Strategic Approach of Reverse Logistics Management for Recyclable Waste and Transportation: A Systematic Review

Abstract

Strategic reverse logistics management is a key driver of sustainability in supply chains, where challenges in recyclable waste must be aligned with transportation systems to achieve optimal outcomes. A systematic review using the PRISMA methodology was conducted in December 2024 by searching Scopus, Google Scholar, and Thai Journals Online to examine the global research landscape and the strategic approaches applied in reverse logistics for recyclable waste and transportation. Analysis of 32 publications shows a steady rise in research, with most studies in Asia and dispersed across multiple journals, reflecting the field’s multidisciplinary nature. Four strategic approaches were identified. Model-driven approaches demonstrate strong capability through mathematical, computational, conceptual, and hybrid models, achieving reductions of 44% in climate impacts and 34% in costs. Technology-driven approaches contribute innovations to enhance battery transport safety. Exploratory approaches reveal contextual policy gaps and financial limitations. Hybrid approaches can improve efficiency and reduce CO₂ emissions. The future development of hybrid approaches still offers substantial room for broader application and deeper integration. This review supports the development of more effective systems, policies, and future research.

1. Introduction

Recyclable waste is a global environmental concern [1,2,3]. In developed countries, recycling is included in national environmental policies [4,5]. Governments establish frameworks, such as Extended Producer Responsibility (EPR) policies, that make producers responsible for waste management across the product life cycle [6,7]. In developing countries, recycling is commonly handled by the private sector. In Thailand, the recycling industry has grown [8], and the recycling and waste management market was valued at USD 9.29 billion in 2021 [9]. Recyclable materials are often treated as commercial goods [8,9,10] rather than part of environmental management. Managing recycled waste involves logistical challenges, especially in transporting materials from the source to the recycling plant [11,12]. Although recycling helps reduce environmental impact, the transportation process can generate pollution [13]. If reverse logistics are poorly managed, recycling may relocate environmental impacts rather than reduce them.

Reverse Logistics (RL) for recyclable waste is the process of managing the backward flow of used materials from consumers to recycling facilities or manufacturers for reuse [14,15,16] (Figure 1). In theory, RL supports the Circular Economy (CE) [17,18] and Global Sustainable Development Goals (SDGs) by promoting resource efficiency and environmental sustainability [19,20]. RL can reduce costs [21], minimize carbon footprints, and enhance resource recovery by optimizing transportation [22,23]. In practice, RL faces challenges in effectively managing recyclable waste and transportation [12].

Recyclable waste poses a challenge, as it must be sorted by type and cleaned at the source before being transported to recycling facilities [24,25]. Failure to properly sort the waste leads to contamination and reduces the efficiency of the recycling process [11]. Recycling plants require a steady and predictable supply of recyclable materials to operate efficiently. However, waste availability often fluctuates, and waste deliveries do not always align with the plant’s production schedule.

The transportation challenge is that recyclable waste depends on transport for its collection and distribution to recycling plants [11,12,26]. Due to the constraints posed by the size and quantity of recyclable waste [26], it is essential to select suitable transportation methods, including trucks, trains, or other vehicles. This process involves factors such as distance, route planning, fuel consumption, and coordination between collection points [11]. Without proper planning, transporting recyclables can waste energy and cause environmental pollution [13].

Recyclable waste management is closely linked to transportation activities, and both challenge the performance of reverse logistics [24,25,26]. Effective logistics and waste management, combined with better transport routing, significantly improve resource recovery [27]. Although reverse logistics is widely recognized as essential for waste reduction and circular economy goals [19,20,22,23], it is still in its infancy in developing countries [28] but has grown valuable [8,9], and RL findings on reverse logistics remain scattered [29]. To address the significant gap in RL, a systematic review combined with bibliometric analysis is necessary to map publication trends, geographic contributions, keyword patterns, and authorship networks, and to identify influential research and reliable sources for further study. Accordingly, this review aims to answer the following research question. RQ1: What is the global research landscape of reverse logistics for recyclable waste and transportation as revealed through bibliometric analysis?

Reverse logistics research is diverse. Many theoretical studies use statistical and mathematical methods to solve the problem of cost-effectiveness or RL management of recycled waste, such as mixed-integer linear programming (MILP) [30,31]. Also, various practical research projects address the problem, such as the smart e-waste reverse system [32] and the Study of Green Procurement Practices [33]. Various studies have sought to overcome the challenges associated with reverse logistics.

Strategic reverse logistics management is recognized as a key driver of sustainability within supply chains [34] and can reduce losses and unplanned revenue [35]. Effective waste logistics management, coupled with improved transport routing, can significantly enhance resource recovery [27]. Despite there being a diverse range of methods in reverse logistics increasingly applied to improve performance and reduce environmental impact [36], they are still not sufficiently integrated within reverse logistics networks [37,38], and systematic classifications in the transportation of recyclable waste remain limited [39]. Innovation supporting reverse logistics, the circular economy, and sustainable supply chains should be further explored [40]. The influence of transportation on sustainability outcomes in recyclable waste logistics also requires deeper examination [41,42]. Furthermore, researchers emphasize that collecting more comprehensive data will open new research pathways for advancing knowledge in this field [43]. Despite these indications, the existing literature has not consolidated strategic approaches or synthesized their achievements and limitations. This limits the understanding of how recyclable waste can be effectively integrated into transportation systems in reverse logistics. Accordingly, this review aims to address the following research question. RQ2: How can the strategic approaches in reverse logistics be classified, and what achievements, advantages, challenges, and limitations are identified in the literature?

A systematic review of previous literature was conducted to situate this study within the existing body of knowledge. Previous systematic reviews have laid the groundwork for understanding the role of emerging technologies in logistics. For instance, Sun et al. [38] highlighted both the potential and challenges of technologies such as IoT, AI, and automation in enhancing logistics performance across economic, environmental, and social dimensions. Their review also proposed future research directions, including smart reverse logistics and semi-autonomous transportation. However, the review treats reverse logistics mainly as a field of technological applications and does not consider other approaches or focus specifically on recyclable waste.

Similarly, the study by Sonar et al. [12] identifies and prioritizes key barriers to implementing reverse logistics within the circular economy framework, providing valuable insights into strategic and operational challenges. Several barriers were initially identified from past academic literature and finalized using the fuzzy Delphi and DEMATEL methods. However, the study does not address the research gap concerning reverse logistics management specifically for recyclable waste and transportation.

This systematic review aims to synthesize the existing literature by mapping the overall research landscape and classifying reverse logistics management for recyclable waste and transportation. The findings are expected to guide future research and support the development of more effective systems and policies, contributing to both academic understanding and practical solutions in sustainable waste management.

2. Methodology

We accomplish this by employing transparent, systematic reviews and the meta-analysis (PRISMA) technique, ensuring the results’ reliability and reproducibility while mitigating bias to address the research gap in reverse logistics management for recyclable waste and transportation. In December 2024, 3 researchers and 2 professors were invited to review selected articles and offer recommendations to confirm that the selected research addresses the research questions and is comprehensive within the database.

2.1. Database

The primary databases utilized for this study included Scopus, Google Scholar, and Thai Journals Online. The largest database, Scopus, aggregates abstracts, references, and works from renowned authors worldwide [44]. Google Scholar is essential for mitigating publication bias because it encompasses both scholarly and gray literature and is the preferred search engine for many researchers [45]. Finally, unlike other studies, Thai Journals Online was chosen as the central electronic journal database system, encompassing all disciplines in Thailand. Utilizing this database provides valuable information for future research on Thailand’s recyclable waste market and reverse logistics, particularly within the context of developing countries. It should also be noted that the research team did not have institutional access to the Web of Science (WOS). Therefore, Scopus was selected as the primary database due to its broad coverage of high-quality international publications.

2.2. Data Collection

The research process follows the PRISMA guidelines, which recommend four crucial steps for reporting publications in systematic literature review procedures aligned with our established research questions (Figure 2 and Supplementary Materials).

• Identification phase:

Based on the identified research gaps and research questions, recyclable waste and transportation are key factors that can enhance the effectiveness of reverse logistics management. The keywords for the literature search include “reverse logistics”, which is the central focus, combined with “recycling waste” or “recycle waste”, which are both commonly used, and “transportation” or “vehicle”, reflecting the connection to related transport aspects.

Scopus was used for searching documents with title + abstract + keywords using keyword search words (reverse logistic) AND ((recycling) OR (recycle)) AND (waste) AND ((transportation) OR (vehicle)). This database provided a total of 205 documents between 2005 and 2024.

In Google Scholar, a search for “secondary source” or the gray literature was conducted to include limited published documents such as reports, theses, academic conference papers, and others that might not be formally published; the search aimed to ensure comprehensiveness. The search criteria involved the terms (reverse logistic) AND ((recycling) OR (recycle)) AND (waste) AND ((transportation) OR (vehicle)) within the title–abstract field. We obtained 33 publications from 2023 to 2024.

Thai Journals Online (ThaiJO) is a national research database in Thailand. Using this database will provide valuable information for future research on reverse logistics management for recyclable waste and transportation. The search was conducted using the criteria of title + abstract, using the keyword search “reverse logistic AND recycling”, “reverse logistic AND recycle”. We obtained 13 publications from 2017 to 2024. Broader keywords were used because preliminary searches showed that adding the terms “waste” or “transportation” returned zero results. Therefore, broader terms “reverse logistic AND recycling” were used to capture all potentially relevant studies, followed by manual screening in the screening phase.

A total of 251 publications were identified from 2005 to 2024. The remaining publications proceeded to the screening process after 15 documents were excluded due to duplication in the Scopus and Google Scholar databases.

2.

Screening phase:

The document search yielded 236 publications spanning from 2005 to 2024, obtained during the identification phase and subsequently considered in the screening stage. A preliminary scan of titles and abstracts was conducted to determine. Following this screening, 140 publications were excluded for not meeting the inclusion criteria, specifically because they were not relevant to reverse logistics, waste, and transportation.

During the retrieval stage, 96 publications remained, while 16 could not be accessed in full text. This was due to unavailability in the databases accessible to the research team, lack of peer-reviewed status, or the use of languages other than English or Thai, which limited interpretability.

In the eligibility assessment phase, 80 publications were retained. A total of 156 publications were excluded after independent full-text review by the research team because they were not related to reverse logistics, recyclable waste, and transportation.

3.

Including phase:

Finally, after meticulous consideration and thorough discussion among all authors, a total of 32 publications were included in this systematic review.

2.3. Data Analysis

To address the first research question, RQ1 (What is the global research landscape of reverse logistics for recyclable waste and transportation as revealed through bibliometric analysis?), descriptive bibliometric analysis was conducted of all 32 selected publications. Articles were categorized by publication year, geographic origin, research source, keyword co-occurrence, co-authorship, and citation count, providing a structured overview of research distribution and scholarly influence.

For the second research question, RQ2 (How can the strategic approaches in reverse logistics be classified, and what achievements, advantages, challenges, and limitations are identified in the literature?), descriptive and comparative analyses were conducted, and a thematic synthesis was used to classify the main and sub-approaches and their reported achievements. The synthesis clarifies their practical and theoretical implications and identifies directions for future research.

3. Results

This section presents the results of the systematic literature review, divided into two parts. Section 3.1. Global Research Landscape of Reverse Logistics for Recyclable Waste and Transportation—a bibliometric assessment that identifies publication trends, the geographic distribution of articles, keyword co-occurrence patterns, co-authorship networks, citation impacts, and the sources of research publication. Section 3.2. Strategic Approaches of Reverse Logistics Management for Recyclable Waste and Transportation Identified in the literature—summarizes main and sub-approaches, achievements, advantages, challenges, limitations.

3.1. Global Research Landscape of Reverse Logistics for Recyclable Waste and Transportation: A Bibliometric Assessment

3.1.1. Publication Trends

This review, based on 32 articles published between 2010 and 2024 (Figure 3), provides an overview of research on reverse logistics for recyclable waste and transportation over the past 14 years. The analysis shows both growth in the number of studies and increasing academic interest in the field.

From 2010 to 2012, there was only one article in 2010. This number dropped to zero in 2011 and 2012, and the topic remained at its nascent stage and had not yet attracted widespread academic attention.

From 2013 to 2015, two to three articles were published per year, and in 2015, that number increased to three. This period marks the beginning of a steady, growing interest in publications. This growth is associated with the introduction of the United Nations’ Sustainable Development Goals (SDGs) in 2015, emphasizing enhanced resource management, waste reduction, and the promotion of circular economy principles.

From 2016 to 2020, the number of articles fluctuated significantly, with two to three articles per year in other years. There were no articles published in 2016 and 2020, reflecting discontinuity.

Since 2021, the publication trend has returned to steady growth. In 2021, there were two papers. In 2022–2023, four papers will be published per year. This number will increase to a maximum of six in 2024. This increase likely reflects growing international attention and the rising adoption of Net Zero policies. This trend also aligns with a broader academic movement toward interdisciplinary research and a growing volume of publications across fields.

3.1.2. Geographic Distribution of Articles

The geographic distribution of the reviewed studies shows that research output is concentrated primarily in Asia (

Table 1. Country of publications.

Country	Number of Publications	Proportion (%)
Thailand	8	25
China	7	21.88
Turkey	3	9.38
Brazil	2	6.25
United Kingdom	2	6.25
Canada	1	3.13
Germany	1	3.13
Hong Kong	1	3.13
India	1	3.13
Iran	1	3.13
Italy	1	3.13
Kenya	1	3.13
Netherlands	1	3.13
Poland	1	3.13
Ukraine	1	3.13

). Thailand contributed the most publications (eight articles), followed by China (seven articles) and Turkey (three articles). Several countries contributed moderately, including Brazil and the United Kingdom (two articles). There were additional articles from Canada, Germany, Hong Kong, India, Iran, Italy, Kenya, the Netherlands, Poland, and Ukraine, which contributed one each.

Of the eight studies from Thailand, seven were indexed in ThaiJo and one in Scopus, reflecting the context of developing countries. Thailand’s growing recycling trade industry is driving recycling activities. Overall, the findings indicate that although a diverse group of countries has contributed to the literature, research in this field remains heavily dominated by Asian regions.

3.1.3. Keyword Co-Occurrence

A keyword trend analysis was conducted using VOSviewer (version 1.6.20). A minimum occurrence threshold of three yielded 21 qualified keywords from 301 extracted terms. Each keyword’s total link strength (TLS) was used to assess its co-occurrence connectivity. (Figure 4,

Table 2. Words that occur at least 3 times in the 32 selected articles.

Word	n (Occurrences)	Total Link Strength (TLS)
logistics	9	50
recycling	10	49
reverse logistics	18	38
waste management	9	38
network design	5	35
transportation	4	32
electronic equipment	3	28
flame retardants	3	28
oscillators (electronic)	3	28
reverse logistics network design	3	28
design	4	25
article	3	20
electronic waste	3	19
solid wastes	4	19
supply chain management	3	19
sustainability	3	18
sustainable development	3	18
optimization	3	13
reverse logistic networks	3	13
sensitivity analysis	3	12
circular economy	3	8

).

Keyword co-occurrence analysis revealed three major thematic clusters. The first cluster (red) comprises core concepts in the field, with dominant keywords such as “reverse logistics”, “waste management”, “recycling”, and “logistics” that frequently co-occur and indicate the field’s central research focus. Analytical terms including “optimization” and “sensitivity analysis” also appear in this cluster, reflecting the prevalence of quantitative modeling approaches.

The second cluster (green) is associated with system and supply chain design, characterized by strong connections among “reverse logistics network design”, “transportation”, “design”, and “supply chain management”. This cluster also includes application-related terms such as “electronic equipment”, suggesting particular emphasis on electronic and hazardous waste flows.

The third cluster (blue) highlights sustainability-oriented studies, integrating keywords such as “network design”, “reverse logistic networks”, and “sustainability”. This cluster reflects research linking reverse logistics with environmental performance and sustainable development objectives. Overall, the co-occurrence map indicates a well-connected research landscape in which core RL concepts, network and supply chain design, and sustainability considerations are closely integrated across the three clusters.

3.1.4. Co-Authorship Analysis

Co-authorship analysis can help illustrate patterns of collaboration among authors. It is commonly used to reflect the shared contributions of multiple researchers involved in producing scholarly work.

The co-author network consisted of nodes representing individual authors (see Figure 5). In terms of total link strength, Ayvaz had the most significant influence in the network (2 documents, 227 citations, total link strength of 5). Bolat ranked second among the most influential authors (2 documents, 260 citations, total link strength of 3). Aydin ranked third (2 documents, 245 citations, total link strength of 4).

3.1.5. Number of Citations

The number of citations was examined to provide an overview of research in the field (Figure 6). Citation counts serve as indicators of a study’s academic significance and influence, helping identify key publications that have shaped discourse. This analysis also highlights trends in scholarly attention, as reflected by the growing volume of research citations. Highly cited studies underscore their value as foundational sources of knowledge, especially in examining reverse logistics and recyclable waste management within the given context.

The provided data indicates that the top 3 most referenced publications out of the 32 considered are those authored by Ayvaz et al. (2015) [46]. Reverse Logistics Implementation in the Construction Industry: Paper Waste Focus [46] develops a multi-level reverse logistics model that manages uncertainty in quantity, quality, and transport cost and provides profit-oriented solutions for waste electrical and electronic equipment. Ayvaz has the highest influence in Section 3.1.4.

Zhao & Ke (2017), in Incorporating Inventory Risks in Location-Routing Models for Explosive Waste Management [47], develop an inventory and routing model for hazardous waste, demonstrating significant reductions in both cost and risk.

Zhou & Zhou (2015), in Designing a Multi-Echelon Reverse Logistics Operation and Network: A Case Study of Office Paper in Beijing [48], contribute a multi-level network design for office paper waste using a nonlinear integer model to determine optimal facility locations and analyze waste volumes, capacities, and transport costs.

These publications are central because they introduce influential frameworks and decision structures that continue to shape the development of reverse logistics research.

Highly cited studies have significantly shaped discourse on reverse logistics, recyclable waste transportation, and broader sustainability strategies. In contrast, less-cited works may reflect their recent publication or limited dissemination, yet they still provide valuable insights by highlighting research gaps, testing novel methods, and addressing overlooked issues. Note that publications indexed in ThaiJo, the central Thai database, show zero citation counts (six of the seven articles). Six papers are written in Thai language, and several ThaiJo journals are not indexed in Google Scholar, which further contributes to their low visibility and citation frequency.

3.1.6. Research Publishing Source

Investigating research publishing sources provides insight into the dissemination patterns and scholarly landscape of the field in reverse logistics Management for recyclable waste and transportation. The research publishing sources identified in this review may serve as a valuable reference for future studies. Among the 32 articles, only Resources, Conservation and Recycling published two papers, while all other journals contributed just one article each. This lack of repetition indicates a significant dispersion of scholarly interest across a wide range of research publishing sources, with no single journal or set of specialized publications dominating the field.

Consequently, research on this topic appears across publications spanning multiple disciplines, reflecting its interdisciplinary nature and suggesting that researchers must consult a variety of research publishing sources to gain a comprehensive understanding of the field. The research source domains of the selected publications were categorized into five source domain categories based on the characteristics of the journals.

Environmental Science and Sustainability (9 journals): Publications in this domain include Waste Management and Research, Resources, Conservation and Recycling (2 articles), Recycling, Environment, Development and Sustainability, Transformation Towards Circular Food Systems, Sustainability, Environmental Modeling and Assessment, IOP Conference Series: Earth and Environmental Science, and Journal of Energy Storage.

Engineering and Industrial Technology (8 journals): Publications in this category include System Science and Simulation Engineering, International Journal of Production Economics, International Journal of Sustainable Engineering, International Journal of Physical Distribution and Logistics Management, Automation in Construction, Production, Engineering and Applied Science Research, and Current Applied Science and Technology.

General/Applied Science (6 journals): Publications in this category include A KKU Research Journal, Songklanakarin Journal of Science and Technology, CIE44 and IMSS’14 Proceedings, Journal of Physics: Conference Series, Annals of Operations Research, and Urban Science.

Social Science, Management, and Education (5 journals): Publications in this category include Doctor of Philosophy in Social Sciences Journal, Ph.D. in Social Sciences Journal, LogForum, International Journal of Social Sciences Management and Entrepreneurship (IJSSME), and Panyapiwat Journal.

Transportation and Logistics (3 journals): Publications in this domain include Frontiers in Traffic and Transportation Engineering, IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC, and The Journal of KMUTNB.

However, the articles’ titles, authors’ names, and other related information for the selected articles discussed in this section are included in Appendix A.

3.2. Strategic Approaches of Reverse Logistics Management for Recyclable Waste and Transportation Identified in the Literature

3.2.1. Strategic Approaches and Achievements Identified in the Literature

To address the existing research gap, this study examines various strategic approaches to reverse logistics management for recyclable waste and transportation. This analysis facilitates a more comprehensive understanding of how recyclable waste can be effectively matched with transportation systems to achieve optimal solutions for reverse logistics management. From 2010 to 2024, the 32 included studies were examined by independent researchers who triangulated and coded keywords in the full texts to identify the strategic approaches used in each article.

The classified codes revealed three main categories and an additional hybrid category (Figure 7). The literature review indicates that strategic approaches to reverse logistics management manifest in diverse forms and operate at varying levels of sophistication. The model-driven approach uses analytical and computational models to optimize reverse logistics [74], including mathematical programming [62] for facility location, transportation routing, and algorithms for efficient problem-solving [75]. Although the model-driven methods span a wide range, the researchers’ coding identified four sub-approaches: mathematical, computational, and conceptual models, as well as hybrid models that integrate multiple modeling. Mathematical models use equations for optimization [62], computational models use algorithms or simulations to solve complex problems [56], and conceptual models map processes and system structures [70]. The technology-driven approach integrates emerging technologies with reverse logistics to improve tracking, monitoring, and management of recyclable waste flows [32,69]. The exploratory approach examines contextual challenges and management practices in reverse logistics through case studies and qualitative analysis to identify practical barriers and opportunities across various regions or industries [45]. The hybrid approach combines elements of the above strategies to create comprehensive reverse logistics management.

From 2010 to 2024, the 32 included studies were examined by independent researchers who triangulated and coded keywords in the full texts to identify the strategic approaches used in each article. The classified codes revealed three main strategic categories and an additional hybrid category (Figure 7). Reverse logistics management for recyclable waste and transportation can be classified into three main categories. 23 articles (most studies) focused on the model-driven approach, 9 articles used mathematical models [31,46,48,51,59,62,63,67,72], 4 articles used computational models [39,52,56,70], 2 articles used conceptual models [55,73], and 8 articles used hybrid models [30,47,54,57,58,61,66,71]. This was followed by 5 articles that used the exploratory approach [33,50,53,65,68] and 1 article that used the technology-driven approach [60]. Hybrid approaches combining elements from these categories were identified, and 2 articles employed a model-driven + technology-driven approach [32,69], and 1 article used a model-driven + exploratory approach [64].

Table 3. Summary of strategic approaches, achievements, advantages, challenges, and limitations of the research.

Strategic Approaches	Ref.	Achievements	Advantages	Challenges	Limitations
1. Model-driven Approach (23 articles)
1.1 Mathematical Models (9 articles)	[62]	The Campo Limpo System cuts environmental impacts in 9 of 13 categories, reducing climate change impacts by 33–44% and overall impacts by 20–79%.	Shows major reductions in environmental impacts. Confirms recycling as the best end-of-life option.	Long-distance transport increases emissions, and improper rinsing raises residual pesticide impacts.	Transport flows are modeled estimates. Hazardous waste modeled without chemical detail.
	[46]	Presents a waste electrical and electronic equipment RLND model that identifies profit-maximizing solutions under uncertainty and offers flexible support for managerial use.	Develops a generic multi-echelon RL model. Manages uncertainty in quantity, quality, and transport cost.	High uncertainty in quality and costs. Limited adoption due to budget/capability constraints.	Focuses on a two-stage stochastic model that depends on sufficient data and needs computational methods to solve.
	[59]	Develops a plastic waste recycling model that minimizes the total reverse logistics cost in Indonesia, yielding approximately USD 11,773.	Develops a cost-minimizing mathematical model for LDPE and HDPE recycling.	Complex decisions involving waste quantities, capacities, and routing.	Based on Indonesia’s data. Accuracy depends on assumed costs and capacities.
	[31]	This study applies the MILP model to design an e-waste recycling network in Turkey. Improves recovery efficiency and reduces system cost through optimized routing and facility planning.	Develops a MILP model for the design of an e-waste RL network. Identifies optimal facility locations and capacities.	RL networks face uncertainty in waste amounts and transport costs. Multi-stage systems add operational complexity.	Deterministic models cannot handle uncertainty. More complex systems may need heuristic methods.
	[51]	Multimodal transport in plastic recycling reduces transportation and emissions costs by nearly 20%, enhancing sustainability and efficiency, especially on long-distance routes.	The MILP model evaluates a multi-modal plastic recycling network with full cost and emission assessments.	Network performance depends on location, distance, and transport modes, with varying channel sensitivities adding complexity.	Use fixed existing facility nodes. Municipal collection is not optimized in the model.
	[63]	The stochastic end-of-life vehicle (ELV) model shows that operational costs dominate, and profitability depends heavily on these costs and material selling prices.	Stochastic model improves ELV network planning under uncertainty. Accounts for variable costs and material composition.	ELV recycling faces high uncertainty and multi-actor complexity. Operational and dismantling costs are significant burdens.	Model tailored to Istanbul. Dependent on uncertain ELV return volumes.
	[48]	A nonlinear integer model determines the number and locations of recycling stations and plants to minimize total cost.	Strategic-alliance model with optimization for locating recycling stations and plants.	Sensitive to waste volume, capacity, and transport cost changes.	Identifies facility locations and analyzes waste volumes, capacities, and transport costs.
	[72]	Identifies Pareto-optimal solutions and offers a practical framework for safe, timely pandemic waste collection and transport.	Optimizes costs and risks associated with pandemic infectious waste using a bi-objective location-routing model.	Managing explosive waste growth and balancing cost–risk trade-offs increase system complexity.	Pandemic uncertainty, multi-party preferences, and objectives like carbon emissions are not included.
	[67]	A dual-objective green closed-loop supply chain model for steel minimizes cost and emissions, showing that quantity discounts cut economic costs by ~10% and total costs by ~30%, with minimal increase in emissions.	Dual-objective model integrates cost, carbon emissions, and quantity discounts.	Requires extensive data across multiple supply chain participants. Balancing short-term costs with long-term environmental goals is difficult.	Long computation time reduces model accuracy. Uncertainty in parameters such as demand and recovery rates is not fully accounted for.
1.2 Computational Models (4 articles)	[56]	A system dynamics model shows that reverse logistics for paper waste in Thailand’s construction industry becomes profitable after 3 years, with an IRR >12% by year 9, recommending recycling and remanufacturing to reduce landfill use.	Evaluates landfill, recycling, and remanufacturing options. Identifies economic and environmental benefits.	Profitability requires long-term implementation. Needs skilled labor and financial support.	Requires parameter changes for other contexts.
	[39]	The simulation shows average stock time drops to 56 h, a 21% reduction from the Current State Map, and processing time decreases from 140 to 110 h, making the system more robust to rising future waste flows.	Provides hybrid optimization for WEEE routing and facility decisions and uses a push–pull system dynamics model to improve waste flow and reduce waiting time.	Traditional algorithms risk getting stuck in local optima and producing impractical routes. Managing waste flows and facility choices is complex under varying logic.	Verified only through example-based simulation. Results depend on assumed facility types, thresholds, and stock behaviors.
	[70]	The study shows that closed-loop strategies increase farmer profit by about 7% and reduce CO2 emissions by 29% for citrus and 26% for juice, while also creating added value through composting and upcycling.	Improves profitability and reduces CO2 emissions through recycling and upcycling in a closed-loop citrus supply chain.	Depends on accurate data processing and must balance economic and environmental outcomes.	Uses expert-estimated essential-oil data, a cold-press process, and a single Jordan case.
	[52]	Demonstrates Genetic Algorithms (GA)’s feasibility for designing computer-waste reverse logistics and supports future real-world applications.	GA efficiently finds low-cost locations for collection and recycling facilities.	The reverse logistics system is complex and lacks comparison with other methods.	Does not benchmark GA against alternative methods for solving location problems.
1.3 Conceptual Models (2 articles)	[73]	Presents IDEF0 as a valuable tool for modeling and improving reverse logistics, helping reduce errors, cut waste, and enhance process alignment.	Highlights the benefits of reverse logistics and clarity using IDEF0 process mapping.	Reverse logistics are complex due to many stakeholders and inefficient without precise process mapping and standardized procedures.	It depends on accurate data and well-developed sub-processes and Standard Operating Procedure (SOP).
	[55]	Presents a new reverse logistics system that boosts computer waste return, increases recovery of valuable materials, and reduces environmental impact.	Provides a structured reverse logistics network for end-of-life computers, from return points to recycling and material recovery.	Current Thai e-waste practices are inefficient and lack clear stakeholder responsibility.	No specific legislation or dedicated recycling facilities for end-of-life computers.
1.4 Hybrid Models (8 articles)	[61]	The study integrates queuing theory, a reverse logistics network, and simulation to improve the management of inert construction waste and to assess multiple delay scenarios.	Scenario-based analysis of queuing delays. More efficient reverse logistics network design.	Collection and transport require careful planning.	Impact assessment is limited to existing parameters. Focused only on inert construction waste.
	[47]	A bi-objective location-routing model for explosive waste using TOPSIS optimizes collection, inventory, and routes, reducing cost by 34% and risk by 57% in Southwest China.	Integrated model minimizing cost and risk. Combines location, inventory, and routing decisions.	Hazardous waste systems face uncertainty and complex operations.	Does not model time-varying waste generation. Lacks simultaneous delivery and pick-up.
	[57]	Cooperation in two-echelon reverse logistics, supported by the Minimum Cost-Remaining Savings (MCRS) model, improves cost efficiency, reduces emissions, and strengthens alliances.	Improves routing and cost performance. Reduces distance, emissions, and system cost.	Multi-echelon routing and alliance management are complex.	Does not model stochastic returns. No real-time traffic or vehicle-sharing stability analysis.
	[66]	Optimized site selection cuts medical waste handling time by 58.4% and costs by 20.8%, improving reverse logistics efficiency during COVID-19 and validating the hybrid method.	Integrates classification, site selection, and routing into a single RL system. Uses hybrid linear programming and K-means.	COVID-19 waste surges heighten system complexity and demand large-scale, real-time multisite management.	Detailed modules are not fully described due to space limits. Relies on available real-world data for verification.
	[54]	Regulatory WEEE reverse logistics supports long-term sustainability and profitability, with ANN improving forecast accuracy and target setting.	Compares regulatory vs. non-regulatory WEEE RL models using MILP and ANN.	High initial investment and low early return volumes hinder feasibility.	Based on one case and fixed return and capacity assumptions.
	[58]	The improved ant colony algorithm provides flexible, practical recycling routes that cut costs and boost logistics efficiency using existing facilities.	An improved ant colony algorithm provides flexible, practical routing solutions.	Traditional methods may find impractical optima and converge slowly.	Cost-focused model verified only by a single example.
	[71]	A green reverse logistics model using electric vehicles is proposed for sustainable construction and demolition (C&D) waste management and validated in the Tehran hospital case study.	Integrates sustainability, electric vehicles, fuzzy decision-making, and robust optimization for C&D waste RL.	EV use has operational limits, and RL networks are complex under uncertainty.	The NP-Hard model is complex for large-scale solving. EV battery disposal and routing are not fully included.
	[30]	MILP with SAL-PSO yields 2.21–7.31% improvement (average 3.25%) and provides a practical framework for lowering transport and driver costs in 3PL reverse logistics.	Improves routing efficiency with an eco-cost model and enhanced SAL-PSO performance.	The problem is NP-hard, making exact optimization slow. SAL-PSO needs more computation time than standard PSO.	Assumes constant vehicle speeds, reducing real-world accuracy.
2. Technology-driven Approach (1 article)
	[60]	Cryogenic freezing safely prevents thermal runaway in damaged lithium-ion batteries, enabling safer transport and reuse with minimal impact on performance, supporting sustainability.	Cryogenic cooling removes thermal runaway risk. Enables reuse and remanufacture with minimal performance loss.	Regulations still require costly explosion-proof containers. Cryogenic handling adds complexity.	Tested only under specific abuse conditions and two chemistries. Long-term effects beyond five freeze–thaw cycles unknown.
3. Exploratory Approach (5 articles)
	[65]	This study identifies woody biomass supply chain processes, highlighting opportunities for efficient logistics and processing, as well as challenges such as seasonality and handling requirements.	Identifies challenges and opportunities. Supports future empirical research.	Seasonal supply issues and varied handling and storage needs. High overall costs.	Based on secondary data/focus group interviews, it requires further empirical modeling.
	[50]	Reverse logistics works well in large areas but is inefficient in small communities, suggesting that recycling is not always optimal and that tailored, cost-effective strategies are needed.	Reveals why compulsory pesticide-packaging reverse logistics fails in small communities.	Recycling is costly and impractical in small areas, with strict hazardous-waste rules adding complexity.	Based on a single exploration case, the study limits its broad applicability.
	[33]	This study shows green procurement improves efficiency, reduces waste, cuts costs, and supports sustainability in motor dealerships.	Improves lead time, inventory turnover, and capacity use through reverse logistics.	Requires reliable systems, technology upgrades, and strong coordination.	Findings apply only to Nairobi dealerships; broader organizational factors were not assessed.
	[68]	Proposes EU-aligned ELV regulations and a comprehensive RL network for Thailand, showing resource recovery and environmental benefits.	Applies sustainability and regulatory insights to improve ELV reverse logistics.	Requires strong coordination and supportive regulations.	Findings depend on foreign models and do not align with Thai regulatory requirements.
	[53]	Using reverse vending machines in Thailand increased PET bottle collection by 21% and improved their quality with fewer contaminants.	Reverse Vending Machines (RVMs) increase PET collection and improve waste quality.	Machine reliability issues and difficulty reading barcodes; cannot remove labels.	Applies only to PET bottles. Results may differ with other users or waste types.
4. Hybrid Approach (5 articles)
1 + 2 Model + Technology (2 articles)	[32]	The smart e-waste reverse system reduces CO2 emissions by optimizing the deployment of collection vehicles. Develop a business model for the system using Business Model Canvas (BMC).	CO2 reduction via optimized routes. Digital/IT supports a circular business model.	Countries’ legislative differences. Need for special driver apps.	Collection-rate impact unknown without pilot testing. Needs more pilot tests and cost–benefit analysis.
	[69]	This study developed an AI-drone-based circular supply system that reduces waste and emissions while optimizing medical deliveries and profits in rural healthcare.	Integrates AI, drones, and green technologies. Reduces emissions and waste and improves delivery speed and efficiency.	Emission-reduction effects need further study. Dependent on government support.	Requires advanced AI, cloud, and data systems. Zero-carbon chain and IoT/blockchain have not been achieved.
1 + 3 Model + Exploratory (1 articles)	[64]	RL-based solar-panel waste management recovers high-value materials and reduces impacts, offering policy and operational guidance for a sustainable system in Thailand.	Analyzes the whole solar panel life cycle using reverse logistics and shows value recovery and sustainability benefits.	Thailand lacks laws, clear roles, and modern recycling technology, resulting in improper waste disposal.	No dedicated recycling plants and weak policy coordination limit effective management.

summarizes the research findings, strategic approaches, achievements, advantages, challenges, and limitations of the research.

1.

Model-driven Approach (23 articles): The model-driven approach forms the foundation of academic inquiry into reverse logistics. These studies primarily focus on developing models to analyze and solve problems related to reverse logistics.

1.1

Mathematical Models/Analytical Models (9 articles): Use mathematical equations or quantitative analysis methods to find the most suitable method, such as locating warehouses or arranging transportation routes. In the model-driven approach, mathematical models/analytical nodels in reverse logistics optimize waste collection, recycling, and transportation by minimizing costs [46,48,59,67], reducing environmental impacts [51,62,67] (e.g., a dual-objective green closed-loop supply chain model for the steel industry reduces economic costs by ~10% and overall costs by ~30% with only ~1% higher emissions [67]), supporting real-case applications [31], improving planning and cost management [63], and balancing cost and risk for safe and efficient disposal [72].

1.2

Computational Models (4 articles): Use algorithms or applied computational techniques. Computational models in reverse logistics support decision-making and efficiency by evaluating the long-term feasibility and profitability of paper waste recycling worth the investment by year 9, with an IRR > 12% [56], simulating WEEE flows to enhance recovery and recycling [39], assessing workforce impacts on profit and product loss reduction [70], and optimizing the locations of collection centers and recycling plants for computer scraps [52].

1.3

Conceptual Models (2 articles): Develop a concept or conceptual framework to describe or manage a reverse logistics system. Conceptual model-driven approaches in reverse logistics emphasize environmental and operational benefits by using tools like IDEF0 flow charts to streamline processes and recover resources [73], and by designing strategic systems for end-of-life computer scrap collection and recycling based on the 3Rs, EPR, and regulations to reduce environmental and human impact [55].

1.4

Hybrid Models (8 articles): These combine multiple modeling approaches. Hybrid models in reverse logistics enhance circularity, cost efficiency, and sustainability by optimizing material recovery from construction [61,71] and medical waste [66]; bi-objective location-routing for explosive waste management achieved a 34% cost reduction and 57% risk reduction [47], two-echelon cooperation and profit-sharing in urban recycling networks [57], regulatory-based WEEE management and forecasting with ANN [54], advanced routing using ant colony algorithms [58], and MILP with SAL-PSO [30] for vehicle logistics.

2.

Technology-driven Approach (1 articles): This approach focuses on the development and application of technologies. Cryogenic freezing safely prevents thermal runaway in damaged lithium-ion batteries, enabling safer transport and reuse with minimal impact on performance, supporting sustainability [60].

3.

Exploratory Approach (5 articles): This approach explores the problems and management approaches of reverse logistics in specific contexts. The exploratory approach in reverse logistics highlights efficient supply chains, cost reduction, and sustainability through optimized woody biomass processing [65], tailored strategies for community-scale reverse logistics in Brazil [50], green procurement in motor dealerships [33], effective ELV management with legal and technological support [68], and improved PET bottle collection and quality via reverse vending machines in Thailand [53].

4.

Hybrid Approach (3 articles):

Model-driven + Technology-driven Approach (2 articles): The hybrid model-driven and technology-driven approach underpins the development of a sustainable Business Model Canvas (BMC). One study presents a smart e-waste reverse system that lowers CO₂ emissions through optimal vehicle deployment [32], while another employs AI-driven drones to reduce waste and emissions, enhance medical deliveries, and improve profitability in rural healthcare [69].

Model-driven + Exploratory Approach (1 article): The hybrid model-driven and exploratory approach finds that Thailand still lacks laws and dedicated solar panel recycling facilities, while the volume is expected to rise in the future. Sustainable management should therefore be guided by product life cycle assessment (LCA) and reverse logistics [64].

3.2.2. Advantages, Challenges, and Limitations of Strategic Approaches

After the classified codes revealed four strategic approaches, a comparative and thematic synthesis of the full-text publications was conducted. Through this process, the advantages, challenges, and limitations of each strategic approach were identified.

Model-driven approaches use mathematical models to reduce environmental impacts, confirm recycling as the optimal end-of-life option, optimize multi-echelon networks, and manage uncertainties in waste quantity, quality, and transport costs. Their challenges include transport emissions, improper rinsing, variable waste characteristics, complex routing, and multi-actor coordination. They are limited by estimated transport data, context-specific inputs, deterministic assumptions, fixed nodes, and high computation.

Computational models evaluate recycling and remanufacturing options, improve waste-flow efficiency, and identify cost-effective facility locations. Difficulties arise from long-term implementation needs, financial constraints, the risks of local optima, and the complexity of managing waste flows. Their limitations include context-dependent tuning, example-based simulations, single-case evaluations, and limited benchmarking.

Conceptual models add value by clarifying reverse logistics processes through IDEF0 mapping and structured network design. Challenges arise from multi-stakeholder complexity, unclear responsibilities, and the need for precise and standardized process frameworks. They are limited by their need for accurate data and the lack of supporting legislation and dedicated facilities.

Hybrid models integrate scenario analysis, network design, cost and risk decisions, improved routing, and reduced emissions and system costs. Their challenges include difficult collection and transport planning, COVID conditions, and high investment demands. They are constrained by their narrow parameters, focus on specific waste, missing time-varying or stochastic returns, limited real-time integration, and case-specific data.

The technology-driven approach shows strong promise. Cryogenic cooling improves safety during transport and allows reuse and remanufacture with minimal performance loss. Its challenges include strict requirements for explosion-proof containers and the overall complexity of cryogenic handling, which increases both costs and operational difficulties. It is limited to testing to only two battery chemistries, with the long-term effects remaining unclear.

The exploratory approach identifies system challenges, opportunities, and future research needs. It reveals failures in small-scale hazardous-waste systems, the need to improve lead times and inventory turnover, and the importance of sustainability. Its key challenges include seasonal variability, diverse handling needs, high costs, strict regulations, and the need for reliable, well-coordinated systems. It is limited by its reliance on secondary data and single-case analyses, rendering the findings highly context-specific.

The hybrid approach with the technology model improves efficiency and reduces CO₂ by optimizing routing, leveraging digital support, and enabling AI and drone-enabled operations. Its challenges include regulatory differences, the need for specialized apps, uncertain collection performance, and reliance on government support. Its limitations include the need for additional pilot tests, advanced data, and integration with zero-carbon or IoT/blockchain technology. The hybrid approach with the exploratory model shows the recovery of high-value materials and reduces impacts, providing policy and operational guidance. Its challenges include the absence of specific laws and recycling technology, and it is limited by the lack of dedicated recycling plants and weak policy coordination.

4. Discussion

Diverse reverse logistics solutions are increasingly used to enhance efficiency and reduce environmental impact [33], but their integration within RL networks remains insufficiently studied [36,37] and lacks classification and synthesis of transportation specifically for recyclable waste [38]. This systematic review of three databases reveals that the overall number of publications on reverse logistics for recyclable waste and transportation remains relatively limited. However, ongoing research demonstrates notable progress and a steady increase in scholarly output within this field. The peak of six articles in 2024 reflects growing global attention to environmental [76] and waste management challenges [77], consistent with the global movement toward Net Zero policies [78,79]. The upward trend in publications has continued to this. If the trajectory follows previous patterns, further growth can be expected in the coming years.

Most studies, primarily based in Asia and distributed across multiple journals, underscore the field’s interdisciplinary nature and its alignment with broader trends toward cross-disciplinary, integrative research [80]. This integration of diverse perspectives and applications enhances the field’s adaptability to complex real-world challenges.

The analysis identified four strategic approaches, with model-driven volume dominating the field, by improving cost, environmental, and performance evaluation. Exploratory and technology-driven approaches provide contextual insights. Hybrid approaches have bridged the gap by combining analytical with practical adaptability, though further integration and broader testing across diverse contexts are still needed.

Future research on model-driven approaches and mathematical models require improved data quality, flexibility, and efficiency. Computational models must be validated in diverse contexts with more rigorous method comparisons. Conceptual models need stakeholder coordination and more supportive regulatory frameworks. Hybrid models should integrate dynamic and stochastic factors, real-time operations, broader datasets, and scalable optimization methods to reflect the complex behaviors of systems. The technology-driven approach remains a significant opportunity to expand and adapt these approaches for broader applications. Exploratory approaches should involve multi-context and thorough evaluation across a broader range of waste types and operational settings. The future development of hybrid approaches still offers substantial room for broader application and deeper integration of model, technology, and exploratory or future unidentified approaches, representing an important research gap.

5. Conclusions

This review of 32 articles between 2010 and 2024 shows that contributions from multiple countries remain heavily concentrated in Asia. Publication activity has steadily increased, accelerating after 2015 and peaking in 2024, driven by SDG and Net Zero agendas and growing interdisciplinary interest. The field’s multidisciplinary character and the absence of a dominant journal are evident, as only Resources, Conservation and Recycling has published more than one article.

Reverse logistics has advanced through strategic approaches. Model-driven approaches provide the strongest analytical tools with mathematical, computational, conceptual, and hybrid models. They achieve 44% reduction in climate change impacts, 34% cost reduction, and a 57% reduction in explosive-waste routing, demonstrating superior capability to optimize large, complex systems. Technology-driven approaches contribute to innovation that models alone cannot produce. Cryogenic freezing for damaged lithium-ion batteries enables safe transport and reuse. Exploratory approaches add contextual depth, revealing real-world constraints in policy gaps, seasonal variability, and financial limitations. These studies uncover operational realities that neither models nor technology can fully capture. Hybrid approaches deliver comprehensive advantages by integrating strengths across domains. AI-enabled drones, smart e-waste routing, and LCA-guided solar panel management show how combining modeling, technology, and context leads to lower CO₂ emissions, higher efficiency, and stronger decision-making. Future work should integrate real-time data, IoT systems, and multi-context validation. However, integrating the combined use of model, technology, and exploratory approaches remains an essential unaddressed research gap.

The study is limited to three databases: Scopus, Google Scholar, and Thai Journals Online, based on the research team’s access capabilities. In addition, studies published after the timeframe of this review may introduce additional approaches. Some model-driven studies also provide insufficient methodological detail, limiting comparability, as some research emphasizes other aspects and does not explicitly describe the models used. This systematic review provides a key for improving reverse logistics management for recyclable waste and transportation. Policymakers and researchers can identify the remaining gaps and develop more integrated approaches. The review supports the development of effective systems, policies, and future research.