A Comprehensive Study on Concrete Produced with Recycled Concrete Aggregate (RCA)

Abstract

It is known that a significant portion of global carbon dioxide (CO₂) emissions originate from concrete production. However, construction and demolition activities result in a considerable amount of construction and demolition waste (CDW). The proper recycling of CDW is important in terms of conserving natural resources and ensuring sustainability. A significant amount of recycled concrete aggregate (RCA) is obtained from the recycling of CDW. Many researchers have contributed to reducing carbon emissions by conducting studies on RCA. However, the fact that recycled aggregates (RAs) are obtained from different construction wastes is the biggest obstacle to generalizing the studies in the literature. This study aims to identify machine learning (ML) models that can reliably predict the compressive strength of concrete produced with recycled concrete aggregates (RCAs) and to evaluate the impacts of their use. In this study, keywords (15) obtained from articles (7953) selected from Web of Science were searched in the Scopus database. The selected studies (397) were analyzed using VOSviewer (version 1.6.20) software to identify leading institutions, countries, authors, sources, fields, gaps, challenges, and trends related to the use of recycled aggregate in concrete. This study not only has a theoretical structure but also makes a significant contribution to the literature by offering practical recommendations for field applications. This is the most important feature that distinguishes this study from other research. This study also promotes the use of RAs in concrete to reduce CO₂ emissions and encourages its sustainable use in the construction sector.

1. Introduction

With the increasing population growth, the demand for concrete is rising. It is estimated that approximately 30 billion metric tons of concrete is produced annually worldwide [1]. However, the disposal of structures that have reached the end of their service life results in a large amount of construction and demolition waste (CDW) [2]. Recycling this waste is crucial for sustainability [3]. The sustainable recycling process of CDW begins with separation at the construction site and is completed in a sequence involving collection and transportation, crushing and screening, magnetic separation for metal materials, and reuse [4,5]. At the construction site, concrete and steel materials are separated using mechanical shears and construction machinery [6]. Materials such as concrete, bricks, asphalt, metal, wood, and plaster are then collected separately. At recycling facilities, powerful magnets are used to separate reinforcing steel from concrete [7]. During the crushing and screening stage, materials such as concrete and bricks are crushed using jaw and impact crushers and classified according to particle size. The resulting recycled materials are reused in areas such as fill material and concrete production, depending on the purpose. However, due to their negative effects on mechanical properties, their use in concrete is currently limited [8].

In spite of this, recycled concrete aggregate (RCA) obtained from CDW has become a popular research topic in recent years [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24]. In particular, significant studies have been conducted in the literature on CDW, RCA, recycled aggregate (RA), recycled concrete (RC), recycled powder (RP), recycled coarse aggregate (RCA), recycled fine aggregate (RFA), and fully recycled aggregate concrete (FRAC) [9,15,18,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52]. In recent studies, researchers have defined classification frameworks for RCA based on parameters such as density and water absorption, which are derived from material and physical properties [53,54,55]. The aim of these frameworks is to reduce heterogeneity and increase the usability of RCA in applications.

Nearly all of these studies aimed to reduce natural resource consumption, decrease the need for land used for waste disposal, reduce carbon emissions, and facilitate the transition to a circular economy model for sustainability [56,57,58,59,60,61]. Furthermore, a critical issue in these studies is that recycled aggregates are derived from different construction wastes and therefore exhibit varying physical and chemical properties. The absence of a unified standard for recycled aggregate concrete leads to inconsistent results across studies. As a result, limited and small-scale laboratory-based studies are often prioritized over real-world applicability. This represents a major obstacle in scientific research on RCA. Consequently, the generalizability and reliability of findings in the current literature are considerably limited. Moreover, research on RCA remains insufficient in both quantity and level of development. As a result, despite significant contributions from researchers, important research gaps still exist in the literature.

In this context, this study aims to identify and comprehensively investigate machine learning (ML) models that reliably predict the compressive strength of RCA. To this end, gaps, challenges, trends, prominent sources, institutions, and countries in the RCA field have been analyzed using VOSviewer software. In addition to the existing literature, this study is not limited to a theoretical framework but also provides practical suggestions for field applications.

In conclusion, the data obtained from this study contribute to identifying the most commonly used machine learning models for predicting the compressive strength of RCA concrete, highlighting the key parameters affecting it, revealing research gaps in the existing literature, and guiding future experimental and modeling studies in concrete technology.

2. Methodology

Studies on RCA have been comprehensively researched in this study. For this purpose, first, the abstracts of 77 SCI-EXPANDED English-language articles with the highest citation counts were selected from a dataset of 7953 publications identified through the Web of Science for the RCA keyword group, and were systematically analyzed to identify the most frequently repeated keywords. The identified keywords and their frequencies of use are presented in

Table 1. Keywords most frequently repeated in the abstract section of the article.

Top Keywords Appearing in Abstracts	Keyword Frequency
Recycled Aggregate (RA)	128
Recycled Concrete (RC)	84
Recycled Powder (RP)	47
Recycled Coarse Aggregate (RCA)	38
Compressive Strength	38
Environment	34
Recycled Fine Aggregate (RFA)	27
Replacement	25
Sustainable	24
Fully Recycled Aggregate Concrete (FRAC)	20
Life Cycle Assessment (LCA)	13
Analyze	12
Machine learning (ML)	12
Construction and Demolition Waste (CDW)	8
Artificial Intelligence (AI)	5

.

The keywords identified in WoS were formulated via Scopus as follows: (TITLE-ABS-KEY (recycled AND aggregate) OR TITLE-ABS-KEY (recycled AND concrete) OR TITLE-ABS-KEY (recycled AND powder) OR TITLE-ABS-KEY (recycled AND coarse AND aggregate) OR TITLE-ABS-KEY (recycled AND fine AND aggregate) OR TITLE-ABS-KEY (fully AND recycled AND aggregate AND concrete) OR TITLE-ABS-KEY (construction AND demolition AND waste) AND TITLE-ABS-KEY (compressive AND strength) AND TITLE-ABS-KEY (replacement) OR TITLE-ABS-KEY (life AND cycle AND assessment) OR TITLE-ABS-KEY (environment) OR TITLE-ABS-KEY (sustainable) AND TITLE-ABS-KEY (analyze) OR TITLE-ABS-KEY (machine AND learning) OR TITLE-ABS-KEY (artificial AND intelligence)). A comprehensive literature search yielded 465 documents. Among these, sources published in English and of the article type were selected, resulting in a total of 397 documents. The purpose of limiting the study to English-language articles recognized at both the national and international levels is to enhance the reliability of the dataset. These articles were analyzed using VOSviewer software to identify trends in RCA research, research gaps, and to serve as a resource for future research. The VOSviewer software can analyze bibliographic data obtained from databases such as Scopus, Web of Science, Dimensions, and PubMed, and enables the creation of different types of networks [62]. With the help of the software, relationships such as citation analysis, co-authorship analysis, co-occurrence analysis of keywords, and bibliographic matching can be linked to reveal scientific structures in the field of research [63]. Furthermore, the clusters displayed in different colors thematically show similar research topics in the literature [64]. The research methodology diagram summarizing the study’s methodology is provided in Figure 1.

3. Findings

3.1. The Intensity of Studies on RCA

The distribution of studies on RCA by year is shown in Figure 2.

According to Figure 2, studies on RCA have increased over the years, showing a polynomial (i.e., dynamic) growth trend. While only one article was published in 2006, the academic interest in the topic gradually increased after 2015, showing a steady growth trend in the literature, and reaching seven publications in 2019. This process can be described as the study’s start-up period. The upward acceleration in the rate of article publication after 2021 is noteworthy. Research on RCA has accelerated, particularly since 2022. The number of articles, which was 43 in 2022, reached 51 in 2023, 97 in 2024, and 121 in 2025. The highest increase (207%) occurred in 2022 compared to the previous year. This period represents a phase of rapid growth. This process, in particular, has supported the increase in publication output by highlighting the growing recognition of RCA’s potential as a sustainable construction material. This strong upward trend can be explained by the increasing relevance of RCA to the global environmental goals, particularly its direct association with sustainability, CO₂ emissions reduction, and the circular economy. As of 1 March 2026, 29 articles have been published for 2026. This figure does not represent the total number of articles for 2026. Therefore, it is anticipated that the number of published articles will increase until 31 December 2026.

Overall, the findings clearly indicate that research on RCA has evolved into an increasingly attractive field of study, driven by critical global concerns such as climate change, CO₂ emissions, and sustainability [9,18,44]. Therefore, it is important to determine the volume and quality of research in the field through scientometric analysis to contribute to the literature.

3.2. Prominent Research Areas

Keywords are important indicators that reflect the fundamental topics of published studies, revealing enduring research themes and primary areas of focus within a field [65,66]. Therefore, the most frequently recurring keywords in the RCA research field were analyzed using VOSviewer software. To identify important research areas related to RCA, a minimum occurrence value of five was selected. A map of the recurring words in the articles is shown in Figure 3.

As shown in Figure 3, only 54 keywords met the minimum threshold value. The nodes on the map represent the frequency of keyword use, whereas the connections between nodes represent the relationship between keywords [67]. Accordingly, the keywords machine learning, recycled aggregate concrete, mechanical properties, and concrete are at the center of the network. This indicates that the research topic of RCA using the ML method has been a popular area of research in recent years. Words grouped in red, such as microstructure, pore structure, and mechanical properties, represent studies that examined the microstructure of RCAs after they had undergone mechanical testing. Words grouped in green, such as sustainability, sustainable concrete, green concrete, and gene expression programming, represent studies that examined the sustainability-focused modeling of RCAs. The blue group, containing words such as waste, concrete, aggregate, and circular economy, referred to studies that examined the waste management aspects of RCA. The yellow group, containing words such as recycled aggregate concrete, compressive strength, and bond strength, referred to studies that examined the mechanical properties of RCA. The grouping of recurring words by research area is presented in

Table 2. Grouping of repeated words by research area.

ID	Keywords	Cluster	Rate (%)	Links	Total Link Strength	Occurrences
205	Construction and demolition waste	Thermo-mechanical analysis of recycled aggregate concrete	59.25	19	29	17
266	Durability			13	25	13
371	Fly ash			17	28	15
547	Mechanical performance			7	7	5
549	Mechanical properties			28	67	46
550	Mechanical property			6	7	5
578	Microstructure			20	33	21
730	Pore structure			8	12	7
764	Random forest			19	30	10
807	Recycled coarse aggregate			19	25	19
813	Recycled concrete			10	18	13
815	Recycled concrete aggregate			17	29	19
818	Recycled concrete aggregates			7	8	6
859	Recycling			8	10	6
928	Self-compacting concrete			16	24	10
1006	Strength prediction			11	13	5
26	Aggregate			6	9	5
144	Circular economy			9	11	6
169	Compressive strength			43	190	95
174	Concrete			29	70	30
368	Flexural strength			10	17	8
452	High temperature			8	9	5
792	Recycled aggregates			22	41	25
1104	Waste			9	12	6
87	Bond strength			8	12	6
173	Compressive strength prediction			11	18	12
286	Elastic modulus			12	20	8
288	Elevated temperature			11	12	7
359	Finite element analysis			6	11	12
786	Recycled aggregate concrete			32	109	61
787	Recycled aggregate concrete (RAC)			8	15	11
1044	Sustainable construction			17	24	10
47	Artificial intelligence	Machine learning and optimization-based prediction	40.75	14	21	9
522	Machine learning			50	225	102
606	Modeling			10	14	5
678	Optimization			8	9	6
744	Prediction			13	27	8
785	Recycled aggregate			24	51	26
886	Response surface methodology			10	12	5
1027	Supplementary cementitious materials			17	37	14
279	Eco-friendly concrete			11	17	6
939	SHAP			8	16	6
1145	XGBoost			10	17	7
629	Multi-objective optimization			5	12	7
49	Artificial neural network			13	16	6
299	Ensemble learning			10	13	5
303	Ensemble models			8	14	6
396	Gene expression programming			13	15	6
434	Green concrete			13	15	7
911	Rubberized concrete			7	9	5
938	Sensitivity analysis			11	15	5
940	SHAP analysis			13	16	7
1035	Sustainability			25	57	23
1043	Sustainable concrete			15	29	13

.

This study has been categorized into two main groups according to Table 2: determining mechanical and thermal performance, and prediction based on optimization using machine learning. The total link strengths of the keywords compressive strength (190), recycled aggregate concrete (109), and machine learning (225) are noteworthy. This indicates the popularity of the topic of compressive strength prediction for RCA. ML techniques (random forest, XGBoost, ANN, ensemble models, etc.) have been frequently used to predict RAC properties. Among these models, RF performs well in capturing nonlinear relationships, particularly when the input variables are complex [68]. On the other hand, XGBoost demonstrates superior predictive performance due to its efficient handling of nonlinear relationships and built-in regularization mechanisms [69]. Its ability to achieve high accuracy is particularly evident in heterogeneous datasets such as those involving RCA [70]. Compared to RF, XGBoost generally provides slightly higher prediction accuracy in complex problem domains [71]. Similarly, ANN yields strong performance in modeling nonlinear relationships. However, it typically requires longer training times, especially with large datasets, and may be prone to overfitting if not properly tuned [72,73]. Therefore, while RF offers relatively more robustness and some level of interpretability, ANN is well-suited for capturing complex patterns, and XGBoost stands out in terms of accurate and reliable generalization performance [74].

However, sustainability is increasingly coming to the fore, particularly from the keywords green concrete and eco-friendly concrete. Again, according to the table, the combination of multi-objective optimization and eco-friendly concrete stands out as an area that has been little studied in the literature and shows promise for the future.

In addition to the information above, it is important to consider the mechanical and microstructural properties of RAC. In particular, the physical, chemical, and mechanical characteristics of RAC are key parameters that directly influence its overall performance. However, compared to natural aggregates, recycled aggregates generally exhibit a weaker and more porous structure, which can lead to reductions in mechanical performance. Due to the increased porosity, adhered mortar, and the presence of microcracks within its internal structure, RAC exhibits greater heterogeneity, especially in the interfacial transition zone, which is considered the weakest region in concrete. Therefore, integrating microstructural parameters of RAC into prediction models can improve the prediction accuracy and provide a more comprehensive understanding of its behavior. This also highlights a significant research gap in the existing literature. However, these microstructural deficiencies have significant consequences for the long-term performance and durability of RAC. Excessive microstructural porosity and a weak interfacial transition zone can facilitate the ingress of harmful substances such as sulfates and chlorides, thereby accelerating reinforcement corrosion. Similarly, the increased susceptibility of RAC to freeze–thaw cycles can negatively affect its long-term mechanical stability and structural safety under harsh environmental conditions. Therefore, durability parameters should also be evaluated to ensure the reliable use of RAC.

3.3. Leading Researchers

The Leading Researcher analysis was conducted to identify researchers who contributed the most to scientific production in the RCA field and guide the development of the literature. The lines in the network maps represent author collaborations, the color clusters represent collaboration networks, and the nodes represent the number of citations and total connection strength of that author [75]. For this purpose, the minimum number of documents per author was set to 2 in the VOSviewer software. The main researchers working in the RCA field are shown in Figure 4.

According to the network map shown in Figure 4, Ahmad Waqas, Ahmad Ayaz, and Aslam Fahid are the leading authors working on RCA, as they have the largest node and the most connections. This indicates that they frequently collaborate. Within the green group, Ahmad Ayaz and Ahmad Waqas have formed a strong collaborative network. Within the blue group, Jameel Mohammed and Kabbo Md Kawsarul Islam have dense connections among themselves. Similarly, authors in the orange, red, yellow, and purple groups have formed separate collaboration groups. Furthermore, Joyklad Panuwat, Onyelowe Kennedy Chibuzor, and Jameel Mohammed stand out as authors who act as bridges between different groups. Citation and article information for the main researchers working in the field of RCA are provided in

Table 3. Citation and article information for the main researchers working in the field of RCA.

ID	Author	Cluster	Rate (%)	Links	Total Link Strength	Documents	Citations
57	Al-Khafaji, Zainab S.	1	18.75	4	4	2	42
129	Arunachalam, Krishna Prakash			6	7	2	12
151	Azize, Noralhuda M.			6	8	2	43
184	Bernardo, Luís Filipe Almeida			4	5	2	189
940	Onyelowe, Kennedy Chibuzor			11	12	5	44
943	Ostrowski, Krzysztof Adam			6	6	2	149
1228	Thoeny, Zainab Abdul Rdha			6	8	2	43
1560	Ibrahim, Majed			2	3	2	97
1565	Imran, Hamza			8	12	5	289
31	Ahmad, Ayaz	2	18.75	7	15	5	492
36	Ahmad, Waqas			7	17	6	508
62	Alabduljabbar, Hisham			2	3	2	3
136	Aslam, Fahid			9	13	4	490
386	Farooq, Furqan			3	3	2	185
561	Javed, Muhammad Faisal			4	4	3	170
623	Khan, Kaffayatullah M.			4	9	3	126
904	Nasir Amin, Muhammad Nasir			6	11	4	130
1564	Iftikhar, Bawar			4	4	2	141
22	Abubakar, Sani Aliyu	3	16.67	4	7	2	17
110	Alzlfawi, Abdullah			5	8	2	32
132	Ashraf, Jawad			4	5	2	47
553	Jabin, Jannat Ara			2	3	2	54
559	Jameel, Mohammed			7	14	5	35
589	Kabbo, Md Kawsarul Islam			6	14	4	60
626	Khan, Md Munir Hayet			5	7	2	43
1157	Sobuz, Habibur Rahman			7	17	6	114
58	Al-Naghi, Ahmed Abdullah Alawi	4	16.67	9	10	3	32
88	Ali, Tariq			6	10	3	1
183	Ben Kahla, Nabil			10	13	4	37
442	Ghazouani, Nejib			9	11	3	30
1002	Qureshi, Muhammad Zeeshan			6	10	3	1
1026	Raza, Ali			4	5	2	62
1073	Salmi, Abdelatif			4	5	2	62
1566	Inam, Inamullah			5	7	2	0
34	Ahmad, Jawad	5	10.42	4	7	2	71
95	Almujibah, Hamad R.			7	7	2	40
815	Manan, Aneel			6	10	3	80
1256	Umar, Muhammad Wasif			6	10	3	80
1485	Zhang, Pu			4	7	2	71
68	Alaneme, George Uwadiegwu	6	10.42	8	11	4	43
649	Kiran, Golla Uday			4	6	2	13
899	Nakkeeran, G.			5	10	4	51
1051	Roy, Dipankar			4	6	2	13
1133	Shinde, Sumant Nivarutti			4	6	3	17
361	Ejaz, Ali	7	8.33	3	5	2	26
547	Hussain, Qudeer			3	5	2	26
585	Joyklad, Panuwat			7	7	2	107
1068	Saingam, Panumas			3	5	2	26

.

Table 3 presents the authors’ impact, collaboration, and status in the literature under seven different groups. These groups were formed based on common research topics and the intensity of collaboration in the literature. The first group comprises authors who contribute to strong collaboration and receive high citations. The main authors in this group are Al-Khafaji Zainab S., Arunachalam Krishna Prakash, Azize Noralhuda M., Bernardo Luís Filipe Almeida, and Onyelowe Kennedy Chibuzor. The authors in Group 2 are in a central research group position because of strong collaboration links (link strength 17). Group 3 authors represent a medium-sized research group with moderate citation and collaboration levels. Group 4 authors focus on emerging research areas and understudied topics. This group has relatively low citation and link counts. Group 5 authors have relatively lower citation counts but are strong in terms of connections. Group 6 authors appear to be moving towards emerging research areas with low citations and document counts. Group 7 consists of smaller authors with less collaboration.

Based on the information presented above, it can be concluded that collaboration patterns play a significant role in directly influencing research quality. Strongly connected groups, in particular, tend to produce more consistent and higher-quality studies. Furthermore, researchers who bridge gaps between different author groups foster interdisciplinary collaboration and contribute to innovation. In this context, integrating emerging research topics such as RCA into the field of sustainability through intergroup collaborations can lead to more advanced and sustainable solutions.

3.4. Top Journals

The primary objective of the Top Journals analysis was to reveal which journals are at the center of scientific knowledge production in the RCA field. To this end, the minimum number of documents per source was set to five in the VOSviewer software to identify the most influential journals publishing in the RCA field. The density and connections of the network map allowed us to determine which journals researchers were turning to and which topics were closely related. A map of the most influential journals publishing in the RCA field is shown in Figure 5.

As shown in Figure 5, the Construction and Building Materials journal, which is in the red group and has strong network connections with other colored groups, stands out as one of the most fundamental publications in the field. However, red group journals are RCA-focused materials journals (Construction and Building Materials, Materials Today Communications, Journal of Building Engineering, Applied Sciences (Switzerland), Buildings, Materials, etc.), green group journals are engineering and sustainability-focused journals (Scientific Reports, Sustainability (Switzerland), Case Studies in Construction Materials. Structures, etc.), while blue-group journals refer to journals focused on the environment, production, and interdisciplinary approaches (Journal of Cleaner Production, Innovative Infrastructure Solutions, Asian Journal of Civil Engineering, Multiscale and Multidisciplinary Modeling, Experiments and Design, etc.). The high connectivity within the red group is evidence of the strong network formed by RCA and its literature. Similarly, the medium connectivity of the red–green and red–blue connection lines indicates a significant degree of interaction between the materials science, sustainability, and environmental journals. Journals grouped according to the number of references and articles are presented in

Table 4. Journals grouped by reference and article count.

ID	Journals	Cluster	Rate (%)	Links	Total Link Strength	Documents	Citations
12	Applied Sciences (Switzerland)	1	40.00	10	30	8	91
17	Buildings			9	42	15	222
32	Construction and Building Materials			14	176	48	2663
56	Journal of Building Engineering			13	81	18	808
76	Materials			14	110	24	765
78	Materials Today Communications			10	37	9	125
18	Case Studies in Construction Materials	2	33.33	14	97	19	297
96	Results in Engineering			10	19	5	15
101	Scientific Reports			12	58	14	130
107	Structures			9	23	6	54
109	Sustainability (Switzerland)			12	29	13	165
15	Asian Journal of Civil Engineering	3	26.67	8	29	9	56
58	Journal of Cleaner Production			12	60	14	947
85	Multiscale and Multidisciplinary Modeling, Experiments and Design			12	51	8	83
120	Innovative Infrastructure Solutions			11	60	8	130

.

As shown in Table 4, journals in the fields of construction, materials, and fundamental engineering (40%) are in the first group, journals in the fields of sustainability and applied engineering (33.33%) are in the second group, and innovative, multidisciplinary, and environment-focused journals (26.67%) are in the third group. Construction and Building Materials, in the first group, is the leading journal in the field, with the highest number of articles (176) and citations (2663). Case Studies in Construction Materials and Sustainability, in the second group, is the most-cited journal in the literature, particularly for practical applications and sustainable material use. Furthermore, the total link strength values indicate that these journals play an important bridging role in interdisciplinary studies. Finally, the Journal of Cleaner Production, which is in the third group, plays a critical role in the literature on sustainable production and environmental impact, with a high number of citations (947). In this context, these journals dominate the field due to their high publication volume, strong citation impact, interdisciplinary engagement, and emphasis on application-oriented research. As a result, they stand out as some of the most influential and widely recognized journals in the literature.

3.5. Prominent Research Institutes

The Prominent Research Institutes analysis was conducted to identify the most influential and productive institutions in the RCA field. For this purpose, the minimum number of documents per organizational value was set to three in the VOSviewer software. This analysis identified the universities and research centers with the strongest academic impact in the RCA field. A map of the major research institutions working in the RCA field is presented in Figure 6.

Figure 6 shows the collaboration network of institutions conducting research in the RCA field. Each circle in the figure represents a department. The main collaboration networks are shown in yellow and green, smaller subgroups are shown in blue, and different collaboration groups are shown in red. The Civil Engineering Department is at the center of the network, and there is intense academic collaboration between institutions. The institutions on the right side have weaker or fewer connections. This suggests that academic collaborations are concentrated around certain institutions. This situation shows that the quality of research in the RCA field is strongly influenced by the most collaborative institutions. However, less collaborative institutions may reveal new research opportunities by highlighting existing gaps in the literature. Furthermore, it can be concluded that dominant institutions tend to shape the direction of research, while peripheral institutions are more likely to develop and expand collaborative networks. Overall, this provides a realistic perspective on how institutional contributions and collaboration patterns influence developments in the RCA field. Accordingly, important research institutions grouped according to citation and article counts are presented in

Table 5. Important research institutions grouped according to citation and article counts.

ID	Organizations	Cluster	Rate (%)	Links	Total Link Strength	Documents	Citations
39	Chongqing University, Chongqing, China	1	20.00	4	5	3	133
59	College of Civil Engineering and Architecture Zhejiang University, Hangzhou, Zhejiang, China			3	4	5	225
70	College of Civil Engineering, Nanjing Tech University, Nanjing, Jiangsu, China			2	2	3	88
72	College of Civil Engineering, Taiyuan University of Technology, Taiyuan, Shanxi, China			3	3	5	52
92	College of Urban and Rural Construction, Zhongkai University of Agriculture and Engineering, Guangzhou, Guangdong, China			3	5	3	289
551	Key Laboratory of Disaster Prevention and Structural Safety, Guangxi University, Nanning, Guangxi, China			2	3	3	200
689	School of Civil Engineering, Shenyang Jianzhu University, Shenyang, Liaoning, China			5	5	3	21
707	School of Design and the Built Environment, Curtin University, Perth, WA, Australia			5	7	3	251
147	Department of Civil & Environmental Engineering, King Faisal University, Al-Ahsa, Ash Sharqiyah, Saudi Arabia	2	17.50	5	9	6	169
152	Department of Civil & Environmental Engineering, Srinakharinwirot University, Bangkok, Thailand			5	8	3	287
197	Department of Civil Engineering, Comsats University Islamabad, Abbottabad Campus, Abbottabad, Khyber Pakhtunkhwa, Pakistan			8	18	9	681
213	Department of Civil Engineering, Ghulam Ishaq Khan Institute of Engineering Sciences and Technology, Topi, Khyber Pakhtunkhwa, Pakistan			5	7	3	35
279	Department of Civil Engineering, Prince Sattam Bin Abdulaziz University, Al Kharj, Riyad, Saudi Arabia			13	24	10	655
485	Faculty of Civil Engineering, Politechnika Krakowska, Krakow, Lesser, Poland			3	6	3	329
605	National University of Sciences and Technology, Islamabad, Pakistan			8	9	5	37
205	Department of Civil Engineering, Dr. Vishwanath Karad Mıt World Peace University, Pune, Mh, India	3	17.50	2	4	3	17
232	Department of Civil Engineering, Kampala International University, Kampala, Uganda			17	24	9	61
238	Department of Civil Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok, Thailand			5	6	3	29
245	Department of Civil Engineering, Madanapalle Institute of Technology & Science, Madanapalle, Ap, India			2	4	4	38
252	Department of Civil Engineering, Michael Okpara University of Agriculture, Umuahia, Abia, Nigeria			8	10	3	31
300	Department of Civil Engineering, Srm Institute of Science and Technology, Kattankulathur, Tn, India			2	2	5	54
415	Department of Environmental Sciences, Al-Karkh University of Science, Baghdad, Baghdad, Iraq			3	3	3	192
24	Center for Engineering and Technology Innovations, King Khalid University, Abha, Asir, Saudi Arabia	4	15.00	10	25	5	31
273	Department of Civil Engineering, Northern Border University, Arar, Al Hudud Ash Shamaliyah, Saudi Arabia			9	16	3	29
301	Department of Civil Engineering, Swedish College of Engineering and Technology, Wah, Pakistan			8	14	3	1
321	Department of Civil Engineering, University of Engineering and Technology Taxila, Taxila, Punjab, Pakistan			9	18	6	66
325	Department of Civil Engineering, University of Ha’il, Ha’il, Ha’il, Saudi Arabia			10	13	3	32
596	Mining Research Center, Northern Border University, Arar, Al Hudud Ash Shamaliyah, Saudi Arabia			10	15	3	30
58	College of Civil and Transportation Engineering, Shenzhen University, Shenzhen, Guangdong, China	5	10.00	2	2	4	66
281	Department of Civil Engineering, Qassim University, Al-Mulida, Al Qasim, Saudi Arabia			2	2	5	65
303	Department of Civil Engineering, Taif University, Taif, Makkah Al Mukarramah, Saudi Arabia			10	12	3	40
703	School of Civil Engineering, Zhengzhou University, Zhengzhou, Henan, China			4	4	5	135
153	Department of Civil & Environmental Engineering, The Hong Kong Polytechnic University, Hong Kong, Hong Kong, Hong Kong	6	10.00	3	6	7	264
501	Faculty of Engineering, Kafr El-Sheikh, Kafr El-Sheikh, Egypt			3	3	3	231
716	School of Engineering, RMIT University, Melbourne, Vic, Australia			2	4	3	141
764	Shenzhen University, Shenzhen, Guangdong, China			3	5	4	196
132	Department of Building Engineering and Construction Management, Khulna University of Engineering and Technology, Khulna, Bangladesh	7	10.00	4	7	7	128
148	Department of Civil & Environmental Engineering, Majmaah University, Al-Majmaah, Riyad, Saudi Arabia			2	4	3	32
230	Department of Civil Engineering, Jouf University, Sakakah, Al Jawf, Saudi Arabia			3	3	3	30
237	Department of Civil Engineering, King Khalid University, Abha, Asir, Saudi Arabia			15	39	12	102

.

Table 5 shows the inter-institutional collaboration network and group structure. The groups, which represent collaboration networks formed according to the institutions’ tendency to publish jointly, consist of seven clusters. Most of the institutions forming Group 1 are universities located in China (Chongqing University, Zhejiang University Civil Engineering College, Nanjing Tech University, Taiyuan University of Technology, Zhongkai University of Agriculture and Engineering, Guangxi University, and Shenyang Jianzhu University). Curtin University, also in this group, is an Australian-based institution. This indicates the establishment of a strong research network in the field of RCA between China and Australia. Institutions in Group 2, particularly universities based in Pakistan and Saudi Arabia (COMSATS University Islamabad, Prince Sattam bin Abdulaziz University, National University of Sciences and Technology, King Faisal University, and Srinakharinwirot University), have made significant contributions to the field of RCA in recent years. Group 3 institutions, particularly universities in developing countries (India, Uganda, Nigeria, and Africa) (SRM Institute of Science and Technology, MIT World Peace University, Kampala International University, and Michael Okpara University of Agriculture), have contributed to RCA research. This situation reveals that the RCA topic has attracted global interest. Group 4 institutions consist of universities in Saudi Arabia and Pakistan. Within this group, King Khalid University stands out with 12 articles, 102 citations, and 39 total link strength. Group 5 institutions constitute approximately 10% of the network. Within this group, Zhengzhou University and Qassim University stand out in terms of publication numbers and citation values. The institutions in Group 6 (Hong Kong Polytechnic University, RMIT University, and Kafr El-Sheikh University) exhibit a more international structure. Group 7 institutions mainly include universities from Bangladesh and Saudi Arabia.

3.6. Leading Countries

The Leading Countries analysis was conducted to determine which countries are pioneers in scientific production and research impact in the RCA research field. To this end, the minimum number of documents per country was set to three in the VOSviewer software to identify the most influential journals publishing in the RCA field. This analysis revealed the global distribution, collaboration networks, and scientific leadership structure in the RCA research field. The leading countries in the RCA field are presented in Figure 7.

As shown in Figure 6, China is the largest node in the blue group with 144 articles and 3612 citations. However, in terms of international collaboration, China (82), Saudi Arabia (87), and Pakistan (67) are among the countries with the strongest networks. Among the Asian countries, India is particularly influential in the field of RCA. Among the European countries, Spain and Portugal produce influential studies with high citation values despite having fewer articles. Similarly, the United States and Australia stand out in terms of scientific impact with their high citation values. It is noteworthy that Afghanistan has only one connection with Pakistan and Saudi Arabia. Furthermore, developing countries such as Turkey, Malaysia, Brazil, and Bangladesh are becoming increasingly prominent in the research network. However, the dense connections formed by different colored groups prove that RCA has created a strong network among countries, leading to significant interactions. The leading countries in the field of RCA are listed in

Table 6. Leading countries grouped by citation and article count.

ID	Country	Cluster	Rate (%)	Links	Total Link Strength	Documents	Citations
3	Australia	1	30.77	12	29	23	934
17	Ethiopia			7	7	4	80
19	France			2	2	4	25
21	Greece			4	4	3	330
23	Hungary			6	8	5	78
24	Japan			5	10	7	261
47	South Korea			12	14	11	265
57	United Kingdom			10	16	9	242
58	United States			15	32	26	805
59	Vietnam			11	12	12	467
60	India			17	24	63	1112
62	Iran			9	16	20	459
1	Afghanistan	2	17.95	2	6	3	1
16	Egypt			16	38	20	738
36	Pakistan			22	67	29	992
42	Russian Federation			13	15	5	152
43	Saudi Arabia			25	87	45	1054
49	Sudan			9	10	3	102
64	Ireland			6	10	4	210
12	Colombia	3	15.38	1	2	5	109
25	Jordan			7	8	4	110
39	Poland			9	12	11	423
40	Portugal			3	5	11	758
48	Spain			12	14	24	1705
63	Iraq			21	30	15	439
6	Bangladesh	4	15.38	7	18	12	320
8	Brazil			5	5	6	169
9	Canada			11	17	12	409
30	Malaysia			13	21	11	267
54	Turkey			10	15	14	242
65	Italy			9	11	7	189
10	Chile	5	12.82	5	6	3	15
33	Nigeria			9	15	5	66
46	South Africa			6	7	3	36
52	Thailand			9	17	8	333
55	Uganda			14	26	10	78
11	China	6	7.69	29	82	144	3612
20	Germany			4	5	3	6
22	Hong Kong			7	17	11	356

.

Table 6 presents the leading countries grouped under seven different categories based on the country collaboration network and scientific productivity. These groups were formed using indicators such as links, total link strength, documents, and citations. The first group comprises countries with advanced research ecosystems. The main countries in this group are Australia, Ethiopia, France, Greece, Hungary, Japan, South Korea, the United Kingdom, the United States, Vietnam, India, and Iran (30.77%). A significant portion of the studies on RCA in the literature have been published by these countries. Group 2 comprises Middle Eastern and South Asian countries. Within this group, Saudi Arabia stands out with 45 documents and 1054 citations. The third group comprises European and Latin American countries. Within this group, Spain has a very high scientific impact with 24 documents and 1705 citations. Portugal, despite its low number of publications (11), has a high impact value because of its high number of citations (758). The fourth group comprises developing countries. This group includes Turkey, Canada, Malaysia, Brazil, Italy, and Bangladesh. The fifth group comprises developing countries such as Africa, Nigeria, South Africa, Thailand, Uganda, and Chile. Thailand stands out with eight documents and 333 citations. Uganda plays an active role in the collaboration network with 14 links and a total link strength of 26. Group 6 comprises China, Germany, and Hong Kong, with a contribution rate of 7.69%. China, with 144 documents, 3612 citations, 29 links, and 82 total link strength, is at the center of the global research network and is by far the most productive and influential country in the field of research.

4. Discussion

In this study, a comprehensive literature review was conducted using VOSviewer software and scientific analysis methods to identify leading institutions, countries, authors, sources, areas, gaps, challenges, and trends in RCA research.

The most frequently recurring keywords in the study can be grouped into four categories: mechanical properties, microstructural properties, sustainability, and ML-based prediction methods. This study indicates an important research direction for achieving a more accurate prediction of RCA performance. Furthermore, sustainability is increasingly coming to the fore, particularly from the keywords green concrete and eco-friendly concrete. This situation emphasizes the importance of CDW management processes and reducing CO₂ emissions. However, the combination of multi-objective optimization and eco-friendly concrete is a promising area that has been little studied in the literature.

Articles within the scope of this study can be grouped around two main scientific approaches: determining mechanical and thermal performance, and prediction based on machine learning and optimization. Studies focusing on determining mechanical and thermal performance use experimental methods to examine the engineering properties of materials, such as compressive strength, flexural strength, modulus of elasticity, and high-temperature behavior. These studies are important because the performance data of structural materials are obtained by direct measurement. In contrast, machine learning and optimization-based studies aim to develop prediction models and optimize material properties using experimental data [76]. This method allows complex material behaviors to be modeled using existing experimental data without the need for new experiments, enabling the effects of different parameters to be determined through faster and lower-cost analyses. Therefore, researchers have frequently used machine learning techniques, such as random forest (RF), extreme gradient boosting (XGBoost), artificial neural networks (ANNs), and ensemble models (EM), to predict RCA properties.

These models, specifically RF, are used to predict and optimize the compressive strength of nonlinear RCA concrete. XGBoost predicts the compressive strength of concrete by learning the complex and nonlinear relationships between parameters, such as dosage, water/cement ratio, fine and coarse aggregate quantities, admixtures, and curing age, using these parameters as input variables [77]. ANN is an effective method for determining the impact of complex RCA-based concrete components on compressive strength [78]. EM, in contrast, uses three fundamental approaches— bagging, boosting, and stacking—to provide a lower error rate and higher prediction accuracy than machine learning algorithms used alone [79].

However, the Construction and Building Materials journal is the most prominent journal in the RCA field in terms of both the number of articles and citations. The main reason for this is that the journal has a high impact factor in the field of building materials and construction engineering. The increasing importance of the RCA research field in recent years has led researchers to prefer high-visibility and reputable journals in the field. This situation demonstrates that the journal Construction and Building Materials plays a central role in the RCA literature in terms of both productivity and scientific impact.

5. Conclusions

RCA has been a topic that researchers have focused on over the past 20 years. Interest in this subject has been increasing, particularly in recent years. Although many studies have been conducted on the subject in the literature, no study has yet been conducted that systematically analyzes and summarizes the existing literature and provides researchers with a comprehensive overview. In this regard, it is believed that the assessments obtained in this study will contribute to the development of more sustainable building materials in the field of RCA and will serve as an important reference for future research.

This study, which examines key parameters related to RCA, contributes to sustainability by reducing natural aggregate usage, lowering the carbon footprint, promoting sustainable building design, and facilitating green building certification. To this end, this study aims to evaluate the potential use of RCA to provide practical contributions to field applications and support practices that offer advantages in terms of sustainability. Furthermore, it aims to promote the use of concrete produced with RCA as a sustainable and environmentally friendly building material. Therefore, the more widespread use of RCA in concrete production can contribute to the development of low-carbon building applications.

5.1. Contributions of the Study

Through an in-depth review of the RCA literature, the study determined that research has focused on three main axes: the investigation of the mechanical and microstructural properties of RCA concrete; sustainability, waste management, and the circular economy perspective; and machine learning and optimization-based prediction models. This indicates that the RCA literature does not solely focus on material properties but is increasingly transforming into an interdisciplinary structure that incorporates environmental sustainability and data-driven engineering approaches. This assessment contributes to a better understanding of RCA’s effects on concrete behavior and the development of data-driven concrete design processes. The beginning of publications related to RCA can be considered to be 2006. Publications related to RCA began in 2006, and it entered a rapid growth trend after 2021 and a period of significant rapid growth from 2022 onwards, indicating that the research area is receiving increasing attention. Furthermore, this trend is inevitable and will continue in 2026 and beyond.

Furthermore, article, citation, and link parameters in the RCA field place China at the center of the global research network, making it far and away the most productive and influential country in the research field. Ahmad Waqas, Ahmad Ayaz, and Aslam Fahid are the most cited authors, while Ahmad Waqas and Sobuz Habibur Rahman are the authors with the most publications. These authors’ connections with Pakistan, Ireland, Saudi Arabia, and Bangladesh highlight that these countries encourage local research efforts in the RCA field and contribute to sustainability. Furthermore, the journal that publishes the most articles and receives the most citations in the RCA field is Construction and Building Materials, published by the Netherlands-based Elsevier publishing house.

In conclusion, the study presents important findings for both researchers and practitioners based on the results obtained. It is believed that the assessments presented in the study will contribute to a better understanding of RCA-containing concretes in terms of both performance and CO₂ emissions reduction and will serve as a guide for future research. Furthermore, the study will contribute to researchers developing publication strategies that integrate materials science, sustainability, and environmental issues through interdisciplinary work.

5.2. Recommendations for Academy and Practice

Based on the data obtained within the scope of this study, the recommendations for both academia and practice are as follows:

• The fact that research on eco-friendly concrete using multi-objective optimization is still limited presents an important research opportunity for the coming period.
• The use of different machine learning algorithms in future studies and validation of these models with experimental data may contribute to more accurate predictions of RCA concrete properties.
• It is recommended that microstructural analyses of RCA-containing concrete be conducted more comprehensively.
• Comprehensive analyses should be conducted to evaluate the effects of RCA use on carbon emissions, energy consumption, and environmental impacts.
• Research on optimizing CDW recovery processes and integrating circular economy principles into concrete production will support the development of this research area.
• Although numerous studies have been conducted on the mechanical performance of RCA-containing concrete, future studies should examine parameters, such as long-term performance and service life, in more detail.
• The development of multidisciplinary and integrated research approaches will contribute to a more comprehensive assessment of RCA from both technical and environmental perspectives.
• Encouraging interdisciplinary collaboration among researchers focusing on emerging and understudied topics can contribute to filling gaps in the literature.
• Interdisciplinary and sustainability-focused projects conducted by leading countries, such as China, the United States, and the United Kingdom, could pioneer the development of new trends and innovative methodologies in RCA research.

In conclusion, this study provides a comprehensive review of the literature on RCA research and contributes to systematizing the existing knowledge in the field. The findings are expected to serve as a reference point for future research and to strengthen the understanding of application processes. However, its limitations should be considered.

5.3. Limitations of This Study

The identification of keywords through 7953 articles determined via Web of Science, followed by the screening of these keywords via Scopus, resulting in a total of 465 documents, and finally the evaluation of 397 articles, constitutes the methodology of the study and can be considered a limitation of the study. However, using only the Web of Science and Scopus databases may have excluded relevant studies indexed in other databases, which could unintentionally introduce bias in database selection. Similarly, deriving keywords from selected publications may restrict and limit the search results. These aspects can be considered subjective limitations of the study. However, RCA research topics have grown in popularity over the past 20 years. Therefore, this study is limited to publications from 2006 to 1 March 2026. This limitation ensures a more detailed examination of modern, current, and advanced studies conducted in recent years and better reflects current research trends. Therefore, the specified limitations enable a clearer and more systematic presentation of the scope of the research, allowing for a more sound evaluation of the findings.

Consequently, this study contributes to understanding the current state of the research field and serves as an important reference and starting point for future research.