A Review of Supply Chain Resilience: A Network Modeling Perspective

Abstract

Against the backdrop of globalization, the complexity of supply chains has been increasing, making supply chain resilience a critical factor in ensuring the stable operation of enterprises, national economies, and international trade. This paper adopts a network modeling perspective to systematically review the theoretical foundations and research progress in supply chain resilience, focusing on the application of network modeling methods. First, the concept of supply chain resilience is defined, and its developmental trajectory is reviewed. Through literature visualization analysis, this study delves into the current state of research on supply chain resilience, addressing challenges and risk management, highlighting the importance of network modeling techniques in this field. Subsequently, it explores supply chain network modeling based on complex networks and agent-based modeling, analyzing their strengths and limitations in simulating the overall evolution of supply chains and the dynamic behavior of individual entities. By integrating network structural characteristics with resilience evaluation methods, this paper suggests potential directions for future research. These include enhancing the description of individual firm behavior, analyzing the dynamics of information networks, and emphasizing task-oriented model design, thereby offering new perspectives and pathways for managing supply chain resilience in a way that can generate significant positive externalities for global economies. This research also indicates that the enhanced resilience of supply chains can produce a multiplier effect, benefiting not only individual firms but also promoting economic stability and growth across multiple countries.

1. Introduction

In the wave of globalization, supply chains, as critical links connecting the global economy, are becoming increasingly complex and dynamic. Managing supply chain resilience has become a critical priority not only for enterprises and industries but also as an essential safeguard for the stability of industrial systems and national economies. However, as supply chains expand and deepen, they face growing challenges and risks. From natural disasters to political conflicts, technical failures to market fluctuations, various internal and external factors can disrupt supply chains at any time, affecting their stability and efficiency [1]. Consequently, supply chain resilience has emerged as a critical focus in both academic research and practical applications.

Current research on supply chain network resilience predominantly focuses on static analyses or evaluations at specific time points, often overlooking the dynamic evolution of supply chains over time. As a result, there is a lack of in-depth studies on how resilience changes across different time scales. To address this gap, it is crucial to develop dynamic models to better capture the resilience fluctuations throughout the supply chain’s life cycle. Moreover, existing resilience assessment methods tend to consider a limited range of factors, primarily focusing on operational aspects while neglecting broader influences, such as policies, market conditions, and technological advancements. To provide a more holistic evaluation, future research should integrate these diverse factors and expand the scope to include financial and market resilience, offering a comprehensive view of the supply chain’s overall resilience. Additionally, there is a noticeable scarcity of cross-industry and context-specific applications of resilience evaluation methods. So, it is necessary to apply these models to various industries and investigate the resilience of supply chains under unique scenarios, such as political conflicts or natural disasters, to refine strategies for coping with such challenges.

In this context, we propose a new idea that combines complex network theory with agent-based modeling, offering a more comprehensive and dynamic framework for supply chain network modeling. This innovative concept enables the analysis of the supply chain’s structure, information flow, and node connectivity at a macro level while simultaneously simulating the autonomous behavior and decision-making processes of individual entities at the micro level. This dual approach not only improves the model’s accuracy but also enables a deeper understanding of interactions and coordination during task execution, especially in uncertain and complex environments. By utilizing agent-based modeling, we can uncover the behavioral characteristics and decision-making logic of individual nodes. In contrast, complex network modeling provides insights into the efficiency, paths, and topology of information transmission. Together, these approaches enhance the precision of task allocation, resource configuration, and information-sharing simulations, ultimately improving emergency response capabilities, resource optimization, and overall resilience. This idea offers a novel perspective on enhancing the task completion capabilities of the supply chain, thus contributing meaningfully to filling the existing research gaps in supply chain resilience assessment.

This paper aims to systematically review the theoretical foundations and research progress of supply chain resilience from the perspective of network modeling while analyzing the application of network modeling methods in this field. As a powerful analytical tool, network modeling enables a comprehensive understanding of supply chain structures and behaviors at both macro and micro levels. It enables the identification of critical nodes and vulnerable links while also facilitating the simulation of the dynamic evolution of supply chains under various conditions [2,3].

First, this paper defines the concept of supply chain resilience and reviews its developmental trajectory. Through a literature visualization analysis, this study explores the current state of research on supply chain resilience, focusing on the challenges faced by supply chains and risk management strategies. It underscores the pivotal role of network modeling techniques within this field. Next, the paper categorizes the methods of supply chain network modeling from top-down and bottom-up perspectives. It explores methods based on complex networks and agent-based modeling, analyzing their strengths and limitations in simulating both the overall evolution of supply chains and the dynamic behaviors of individual entities. Furthermore, it integrates traditional resilience evaluation methods with those based on network structural characteristics, providing a detailed discussion of the latter. Finally, the paper outlines potential future research directions, including enhancing the description of individual firm behavior, analyzing the dynamics of information networks, and emphasizing task-oriented model design, offering new perspectives and pathways for managing supply chain resilience. The framework is shown in Figure 1.

2. Overview of Supply Chain Resilience Development

2.1. Fundamental Concept of Resilience

2.1.1. The Concept and Origin of Resilience

In today’s interconnected world, different systems can interact and connect to generate functions that are unattainable by individual systems alone. This phenomenon is known as emergence, and systems possessing this characteristic are called Systems of Systems (SoS) [4] or multi-system integrations. SoS is widely applied across various domains, encompassing infrastructure such as urban underground pipelines and power grids, networked military organizations, and supply chain systems of large enterprises or e-commerce platforms [5].

In multi-system integrations, the interconnectivity between subsystems is tighter, but the complexity and uncertainty of such systems pose potential threats that can lead to subsystem failures, consequently degrading the performance of the entire integrated system. Maintaining the regular operation of multi-system integrations thus presents increasingly significant challenges, and mitigating the impact of these potential threats on system performance has become ever more critical [6].

The typical approach to addressing threats or disruptions faced by systems is to design robustly integrated systems. Traditional robust design typically employs probabilistic methods to optimize system parameters, making the system less sensitive to external disturbances [7] or utilizing over-specification to reduce the likelihood of failure [8]. Moreover, redundancy strategies, such as backup systems or functions, are often implemented to prevent system failures [9]. While robust design has been widely applied, its application in multi-system integration presents certain limitations.

Firstly, robust design is typically suitable for single systems. For large-scale integrated systems, over-specification and redundancy strategies can impose significant economic burdens. Secondly, due to the complex emergent behaviors of multi-system integration, the threats and uncertainties it faces are more severe, making it increasingly challenging to enhance its resistance to disruptions.

Rather than merely striving to improve the disturbance resistance of multi-system integration, a more effective approach is acknowledging that performance degradation in the face of disruptions is inevitable [10]. Under such circumstances, research should focus on adapting to disruptions and improving system performance post-disturbance. Resilience, which emphasizes a system’s capacity to adapt and recover from disruptions or failures, provides a promising framework. By quantifying changes in system functionality before and after disruptions, resilience is particularly well-suited for multi-system integration.

The ecologist Holling [11] was among the first researchers to propose the concept of system resilience, defining it as “a measure of a system’s ability to absorb changes and disturbances while maintaining relationships among key species and state variables.” This concept has since been widely applied in fields such as transportation, power infrastructure, economics, and social systems.

The diversity of research subjects has led to multiple definitions of resilience. For instance, the U.S. Department of Homeland Security defines resilience as the ability of assets, systems, or networks to sustain functionality or rapidly recover during terrorist attacks [12]. The American Society of Mechanical Engineers describes it as the ability of a system to continue functioning under external or internal disruptions or to recover following failure quickly [13]. The U.S. National Infrastructure Advisory Council views resilience as the capacity of infrastructure systems to anticipate, absorb, adapt to, and recover from disruptive events such as natural disasters [14]. In the field of systems engineering, resilience is often defined as the ability of a system to adapt and recover to normal operating conditions after experiencing disruptions [15].

The concept of resilience in China has developed through three main perspectives: initially focusing on engineering resilience, followed by ecological resilience, and later advancing to evolutionary resilience. Each adjustment and refinement has expanded the connotation and denotation of the resilience concept, reflecting the academic community’s progressively more profound understanding of resilience [16,17], as illustrated in

Table 1. Comparison of resilience concepts.

Type	Equilibrium State	Research Focus	Connotation	Target System
Engineering Resilience	Pursues a single equilibrium state	Stability, robustness	The system adapts and recovers to its normal state after disturbances, emphasizing adaptability and recovery speed and extent.	Physical systems
Ecological Resilience	Pursues multiple equilibrium states	Maximum disturbance a system can withstand before changing its equilibrium state	The system shifts from one equilibrium state to another after disturbances, emphasizing the stability of the system structure and function.	Ecological systems
Evolutionary Resilience	Abandons equilibrium pursuit, emphasizes continuous evolution	Complex adaptive systems, dynamic evolution	Resilience is a continuously evolving process, with the potential to create new development pathways, emphasizing the adaptive transformation of the system structure and function.	Economic systems

.

Although the definition of resilience remains inconsistent across disciplines, there are some commonalities among the various definitions. Most emphasize the system’s ability to “absorb” and “adapt” to disturbances, with recovery capacity being a key component of resilience. The primary differences lie in the required level of recovery across systems: some systems prioritize robustness, while others focus more on recovery capacity or balance both aspects.

2.1.2. The Relationship Among Resilience, Flexibility, Robustness, Vulnerability, and Reliability

Resilience, flexibility, robustness, vulnerability, and reliability are commonly used concepts in systems science and engineering. Each has distinct definitions and application contexts, addressing various system characteristics in response to external shocks, internal failures, or environmental changes. While these concepts describe different aspects of a system’s adaptability and recovery capacity from different perspectives, they also exhibit the following specific intrinsic interconnections [18]:

• Reliability refers to a system’s ability to perform its specified functions under predetermined conditions within a specified time frame. A reliable system maintains regular operation during the expected period, with a low probability of failure;
• Vulnerability describes the degree to which a system is susceptible to damage or functional failure when exposed to external or internal threats. A highly vulnerable system lacks resistance to disturbances or shocks and is prone to failure. High vulnerability indicates insufficient protective mechanisms or response strategies, making the system more susceptible to harm;
• Robustness is the ability of a system to maintain its normal functions in the face of uncertainty, disturbances, or environmental changes. A highly robust system can withstand a certain degree of shocks or failures without losing functionality. Robustness is often associated with system stability and strength, emphasizing the capacity to operate continuously even when subjected to certain levels of disruption;
• Flexibility refers to a system’s ability to adapt to changes in the environment or demands. A highly flexible system can adjust to maintain functionality or improve efficiency amidst change. Flexibility does not necessarily imply maintaining original functionality after a shock but involves reconfiguring strategies, structures, or configurations to adapt to new conditions;
• Resilience is the ability of a system to recover from disturbances, shocks, or failures and return to its original state or adapt to a new environment. Resilience emphasizes recovery capacity and adaptability. It relates to a system’s self-repair and recovery mechanisms, particularly the ability to quickly return to normal or establish a new equilibrium after partial failures or disruptions [19,20,21].

The distinctions and similarities among resilience, flexibility, robustness, vulnerability, and reliability primarily manifest in the following aspects: Reliability pertains to the probability of regular operation, emphasizing the absence of system failures. Vulnerability reflects the weaknesses of a system under threats, indicating the degree to which the system is prone to damage. It is conceptually opposite to reliability, as high vulnerability often implies lower robustness, flexibility, and resilience. Robustness focuses on a system’s ability to continue functioning under disturbances without requiring adjustments or adaptations. Systems with high robustness typically exhibit low vulnerability but may not possess flexibility or resilience. Flexibility emphasizes a system’s adaptability, allowing it to adjust to meet new demands or environments without necessarily maintaining its original functionality. Highly flexible systems can cope with dynamic environments but may not consistently demonstrate strong robustness. Resilience highlights a system’s capacity to adapt and recover after experiencing damage. A system may fail to function during disruptions (indicating low robustness) but can restore normal operations through rapid repair or adaptive adjustments (indicating high resilience) [22,23].

For instance, a system can be robust (capable of operating under external disturbances) but vulnerable (susceptible to specific types of attacks). Likewise, a highly flexible system is better equipped to adapt to future changes, while a resilient system can recover quickly from disruptions. Furthermore, flexibility generally pertains to operational-level management, whereas resilience and robustness are more relevant to strategic considerations [24]. These system characteristics overlap yet have distinct focal points.

A simple analogy can help illustrate these system characteristics. Consider a tree as a system: when strong winds blow, a tree with high flexibility bends with the wind and survives. A resilient tree also bends but returns to its upright position once the wind subsides. In contrast, a robust tree remains upright during the windstorm without significant impact [25].

In summary, vulnerability represents a system’s weaknesses, robustness denotes its ability to resist disturbances, flexibility refers to its capacity to adapt to changes, and resilience describes its ability to adapt and recover. These characteristics can coexist, but the design and requirements of different systems may emphasize one trait over others. A conceptual comparison of these properties is shown in Figure 2.

2.2. Current State of Research on Supply Chain Resilience

Supply chain resilience refers to the ability of a supply chain to maintain its functions, quickly recover, and adapt to changes when facing various disruptions and uncertainties. In today’s VUCA era (characterized by Volatility, Uncertainty, Complexity, and Ambiguity), the challenges confronting supply chains have become increasingly prominent, making research and practice on supply chain resilience particularly critical [26].

Moreover, against the backdrop of global trade growth and the expansion of e-commerce, supply chains have become more complex and, consequently, more vulnerable to various natural and human-induced disruptions, such as natural disasters, accidents, trade conflicts, wars, and strikes [27]. The fragility of supply chains is particularly pronounced within the constantly shifting global economic, commercial, and political landscape. Therefore, enhancing supply chain resilience has emerged as a key strategy for maintaining supply chain stability [28].

Supply chain resilience has gradually become a focal point in academic research, particularly in recent years, as global supply chains have faced significant disruptions due to events such as the COVID-19 pandemic and the Russia–Ukraine conflict. These events have heightened awareness among scholars regarding the importance of studying supply chain resilience [29]. A systematic and comprehensive analysis of the current state of supply chain resilience research was conducted using bibliometric analysis tools to explore the research frontiers and developmental trajectory in this field.

Co-citation analysis enables exploring the development and evolutionary dynamics within a specific research domain. By constructing a co-citation network, stronger citation relationships can be observed among studies within the same research direction. Through clustering, the primary research areas within the field can be identified.

Using the search term TS = “Supply Chain Resilience” in the Web of Science database, searches were conducted in both the “SCI-E” and “SSCI” databases to ensure data accuracy. After filtering for “Article” and “Review” document types, 1170 papers were identified. These were imported into CiteSpace (6.2R6) for co-citation analysis. The clustering results are shown in Figure 3, with cluster labels generated based on the keywords of the clustered documents. The reddest clusters represent the current research hotspots.

It is evident that research on supply chain resilience has emerged as a rapidly growing field in recent years. Early studies primarily focused on models and methods using supply chain resilience as a context, such as system dynamics and directed graph matrix methods [30,31], as well as conceptual explorations related to supply chain resilience [32]. Subsequently, studies addressed organizational resilience at the individual enterprise level within supply chains [33]. However, these efforts did not coalesce into well-defined and focused research directions.

The sudden outbreak of the COVID-19 pandemic, which caused widespread disruptions to global supply chains, has highlighted that supply chain resilience, once considered a concern at the enterprise level, has now escalated to an industry-wide, regional, and even national strategic issue. Supply chain resilience is closely tied to national security and economic development [34].

At the same time, the increasing complexity of supply chains and the rise of artificial intelligence (AI) have introduced new pathways for enhancing supply chain resilience. Modgil et al. noted that AI can help supply chains respond to unexpected events by enhancing transparency, providing personalized services, reducing the impact of disruptions, and promoting flexible procurement strategies. Furthermore, AI can identify risks, analyze potential issues, reconfigure networks, and activate response measures to maintain supply chain resilience under extreme disruptions [35,36]. Belhadi et al., through empirical research methods, concluded that the integration of AI technologies not only enhances short-term supply chain performance but also strengthens long-term resilience, thereby ensuring stable operations in uncertain environments [37]. Integrating AI to strengthen supply chain resilience has thus become a critical direction for future research and practice.

The ripple effect describes a phenomenon in which throwing stones into a calm lake creates waves that gradually spread to distant areas [38]. In risk management, the ripple effect refers to the negative impact that a risk event can have on its surroundings, akin to the ripples caused by a stone thrown into the water. If left uncontrolled, the ripple effect of an initial risk can often trigger more significant crises [39].

Although the concept of the ripple effect was introduced long ago, it is evident from Figure 3 that research related to this effect in the context of supply chain resilience surged following the outbreak of the COVID-19 pandemic. The pandemic’s impact on global supply chains exposed numerous vulnerabilities in modern supply chains, making the ripple effect—referring to the propagation and impact of risks within the supply chain—one of the core issues in current supply chain resilience research [40].

In recent years, supply chain risk management has garnered significant attention in academic research, with methodologies spanning mathematical optimization, simulation, and game theory. When studying risk propagation within supply chains, Li et al. identified network characteristics as a critical factor influencing supply chain resilience. They argued that specific combinations of network features could enhance the overall risk-resistance capacity of the network [41]. Zhao et al. found that the topological structure of supply chains directly affects the speed and extent of risk propagation under different types of disruptions. Adjusting the network structure can effectively mitigate the impact of the ripple effect [3]. By strategically selecting network structures and characteristics, it is possible to impede risk propagation within the supply chain network, thereby improving resilience.

Furthermore, existing assessment frameworks of supply chain resilience often overlook the ripple effect, leading to biased evaluations. To address this, Habibi et al. introduced resilience-related indicators, such as absorptive capacity, adaptive capacity, and recovery capacity, to comprehensively assess the supply chain’s ability to respond to risk propagation [42]. Building on this, Saisridhar underscored the importance of adopting a complex adaptive systems perspective for supply chains and utilizing simulation techniques to enhance the understanding and management of risk propagation, thereby exploring novel paradigms in supply chain management [43].

However, current research on resilience often neglects the dynamic changes and uncertainties inherent in real-world supply chains. Real-world supply chain systems are typically characterized by asymmetric information, conflicting interests, and complex nonlinear relationships, which existing models fail to simulate effectively [44]. Additionally, risk propagation in supply chains involves multiple layers, including upstream and downstream risk transmission pathways and the interdependencies among different stages of the supply chain. Most studies focus on single risk propagation pathways, overlooking the complex interactions across multiple tiers and various types of risks [45].

With the intensification of globalization and uncertain environments, supply chain resilience is no longer merely the ability of an isolated system to respond to disruptions. It also encompasses enhancing the entire supply chain system’s responsiveness and adaptability to unexpected events through network structure optimization and multi-level risk propagation analysis [42]. Future research will increasingly focus on how supply chain risks propagate across hierarchical levels and how these propagation pathways impact overall resilience. This will involve exploring multidimensional risk propagation mechanisms to provide more precise and systematic theoretical support for improving supply chain resilience.

3. Current Status of Supply Chain Network Modeling Research

The supply chain network is a critical component of modern economic systems, encompassing multiple nodes, such as suppliers, manufacturers, warehouses, distribution centers, and retailers. Given the escalating complexity and globalization of supply chain management, modeling techniques for supply chain networks are becoming increasingly important for optimizing resource allocation, reducing operational costs, and enhancing supply chain efficiency. Current modeling approaches for supply chain networks can be broadly categorized into two main types:

• Complex Network-Based Modeling: This approach adopts a top-down perspective, focusing on the static structure and relationships within the supply chain network;
• Agent-Based Modeling: This method takes a bottom-up approach, emphasizing the dynamic behaviors and decision-making processes within the supply chain [46,47,48,49].

In the study of supply chain resilience, network modeling methods offer numerous advantages over empirical studies. Network modeling provides a macro perspective, allowing analysis of the supply chain network’s topological structure and its impact on system performance. This approach helps identify critical nodes and connection patterns, enabling the anticipation of potential vulnerabilities within the supply chain [50]. Unlike static empirical studies, network modeling dynamically simulates the evolution of networks under varying conditions, accounting for the temporal evolution of node connections and the integration of new nodes. By applying concepts from graph theory and network science, this approach facilitates a higher level of abstraction in understanding the internal relationships and performance dynamics of supply chains.

Moreover, network modeling supports system optimization and decision-making by identifying which node failures might lead to the collapse of the entire supply chain, providing scientific guidance for designing resilience strategies [51]. With its ability to offer a macro-level perspective, dynamic simulation, complexity analysis, and system optimization support, network modeling comprehensively and profoundly reveals the resilience of supply chain systems from multiple angles.

3.1. Supply Chain Network Modeling Based on Complex Networks

In recent years, significant progress has been made in supply chain network modeling using complex network theory. A literature search using the query TS = “supply chain” AND TS = “complex network” in the “SCI-E” and “SSCI” databases yielded 187 publications. These studies were analyzed with CiteSpace to track the evolution of research hotspots, as shown in Figure 4. Terms such as “complex network theory”, “resilience”, “cascading failure”, and “risk transmission” have shown notable citation bursts in recent years, highlighting their growing importance in supply chain resilience research. For instance, scholars like Hearnshaw have advanced supply chain network theory by applying complex network theory. Their work includes understanding the network properties of supply chains as complex adaptive systems and analyzing the robustness of supply chains through these properties [52].

Complex network theory abstracts nodes and edges within a system to describe and analyze the structure and behavior of complex systems. In supply chain networks, nodes represent various entities within the supply chain (e.g., suppliers and manufacturers), while edges denote the relationships between these entities (e.g., logistics and information flows). Many characteristics of supply chain networks can be represented using complex network theory, such as the concentration of a significant portion of supply tasks on a few core suppliers or logistics hubs, the small-world effect among enterprises, and the local clustering and hierarchical organization of nodes within the supply chain network [53,54].

From a functional perspective, the supply chain can be divided into multiple layers, such as the supply layer, information layer, and financial layer. The supply layer primarily focuses on the flow of materials and products within the supply chain, the information layer is responsible for information transmission, communication, and decision support, while the financial layer emphasizes payment, financing, and cost control [55,56]. These layers are interwoven, collectively forming the structural foundation of the supply chain network.

The complex network model offers unique advantages in analyzing risk transmission mechanisms across these layers. The intense citation burst of the keyword “risk transmission” between 2021 and 2024 highlights the growing academic focus on the diffusion of risks across different layers within supply chains. For instance, Yao et al. observed that risk propagation in supply chains is influenced by issues such as information asymmetry between enterprises, leading to the rapid transmission of risks to other nodes and threatening the stability of the entire network [57]. Research advancements suggest that analyzing the interactions between supply, information, and other layers can provide deeper insights into optimizing the overall performance of supply chain networks.

From a structural perspective, supply chains can be categorized into two types: hierarchical and non-hierarchical structures (Figure 5). Hierarchical structures are typically organized by layers based on different roles, such as suppliers, manufacturers, distributors, retailers, and customers. These layers collaborate, emphasizing the functions and positions of various roles within the supply chain. For example, suppliers handle raw material provision, manufacturers focus on production and processing, distributors manage inventory, and retailers oversee end-user sales. In contrast, non-hierarchical structures treat the supply chain as an integrated whole, highlighting cross-functional coordination and integration without explicit hierarchical distinctions. Relationships among nodes in non-hierarchical networks tend to be equal, making this structure suitable for distributed or decentralized supply chain networks [58,59].

Whether the perspective is hierarchical or non-hierarchical, complex network theory can be employed to uncover the dynamic behaviors and optimization strategies within supply chain networks. The emergent trends of the keywords “cascading failure” and “disruption” in Figure 4 suggest that using such models to study the resilience and vulnerabilities of supply chain networks has become an increasingly prominent research focus. For instance, Basole et al. explored computational methods to analyze supply chain network structures, visibility, and risk propagation, offering deep insights into the functions and roles of different actors within the supply chain, as well as strategies for understanding and optimizing overall network performance [60].

Figure 5. (a) Hierarchical supply chain network structures [61] and (b) non-hierarchical supply chain network structures.

Figure 5. (a) Hierarchical supply chain network structures [61] and (b) non-hierarchical supply chain network structures.

In the study of applying complex network theory to supply chain networks, the cascading failure model and epidemic model are among the core methods for analyzing supply chain resilience and vulnerability [62]. The cascading failure model is primarily used to simulate chain reactions triggered by the failure of a single node within a supply chain network. Guo et al. [63] developed a cascading failure model based on a project subtask network to analyze the chain reactions initiated by the failure of individual tasks and their impact on the overall project. As shown in Figure 6, the author constructed a project cascading failure network model using the NetworkX library. In the experiment, the parameters β and α were varied to analyze the impact of node load (L) and capacity (C) on the project. Specifically, β changes from 0.1 to 1.5, while α takes on four specific values. They also proposed corresponding risk management measures. Similarly, disrupting critical nodes in a supply chain, such as core suppliers or distributors, can lead to cascading collapses across other nodes, thereby undermining the stability of the entire network. Liu et al. [64] further revealed that such models not only identify the potential risks associated with specific nodes but also offer guidance for optimizing the network structure of supply chains. For example, resilience can be enhanced by adding redundant nodes or increasing the capacity of critical nodes to mitigate risks.

On the other hand, epidemic models are applied to analyze the propagation characteristics of information or risks within supply chains, analogous to the spread of diseases. For instance, when an issue arises at an information node in a supply chain, it may rapidly disseminate through the information or financial layers, thereby affecting the decision-making and resource allocation of upstream and downstream nodes. Yao et al. [57] investigated the influence of risk awareness and information disclosure awareness on the spread of supply chain risks based on a two-layer partially mapped SIS model, considering factors such as network topology, information disclosure consciousness, and risk awareness. Similarly, Guo et al. [56] employed a two-layer SIR model that incorporates group psychology and risk preferences to analyze how these factors influence risk propagation in the presence of different early warning scenarios. As shown in Figure 7, the author constructed a risk propagation model of the supply chain based on the SIR model. The enterprises were classified into three states: Susceptible (S), Infected (I), and Recovered (R). The transitions of enterprise states and the risk propagation process were jointly determined by the parameters of infection rate β, recovery rate γ, risk information α, and enterprise preference ρ. Using the microscopic Markov chain approach (MMCA), they derived the threshold for risk propagation. This model helps researchers understand how risks or uncertainties gradually permeate the network. Particularly in decentralized or distributed supply chain structures, epidemic models effectively reveal the pathways of information dissemination and the potential scope of diffusion [65].

Although modeling methods based on complex networks offer significant advantages in describing the structural characteristics of supply chain networks, they exhibit certain limitations in simulating the dynamic behaviors of supply chains and the heterogeneity among individual enterprises. Relying solely on complex network models is insufficient to comprehensively capture the diverse behaviors and complex interactions within supply chain networks.

3.2. Agent-Based Modeling of Supply Chain Networks

The agent-based modeling (ABM) approach treats each entity within a system as an independent agent, defining the behaviors of each individual and its interactions with the environment to simulate and analyze the emergent behaviors of the entire system [66]. This modeling method is well-suited for simulating complex systems, especially those involving multiple autonomous decision-makers.

The multi-agent modeling approach offers a high degree of flexibility and scalability, enabling analysis at both the micro and macro levels. The analysis can either focus on the behavior of an individual agent or explore the collaboration between multiple agents, as well as the collective behavior of the system. Its core principle is “divide and conquer,” where a complex system is decomposed into several relatively independent subsystems. Through the collaboration of agents, the overall system can achieve optimization and accomplish tasks [67].

Agent-based network modeling has been under development for an extended period. With the rise of deep reinforcement learning, this approach has regained significant attention in the academic community. A search in the databases “SCI-E” and “SSCI” using the query TS = “supply chain” AND TS = “agent” yielded 1378 relevant articles after filtering out the unrelated literature. These articles were analyzed using CiteSpace, resulting in a research hotspot evolution analysis, as shown in Figure 8. Keywords such as “multi-agent systems”, “simulation”, “agent-based systems”, and “supply chain management” have exhibited notable citation bursts over the past few years, highlighting their importance in research on supply chain resilience.

A supply chain network, as a complex, multi-layered, and multi-faceted system, is far more than a simple linkage of multiple enterprises. It consists of numerous entities with autonomous decision-making capabilities, information transmission abilities, and interaction mechanisms. Within this network, each node, such as suppliers, manufacturers, and distributors, can be viewed as an agent with independent decision-making, behavioral capabilities, and the ability to interact with other nodes [46]. By simulating and analyzing the behaviors and collaborative patterns of these agents, researchers can gain deeper insights into the complex dynamic processes, information flows, and resource allocation within supply chain networks.

At the micro level, agent-based modeling of supply chain networks primarily focuses on modeling individual behaviors and decision-making processes. As seen in Figure 8, with keywords such as “coordination” and “decision making”, this line of research emphasizes analyzing the interaction patterns, coordination mechanisms, and how agents make decisions based on their goals and the environment. These studies have been widely applied to optimize tasks such as resource allocation, inventory management, route planning, and scheduling within supply chains.

For instance, Du et al. [68] proposed an ontology-based multi-agent decision support framework (Figure 9) to address data heterogeneity issues in the prefabricated component supply chain, aiming to enhance decision-making and resource allocation efficiency. Additionally, Zhao et al. [69] investigated firms’ adaptive strategies in response to disruptions within supply chain networks using an agent-based model, exploring the roles of reactive and proactive strategies for mitigating the impact of supply chain interruptions. By simulating various supply chain scenarios, researchers can test and validate different decision-making mechanisms to optimize operational efficiency at each stage, thereby improving the performance of the overall system.

At the macro level, research on network modeling methods based on agents primarily focuses on multi-agent systems, which examine the emergent and self-organizing behaviors of a large number of agents at the macro level, particularly in developing strategies to address dynamic changes and uncertainties. Moncada et al. [70] utilized an agent-based model to demonstrate how Brazil, under specific policy incentives, could naturally develop a bio-aviation fuel supply chain. This emergence is driven by the interactions and decision-making among various participants in the existing sugarcane–ethanol supply chain (e.g., farmers, factory owners, and fuel suppliers) and policies (e.g., subsidies and taxes) (Figure 10). Maltseva et al. [71], through a self-organizing network dynamic systems model, analyzed technology-driven production systems and socially driven social networks, emphasizing the unique characteristics of applying continuous models at the macro level. Shim et al. [72], leveraging agent-based modeling and simulation, explored how micro-level heterogeneity in thresholds and consumer activity levels influences the speed and scope of new product diffusion at the macro level. Their research highlighted the roles of advertising and negative word-of-mouth in the diffusion process. Furthermore, Bertani et al. [73] combined agent-based modeling with the macroeconomic model Eurace to reveal how intangible digital technologies (e.g., software, artificial intelligence, and algorithms) lead to phenomena such as increasing returns, winner-takes-all dynamics, and market lock-in. These, in turn, trigger macro-level effects, such as employment declines driven by productivity improvements. However, it is worth noting that studies explicitly targeting the emergent properties of supply chain networks remain relatively scarce.

In summary, agent-based modeling techniques for supply chain networks have made considerable advancements in simulating the decision-making, coordination, and collaboration processes among supply chain participants. These methods demonstrate substantial application potential in enhancing supply chain flexibility, reducing risks, and optimizing resource allocation. With the continuous advancement of agent-based technologies, future supply chain models are expected to more accurately simulate dynamic behaviors and system responses in complex environments, providing more reliable decision support for enterprises in practical operations.

Despite the advantages of agent-based modeling in simulating the dynamic behavior and task execution of supply chains, challenges remain in terms of model complexity and computational resource requirements. Multi-agent models must account for the interactions and decision-making processes of numerous individual agents, resulting in high computational complexity and difficulties in capturing a comprehensive, global perspective on the supply chain network.

3.3. Comment on Supply Chain Resilience Based on Network Modeling Methods

In the study of supply chain resilience, the primary goal of modeling is to accurately simulate the dynamic behavior of supply chain networks and their task execution capabilities to address the growing demands of uncertain and complex environments. By focusing on the following three aspects, the integrated application of complex network theory and agent-based modeling methods can significantly enhance the effectiveness of supply chain modeling:

• Behavioral Modeling of Business Entities

Supply chain networks are not merely simplistic structures of nodes and connections; instead, the nodes represent business entities with autonomous behaviors and decision-making capabilities. Businesses within the supply chain, such as suppliers, manufacturers, and distributors, function as intelligent agents with the ability to adjust and make decisions based on external environmental factors and internal objectives. Consequently, supply chain modeling must focus on these behavioral characteristics to accurately depict the dynamic performance of the supply chain.

Agent-based modeling is particularly effective for simulating the independent decision-making behaviors and task execution processes within organizations. By defining the roles, objectives, and interaction mechanisms of businesses within the supply chain, ABM enables the realistic simulation of task execution within the supply chain. This is especially valuable for analyzing scenarios involving task allocation, inventory management, and resource distribution. ABM facilitates an in-depth exploration of decision-making logic and collaboration patterns under varying conditions, thereby enhancing the task-oriented nature and practical applicability of the model.

2.

Information Network Modeling as a Key Factor

The information network within a supply chain plays a critical role in facilitating information transmission, sharing, and decision-making support among businesses, serving as the core of the supply chain’s coordination and responsiveness. Information network modeling must not only address the mechanisms and efficiency of information transmission but also encompass mechanisms for information sharing and coordination, as well as their impact on business decisions. Such an information network is more than a data transmission channel—it constitutes a decision-support and collaboration-oriented information flow system.

Applying complex network modeling methods can effectively describe the structural features of supply chain information networks. These methods facilitate the analysis of information transmission pathways, topological structures, and the efficiency and reliability of information flow between business nodes. By employing complex network modeling, researchers can uncover the propagation mechanisms of information within supply chains and evaluate the importance of various nodes within the information network and their influence on supply chain collaboration. This approach is critical for optimizing information-sharing strategies and enhancing the collaborative and responsive capabilities of the supply chain.

3.

Task Fulfillment as the Ultimate Objective

The ultimate goal of supply chain network modeling is to support the efficient execution of supply chain tasks, particularly by ensuring rapid response and effective resource allocation in mission-critical scenarios. The task execution capabilities of a supply chain stem from the collaboration among its business nodes and the support provided by the information network. By embedding business behavior characteristics and information network structures into the model, it becomes possible to simulate the interactions and coordination between different businesses during task execution, optimizing resource allocation and reducing the risks of supply disruptions.

In summary, the two main approaches currently used in supply chain network modeling are agent-based modeling and complex network modeling.

Complex network modeling is well-suited for analyzing the overall structure, information transmission, and node connectivity of supply chain networks from a macro perspective. It helps identify critical nodes and vulnerable links but struggles to reflect the individual behavioral characteristics of businesses within the supply chain.

Agent-based modeling, on the other hand, adopts a bottom-up approach, focusing on individual entities. It emphasizes autonomous behaviors and local interactions, making it practical for capturing the dynamic processes of task execution within the supply chain. However, it is less effective for systematic design from a global perspective.

By integrating these approaches—viewing nodes in complex networks as intelligent agents with autonomous decision-making capabilities and incorporating the structural characteristics of information networks—it becomes possible to simulate task execution processes in dynamic environments better. This integration enhances the overall task fulfillment capacity of the supply chain, improving its resilience and operational reliability.

4. Resilience Assessment Methods: Current Research Status

4.1. Traditional Resilience Assessment Methods

The process of resilience assessment can be categorized into two main types: qualitative description and quantitative description [74]. Qualitative descriptions typically focus on further refining and elaborating the concept of resilience. Quantitative descriptions, on the other hand, can be divided into two subtypes: specialized assessments tailored to specific domains and general assessments applicable across systems. Specialized resilience assessments are designed for the structural characteristics of specific systems and lack broad applicability. Therefore, this paper does not delve into these in detail.

General resilience assessment refers to constructing a universal evaluation framework that uses system performance data as input to produce quantitative results. These metrics allow for intuitive comparisons of resilience levels between different systems or within the same system under varying conditions [75]. Since the results of general assessments rely solely on system performance data and are independent of the specific structure of the system, they can be widely applied to resilience analyses across diverse systems.

In the field of resilience metrics research, Bruneau et al. [76] were the first to propose a triangular model (Figure 11) to construct a resilience assessment framework for analyzing the impacts of earthquakes on community losses. The resilience calculation method is detailed in Equation (1).

$$RL={\displaystyle {\int }_{t_0}^{t_1}[100-Q(t)]dt}$$

(1)

Here, t₀ represents the moment when the disturbance occurs, t₁ marks the time when the system recovers to normal, and Q(t) denotes the system’s quality during the disturbance period. Q(t) can represent various performance metrics. A higher RL value indicates lower resilience, whereas a smaller RL value reflects higher resilience.

The assumptions underlying this method have been a topic of debate. First, setting the initial performance state of all systems uniformly to 100 may be problematic, as the normal operating state of different systems is difficult to quantify accurately as 100. Second, the assumption that system performance immediately drops to its lowest point upon disturbance is not universally applicable, as some systems experience gradual declines in performance. Third, the assumption that recovery actions commence immediately after a disturbance may not be entirely realistic, as certain systems may require an adaptation period before recovery efforts begin. Finally, the assumption that system performance can fully recover to its original state is also questionable, as some systems may struggle to return to their pre-disturbance performance levels.

Henry et al. [78] proposed a hypothesis addressing the aforementioned deficiencies, suggesting that a system undergoes three distinct steady states during a single disturbance or attack, as illustrated in Figure 12. The first stage is the initial steady state, representing the system’s condition before the disturbance (from t₀ to t_e). The second stage is the post-disturbance steady state, which reflects the state of the system under the influence of the disturbance (from t_d to t_s). The third stage is the recovered steady state, indicating the new state the system reaches after recovery operations are completed (after t_f).

Some studies have conducted an in-depth exploration of resilience quantification metrics to more comprehensively evaluate system resilience. Ayyub et al. [79] categorized resilience into two dimensions: robustness and adaptive capacity. They defined system resilience as the ratio of system strength to the applied load, with the calculation formula shown in Equation (2).

$$R_e={\displaystyle \frac{T_i+F\Delta T_f+R\Delta T_r}{T_i+\Delta T_f+\Delta T_r}}$$

(2)

Here, T_i represents the time when the event occurs, T_f denotes the time of failure, and T_r indicates the recovery time. ΔT_f = T_f − T_i refers to the duration of failure, while ΔT_r = T_r − T_f represents the recovery duration. F and R stand for robustness and redundancy, respectively. This metric is relatively comprehensive as it integrates reliability and sustained recovery strategies. By introducing F and R, it establishes the ratio of robustness to redundancy as well as the ratio of adaptive capacity to recovery speed.

In summary, system resilience can be described through dimensions such as damage resistance and recovery capability. These dimensions can be quantified based on system performance data and integrated into a comprehensive resilience metric through systematic analysis. This quantification approach is intuitive, concise, and highly generalizable, making it suitable for quantitatively comparing the resilience levels of different systems or the same system under varying conditions. However, current system resilience evaluations still face the following challenges:

• Stage Division: From the perspective of stage categorization, different systems may exhibit variations in the refinement or overlap of stages. For instance, some studies divide the system task process into three stages: disruption, adaptation, and recovery, while others categorize it into only two stages: disruption and recovery;
• Capability Definition: From the perspective of capability categorization, there is no consensus on which specific capabilities resilience should include. Moreover, the qualitative descriptions and quantitative representations of various capabilities may overlap or intersect;
• Metric Construction: the construction of resilience metrics still faces unresolved issues, such as unclear quantification methods and the absence of rational integration approaches for various capability factors.

4.2. Resilience Assessment Methods Based on Network Structural Characteristics

In the study of supply chain resilience, evaluating resilience based on network structural characteristics has become a key research direction. This approach primarily focuses on analyzing the structural features of supply chain networks to measure their adaptive capacity and recovery efficiency when disruptions occur. The conceptual foundation of this evaluation originates from social network analysis (SNA) in the social sciences. However, in this context, it is applied to supply chain network research to investigate the specific impacts of network structure on supply chain performance.

Resilience evaluation based on network structural characteristics emphasizes analyzing the connections and interactions among various entities within a supply chain. This approach aids in understanding the mechanisms by which structural relationships between supply chain partners influence corporate performance. For instance, Basole et al. [80] demonstrated that structural significance and network density are critical factors affecting operational performance. Similarly, Akgul et al. [81] further validated that social network analysis provides more precise insights into the characteristics and patterns of supply chain collaboration networks across different markets and scales.

This resilience evaluation approach offers theoretical support for studying network factors and features. It enables researchers to identify key nodes within networks, analyze connection patterns, and compare different networks using various standardized metrics.

Table 2. Typical network structural metrics [82].

Type	Metric	Definition
Diameter	Network Diameter	The maximum shortest path length between any two nodes in the network.
	Average Path Length	The average number of steps along the shortest paths for all pairs of nodes.
Degree	Node Degree	The number of links associated with a node.
	Network Density	Defined as the ratio of total actual links to the total potential links in the network.
	Flow Complexity	Represents the average number of outgoing flows from any node.
Centrality	Closeness Centrality	Measures how close a given node is to all other nodes in the network.
	Degree Centrality	A metric of the positional significance of a given node based on its degree.
	Betweenness Centrality	How often does a node act as a bridge on the shortest paths between other nodes.
Clustering	Clustering Coefficient	A measure of the tendency of nodes to form tightly knit groups, calculated as the ratio of the number of closed triplets (triangles) to the total triplets.

summarizes commonly used network structural characteristics and their definitions, whose applications in the field of supply chain management are increasingly gaining attention.

Traditional supply chain management typically focuses on the direct linear relationships between buyers and suppliers. However, this linear perspective fails to capture the complex dynamics required for corporate strategies and actions. From a network perspective, the relative position of nodes within a network significantly influences overall strategies and behaviors. For example, Zhang et al. [83] analyzed the topological structure of transportation networks to explore their adaptability in responding to disruptions, using graph theory metrics to assess the resilience levels of network structures.

Although the number of empirical studies in this area remains limited, the potential applications of resilience evaluation based on network structural characteristics in supply chain management are gradually gaining recognition. Kao et al. [84] analyzed inter-firm relationship data to identify key network structural metrics correlated with supply chain efficiency. Shao et al. [85] proposed a data analysis framework that integrates network centrality metrics to classify and identify associated suppliers. Zhao et al. [86] discussed a framework that provides managers with a method to evaluate the topological characteristics of supply chain networks under various disruption scenarios. This approach allows for the early identification of network vulnerabilities and facilitates proactive management measures.

Dixit et al. [87] proposed Equation (3) based on a logarithmic linear weighted combination of factor functions, incorporating structural metrics such as average path length, in-degree, and out-degree, to quantify supply chain resilience. They integrated these structural characteristics with supply chain elasticity into a computational framework, providing a novel methodology for evaluating supply chain resilience.

$$R={\displaystyle \frac{CV\times NS}{D\times CT}}$$

(3)

Here, D represents the density of a given network, CT denotes the centrality of the network, CV refers to the connectivity of the network, and NS signifies the scale of the network.

Resilience evaluation based on network structural characteristics offers an effective tool for deepening the understanding and enhancement of supply chain resilience. By quantifying the complex dynamics of modern supply chain relationships, this approach aids managers in identifying critical nodes within the network and optimizing its structural design. Consequently, it strengthens the overall risk resistance and adaptability of supply chains.

5. Summary and Future Outlook

Current research on supply chain resilience often overlooks its dynamic evolution over time, focusing mainly on static analyses. Future studies should develop dynamic models to capture resilience fluctuations throughout the supply chain’s life cycle. Additionally, resilience assessments should include broader factors like policies, market conditions, and technological advancements. There is also a need for cross-industry applications that explore resilience in specific scenarios, such as political conflicts or natural disasters, in order to enhance coping strategies.

We propose an integrated approach that combines complex network theory with agent-based modeling, offering a more comprehensive and dynamic framework for supply chain network modeling. Through this integrated approach, the model is capable of analyzing the supply chain’s structure, information flow, and node connectivity at a macro level while also simulating the autonomous behavior and decision-making processes of individual business entities at a micro level. This combination allows for a more accurate representation of the interactions and coordination during task execution within the supply chain, especially in uncertain and complex environments. Specifically, the agent-based modeling approach helps reveal the behavioral characteristics and decision logic of individual nodes in the supply chain. In contrast, complex network modeling effectively describes the efficiency, paths, and topology of information transmission. By combining these two methods, researchers can more precisely simulate task allocation, resource configuration, and information-sharing processes within the supply chain, thereby improving the supply chain’s emergency response capability, resource optimization, and overall resilience. This innovative modeling approach (Figure 13) provides a new perspective for enhancing the task completion capability of supply chain networks.

In addition, by systematically comparing and distinguishing concepts such as resilience, flexibility, robustness, vulnerability, and reliability, the intrinsic relationships and interactions between them in response to external shocks and internal failures are revealed, providing a new perspective for understanding the adaptability and recovery capabilities of complex systems.

The multiplier effect of supply chain resilience is a key takeaway from this research. Strengthening resilience within supply chains generates far-reaching benefits that extend beyond the immediate operational stability of firms. These effects cascade across industries, regions, and economies, creating a ripple effect that drives broader economic growth, mitigates risks, and fosters national and global stability. By enhancing supply chain resilience, this research illustrates how firms, industries, and even nations can experience compounded positive outcomes, reinforcing the interconnectedness of global economies.

5.1. Key Points to Pay Attention to in the Study of Supply Chain Resilience

Two core aspects must be addressed to conduct an in-depth exploration of supply chain resilience: first, the establishment of an appropriate supply chain network model, and second, the development of resilience evaluation methods tailored to supply chains.

In terms of supply chain modeling, current research often struggles to capture the complexity and uncertainty inherent in supply chains. Many studies assume that supply chains operate under relatively ideal conditions, neglecting potential external threats and corresponding recovery strategies and failing to account comprehensively for real-world environments. Although some recent studies have begun to consider external threats, their solutions predominantly focus on enhancing the threat-avoidance capabilities of supply chains. However, in increasingly complex external environments, such approaches fail to provide comprehensive guidance.

In terms of resilience evaluation, existing methods are primarily applied to systems such as power grids and transportation networks during emergencies. These systems often employ relatively simple performance metrics. However, due to the inherent complexity of supply chains, their components, network topologies, coordination strategies, and interaction mechanisms vary with tasks, making it challenging to establish unified evaluation standards at the task capability level. Current studies often rely on simulation models to assess the performance of supply chains under different attack and recovery strategies, but these efforts primarily focus on system-level performance while neglecting other influential factors, such as network topology and coordination strategies.

Overall, research on supply chain resilience is still in its infancy. On the one hand, existing supply chain network modeling and resilience evaluation methods are insufficiently integrated, making it difficult to reflect the complexity and dynamism of supply chains comprehensively. On the other hand, current resilience evaluation approaches are not directly applicable to assessing supply chains in complex environments. Therefore, there is an urgent need to integrate complex network and agent-based modeling approaches to develop a comprehensive resilience evaluation framework, enhancing the ability of supply chains to respond to emergencies.

5.2. Further Directions

Exclusive reliance on complex network modeling to analyze supply chain networks presents several limitations. Although complex networks effectively uncover the network’s topological structure and the relationships between nodes, they fail to fully capture the multi-layered, multi-tiered, and highly heterogeneous nature of supply chains. Moreover, they are inadequate in simulating the dynamic evolution of supply chains and the impact of external factors. In addition, complex network models often neglect the decision-making behaviors of stakeholders within the supply chain and the mechanisms underlying risk propagation, thereby hindering a comprehensive analysis of supply chain stability and emergency response. Consequently, sole dependence on complex network modeling is insufficient to represent the full complexity of supply chains. It must be complemented by other modeling techniques for a more holistic analysis.

This study provides a systematic review of supply chain resilience from the network modeling perspective. This study conducts a comprehensive literature review and visual analysis to explore the theoretical foundations, research advancements, and both the applications and limitations of network modeling methods in examining supply chain resilience. The main conclusions and contributions of this paper are as follows:

• Theoretical Foundations and Concept Definition: The concept of supply chain resilience is clarified, and its developmental trajectory is reviewed. Supply chain resilience encompasses not only the system’s ability to adapt to and absorb shocks but also its capability to recover to a normal state;
• Research Progress and Network Modeling Applications: the importance of network modeling techniques in supply chain resilience research is highlighted, particularly in the areas of risk propagation, simulation of supply chain evolution, and dynamic behavior analysis of individual entities;
• Methodological Advantages and Limitations: This paper examines the methods of supply chain network modeling based on complex networks and agent-based modeling, analyzing their strengths and limitations. While complex network models are suited for macro-level analyses, agent-based models focus on micro-level dynamic behavior simulations;
• The application of interdisciplinary methods in supply chain resilience research remains at an early stage. Future research can focus on the following directions [88,89,90]:
  • Enhanced Representation of Individual Firm Behavior: as the complexity of supply chain networks increases, accurately depicting firm-level behaviors will become a critical research direction;
  • Dynamic Analysis of Information Networks: The role of information networks in supply chains extends beyond information transmission. They also serve as critical enablers for decision-making and collaboration;
  • Task-Oriented Model Design: in dynamic and highly uncertain environments, task-driven supply chain modeling, particularly for managing emergencies and ensuring the fulfillment of urgent demands, will be a key area of focus.

This study offers new perspectives and methods for researching supply chain resilience, providing theoretical support and guidance for supply chain management practices. By adopting network modeling approaches it not only aims to help researchers design more resilient and sustainable supply chain networks capable of thriving in complex and uncertain environments but also highlights the multiplier effect of supply chain resilience. This research underscores how strengthening supply chain resilience can have far-reaching benefits, enhancing not only the operational stability of individual firms but also contributing to the economic stability and growth of nations. As global supply chains become increasingly interconnected, the insights from this study can foster greater international cooperation and offer strategic pathways to mitigate risks, manage disruptions, and build robust systems that generate positive economic externalities for countries and regions worldwide. Therefore, the findings of this study are of significant relevance for both practitioners and policymakers, offering new avenues for global economic resilience.