Approaches, Challenges, and Opportunities in Humanitarian Logistics Integrated with Reverse Logistics and Sustainability

Abstract

Background: Reverse Logistics (RL) can reduce waste, mitigate environmental impacts, and enhance operational efficiency in Humanitarian Operations (HOs). Nevertheless, it remains weakly institutionalised and insufficiently recognised as a structuring component of Humanitarian Logistics (HL), operating implicitly and reactively. Methods: A Systematic Literature Review (SLR) was conducted in accordance with the PRISMA protocol, examining disaster types, methodological approaches, the roles of RL, and its integration with sustainability and digital technologies. Results: Scientific output increased after 2020, accounting for 66% of the studies. Practices aligned with RL exceed 80% of the publications; however, only 55% explicitly use the term, indicating a disconnect between concept and practice. Sustainability is explicitly addressed in 56% of the studies. Modelling, optimisation, and simulation predominate (27%), followed by reviews (25%) and conceptual analyses (19%). Four roles of RL were identified: environmental (33%), operational (19%), institutional (12%), and technological (19%). Conclusion: RL operates within HOs as a latent and non-formalised capability. As a contribution, this SLR proposes an integrated typology of RL roles in humanitarian contexts. It is recommended to formalise sorting and return protocols, incorporate indicators for reverse flows, assess the use of artificial intelligence to support decision-making, and expand empirical studies that quantify its impacts.

1. Introduction

The intensification and growing complexity of disasters, driven by climate change, public health emergencies, armed conflicts, and population displacement, have substantially increased the demands placed upon Humanitarian Operations (HOs). Within this context, Humanitarian Logistics (HL) plays a central role by enabling the mobilisation and coordination of critical resources under conditions of profound uncertainty, severe time pressure, and compromised infrastructure [1,2,3]. Consolidated evidence indicates that logistical activities account for the largest share of humanitarian expenditure and directly influence the effectiveness of emergency responses and recovery processes [4].

Despite this recognition, the HL literature remains predominantly focused on forward flows, such as procurement, transportation, and the distribution of supplies. In contrast, reverse flows, including surplus or inadequate donations, the return of expired medicines, packaging waste, debris from damaged buildings, and healthcare waste, are recurrent in real operations yet continue to receive peripheral treatment in academic studies [5,6]. These flows simultaneously affect logistical efficiency, operational costs, public health, and environmental impacts, indicating that they are not marginal issues but structural ones.

Reverse Logistics (RL), widely developed within commercial supply chains, provides frameworks for the return, sorting, recovery, reuse, recycling, and environmentally appropriate disposal of materials [7,8]. In humanitarian settings, practices compatible with RL arise predominantly on an ad hoc basis, such as the redistribution of surplus donations, post-disaster waste management, and the removal of debris to reopen access routes [6]. Empirical evidence from different regions and disaster types demonstrates that these reverse flows are decisive for both the continuity of operations and the mitigation environmental impacts, although they are rarely formalised, planned, or assessed as part of an integrated logistics strategy.

The absence of consolidated RL structures in HOs generates significant operational and environmental consequences, including waste accumulation, resource inefficiencies, increased costs, and public health risks. At the same time, sustainability has become a central concern within humanitarian supply chains, driven by the increasing scale of disasters, resource scarcity, and the alignment of humanitarian action with the Sustainable Development Goals, particularly SDGs 11, 12, and 13 [9]. Nevertheless, sustainability is often operationalised in a fragmented manner, and its integration with RL remains insufficiently explored systematically.

Although academic interest in sustainability, HL, and disaster management has risen markedly, especially since 2020, the literature continues to treat these themes in a fragmented fashion [10]. Some studies concentrate on humanitarian distribution models [11] whilst others address post-disaster waste management or specific environmental indicators [12]. Meanwhile, the concept of RL appears implicitly or only marginally, even when reverse flows are central to the analysis [13]. Existing literature reviews have yielded important advances but generally (i) prioritise forward flows, (ii) address reverse activities in a sectoral or episodic manner, or (iii) do not provide a systematic classification of RL within HOs from a sustainability perspective.

This fragmentation reveals both conceptual and empirical gaps. Scientifically, it limits the cumulative development of knowledge on reverse flows in humanitarian contexts. In practice, it restricts organisations’ and policymakers’ ability to design more efficient, resilient, and environmentally responsible logistical systems. Given the substantial recent growth in publications and the increasing demand for sustainable Humanitarian Operations, a Systematic Literature Review (SLR) is both timely and necessary.

In this context, this study undertakes an SLR grounded in the PRISMA protocol to address the following research question: How is Reverse Logistics conceptualised, applied, and assessed within the literature on HOs from a sustainability perspective? Specifically, this review seeks to (i) identify the principal forms of conceptualisation of RL in HOs; (ii) classify its applications according to disaster types, methodological approaches, and geographical contexts; and (iii) examine how the dimensions of sustainability are incorporated into reverse flows. As its central contribution, this study proposes an integrated and evidence-based typology of RL roles in HOs, structured around four interrelated functions: environmental, operational, institutional, and technological. This typology consolidates fragmented practices reported in the literature into a coherent analytical framework, repositioning RL as a structuring capability for sustainable and resilient humanitarian systems rather than as an ad hoc or auxiliary activity.

2. Theoretical Framework

Humanitarian Operations (HOs) may be understood as complex adaptive systems characterised by high levels of uncertainty, multiple interdependent actors, frequently damaged infrastructure, and highly volatile demand patterns [13,14]. In light of complex systems theory and resilience theory [15], the performance of such systems does not depend solely on the speed or volume of aid delivered but rather on their capacity to adapt, reorganise, and restore functional performance in the face of successive shocks [16,17,18]. Within this context, HL constitutes the operational backbone of HOs, as it is responsible for coordinating material, informational, and financial flows under severe temporal, environmental, and institutional constraints [14,19]. The way in which HL is structured directly influences systemic resilience, continuity of operations, and the effectiveness of response, recovery, and reconstruction phases [14,20,21].

Traditional HL frameworks, however, remain anchored in a linear logic with an almost exclusive focus on forward flows such as procurement, transport, warehousing, and distribution [17]. Although indispensable for immediate emergency relief, this approach fails to adequately capture the actual systemic dynamics of HOs. In disaster settings, reverse flows inevitably emerge, including surplus or inappropriate donations, expired medical supplies, disposable equipment, packaging waste, debris from damaged infrastructure, and materials from temporary shelters [8,22,23]. From a resilience perspective, neglecting these flows compromises overall system performance, generating logistical bottlenecks, warehouse saturation, road blockages, public health risks, and significant loss of material value [24,25].

It is at this juncture that Reverse Logistics (RL) acquires theoretical relevance for HOs. RL may be conceptualised as an organising mechanism for return flows, aimed at reuse, redistribution, recycling, recovery, or environmentally sound disposal of materials [8]. Although well-established in commercial supply chains, its importance in humanitarian contexts can be theoretically explained through the lens of the circular economy (CE) [26,27]. CE emphasises value retention and regeneration in closed-loop systems and is particularly pertinent in environments characterised by resource scarcity, severe disruptions, and environmental pressures [28,29]. In disaster scenarios where supply constraints paradoxically coexist with material surpluses and waste accumulation, RL enables the reintegration of these resources into humanitarian supply chains, reducing waste, operational costs, and environmental impacts while simultaneously increasing the effective availability of assets [18,20,24].

Beyond its operational and environmental contributions, the incorporation of RL in HOs may be interpreted through the lens of sustainability transitions theory [30], which frames it as part of a broader structural shift from predominantly reactive, linear emergency responses to more resilient, adaptive, and sustainable humanitarian systems [18,20]. This transition unfolds across multiple levels. At the operational level, it involves the coordinated integration of forward and reverse flows throughout the entire disaster cycle [22,24]. At the organisational level, it demands new coordination arrangements among humanitarian organisations, governments, donors, and logistical partners [19,31]. At the institutional level, it requires regulatory frameworks, performance metrics, and incentives that legitimise decisions oriented not only towards speed but also towards sustainability and long-term recovery [21,23]. In this regard, RL acts as a strategic enabler of sustainability, bridging immediate response objectives with long-term goals of resilience, reconstruction, and development [32,33,34].

Institutional theory [35] helps explain why, despite its theoretical and practical relevance, RL remains weakly institutionalised within HOs. Humanitarian organisations operate under institutional logics strongly oriented towards urgency, visibility of actions, and donor accountability, privileging short-term decisions centred on forward flows [36]. Activities associated with reverse flows, waste management, and post-consumption processes tend to be perceived as peripheral, costly, or outside organisational mandates [22,23]. As a consequence, fragmented governance, low standardisation, and weak incentives for systematic adoption of RL are commonly observed [13,21]. Theoretically, however, integrating RL strengthens transparency, organisational legitimacy, and alignment with regulatory frameworks and international sustainability agendas, such as the SDGs, thereby enhancing the overall effectiveness of HOs [20].

Humanitarian supply chains may also be conceptualised as sociotechnical systems [37] in which technologies, human decisions, organisational structures, and social norms interact dynamically [18,20]. From this perspective, the effectiveness of RL depends not only on isolated technical solutions but also on the articulation of technological capabilities, human competencies, and institutional arrangements [21,31]. Digital technologies, such as tracking systems, decision-support models, the Internet of Things (IoT), blockchain, and artificial intelligence, enhance the visibility and coordination of reverse flows, supporting decision-making under high uncertainty [25,38,39].

Within this analytical framework, reinforcement learning emerges as an exploratory and emerging decision-support approach, rather than a consolidated operational solution in HL [40]. Existing studies conceptualise or simulate reinforcement learning mechanisms to explore their potential under conditions of uncertainty, data scarcity, and dynamic disruptions that characterise humanitarian contexts. Its practical application remains limited and contingent upon institutional maturity, data availability, and technological infrastructure. In humanitarian contexts, marked by incomplete information, rapid changes, and multiple potentially conflicting objectives, this computational learning approach can support decisions related to routing of forward and reverse flows, prioritisation of returns, dynamic allocation of scarce resources, and post-disaster waste management [41,42].

When integrated with human and organisational learning processes, reinforcement learning operates as a cognitive support mechanism, incorporating operational experience, tacit knowledge, and feedback from past operations [39]. This combination of computational and human learning strengthens the adaptive capacity of humanitarian organisations, fostering continuous organisational learning and improved alignment between operational efficiency, systemic resilience, and sustainability objectives. From this perspective, reinforcement learning-based approaches do not replace human decision-making capacity but rather enhance it, contributing to the simultaneous improvement of HL performance, management of reverse flows, and sustainable outcomes in HOs.

The coding of studies is structured around theoretically informed constructs, including resilience and adaptability of humanitarian systems; circular practices (reuse, redistribution, recycling, and recovery); institutional facilitators and barriers; technological and learning mechanisms; and sustainability outcomes [18,20,21]. Thus, the systematic review moves beyond a merely descriptive exercise and becomes an analytical instrument capable of explaining why, how, and under what conditions Reverse Logistics contributes to the effectiveness, resilience, and sustainability of Humanitarian Operations.

3. Materials and Methods

The present study was conducted through a Systematic Literature Review (SLR), a methodological approach that requires prior planning and rigorous criteria in order to address the proposed research question in a structured manner. This review followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [43,44], encompassing the stages of identification, screening, and inclusion. The Prisma Protocol 2020 is available in the Supplementary Material.

The review protocol (Appendix A) defined the research question, specific objectives, inclusion and exclusion criteria, search strategies, screening procedures, coding processes, and methods for results synthesis. The entire process was operationalised using the StArt (version 3.0.3 Beta) software (State of the Art Review System), which enabled the structured and traceable documentation of every decision taken throughout the review [45,46].

The bibliographic search was conducted in the Web of Science and Scopus databases. The selection of these databases was guided by a deliberate balance between coverage and analytical coherence. These databases offer broad interdisciplinary indexing, high-quality peer-reviewed content, and strong representation of journals in the fields of Humanitarian Logistics, sustainability, and operations management. Although other repositories, such as IEEE, PubMed, or ScienceDirect, could capture contributions more specific to technology for health-related domains, their inclusion could increase thematic dispersion and methodological heterogeneity, thereby reducing the comparability of results within the scope of this review. The restriction to two databases was mitigated through complementary search strategies, including backward and forward snowballing, as well as an exploratory search in Google Scholar aimed solely at the initial identification of studies, rather than final selection.

The search strategies were constructed using Boolean operators (OR and AND), quotation marks, truncation, and logical groupings in order to capture the full range of terminological variation within the fields of Reverse Logistics (RL), Humanitarian Operations (HOs), and sustainability. The final combination of search strings, identical across both databases, was as follows: (“reverse logistic*” OR “returns management” OR “closed-loop supply chain*” OR “reverse flow*” OR “end-of-life product*” OR “waste management” OR “product recovery”) AND (“humanitarian” OR “disaster relief” OR “emergency response” OR “crisis management” OR “relief supply chain*” OR “aid distribution” OR “humanitarian operation*”) AND (“sustainability” OR “sustainable development” OR “resilience” OR “circular economy” OR “green logistic OR “environmental management”). Quotation marks were used to ensure the retrieval of exact expressions, truncation allowed the identification of morphological variations, and parentheses ensured logical coherence in the combination of operators.

The application of the search strategies resulted in 136 records (64 from Web of Science and 72 from Scopus). After importing the records in BibTeX format into StArt, 27 duplicates were automatically removed, leaving 109 studies for the screening phase. During the independent screening of titles and abstracts, 12 studies were excluded due to a lack of thematic relevance. Subsequently, full-text retrieval was attempted, and 27 studies were excluded due to the unavailability of full-text access (Figure 1).

The remaining 70 studies were assessed in the eligibility phase through full-text reading. Of these, 34 studies were excluded for failing to meet the inclusion criteria. In addition, among the 18 studies read in full that were identified through complementary methods, 7 were excluded specifically because they did not demonstrate integration between HL and RL (Figure 1).

In the final inclusion phase, 47 studies were considered in the qualitative synthesis. Of these, 36 studies were sourced directly from the databases, while 11 were identified through other methods. Among these additional 11 studies, 10 were obtained from websites, including 4 from institutional organisations (such as international organisations and humanitarian agencies), and 4 were identified through citation tracking (backward and forward snowballing) (Figure 1).

The inclusion criteria were as follows: (i) explicitly addressing RL applied to HOs; (ii) addressing practices conceptually aligned with RL, even without explicitly using the term, such as product recovery, supply returns, recycling or reuse, post-disaster waste management, and surplus redistribution; (iii) falling within the thematic scope of HL, sustainability, and disaster waste; and (iv) being a peer-reviewed scientific article.

Studies excluded during the eligibility phase were classified according to the predominant exclusion criterion, ensuring that each article was assigned to only one category: (i) does not address HOs: includes studies that discuss logistics, sustainability, reverse flows, or sectoral applications without any connection to humanitarian contexts or disaster response (25 articles); (ii) does not address RL, explicitly or implicitly: includes studies focused on Humanitarian Operations or Humanitarian Logistics that do not consider return flows, recovery, reuse, or waste management (5 articles); (iii) does not consider sustainability: No studies were excluded exclusively on this basis. In all cases analysed, the absence of sustainability was always associated with the absence of HOs or RL and therefore did not constitute an independent exclusion criterion (0 articles); (iv) does not present integration between HL and RL: includes studies addressing related themes, such as waste or donations, but without conceptual or operational articulation between HL and RL (4 articles) (Figure 1).

Two authors independently conducted the inclusion and exclusion of studies using StArt, and disagreements were resolved by consensus. Screening consistency was assessed using Cohen’s kappa coefficient, which exceeded 0.80, indicating excellent inter-rater agreement.

To extract and organise the data, StArt provided standardised forms to support the compilation and synthesis of the articles, as presented in Appendix B, Appendix C and Appendix D. Appendix D contains the table of the analysed journals, the number of citations, and the methodological approaches adopted in the studies. Appendix C systematises the main challenges and opportunities associated with Reverse Logistics in Humanitarian Operations, while Appendix D presents the keyword clusters identified through the co-occurrence analysis.

The thematic analysis was conducted using a structured coding protocol combining open and axial coding. Initially, two reviewers independently analysed the full texts of the included studies, identifying excerpts related to concepts, practices, mechanisms, and challenges associated with RL in HOs from a sustainability perspective. These excerpts were assigned initial descriptive codes. Subsequently, the codes were consolidated and grouped into broader analytical categories aligned with the objectives of the review. This process resulted in the identification of core themes that guided the synthesis of the results, including the environmental and operational roles of RL, institutional mechanisms, enabling technologies, and structural challenges. Disagreements were resolved by consensus between the reviewers, with the support of the StArt software, ensuring traceability, analytical consistency, and process reliability.

The methodological quality of this review was assessed using the AMSTAR-2 instrument (A Measurement Tool to Assess Systematic Reviews), which is used for the critical appraisal of systematic reviews. In this study, AMSTAR-2 was applied as a transparency and methodological control tool, rather than as a criterion for study exclusion or validity judgment. According to the instrument’s criteria, the overall rating indicated critically low confidence. This classification primarily results from the absence of a formal risk-of-bias assessment of individual studies, the lack of data extraction by two independent reviewers, and the non-performance of a meta-analysis—items considered critical within the original scope of AMSTAR-2, which was developed mainly for systematic reviews of health interventions with clinical and experimental designs.

It should be noted, however, that these criteria are not fully applicable to systematic reviews of a conceptual, exploratory, and interdisciplinary nature, such as the present study, which encompasses qualitative, modelling, and theoretical research, involves high methodological heterogeneity, and does not rely on comparable effect measures. Accordingly, the “critically low” rating reflects structural limitations of the AMSTAR-2 instrument when applied outside the clinical context, rather than weaknesses related to the transparency, traceability, or rigour of the review process. Despite these limitations, this review satisfied several relevant non-critical domains of AMSTAR-2, including a clearly defined research question and inclusion criteria, the use of a predefined protocol, justified selection of study designs, duplicate study selection with high inter-reviewer agreement, and a systematic analytical synthesis of heterogeneous evidence. Therefore, AMSTAR-2 was used as a framework for methodological control and transparency rather than as a criterion for invalidating the findings, which remain consistent with and appropriate to the analytical and theoretical objectives of the review.

4. Results

The years 2021 and 2023 accounted for the highest volumes of publications, each representing 26%, constituting a peak associated with broader global contextual shifts. Considering the period between 2020 and 2025, it is observed that 25 of the 47 studies fall within this interval, which confirms a significant inflection in the trajectory of the literature. This growth may be associated with the COVID-19 pandemic, which exposed structural vulnerabilities in humanitarian supply chains [48] and heightened concerns regarding medical waste management, the return and redistribution of supplies, the reuse of personal protective equipment (PPE), and the need for more sustainable logistical solutions. As a result, practices aligned with RL began to be explored more intensively, often in an emergent manner, thereby stimulating both scientific production and the debate surrounding its strategic relevance in crisis contexts [49]. Figure 2 shows these results.

Although the peak in publications may be associated with the global impacts of the COVID-19 pandemic, the findings of this review indicate that the relationship between RL, HOs, and sustainability is not confined to health crises; the studies analysed encompass a broader and more diverse set of disaster contexts, enabling a cross-cutting examination of RL practices across different humanitarian settings. Accordingly, each article was classified according to its primary disaster context, as summarised in

Table 1. Distribution of studies by disaster type and focal areas of Reverse Logistics and sustainability.

Disaster Type	Percentage of Studies (%)	Focal Areas of Reverse Logistics and Sustainability
Pandemics and public-health emergencies	17	Management of medical waste (hazardous and recyclable), return/donation of pharmaceuticals, redistribution of surplus supplies (food and essential items), and equity in access to resources, with emphasis on reducing environmental impacts and prioritising vulnerable groups.
Natural disasters	27	Management and recycling of disaster waste (debris, municipal waste, and radioactive materials), repurposing of materials (plastics, packaging, and organic waste), and integration of environmental criteria into logistical decisions to enhance urban resilience and reduce sanitary and environmental risks.
Complex emergencies	21	Reuse of packaging and other waste in refugee camps, management of hazardous/contaminated waste, and circular solutions such as biogas production from human waste, linking public-health protection, clean energy and long-term environmental recovery.
Multi-hazard/unspecified	36	Reverse Logistics addressed transversally: control of surplus items (medicines, food, and textiles), reuse and recycling of materials, and structural support for disaster-waste management, often connected to green logistics, social responsibility, and circular economy frameworks within humanitarian supply chains.

, allowing for a differentiated interpretation of the results.

A significant proportion of the reviewed literature focuses on pandemics and public-health emergencies [49,50,51,52,53,54]. In contrast, studies addressing natural disasters [6,20,55,56,57,58,59,60,61,62,63,64,65], including earthquakes, floods, hurricanes, tsunamis and hydrometeorological events, emphasise post-disaster waste management, debris removal, the recovery of construction and demolition materials, and the reorganisation of transport and collection routes during the response and recovery phases.

Research dealing with complex emergencies [66,67,68,69,70,71,72,73,74,75], such as armed conflicts, protracted displacement, terrorism, and radiological or nuclear incidents, is less prevalent and tends to report decentralised, informal or locally adapted RL practices, including textile reuse in refugee camps, small-scale material recovery initiatives and the management of hazardous waste under severe operational constraints.

A substantial share of the publications adopts a multi-hazard or unspecified scope [22,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90], comprising conceptual articles, optimisation models, and systematic reviews that do not focus on a single type of disaster but instead propose analytical frameworks intended for application across heterogeneous humanitarian contexts.

The studies analysed reveal a strong geographical concentration of research conducted in countries of the Global South and in regions frequently exposed to natural disasters, health crises, and complex emergencies, alongside a significant body of research developed in Europe and North America, particularly with a methodological and normative orientation.

In Brazil, empirical investigations focused on the state of São Paulo during the COVID-19 pandemic [49], examined natural disasters such as floods and landslides [6], and investigated sustainability within humanitarian medical supply chains [53]. Collectively, these studies reflect concerns related to equity, waste management, and the integration of humanitarian and Reverse Logistics in real-world contexts. In Latin America, in addition to Brazil, research conducted in Mexico strongly emphasises multi-objective modelling, the use of the Internet of Medical Things (IoMT), and the integration of direct and reverse flows during pandemics, while studies carried out in Peru focus on post-COVID-19 economic reactivation through spatial modelling approaches [48,50].

North America predominantly concentrates on normative and methodological studies, especially in the United States, addressing topics such as post-radiological recovery, terrorist attacks, governance, transport contracts, and sustainability in Humanitarian Operations [65,67,69,70,91].

Italy stands out as one of the principal hubs of applied research, largely associated with the University of Bologna, ranging from experimental prototyping to applications in humanitarian camps, with a focus on packaging reuse, sanitation, biogas production, and life cycle assessment, often developed in collaboration with international organisations such as the United Nations and UNHRD [56,66,71,72,92].

In France, it was possible to observe studies that specifically addressed relief situations that were carried out only in the country [63,93] and joint investigations involving France, the United Kingdom, and Mexico [64,78] addressed issues of agility, resilience, and governance in humanitarian supply chains, explicitly incorporating post-disaster waste management as a structuring element of urban recovery.

Spain appears in simulations related to rapid prototyping in HOs [22], while Portugal was observed in studies developed with modelling approaches to control excess medicines in humanitarian contexts [81].

In the United Kingdom, contributions predominantly take the form of systematic reviews and multi-criteria analyses, frequently developed through international collaboration [64] and addressing coastal flooding and multi-risk contexts [78].

Across Asia, a wide geographical and thematic diversity is observed, with applied studies conducted in the Philippines (Cavite) focused on locating temporary post-disaster waste storage areas [61]. In China, investigations analyse Reverse Logistics systems in Xiamen, as well as global-scale studies with particular emphasis on the Chinese context [82,86]. South Korea provided the empirical setting for research cantered on floods in Seoul and the use of geographic information systems technologies, artificial intelligence, and the Internet of Things [55,80]. Japan appears in simulation-based studies and analyses related to Typhoon Hagibis [88].

India stands out as one of the main focuses of systematic reviews and modelling studies, addressing health supply chains, sustainability, governmental policies, and ecological stockpiling during pandemics [68,79,83]. In the Middle East, research conducted in the United Arab Emirates, focusing on the redistribution of food surpluses, and in Iran, addressing the acquisition of humanitarian aid in post-conflict contexts, is noteworthy [54,75].

Appendix B highlights the methodological heterogeneity and the dispersion of publication outlets that characterise the literature at the intersection of RL, HOs, and sustainability. The corpus includes studies published both in well-established, high-impact journals in operations and logistics management and in more specialised or regionally oriented journals. This diversity reflects the interdisciplinary nature of the field, situated at the convergence of logistics, engineering, environmental management, sustainability studies, and humanitarian research.

From a methodological standpoint, a predominance of quantitative and analytical approaches is observed in the literature, with particular emphasis on mathematical modelling, optimisation, and simulation studies, which together account for 27% of the total publications analysed (

Table 2. Methodological approaches and Reverse Logistics framings identified in the literature.

Methodological Approach	Percentage of Studies	RL Approach
Mathematical modelling, optimisation, and simulation	27	RL is conceptualised as reverse flows (waste, returns, surpluses, and recycling) integrated into the humanitarian supply chain.
Systematic reviews/literature reviews	25	RL is predominantly associated with waste management and sustainability, often implicitly and without explicit reference to the term “RL”.
Analytical studies	19	RL is addressed as a normative and territorial mechanism for managing post-disaster waste (disposal, decontamination, and storage), supporting recovery, and resilience processes.
Experimental studies and prototyping	15	RL is framed as a circular innovation strategy, whereby waste materials (plastics, packaging, and human waste) are transformed into resources and energy for humanitarian use.
Case studies and applied studies	8	RL is examined as a mechanism for post-crisis recovery and logistical reorganisation through waste management, stock replenishment, and surplus redistribution.
Surveys	6	RL is investigated as a concrete practice of material reutilisation (food, textiles, and medical supply chains), with emphasis on operational barriers and on the distribution of costs and benefits.

). In these studies, RL is predominantly operationalised as reverse flows, involving waste, returned supplies, surplus donations, and recycling, integrated into the design and management of humanitarian supply chains. This orientation reflects a strong emphasis on decision optimisation and the structuring of efficient logistical systems in crisis contexts. Such studies generally exhibit the highest citation counts, as exemplified by articles published in high-impact journals such as the International Journal of Physical Distribution and Logistics Management (178 citations) [93] and Renewable and Sustainable Energy Reviews (57 citations) [66], suggesting greater visibility and academic influence of these approaches within the field.

In parallel, systematic and bibliographic reviews account for 25% of the studies, associating RL predominantly with waste management and sustainability, often implicitly and without the explicit use of the term “reverse logistics”. Conceptual and normative analyses (19%) treat RL as an instrument of territorial and environmental governance, directed towards disposal, decontamination, post-disaster waste storage, and support for recovery and resilience processes. More applied and empirical approaches appear less frequently: Experimental and prototyping studies (15%) explore RL as a form of circular innovation, transforming waste, such as plastics, packaging materials, and human residues, into resources and energy for humanitarian use; case studies and applied studies (8%) focus on RL as a mechanism for logistical reorganisation and post-crisis recovery, whereas surveys (6%) examine concrete reuse practices in food, textile, and medical supply chains, with emphasis on operational barriers and on the distribution of costs and benefits. Although empirically relevant and contextualised in specific events, such as floods, earthquakes, and pandemics, these studies tend to show lower bibliometric impact, possibly due to their dependence on local contexts and their reduced theoretical generalisability (Table 2).

Overall, Appendix B and Table 1 reveal a research field that is methodologically sophisticated yet fragmented, in which modelling and simulation studies attract greater academic recognition, whereas empirical research remains less cited despite its practical relevance. This imbalance suggests an important gap between theoretical–analytical advancement and the empirical validation of RL practices in humanitarian contexts, reinforcing the need for more applied investigations that engage directly with the operational complexity of HOs and contribute to the consolidation of RL as a structuring component of these systems, rather than merely an ad hoc practice.

The keyword co-occurrence analysis (Appendix D) revealed three conceptual clusters that structure the literature on Humanitarian Logistics (HL), sustainability, and reverse flows. These groupings were formed based on the frequency and association of recurrent terms across the 47 analysed articles, demonstrating patterns of thematic proximity and revealing how the field is organised around distinct analytical axes. Overall, a predominance of terms related to Humanitarian Logistics and response operations was observed, alongside a growing body of studies addressing sustainability and waste management. Reverse Logistics emerged as a set of transversal practices, albeit still weakly formalised at the conceptual level.

The first cluster, centred on the terms sustainability, environmental management, and life cycle assessment, comprises articles that investigate environmental impacts, mitigation strategies, and circularity initiatives in disaster contexts. This is a cohesive cluster, characterised by an emphasis on practices that contribute to post-disaster waste reduction, impact assessment, and the integration of circular economy principles into Humanitarian Operations. The presence of this cluster confirms that sustainability has become an emerging and relevant axis within the contemporary literature on Humanitarian Operations.

The second cluster concentrates the most frequent terms in the sample, such as Humanitarian Logistics, transportation, disaster, and humanitarian supply chain. This grouping reflects the historical foundation of the field and its orientation towards operational efficiency in crisis scenarios. Studies within this cluster address transportation, resource allocation, facility location, logistical performance, and rapid response to disruptive events. The strong co-occurrence among these terms evidences the predominance of approaches that treat forward logistics and supply flows as the central components of Humanitarian Operations.

The third cluster brings together terms such as Reverse Logistics, donations, operational costs, and recovery, indicating a body of studies focused on return flows, donation sorting, and material recovery. Although Reverse Logistics appears in 53% of the articles, its role as a structuring concept is less frequent, and many practices associated with RL emerge implicitly and remain unnamed. This cluster represents the link between environmental and operational dimensions, as it addresses both the mitigation of environmental impacts and the reduction of logistical inefficiencies caused by surplus or returned materials.

The comparison across the three clusters (

Table 3. Comparative synthesis of the three identified clusters.

Cluster	Primary Focus	Predominant Terms	Role in the Literature
Cluster 1—Sustainability and Waste	Environmental impacts, circularity, and post-disaster waste management	sustainability; environmental management; life cycle assessment (LCA)	Introduces the environmental dimension into Humanitarian Operations; broadens the debate on circularity yet remains weakly integrated with the operational core.
Cluster 2—Humanitarian Logistics and Operations	Efficiency, transportation, disaster response, and logistical performance	Humanitarian Logistics; transportation; humanitarian supply chain; disaster; Humanitarian Operations; supply chain management; location; critical success factors	Represents the dominant core of the field; consolidates operational and rapid-response models, albeit with limited attention to reverse flows.
Cluster 3—Reverse Flows and Recovery	Material returns, donation sorting, recovery, and operational costs	Reverse Logistics; donations; operational costs	Functions as a bridge between operations and sustainability; practices are frequent but remain weakly theorised as formal Reverse Logistics.

) reveals a field that remains fragmented. While sustainability and RL practices are present in the literature, they are weakly integrated into the dominant theoretical core of HL. Whereas Cluster 2 maintains a predominantly operational orientation, Clusters 1 and 3 highlight practices and concerns that extend beyond the immediate efficiency of response, thereby exposing a gap in conceptual integration. This thematic division is consistent with recent literature and reinforces the need for models capable of coherently connecting operational, environmental, and institutional performance.

The qualitative analysis of the studies (Appendix B) followed a process of open and axial coding, enabling the consolidation of findings into four central themes that collectively explain the role of RL within the context of HOs. These themes reflect recurring patterns in the literature and highlight both the contributions and the limitations of incorporating RL into environments characterised by high operational pressure and structural constraints.

Table 4. Roles of Reverse Logistics (RL) in Humanitarian Operations (HOs) and percentage of corresponding studies.

Role of RL in HOs	Percentage of Studies	Description
Environmental/sustainability	33%	RL employed to manage disaster and medical waste; recycle or repurpose donations and packaging; reduce waste and environmental impacts; support circular economy principles; and contribute to SDGs through practices such as life cycle assessment (LCA), material recycling, food recovery, solar cookers, and biodigesters.
Operational/clearance and support to response	19%	RL functions as direct support for response and recovery activities, including debris clearance, stock replenishment, redistribution of surplus items, release of logistical capacity, and enhancement of agility and resilience within humanitarian supply chains.
Institutional/governance	12%	RL linked to public policies, regulations, and guidelines (particularly in nuclear and radiological contexts), decision-making frameworks, stakeholder coordination, and integration of waste management as a structural component of disaster governance and planning.
Technological/innovation	19%	RL operationalised through technological solutions such as IoMT, 3D printing/EcoPrinting, rapid prototyping, solar cookers, multi-objective models, machine learning, and digital applications, enabling new forms of data collection, material recovery, and the design of sustainable humanitarian supply chains.
No defined RL role (focus on HL/sustainability only)	17%	Studies addressing Humanitarian Logistics, network optimisation, performance, sustainability, or general waste management, but without explicitly or implicitly classifying these practices under the RL framework.

complements this analysis by quantifying the distribution of these roles, indicating that 33% of the studies address RL as an environmental mechanism, 19% link it to operational functions, 12% associate it with institutional aspects, and 19% relate it directly to technological innovation; additionally, 17% do not assign any explicit or implicit role to RL, thereby reinforcing the conceptual fragmentation observed.

The first theme identifies RL as a mechanism for promoting sustainability in disaster contexts. Approximately 78% of the articles analysed associate RL with practices such as recycling, reuse, environmentally sound disposal, post-disaster waste management, and the treatment of personal protective equipment (PPE), packaging, and construction and demolition waste. Table 3 reinforces this role by showing that 33% of the studies frame RL within environmental functions, including the recycling of materials, the recovery of food, the use of solar cookers and biodigesters, applications of life cycle assessment (LCA), and circular economy initiatives that directly contribute to SDGs 11, 12 and 13. From this perspective, RL operates as a fundamental strategy for reducing environmental impacts, reinserting materials into productive chains, and mitigating risks of contamination and public health issues.

The second theme refers to RL as an element of operational efficiency and logistical decongestion, appearing in around 64% of the qualitative studies. The most frequently associated codes include the clearing of access routes, the redistribution of surpluses, stock decommissioning, and the management of reverse routes. Table 3 corroborates this role by demonstrating that 19% of the studies position RL as direct support to response and recovery, emphasising activities such as debris removal, stock replenishment, the release of logistical capacity, and increased agility within the humanitarian supply chain. Empirical evidence from studies conducted in countries such as France, India and Brazil shows that the absence of structured RL systems tends to generate logistical bottlenecks, delays in humanitarian response, and additional operational risks, thereby compromising the effectiveness of operations.

The third theme highlights RL as a practice that is largely implicit and weakly institutionalised within HOs, appearing in approximately 70% of the studies. The associated codes include dispersed practices, the absence of formal regulations, inconsistent terminology, and improvised solutions in the field. Table 3 shows that 12% of the studies explicitly link RL to public policies, standards, guidelines, and decision-making processes, particularly in nuclear and radiological contexts where regulation is more stringent. However, in most cases, even when practices consistent with RL are observed, they are rarely formalised in policies, operational plans, or institutional roles, and they lack incentives, metrics, and clearly defined responsibilities. This institutional gap explains why many initiatives related to reverse flows occur reactively and ad hoc rather than being strategically integrated into Humanitarian Logistics structures.

Finally, the fourth theme positions RL as a vector of technological innovation and structural transformation, identified in around 32% of the qualitative articles. Within this group, codes related to the use of the Internet of Things (IoT), blockchain, multi-objective models, the Internet of Medical Things (IoMT), artificial intelligence (AI), and 3D printing are particularly prominent. Table 3 complements this perspective by indicating that 19% of the studies operationalise RL through technologies such as EcoPrinting, rapid prototyping, machine learning, apps for data collection and tracking, and energy solutions such as solar cookers, broadening the possibilities for monitoring, traceability, and material recycling. Despite the high transformative potential of these technologies, their adoption remains concentrated in countries with greater technological and institutional capacity, which tends to widen inequalities and limit replicability in humanitarian contexts with low levels of infrastructure.

The integrated synthesis of the quantitative, bibliometric, and thematic results confirms that RL performs four structural roles within the context of HOs, contributing simultaneously to the environmental, operational, institutional, and technological dimensions of these systems. However, the presence of the 17% group of studies that do not assign any role to RL reveals that, although RL-related practices permeate a substantial part of HOs, their conceptual and terminological recognition remains inconsistent. This absence of an explicit analytical framing constitutes one of the main factors hindering the cumulative development of knowledge and the institutionalisation of RL as a strategic component of Humanitarian Operations.

When combined, these findings show that RL is not merely an auxiliary component of HOs but rather a structuring element capable of integrating sustainability, efficiency, governance, and technological innovation. Nonetheless, its uneven distribution across roles, predominant in the environmental dimension (33%), relevant in the operational and technological dimensions (19% each), and still incipient in the institutional dimension (12%), demonstrates the need to deepen its formalisation, standardisation, and strategic integration in order to strengthen more effective and resilient humanitarian systems.

Through the qualitative analysis (Appendix D), it was possible to identify the challenges and opportunities associated with the integration between RL and HOs. Initially, two researchers independently extracted all passages describing barriers, limitations, risks, recommendations, and solutions related to reverse flows in humanitarian contexts. These excerpts were organised into a data matrix and coded into initial categories (open codes). Subsequently, an axial coding phase was undertaken, in which the codes were grouped into broader thematic blocks aligned with the theoretical constructs of system resilience, circularity, institutional barriers, and sociotechnical support. Based on the qualitative synthesis. the challenges were organised into four thematic blocks: (i) operational challenges and system resilience; (ii) institutional and governance-related challenges; (iii) technological and data-related challenges; and (iv) local capacity and geographical inequality. These blocks are directly linked to the constructs of system resilience, circularity, institutional barriers, and sociotechnical support (

Table 5. Challenges and opportunities in integrating RL and HOs across thematic blocks.

Thematic Block	Challenges	Percentage of Studies That Mention the Impacts	Classification of the Impacts of Challenges	Opportunities
Operational and system resilience	Uncertainty, disruptions, and low integration between flows	64	Critical	Integrated HL–RL models and resilient logistics networks
Institutional and governance	Institutional fragmentation and weak coordination	70	Critical	Integrated governance and permanent policy frameworks
Technological and data-related	Data scarcity and low technological adoption	32	Moderate	Digital technologies for real-time decision-making
Local capacity and geographical inequality	Limited infrastructure and unequal access	46	Critical	Strengthening local capacities and circular economy solutions

).

Table 5 presents a synthesis of the challenges, indicating the percentage of studies that mention each thematic block as well as their impact classification. It should be noted that a single study may refer to challenges belonging to more than one block; therefore, the percentages are not mutually exclusive. The first block encompasses challenges associated with logistical overload, flow fragmentation, and the difficulty of reconciling rapid response with structured return management, being mentioned in approximately 64% of the analysed studies. These challenges recur frequently in the literature and are classified as critical due to their direct impact on operational continuity and the recovery capacity of humanitarian systems. Examples include the obstruction of access routes by post-disaster waste, the occupation of warehouses by unsuitable donations, and the absence of planned reverse routes. Case studies on earthquake waste management in Italy, for instance, demonstrate that the lack of Reverse Logistics planning compromises the clearance of critical areas and delays the restoration of essential services. From a theoretical standpoint, these challenges weaken the resilience of humanitarian supply chains by reducing their capacity for adaptation and recovery following disruptive events.

The second block refers to the absence of regulatory frameworks, metrics, and clearly defined responsibilities for Reverse Logistics in Humanitarian Operations, as well as the low prioritisation of the topic by donors and organisations, being mentioned in approximately 70% of the studies. Several studies indicate that, although practices compatible with Reverse Logistics are carried out in the field, they are rarely formalised in policies, operational plans, or performance indicators, resulting in a reactive and ad hoc pattern of action. These challenges were classified as critical, with no low-impact challenges identified within this block, as they structurally affect governance and the institutionalisation of Reverse Logistics. Studies addressing sustainability barriers in humanitarian medical supply chains, for example, highlight the lack of institutional incentives and the exclusive focus on delivery speed, to the detriment of return and waste management. This set of challenges is directly linked to the institutional barriers described by institutional theory, in which norms, routines, and legitimacy pressures hinder the adoption of innovative practices.

The third block groups challenges related to insufficient digital infrastructure, low traceability of reverse flows, and difficulties in integrating data across multiple humanitarian actors, appearing in approximately 32% of the studies. In several contributions, solutions based on the Internet of Things, blockchain, collaborative platforms, or optimisation models are proposed; however, their implementation is constrained by costs, lack of interoperability, and limited local technical capabilities. These challenges occur with intermediate frequency and are classified as moderate. No low-impact challenges were identified in this block. They are closely associated with the sociotechnical systems construct, in which technologies and organisational arrangements must evolve in a coordinated manner.

The fourth block concerns limitations in technical, financial, and institutional capacity in regions characterised by fragile infrastructure or political instability, being mentioned in approximately 46% of the studies. The literature shows that countries with higher technological capacity and consolidated regulatory frameworks have progressed more rapidly in integrating Reverse Logistics into Humanitarian Operations, whereas contexts such as Venezuela and parts of Africa and Latin America face significant implementation barriers. These challenges are classified as critical, with no low-impact challenges identified in this block from a geographical equity perspective, as they reinforce asymmetries in the adoption of circular and resilience-oriented practices. From a theoretical perspective, this block articulates institutional barriers, system resilience, and sustainability transitions, highlighting that the diffusion of Reverse Logistics depends on public policies, inter-organisational cooperation, and the strengthening of local capacity.

Taken together, these four blocks demonstrate that the challenges are not merely operational but systemic, involving interactions among material flows, institutional structures, technologies, and geographical contexts, consistent with the understanding of Humanitarian Operations as complex adaptive systems. Notably, no challenges classified as low impact were identified in the analysed literature, reinforcing the structural and systemic nature of the barriers to integrating Reverse Logistics into Humanitarian Operations. The opportunities mapped in the literature represent pathways for structural transformation grounded in circularity, resilience, and sociotechnical change. Similarly, they were grouped into four thematic blocks:

(i)

Opportunities for circularity and sustainable waste management: Several studies demonstrate that RL enables recycling, reuse, energy recovery and the appropriate disposal of post-disaster waste, including construction debris, packaging, food waste and medical waste. These initiatives reduce environmental impacts, prevent contamination, and reintroduce materials into productive cycles, linking RL to the circular economy and SDG 12. Life cycle assessment (LCA) studies in disaster settings, for example, show that early planning of segregation and recycling strategies can significantly reduce emissions and end-of-life disposal costs.

(ii)

Opportunities to strengthen operational resilience: The literature indicates that, when planned, RL contributes to clearing access routes, freeing warehouse capacity, optimising inventories, and redesigning routes, thereby strengthening the response and recovery capacity of humanitarian supply chains. Pre-positioning models that incorporate returns, as well as systems for collecting inappropriate donations, exemplify how reverse flows can reduce uncertainty, mitigate risks, and support decision-making under pressure. Theoretically, such opportunities link RL and systems resilience, showing that structured reverse flows enhance flexibility and the ability of HOs to reorganise.

(iii)

Opportunities for sociotechnical innovation and digitalisation: Recent studies explore the use of blockchain for donation traceability, IoT and IoMT for monitoring inventories and medical waste, 3D printing for transforming plastic waste into useful inputs and multi-objective algorithms to optimise reverse routing and resource allocation. These initiatives are classified as opportunities for sociotechnical innovation, as they integrate emerging technologies, new governance arrangements, and RL practices oriented towards sustainability. Experiences with integrated digital platforms in humanitarian medical supply chains illustrate the potential to monitor returns in real time, reduce losses, and support data-driven decisions.

(iv)

Institutional opportunities and pathways for sustainability transitions: Finally, several studies highlight opportunities to incorporate RL into public policies, waste management regulations, donor guidelines, and socioenvironmental responsibility strategies of humanitarian organisations. Such initiatives create formal incentives for circular practices, enable the standardisation of procedures, and foster coordination among actors. In theoretical terms, these opportunities represent drivers of sustainability transitions, shifting RL from an occasional practice to a structural component of sustainable and resilient humanitarian supply chains.

5. Discussion

This systematic review addresses the research question by demonstrating that, although practices associated with Reverse Logistics (RL) are present in Humanitarian Operations (HOs), RL does not constitute a formally institutionalised field, operating predominantly as an implicit logic activated by environmental, operational, and contextual pressures. Based on the thematic coding of the 47 analysed studies, the discussion is structured around four central themes that emerge recurrently and interdependently: (i) sustainability as the main entry vector for RL; (ii) the operational role of RL in the resilience of humanitarian systems; (iii) the institutional paradox that constrains its consolidation; and (iv) the uneven role of technologies as enablers of reverse flows.

The empirical findings are interpreted through four complementary theoretical lenses: resilience and complex adaptive systems, circular economy, institutional theory, and a sociotechnical systems perspective. Together, these perspectives enable a systematic explanation of how Reverse Logistics contributes to adaptive capacity and recovery in Humanitarian Operations, how it operationalises circularity under conditions of scarcity, why its consolidation remains constrained by dominant institutional logics and governance structures, and how digital technologies and learning mechanisms condition its effective implementation in practice.

The first theme indicates that sustainability constitutes the primary entry point for RL in HOs. The predominant association between RL, post-disaster waste management, recycling, reuse, and environmentally sound disposal—observed in approximately 75% of the studies—converges with previous reviews that emphasise the centrality of environmental concerns in humanitarian supply chains. However, this review advances the literature by demonstrating that, although such practices are recurrent, they are rarely explicitly conceptualised as RL. Instead, they are analysed under diverse labels, such as waste management, circular economy, post-disaster recovery, or sustainability.

This systematic dissociation between practice and terminology partially explains the fragmentation of the field and the difficulty in building cumulative knowledge. The evidence suggests that the problem is primarily conceptual: RL exists in practice but not as a consolidated analytical construct. This finding constitutes a central theoretical contribution of the study, as it repositions RL as an integrative lens capable of connecting sustainability, logistics, and systemic recovery within humanitarian contexts. From a circular economy perspective, these results empirically indicate that humanitarian systems already operate with partial closed-loop practices, particularly in waste management and material recovery. However, the absence of an explicit Reverse Logistics framework limits the consolidation of these practices into structured circular cycles, constraining value retention, scalability, and long-term environmental gains.

The second theme reveals that RL plays a relevant operational role that remains under-theorised in the literature. The structured management of reverse flows contributes to the clearance of access routes, the release of storage capacity, the redistribution of surplus donations, and the mitigation of logistical congestion, functions directly associated with the operational resilience of HOs. These results challenge traditional HL approaches that privilege the efficiency and speed of forward flows as the primary performance criteria.

By demonstrating that operational continuity and the recovery capacity of humanitarian systems depend on the integrated coordination of forward and reverse flows, this study extends beyond still-dominant linear perspectives. RL thus emerges not as a peripheral or post-response activity but as a structural component that affects operational performance, adaptability, and sustainability. This interpretation resonates with theories of complex adaptive systems and resilience, offering a more comprehensive understanding of logistical performance in disaster contexts. Interpreted through the lens of resilience theory, these operational findings demonstrate that Reverse Logistics enhances the adaptive and recovery capacities of Humanitarian Operations. By enabling debris clearance, warehouse decongestion, and surplus redistribution, RL supports system reconfiguration after shocks, confirming that resilience depends on the coordinated management of both forward and reverse flows rather than on response speed alone.

A third central theme concerns the institutional role of RL. The synthesis of the studies reveals a recurring paradox: Although practices aligned with RL are observed in HOs, they are rarely formalised in policies, operational plans, performance metrics, or governance structures. This lack of institutionalisation results in the predominance of ad hoc solutions, dependent on local capacities, individual initiatives, or circumstantial pressures.

Drawing on institutional theory, the findings suggest that dominant logics within the humanitarian field—marked by urgency and the visibility of actions, tend to marginalise post-consumption processes and reverse flows. Accordingly, the barriers to integrating RL are not merely technical but organisational and normative. This theoretical contribution helps to explain why, despite its environmental and operational potential, RL remains at a low level of maturity within the humanitarian sector. From an institutional theory perspective, the results reveal a structural decoupling between practice and formalisation. Although RL activities are widely observed, dominant humanitarian logics prioritising urgency, visibility, and short-term accountability inhibit their translation into formal rules, metrics, and organisational roles. This explains why RL remains a latent capability rather than an institutionalised function within HOs.

The fourth theme relates to the role of technology as an enabler of the integration between RL and HOs. Although only a subset of studies explicitly explores digital technologies, the evidence indicates that solutions such as the Internet of Things, blockchain, artificial intelligence, multi-objective modelling, and 3D printing enhance traceability, coordination, and decision support in reverse flows, particularly in contexts involving medical waste, packaging, and plastics.

The regional disparities observed in the results should not be interpreted merely as geographical variation in case studies but rather as an indication that RL manifests unevenly according to institutional and sociotechnical capacities. In Global South contexts, where natural disasters and health crises tend to generate greater operational pressure and infrastructure constraints, RL frequently appears as a functional field practice—such as sorting, reuse, and waste handling, activated to mitigate bottlenecks and sanitary risks. In contrast, in Global North countries, there is a greater recurrence of normative, methodological, and technological approaches (e.g., guidelines, life cycle assessment, prototyping, and decision frameworks), suggesting that a significant share of scientific production is devoted to the formalisation and modelling of the phenomenon, not always accompanied by empirical validation in highly vulnerable settings.

These differences also help to explain why RL remains weakly institutionalised in the humanitarian field: There is a misalignment between where practices become critical, often in contexts with lower state and organisational capacity, and where knowledge tends to be systematised and translated into frameworks, which is more common in environments with stronger scientific and regulatory infrastructure. Thus, regional disparities reflect not only where research is conducted but how RL is understood: as an adaptive and local response in many vulnerable scenarios or as a formalised and technically structured object in ecosystems with higher institutional maturity.

From a sociotechnical systems perspective, the findings indicate that digital technologies act as enablers of RL only when embedded within appropriate organisational, human, and institutional arrangements. Although tools such as the IoT, blockchain, artificial intelligence, and decision-support models enhance the visibility, coordination, and traceability of reverse flows, their effectiveness remains uneven across humanitarian contexts. This variability reflects differences in local infrastructure, institutional capacity, and governance maturity rather than purely technical limitations.

Within this context, reinforcement learning emerges not as a standalone technological solution but as an adaptive decision-support mechanism capable of strengthening learning processes under uncertainty. When integrated with human and organisational learning, reinforcement learning contributes to the dynamic coordination of forward and reverse flows, resource prioritisation, and post-disaster waste management, thereby enhancing the adaptive capacity and resilience of HOs without replacing human judgment.

Importantly, the reviewed evidence does not support the interpretation of these technologies as established solutions within humanitarian practice. Rather, they operate as exploratory tools, tested through modelling, pilot initiatives, or conceptual proposals. Their effectiveness is mediated by sociotechnical arrangements, organisational learning, and governance capacity, reinforcing that technological adoption alone is insufficient to structurally integrate RL into HOs.

Based on these four themes, Figure 3 is presented as a conceptual synthesis derived from the evidence, rather than as a normative or empirically validated model. The framework organises the findings around four structural roles of RL in HOs, environmental, operational, institutional, and technological, highlighting how they interact throughout the disaster cycle. Sustainability emerges as the main activation vector of RL, while operational gains result from the integrated management of flows. At the same time, institutional fragility and technological asymmetries explain why RL remains an implicit practice rather than a formalised function.

The directional arrows, in Figure 3, illustrate how each dimension contributes to the integration of Reverse Logistics with Humanitarian Operations: institutional cooperation facilitates coordination and governance; local training strengthens community capabilities and resilience; and technological solutions enhance tracking, planning, and monitoring. These flows converge into the central integration mechanism and subsequently generate economic, environmental, and social benefits throughout humanitarian supply chains.

Based on this synthesis, it is possible to explicitly articulate how RL functions are distributed across different decision levels and sustainability dimensions in HOs. At the operational level, environmental and operational RL functions predominate, encompassing activities such as waste sorting and disposal, debris removal, route clearance, and surplus redistribution. These functions provide direct support to environmental sustainability and public health while simultaneously strengthening operational sustainability through the release of logistical capacity and improvements in system responsiveness.

At the strategic and institutional levels, institutional and technological RL functions become more prominent. These functions include the definition of protocols, responsibilities, performance indicators, and decision-support systems that enable the integration of reverse flows into humanitarian planning and governance structures. Such functions are associated with organisational, economic, and systemic dimensions of sustainability, as they strengthen accountability, coordination, legitimacy, and long-term efficiency.

Across these levels, RL operates as a transversal mechanism that connects immediate operational decisions to strategic objectives of sustainability, resilience, and institutional legitimacy. Rather than functioning as a post-operational or auxiliary activity, RL mediates the relationship between short-term response pressures and long-term recovery and development objectives, reinforcing its role as a structuring capability within sustainable and resilient humanitarian systems. From a methodological perspective, the distribution of studies reinforces a relevant analytical implication: The predominance of modelling, optimisation, and simulation supports advances in representing reverse flows and assessing trade-offs among cost, time, and environmental impacts but tends to capture institutional barriers, actor conflicts, and operational improvisation—hallmarks of HOs—only to a limited extent. By contrast, empirical and applied studies, which remain in the minority, offer greater explanatory power regarding coordination, governance, and local capacity, albeit with lower generalisability and weaker cumulative knowledge when they do not engage with shared analytical frameworks.

Accordingly, the identified methodological trends indicate that the conceptual fragmentation of the field is not merely a terminological issue but also an effect of the imbalance between analytical formalisation and field validation. Consolidating RL as a structuring component of HOs therefore requires advances in hybrid methodological approaches—such as models informed by empirical evidence and case studies that test simulation assumptions—capable of narrowing the gap between technically elegant solutions and the real conditions of implementation in humanitarian environments.

Thus, the principal theoretical contribution of this study lies in transforming a dispersed set of empirical findings into an integrated interpretation that clarifies what RL is, how it operates, and why it remains marginalised within HL. By articulating these elements empirically, this review not only answers the research question but also provides an analytical framework capable of guiding future research, supporting empirical comparisons across disaster contexts and informing the development of more systematic RL policies and practices within HOs.

To synthesise and clarify the complex set of relationships identified in this Systematic Literature Review, a Sankey-based interaction map is presented in Figure 4. Given the multidimensional nature of the findings and the coexistence of structural challenges, intermediate impacts, mediating mechanisms, and long-term outcomes, a purely textual synthesis would be insufficient to fully convey the dynamics observed in the literature. The Sankey representation enables an explicit visual articulation of how challenges propagate through humanitarian systems and how Reverse Logistics mechanisms operate as mediating pathways that transform these pressures into opportunities for resilience, sustainability, and improved governance. By organising the evidence into analytically distinct yet interconnected layers, the figure serves as an integrative interpretative tool that complements the qualitative analysis and enhances the transparency of this review’s conceptual contribution.

The Figure 4 uses a colour-coded layered structure to guide interpretation. Layer 1 (dark red/grey) represents structural challenges and the origin of systemic problems in Humanitarian Operations. Layer 2 (orange) depicts intermediate systemic impacts, such as logistical congestion, inefficient resource use, and environmental and public health risks; these effects are not Reverse Logistics activities themselves. Layer 3 (blue) represents Reverse Logistics mechanisms and constitutes the integrative core of the model, mediating between challenges and outcomes. Layer 4 (green) illustrates opportunities and structural gains, including resilience, sustainability, operational efficiency, and governance strengthening.

Figure 4 provides an evidence-based visual synthesis of the relationships identified in this Systematic Literature Review, organising the findings into four analytical layers arranged from left to right.

Layer 1 (structural challenges) represents the main barriers reported in the literature, including operational fragmentation and uncertainty (64% of studies), institutional fragmentation and weak governance (70%), data scarcity and low technological adoption (32%), and local capacity constraints and geographical inequality (46%). These elements constitute the structural origins of inefficiencies and vulnerabilities in Humanitarian Operations.

Layer 2 (systemic impacts on Humanitarian Operations) depicts the intermediate negative effects generated by these challenges, namely logistical congestion and blocked access routes, inefficient use of resources and surplus accumulation, and increased environmental and public health risks. These impacts are not Reverse Logistics activities themselves but rather systemic consequences that emerge when humanitarian systems operate without structured management of reverse flows.

Layer 3 (Reverse Logistics mechanisms) forms the core integrative layer of the model. It shows how Reverse Logistics operates through four distinct yet interconnected mechanisms:

(i)

Environmental mechanisms, such as waste segregation, recycling and reuse, and environmentally sound disposal;

(ii)

Operational mechanisms, including debris clearance, surplus redistribution, and reverse routing with stock release;

(iii)

Institutional mechanisms, comprising protocols and guidelines, integration into disaster governance, and the use of performance indicators;

(iv)

Technological mechanisms, involving IoT/IoMT, artificial intelligence and reinforcement learning, and blockchain-based traceability systems.

This layer visually emphasises Reverse Logistics as the main mediating axis between structural problems and positive system-level outcomes.

Layer 4 (opportunities and structural gains) illustrates the outcomes enabled by the systematic activation of Reverse Logistics mechanisms. These include enhanced system resilience (faster recovery and adaptive capacity), circularity and sustainability outcomes (reduced waste and alignment with SDGs 11, 12, and 13), operational efficiency and capacity release (freed warehouses and improved routing), and governance strengthening and organisational learning (institutional legitimacy and data-driven decision-making).

Across all layers, flow thickness qualitatively represents the relative recurrence and centrality of themes in the analysed studies, rather than quantitative volumes. Thicker flows highlight dominant pathways identified in the literature, particularly the links between institutional fragmentation, institutional Reverse Logistics mechanisms, and governance strengthening, as well as between operational fragmentation, operational Reverse Logistics mechanisms, and system resilience. Thinner flows reflect less recurrent but still relevant pathways, such as those associated with technological challenges and solutions.

Overall, the Sankey-based interaction map demonstrates that Reverse Logistics should not be interpreted as an auxiliary or post-operational activity within Humanitarian Operations but rather as a structuring and integrative capability. The visual synthesis highlights how environmental, operational, institutional, and technological mechanisms jointly mediate the transition from fragmented and reactive humanitarian systems towards configurations characterised by greater resilience, circularity, operational efficiency, and governance maturity. In doing so, the figure consolidates the central argument of this review, namely that the effectiveness and sustainability of Humanitarian Operations depend not only on the optimisation of forward flows but also on the deliberate and systematic integration of reverse flows into Humanitarian Logistics planning and decision-making.

6. Conclusions

This study analysed how RL is conceptualised, applied, and evaluated in the literature on HOs, based on a systematic analysis of 47 articles published between 2005 and 2025. This review addresses the research question by demonstrating that, despite its recurrent empirical presence, RL has not yet become an institutionalised or fully recognised component of Humanitarian Operations. The main contribution of this study lies in the proposal of an integrated, evidence-based typology of RL functions in HOs, which systematises and articulates four interrelated roles. By distinguishing these functions and explicitly clarifying their interactions, the proposed typology advances conceptual consolidation in the field and establishes a robust analytical foundation for both future empirical investigations and the practical implementation of policies, models, and decision-support tools.

This analysis further reveals that only a limited share of the reviewed studies explicitly employs the term “Reverse Logistics”, even though conceptually aligned practices—such as the management of surplus donations, waste sorting and disposal, material recovery and recycling, and post-disaster debris management—are widely documented. This dissociation between practice and terminology indicates that RL operates as an implicit and reactive operational logic, activated by contextual and environmental pressures, rather than as a planned, structured, and strategically integrated function within humanitarian supply chains.

The synthesis of the literature indicates that RL contributes to HOs through four interconnected mechanisms. The first is environmental, whereby RL functions as an instrument of sustainability by supporting waste management, circularity and the mitigation of environmental impacts; this represents the primary pathway through which reverse practices enter HOs. The second is operational, given that, when deliberately planned, RL contributes to unblocking access routes, reducing logistical congestion, freeing warehouse capacity and redistributing surplus supplies, thereby strengthening system resilience. The third mechanism is institutional, as the effectiveness of these practices is hindered by the absence of metrics, protocols, clearly defined responsibilities and inter-organisational coordination. The fourth mechanism is technological: Although digital solutions such as AI, reinforcement learning, IoT, IoMT, blockchain, and advanced decision-support systems offer considerable potential to enhance traceability, waste forecasting, sorting, and decision-making under uncertainty, their adoption remains uneven and dependent on local technological and institutional capacity.

The findings also indicate that the barriers to integrating Reverse Logistics (RL) into Humanitarian Operations (HOs) are structural and systemic. The identified challenges are concentrated in blocks of critical impact, particularly within the operational, institutional, and capacity inequality domains, and in one block of moderate impact associated with data and technology. No challenges classified as low impact were identified in the analysed literature. This result reinforces the notion that isolated or incremental improvements are unlikely to be sufficient when not accompanied by coordinated changes in governance, local capacities, and informational infrastructure, given that the observed constraints simultaneously affect operational effectiveness, circularity, and the institutional legitimacy of RL initiatives in humanitarian settings.

Taken together, these results indicate that RL in HOs should be understood as a latent and sub-formalised capability that gains relevance when explicitly integrated into logistical planning. The conceptual model proposed in this study synthesises these relationships by positioning RL as an integrating axis between HL, sustainability, governance, and technological support. It is an exploratory model, not empirically validated, designed to guide future research and support practical experimentation.

The recommendations derived from this review indicate that humanitarian organisations may enhance their performance by incorporating formal protocols for managing reverse flows, particularly structured donation sorting, waste segregation, and post-operation recovery, while also investing in staff training on safe disposal practices and health literacy. Governments and public authorities can promote the institutionalisation of RL by integrating circularity principles into disaster management, waste management, and recovery policies, as well as by strengthening regulatory frameworks and environmental incentives. Donors play a strategic role in embedding sustainability criteria and performance indicators related to reverse flows in funding guidelines. For logistics managers, the exploration and controlled testing of decision-support tools, particularly those based on artificial intelligence and reinforcement learning, are recommended as complementary mechanisms to support adaptive routing, sorting, and forecasting under uncertainty, rather than as standalone or fully operational solutions

Several limitations should be acknowledged. This review was restricted to publications indexed in Web of Science and Scopus. The methodological heterogeneity of the included articles limited the application of formal bias assessment tools, and the findings therefore reflect the academic discourse, which may not capture the full scope of practical knowledge present within HOs.

This study presents some limitations. The search was restricted to the Web of Science and Scopus databases, which may have excluded relevant studies from other sources or from the grey literature, although complementary searches were undertaken. The methodological heterogeneity of the analysed studies precluded the conduct of meta-analyses, resulting in a qualitative synthesis. In addition, the critically low confidence rating assigned by AMSTAR-2 reflects limitations of the instrument when applied to conceptual and interdisciplinary reviews, rather than shortcomings in the rigour of the adopted process. Finally, many Reverse Logistics practices are described implicitly in the literature, without standardised impact measurement, which constrains direct comparisons across studies.

Based on the gaps and limitations identified in this review, future research is likely to benefit from empirical and applied investigations that advance the consolidation of RL as a structuring component of HOs. First, there is an evident need for empirical studies that quantitatively assess the environmental, operational, and social impacts of RL practices across different disaster types and phases of the humanitarian cycle, enabling comparative analyses and supporting evidence-based decision-making. Second, future studies may explore how institutional capacity, governance arrangements, and regional contexts influence the adoption, formalisation, and effectiveness of RL, particularly in low- and middle-income countries, where reverse flows tend to be more operationally critical yet less institutionalised.

Third, applied research may focus on the development, testing, and validation of decision-support tools that integrate forward and reverse flows, including optimisation models, simulation-based approaches, and digital solutions grounded in real operational data. In this context, artificial intelligence and reinforcement learning show potential to be investigated as adaptive decision-support mechanisms—particularly for routing, sorting, waste forecasting, and resource prioritisation under uncertainty—rather than as autonomous or fully automated solutions. Fourth, future studies may examine cost–benefit trade-offs and propose performance metrics capable of simultaneously capturing efficiency, resilience, and sustainability outcomes, thereby facilitating the integration of RL into humanitarian planning, donor frameworks, and policy instruments.

Additionally, there is scope for applied and participatory research involving humanitarian practitioners, local authorities, and affected communities in the co-development of RL protocols, capacity-building strategies, and scalable circular solutions. Such efforts contribute to narrowing the gap between analytical modelling and field implementation, strengthening organisational learning processes, and supporting the transition from reactive reverse practices towards more systematic, governable, and sustainable Humanitarian Logistics systems.