 years of use while maintaining a “zero discharge” goal throughout the process. This comprehensive strategy has proven remarkably successful, with construction waste resource utilization rates reaching nearly 100% by 2019. The significance of this achievement was further recognized when the recycling industry was designated as one of the 14 key areas in Japan’s Green Growth Strategy in 2021 [39]. Japan’s success can be attributed both to its implementation of advanced technologies and to one of the most comprehensive regulatory and management systems among developed countries, establishing a strong foundation for construction waste resource utilization.

3.1.4. Current Status of Construction Waste Management in China

Considering the fast pace of urbanization in China, and hence the rapid growth of the construction industry, it contributes significantly to a substantial increment in the volume of generated construction waste. According to estimates by the Ministry of Housing and Urban–Rural Development, in 2021, waste from construction constituted 30–40% of urban waste, approximately eight times that of household waste, and was the most significant category of waste [40]. It all started in China’s regulatory framework with the “Law of the People’s Republic of China on the Prevention and Control of Environmental Pollution Caused by Solid Wastes” of 1995 laying down requirements for proper disposal of solid wastes including construction waste. This was followed by the “Urban Construction Waste Management Regulations”, promoting recycling and resource utilization since 2005 [41]. To enhance the promotion of this process, other incentive policies the government successively introduced include a three-year plan from 2018 onward to increase resource utilization in 35 pilot cities, including Beijing, Guangzhou, Shenzhen, Chengdu, and Chongqing. Policies by the General Office of the State Council and Ministry of Ecology and Environment in 2019–2021 for the promotion of comprehensive utilization of waste followed suit, as did the 14th Five-Year Plan for Circular Economy Development that set 2025 up with a 60% comprehensive utilization rate.

In parallel with national initiatives, local governments have progressively developed and refined their regulatory frameworks for construction waste management, establishing comprehensive systems encompassing waste transfer, disposal, and resource utilization. Beijing’s 2020 “Administrative Provisions on Construction Wastes Disposal” established clear principles, objectives, responsibility allocation, and management protocols. Shanghai’s 2024 “Municipal Solid Waste Management Regulations” emphasized waste reduction and resource utilization in pursuit of waste-free city development. Guangdong and Sichuan provinces implemented comprehensive regulations in 2022 and 2024, respectively, delineating protocols for the entire waste management cycle from generation to disposal while defining stakeholder responsibilities. Similar regulatory frameworks were established in 2024 by Zhejiang, Jiangsu, and Hebei provinces and Tianjin municipality.

Although China currently lacks a standardized national management protocol, analysis of various provincial and municipal practices reveals six fundamental components of construction waste management, as illustrated in Figure 2.

(1)

Establishment of Full-Process Management Procedures

The first step in construction waste management is to establish a complete full-process management procedure. As of 2024, Hainan Province and Shandong Province have begun trial implementation of a full-process data platform or the “Triptych” system, which aims to monitor the generation, transportation, and disposal of construction waste in real time [42,43]. Construction waste generators, transporters, and disposers are required to record and report relevant data, ensuring the traceability and transparency of the entire process of waste flow. This approach not only helps prevent illegal dumping but also provides strong support for the refined management of construction waste.

(2)

Source Reduction and Classification Management

Source reduction and classification management has been reinforced through the 2020 Ministry of Housing and Urban–Rural Development guidelines, with cities like Wuhu and Hainan Province mandating waste reduction priorities in construction design and execution [42,44]. Implementation includes optimization of building design schemes for material efficiency and development of pre-construction waste management protocols. Systematic waste classification into project, demolition, and renovation categories facilitates efficient resource utilization through optimized processing pathways.

(3)

Resource Utilization and Technological Innovation

Resource utilization is a key link in construction waste management, and the focus is on technological innovation and the improvement of processing facilities. Inert materials in construction waste, especially discarded concrete and masonry waste, can be processed into recycled aggregates and stabilized materials, which are widely used in municipal engineering, construction, and road construction. To improve the resource utilization rate, various regions are actively promoting new technologies, such as concrete recycling technology and efficient sorting equipment, to ensure the quality and stability of recycled products.

(4)

Policy Incentives and Mandatory Measures

Local governments are using a series of policy incentives to promote construction waste management, such as providing financial subsidies, preferential loans, and green procurement policies to encourage enterprises to participate in the resource utilization of construction waste. Some regions have also adopted mandatory measures, requiring the use of recycled construction waste products in specific projects. Additionally, governments are accelerating the construction of waste processing facilities, such as recycling plants and transfer stations, to ensure that sufficient capacity is in place to handle the increasing amounts of construction waste.

(5)

Facility Construction and Enhanced Supervision

Various regions have accelerated the construction of resource utilization facilities for construction waste, including recycling plants, transfer stations, and landfills. Cities such as Beijing and Sichuan Province have strengthened the monitoring of construction waste transportation and disposal to ensure that waste is effectively managed throughout the entire process.

(6)

Industrial Development and the Promotion of Recycled Products

Through the combined promotion of policy and market forces, the construction waste recycling industry is gradually maturing. Local governments are encouraging construction waste disposal enterprises to participate in the entire chain, from collection and sorting to recycling. Recycled products, such as recycled aggregates and recycled bricks, are gradually being applied in municipal construction projects.

These interconnected management components collectively establish an effective framework for waste reduction, recycling, and treatment, advancing China’s transition toward waste-free urban development. The systematic approach ensures comprehensive coverage of the waste management cycle while promoting sustainable resource utilization practices.

Despite progress in regulatory frameworks and management initiatives, China’s construction waste recycling efforts have significantly lagged behind due to the absence of a unified, comprehensive management system and scientifically effective, economically feasible disposal technologies. In addition, recent studies have highlighted the critical importance of improving resource recovery systems while addressing potential environmental risks in the recycling process [45]. As of 2022, the national average resource utilization rate remained below 10%, substantially lower than rates achieved in developed countries and regions such as Europe and the United States [46]. Several critical challenges persist in China’s construction waste management system.

(1)

Inadequate Statistical Data Collection

The foremost challenge lies in incomplete statistical data on construction waste generation. Many regions lack accurate, timely data collection mechanisms, failing to provide essential baseline information for effective management and resource utilization planning. The absence of comprehensive monitoring systems impedes accurate forecasting of future waste generation, hampering authorities’ ability to plan processing facilities, allocate utilization capacity, and evaluate policy effectiveness.

(2)

Weak Source Reduction and Classification Management

Implementation of source reduction measures remains weak despite policy emphasis on waste reduction during the design and construction phases. Construction enterprises frequently fail to incorporate effective reduction measures in their design and construction processes, resulting in widespread material waste and significant waste generation at the source. Similarly, the classification management system remains underdeveloped, particularly for renovation and demolition waste, where inadequate classification standards and enforcement prevent effective resource utilization. The commingling of household, construction, and hazardous waste further complicates subsequent treatment processes.

(3)

Limited Resource Utilization Capacity

Resource utilization faces multiple challenges despite the establishment of processing facilities in some regions. The overall utilization rate remains low due to immature processing technologies, insufficient processing capacity, and limited market demand for recycled materials. Specific materials present particular challenges—asbestos-containing materials primarily undergo high-temperature incineration before landfilling, while printed circuit boards (PCBs) are limited to copper recovery through physical separation before incineration and landfilling, indicating significant unrealized potential for resource utilization [47].

(4)

Policy Implementation Gaps

Policy implementation and regulatory mechanisms show significant gaps despite government initiatives promoting resource utilization through financial subsidies and mandatory recycled product use. Illegal dumping persists in some regions, while incomplete information technology supervision systems fail to provide comprehensive process coverage, impeding the formation of effective closed-loop management systems [48,49].

3.2. Estimation and Prediction of Construction Waste Generation in China

3.2.1. Annual Construction Waste Generation of China in 2000–2022

Utilizing the building area estimation method, the annual production of construction waste in China from 2000 to 2022 and the production of construction waste in each region are illustrated in Figure 3. The analysis of regional construction waste generation from 2000 to 2022 revealed distinct patterns across China’s geographical regions. Our findings demonstrate significant regional disparities driven by multiple interconnected factors.

The East China region emerged as the dominant contributor, accounting for 41–55% of the national total waste generation. This substantial contribution stemmed not only from the region’s robust economic growth but also from its distinct urbanization patterns, characterized by intensive urban renewal projects, frequent infrastructure upgrades, and high-density development initiatives. Central China’s contribution of 10–16% reflects its position as an emerging economic center, where waste generation primarily originated from new infrastructure development rather than urban renewal. This distinction significantly influenced waste composition and management requirements. Similarly, the Southwest region’s 9–14% contribution demonstrates how geographical constraints shape both construction practices and waste generation patterns, with mountainous terrain creating unique challenges for construction methods and waste management. The South China region’s evolving contribution of 6–11% reflects its rapid urban transformation, particularly driven by the Greater Bay Area development initiative and intensive transportation infrastructure construction. The North China region maintained a stable contribution of 10–12%, balancing urban development with environmental constraints. The Northwest and Northeast regions showed more modest contributions of 3–5% and 2–6%, respectively, reflecting their slower economic development and distinct geographical challenges.

Statistical analysis revealed strong correlations between regional GDP and construction waste generation, with correlation coefficients exceeding 92% in most regions (

Table 2. Correlation analysis between regional GDP and annual production of construction waste in China.

Region	Eastern	South	North	Central	Southwest	Northwest	Northeast
R2	0.9273	0.9862	0.9868	0.9654	0.9273	0.942	0.4298

). However, this relationship manifested differently across regions due to varying development stages and local conditions. The Northeast region, notably, showed a weaker correlation (R² = 0.4298), indicating that factors beyond economic growth, such as industrial transformation and population dynamics, significantly influence waste generation patterns. These regional variations have profound implications for waste management strategies and China’s broader sustainability objectives, particularly the goals outlined in the 14th Five-Year Plan for Circular Economy Development. The East China region’s high waste volume required sophisticated management systems capable of handling large-scale operations, while the South China region’s rapid growth necessitated innovative solutions for limited land resources. Central China’s emerging status presented opportunities to implement advanced waste management systems during its development phase, potentially avoiding the challenges faced by more developed regions.

This analysis demonstrates that effective construction waste management requires understanding not just the volume of waste generated but also the underlying drivers and regional characteristics that shape generation patterns. Such understanding is crucial for developing targeted management strategies that align with both local conditions and national sustainability goals.

3.2.2. Prediction of Construction Waste Generation in China and Seven Regions

In this study, we employed the Grey Prediction GM (1,1) model to forecast China’s construction waste generation from 2023 to 2032, utilizing historical data from 2013 to 2022 as the baseline for our predictions. The selection of this ten-year historical period provided a robust foundation for capturing recent trends while accounting for the dynamic nature of China’s construction sector development. To validate the model’s applicability, we first conducted a quasi-smoothness test on the original data sequence, followed by a quasi-exponential regularity test on the derived data sequence. The results of these preliminary tests are presented in

Table 3. Smoothness and quasi-exponential test values.

t	1	2	3	4	5	6	7	8	9	10
$p(t)$	0	1.10	0.52	0.35	0.27	0.22	0.19	0.16	0.15	0.13
${\sigma }^{(1)}(t)$	0	2.10	1.52	1.35	1.27	1.22	1.19	1.16	1.15	1.13

. These tests were essential to ensure the reliability and suitability of the Grey Prediction Model for our forecasting purposes. The computational outcomes, when $t>3$, $p(t)<0.5$, and 1 < ${\sigma }^{(1)}(t)$ < 1.5, indicated that both the original data $X^{(0)}$ and the newly generated sequence $X^{(1)}$ passed the quasi-exponential regularity test and the quasi-smoothness test, respectively. The test results confirmed that the data sequence of China’s annual construction waste production satisfied the necessary conditions for Grey Prediction Model application, thereby validating its suitability for forecasting future trends.

Following the establishment of sequence $X^{(1)}$, we constructed a differential equation and applied the least squares method for parameter estimation. Using Matlab’s matrix operations, we derived the general solution of the GM (1,1) forecasting model’s differential equation (Equation (16)). The analysis yielded two critical parameters: a development coefficient (a) of 0.0275 and an internal control coefficient (u) of 178,071.014. These parameters were subsequently incorporated into the specific response equation for forecasting China’s annual construction waste production, as presented in Equations (16) and (17):

$$\widehat{U}=\left[\begin{array}{c}\widehat{a}\\ \widehat{u}\end{array}\right]={\left(B^{T}B\right)}^{-1}{B}^{T}y = [0.0275, 178,071.01]$$

(16)

$${\displaystyle \frac{d{X}^{(1)}}{dt}}-0.027479X^{(1)}=178,071.014$$

(17)

The model’s reliability was validated through rigorous error analysis of the forecasted values against actual construction waste production data from 2013 to 2022. The analysis yielded a mean relative error of 0.014 (less than 0.1), a variance ratio (C) of 0.03, and a minimum error probability (P) of 1. When evaluated against the forecasting model’s accuracy standard-level criteria, these parameters classified our model as Grade 1, indicating excellent predictive accuracy.

Based on this validated model, we utilized MATLAB 2022b to project China’s annual construction waste production from 2023 to 2032 (Figure 4). The forecasting results revealed a consistent upward trend in waste generation over the next decade, with projected volumes increasing from 2.39 billion tons in 2023 to 3.084 billion tons by 2032, representing approximately a 1.3-fold increase. This substantial growth trajectory, driven by continued urbanization, suggests mounting challenges for waste management and resource utilization in the coming years. These findings underscore the pressing need to enhance low-carbon treatment approaches and resource utilization strategies to address the environmental and resource pressures associated with the projected increase in construction waste production.

Regional forecasting analysis was conducted using the GM (1,1) model for five of China’s seven geographical regions: North China, East China, South China, Southwest, and Central China, projecting construction waste production from 2023 to 2032. The Northeast and Northwest regions were excluded from this forecast analysis as their annual waste production patterns did not exhibit the requisite increasing time series trend, thus failing to meet the fundamental conditions for grey prediction modeling. For the five analyzed regions, we performed rigorous data validation using the 2013–2022 historical data. Both quasi-exponential regularity tests and quasi-smoothness tests were applied to $X^{(0)}(t)$ and $X^{(1)}(t)$ sequences. The test results confirmed that all five regions’ data sequences satisfied the necessary predictive conditions, validating their suitability for grey prediction modeling. Subsequently, utilizing MATLAB’s matrix operations and the least squares method, we calculated region-specific parameters including development coefficients, internal control coefficients, and corresponding time response equations, as detailed in

Table 4. GM (1,1) model values for different regions.

Region	a	u	Time Response Equation
North China	0.03	17,947.08	${\displaystyle \frac{d{X}^{(1)}}{dt}}-0.025543X^{(1)}=17,947.0835$
East China	0.02	95,280.26	${\displaystyle \frac{d{X}^{(1)}}{dt}}-0.020764X^{(1)}=95,280.2598$
South China	0.08	9470.51	${\displaystyle \frac{d{X}^{(1)}}{dt}}-0.082082X^{(1)}=9470.5071$
Southwest China	0.03	17,197.36	${\displaystyle \frac{d{X}^{(1)}}{dt}}-0.027987X^{(1)}=17,197.3591$
Central China	0.05	23,718.44	${\displaystyle \frac{d{X}^{(1)}}{dt}}-0.046602X^{(1)}=23,718.438$

.

The model’s regional forecasting accuracy was evaluated through comprehensive error analysis, yielding region-specific metrics including average relative error ($\overline{\phi (i)}$), variance ratio (C), and minimum error probability (P), as presented in

Table 5. The average relative error, variance ratio (C), and minimum error probability (P) of GM (1,1) model for different regions.

Region	$\overline{\mathit{\phi }(\mathit{i})}$	C	P	Accuracy Level
North China	0.03	0.06	1	First grade
East China	0.01	0.05	1	First grade
South China	0.03	0.03	1	First grade
Southwest China	0.01	0.03	1	First grade
Central China	0.03	0.06	1	First grade

. The evaluation results demonstrated the robust predictive capabilities of our GM (1,1) model implementation. All five analyzed regions achieved first-grade accuracy levels, with variance ratios (C) ranging from 0.03 to 0.06, significantly below the excellence threshold of 0.35. This consistently low C value indicated minimal deviation between the predicted and actual values, suggesting strong model reliability across different regional contexts. The probability of small errors (P) reached the optimal value of 1 across all regions, exceeding the 0.95 threshold for excellent prediction accuracy. The average relative errors remained below 0.03 for all regions, with East China and Southwest China showing particularly strong performance with errors of 0.01. This exceptional accuracy level for the East China region was especially significant given its dominant contribution to national waste generation. The slightly higher relative errors of 0.03 observed in North China, South China, and Central China regions, while still well within first-grade accuracy standards, might reflect these regions’ more dynamic construction activity patterns. The model’s strong performance across diverse regional contexts validated its suitability for construction waste prediction and supports the reliability of our forecasting results. However, it is important to note that these accuracy levels were achieved for the model’s training period (2013–2022), and actual prediction accuracy for future periods might show some variation, particularly in regions experiencing rapid development changes.

Using the validated GM (1,1) model, we projected construction waste generation for five major regions from 2023 to 2032 (Figure 4b–f), illustrating the trends in construction waste production over the next decade for each region. The forecasting results revealed that the annual production of construction waste in these five major regions is expected to show a steadily increasing trend through 2032. The East China region’s annual production of construction waste is significantly higher than that of other regions, with projections reaching 1.426 billion tons by 2032, suggesting that the construction activities and urbanization process in this area have a substantial impact on waste generation. Both North China and Central China regions demonstrated consistent upward trends in construction waste production, with projected figures reaching 367 million tons and 585 million tons respectively by 2032, reflecting the sustained high level of activity in the construction industry in these regions. The South China region exhibited a more rapid increase, with waste generation projected to grow from 226 million tons in 2023 to 473 million tons by 2032, indicating an acceleration of construction activities in the future. In contrast, the Southwest region’s annual production of construction waste showed steadier growth, with projections reaching 296 million tons by 2032; although the total volume is smaller, the growth rate remains stable.

The proportional distribution of regional waste generation through 2032 (Figure 5) revealed evolving patterns in the regional contribution to total national construction waste production. The East China region, while leading in construction waste production, is expected to see its proportion gradually decline over the next 10 years. Despite the large economic scale of the East China region, the rapid development in the past may lead to a gradual saturation of the construction industry in the future, resulting in a decreased proportion of construction waste. Meanwhile, with the proposal of large-scale projects such as the “Greater Bay Area”, the construction industry in the South China region will continue to grow over the next decade, leading to a steady annual increase in its proportion of construction waste production. These growth trends reflect the ongoing development of the construction industry and the acceleration of urbanization processes across various regions in China while also indicating that the pressure of managing construction waste will gradually increase. Therefore, it is necessary to further strengthen the low-carbon management and resource utilization of construction waste to cope with the continuous increase in future waste production.

3.2.3. Sensitivity Analysis of the Grey Prediction Model

The reliability and robustness of the Grey Prediction Model were systematically evaluated through a comprehensive sensitivity analysis of its key parameters. Following established methodologies [50,51], we examined the model’s response to perturbations in the development coefficient (a) and internal control coefficient (u) [52]. The analysis employed both moderate (5%) and substantial (10%) parameter perturbations to assess model stability across different regions and prediction timeframes. Table S1 presents the perturbed model parameters for different regions, while Tables S2–S7 detail the resulting predictions and relative errors for the whole country and each region over the period 2023–2032. These comprehensive results enable systematic evaluation of the model’s sensitivity across different geographical contexts and time horizons.

Specifically, the sensitivity analysis revealed distinct patterns in regional responses to parameter perturbations. When model parameters were perturbed by 5% and 10%, predicted construction waste generation consistently exceeded original forecasts across all regions, with discrepancies increasing over the prediction timeframe. The South China region exhibited the highest sensitivity to parameter variations, with relative errors between new and original predictions reaching 13.29% under 5% perturbation and 28.05% under 10% perturbation by 2032. This heightened sensitivity indicated that predictions for this region required particularly careful interpretation and validation.

The analysis also demonstrated a clear temporal pattern in prediction reliability. Relative errors showed a consistent increasing trend over longer prediction horizons across all regions, though with varying magnitudes. For instance, national-level predictions showed relative errors increasing from 6.42% to 7.79% under 5% perturbation from 2023 to 2032, while the Central China region demonstrated a more pronounced increase from 7.35% to 9.63%. This pattern suggests a general decrease in model robustness for longer-term predictions, particularly evident in regions with more dynamic construction activity.

The comprehensive error metrics, including the posterior difference ratio (C) and small error probability (P), were calculated to validate these findings. Despite the observed sensitivities, all regional models maintained first-level accuracy according to these standardized metrics, confirming the overall reliability of the predictions within the studied timeframe while acknowledging the need for careful interpretation of longer-term forecasts in more sensitive regions.

3.3. Suggestions for Resource Management of Construction Waste in China

3.3.1. Recommendations for Construction Waste Management at the National Level

Based on our analysis and the identified challenges in China’s construction waste management, particularly the large waste volume and low resource utilization rate, we propose several key management recommendations drawing from successful experiences of developed Western countries and pioneering pilot cities.

First, strengthening the construction waste production statistics system is crucial. A comprehensive statistical system would provide accurate and detailed data on waste generation, creating a solid foundation for government decision-making, industry planning, and corporate operations. With reliable statistical data, governments can develop more scientifically sound policies for resource utilization, including financial subsidies and tax incentives, to encourage enterprise participation. Furthermore, accurate production data enable better assessment of current situations and resource utilization potential, leading to optimized technological routes and processes that can enhance resource utilization rates.

Second, emphasis should be placed on on-site classification and processing of construction waste. For large construction sites with significant waste generation, implementing proper source classification is essential for achieving higher recycling rates and reducing processing costs. Drawing from European Union experiences, on-site classification not only reduces transportation costs but also increases material recycling rates. Modern equipment can facilitate real-time sorting, recycling, and reuse of construction waste on-site. For instance, mobile rapid processing machinery can be employed to sort and crush old bricks, tiles, and concrete, grade them by particle size, and transform them into immediately usable construction materials [53,54].

Third, the level of in-depth resource utilization processing requires significant enhancement. Currently, China’s construction waste classification and processing levels remain relatively low, with most waste still collected in a mixed state, increasing the difficulty of resource utilization and harmless treatment. Although China introduced the concept of “construction waste resource utilization” relatively early, its implementation lags behind developed Western countries. Learning from EU countries’ waste management procedures, pre-treatment should prioritize the separation of hazardous waste materials, followed by specialized resource utilization or physical and chemical treatments for transformation into harmless substances. Common treatment methods include converting asbestos-containing materials into glass powder through high-temperature melting and recovering metals from PCB-containing equipment through thermal decomposition or microwave treatment. Additionally, China’s significantly lower landfill rate compared to developed countries like the United States and Japan is primarily due to the predominant use of landfill methods for disposing of inert materials such as waste concrete and old asphalt, indicating insufficient in-depth resource utilization. Advanced processing methods could transform waste concrete through extrusion crushing into recycled aggregate, which can be further processed into various construction materials such as recycled aggregate bricks, concrete blocks, permeable concrete, and mortar, suitable for diverse construction applications [55]. For example, these materials are particularly valuable for sponge city applications, where enhanced water absorption and retention capabilities are crucial for managing urban flooding and improving water resource utilization [56].

3.3.2. Recommendations for Construction Waste Management at the Regional Level

Our model predictions indicate distinct regional patterns in construction waste generation that necessitate targeted management approaches. Based on these quantitative findings and successful international practices, we propose region-specific recommendations that address each area’s unique challenges and projected waste volumes.

The East China region, projected to generate 1.426 billion tons of waste by 2032 and contributing over 50% of the national total, requires immediate implementation of comprehensive management systems. Drawing from the EU’s successful experience, we recommend establishing an advanced digital monitoring platform integrating real-time waste tracking with blockchain technology for data verification. This system should be implemented initially in Shanghai by 2025, with systematic expansion to other eastern provinces by 2027. Additionally, the region should develop large-scale processing facilities equipped with automated sorting technology, focusing on areas with the highest construction activity density.

For the South China region, where waste generation is expected to reach 473 million tons by 2032 with the fastest growth rate among all regions, we propose adopting Japan’s detailed classification approach. Given the region’s limited land resources and diverse construction activities, establishing three regional recycling centers in Guangdong Province by 2025 is crucial. These centers should be equipped with advanced sorting technology capable of processing 50,000 tons annually, with particular emphasis on concrete recycling, which comprises approximately 60% of the region’s construction waste.

The Central China region’s projected 585 million tons by 2032 presents an opportunity to implement comprehensive waste management systems during its development phase. Following the US “four-pronged management” approach, we recommend a three-phase implementation strategy: establishing standardized waste classification systems in major cities (2024–2026), developing specialized processing facilities for masonry and concrete recycling (2026–2028), and implementing a regional trading platform for recycled materials (2028–2032).

For the Southwest region, with its projected 296 million tons and challenging terrain, innovative solutions are essential. We recommend developing mobile processing units capable of on-site waste recycling, reducing transportation costs and environmental impact. These units should be operational in Chongqing and Sichuan by 2025, with expansion to other southwestern provinces by 2027, addressing the unique geographical constraints while maintaining processing efficiency.

The North China region, facing 367 million tons of waste by 2032, requires adapting management systems to severe weather conditions. We recommend developing indoor processing facilities equipped with heating systems to ensure year-round operation, strategically located near major urban centers. The first facility should be operational in Beijing by 2025, incorporating lessons learned from Japan’s all-weather waste processing facilities.

For regions showing unstable waste generation patterns, particularly the Northeast and Northwest, we recommend focusing on optimizing existing infrastructure and implementing strict source reduction measures. These regions should prioritize developing basic processing capabilities while maintaining flexibility to adapt to changing waste generation patterns.

To support these regional initiatives, establishing a national coordination center as quickly as possible is essential. This center would monitor implementation progress, facilitate technology transfer between regions, standardize reporting protocols while accommodating regional variations, and coordinate cross-regional resource sharing and knowledge exchange. Regular evaluation and adjustment of these strategies would ensure their effectiveness as actual waste generation patterns emerge.

3.3.3. Analysis of Resource Management Strategy Implementation

The technical feasibility of our proposed strategies is supported by successful precedents in pilot programs. The digital monitoring platform recommended for the East China region builds upon existing systems, with a modular implementation approach beginning in Shanghai by 2025 before expanding to other eastern provinces [57]. This phased approach allows for continuous technological refinement informed by initial operational experiences. However, several challenges must be carefully addressed to ensure successful implementation. The proposed regional recycling centers in the South China region will require significant infrastructure investment and effective coordination among a diverse range of stakeholders. Similarly, while the phased implementation strategy for the Central China region (2024–2032) allows for gradual capacity building through distinct stages (from waste classification systems to specialized processing facilities), it demands sustained commitment and resource allocation.

Our recommendations have been tailored to account for regional disparities in both economic conditions and technical capacities. For instance, the mobile processing units proposed for the Southwest region are designed to address geographical limitations while optimizing operational efficiency. In contrast, the indoor processing facilities planned for the North China region are specifically adapted to integrate seamlessly with existing technological infrastructures.

From an institutional perspective, successful implementation requires careful alignment with existing administrative structures. The proposed national coordination center, while essential for implementation, necessitates robust inter-regional cooperation mechanisms and coordinated policy frameworks. These considerations suggest that while our proposed strategies are technically feasible, their successful execution depends on systematic planning and strong institutional support.

4. Conclusions

This study provides critical insights into China’s construction waste management challenges and opportunities through comprehensive quantitative analysis. Our findings reveal that construction waste will reach 3.084 billion tons by 2032, with significant regional disparities, as East China contributes over 50% of the national total. The current resource utilization rate below 10% represents a substantial gap compared to international benchmarks in developed countries, indicating significant room for improvement. Our analysis supports implementing a tiered regulatory framework where high-waste-generating regions face stringent requirements, while developing regions receive targeted infrastructure support. For industry stakeholders, processing solutions must align with regional characteristics. While mobile processing units offer flexibility for areas with dispersed construction activities, permanent large-scale facilities are better suited for high-density urban regions with concentrated waste generation. This study contributes to policy development by providing quantitative forecasts, establishing frameworks for regional-specific strategies, and identifying critical gaps in current practices. Success in transitioning to effective management requires coordinated technological innovation, policy reform, and regional cooperation, supported by clear responsibilities, adequate resources, and consistent monitoring mechanisms across all governance levels.

Future perspectives: This study utilized the building area estimation method and Grey Prediction Model for forecasting construction waste generation. However, we acknowledge that waste generation is influenced by multiple factors beyond construction area. Due to current limitations in data availability and consistency across regions, as well as the complex interactions between variables, our study adopted a focused approach. As regional data collection systems improve and standardization advances, future research should develop more comprehensive prediction models incorporating economic indicators, technological advancement impacts, and policy effects. Such multi-variable models will enhance prediction accuracy and provide more robust support for waste management planning.