Freight transport networks are vulnerable not only because links or nodes may be physically disconnected, but also because residual capacity may become insufficient even when origins and destinations remain topologically connected. This paper develops an analytically tractable framework for freight network resilience that treats connectivity and redundancy as explicit structural economic variables. We represent a multimodal freight system as an attributed spatial graph whose links and nodes may differ by mode, capacity, generalized cost, and transfer function. Connectivity is defined in two complementary ways: topological connectivity through edge-disjoint paths and capacity-mediated connectivity through max-flow/min-cut capacity and cut slack. Disruptive shocks are modeled as link or node removal, capacity degradation, and cost jumps, capturing freight-specific disruptions such as port closures, bridge and rail-corridor failures, border delays, canal blockages, and intermodal terminal disruptions. The analysis derives cut-based vulnerability conditions, regime boundaries between smooth degradation, capacity infeasibility, and disconnection, and spatial incidence results governed by binding cuts and corridor shadow prices. A computational experiment on a synthetic bottleneck-corridor freight network illustrates the operability of the proposed measures and shows that binary connectivity can substantially overstate resilience when residual capacity is depleted. The framework bridges network theory and spatial economics by clarifying when redundancy yields systemic resilience benefits rather than marginal capacity improvements.
Nowadays, the integrated energy system (IES) faces significant security challenges from both natural occurrences and intentional disruptions, even minor disruptions can impact equipment operation and the system’s energy supply. To mitigate such disruptions in IES, a protective decision optimization model (PDOM) has been proposed based on complex network theory, which combines three protective measures: backup protection, physical protection, and new lines connection to minimize the weighted betweenness loss expectation. The result of the calculation, i.e., an optimized pre-disaster protection scheme, aims to determine which nodes and lines need to be protected (and different-type protection) within the specified budget. Furthermore, EvoRein-mixed integer linear programming (EvoRein-MILP) algorithm is presented to optimize the proposed model’s scheme. Simulation results demonstrate the effectiveness and feasibility of the obtained scheme, as verified by three validation indicators (system fragmentation, energy transmission efficiency, and load losing rate). Finally, compared to the genetic algorithm (GA)-MILP algorithm and adaptive GA-MILP algorithm, the EvoRein-MILP algorithm exhibits superior optimization and operation time performance under identical conditions.
This study presents a novel correlation network model that integrates the Mahalanobis distance and complex network theory to explain the dynamic relationship between geopolitical events and crude oil. The geopolitical risk index and three crude oil futures prices are selected as sample data, and a correlation network is established. The node represents a type of correlation pattern, which is a combination of symbolic correlations between a geopolitical risk index and any crude oil price on a given day. The frequency and direction between two nodes are expressed in terms of weighted edges. The evolutionary properties of correlations can be described by the network’s topological structure. The results show that the correlation pattern in the network changes with time. The correlation between a geopolitical risk index and any crude oil price is strong and has long-term consistency. Furthermore, the correlation between China crude oil futures price and the geopolitical risk index slightly lags behind that between China crude oil futures price and the other two indices, and the influencing factors are discussed. This study explores the evolution of the dynamic mechanism of correlation, which can provide valuable implications for market investors and policymakers.
Accurate forecasting of daily demand in network systems is critical for efficient resource management, yet traditional forecasting methods often struggle to capture complex and dynamic demand patterns. Despite numerous studies on demand prediction, there remains a lack of systematic evaluation of nature-inspired optimization algorithms in field-based settings. This study addresses this gap by developing and assessing a daily demand forecasting model using four Swarm Intelligence algorithms: Black Hole Algorithm (BHA), Cuckoo Optimization Algorithm (COA), Shuffled Complex Evolution (SCE), and Stochastic Fractal Search (SFS). Historical demand data were collected through field surveys, and the forecasting model was constructed by integrating these algorithms with a Multilayer Perceptron (MLP) network. The model’s performance was evaluated using R² (Coefficient of Determination) and RMSE (Root Mean Square Error), benchmarked against traditional approaches. Results demonstrate that incorporating Swarm Intelligence significantly enhances forecasting accuracy. In particular, the SCE-optimized MLP achieved the highest R² values (0.99785 in testing, 0.99986 in training) and the lowest RMSE (0.01805 in testing, 0.00658 in training), indicating exceptional predictive reliability. Similarly, SFS yielded high R² (0.99554 in testing, 0.99995 in training) and low RMSE (0.02599 in testing, 0.00389 in training), confirming its effectiveness for daily demand prediction. These findings highlight the practical potential of SCE and SFS in improving network system forecasting accuracy and adaptability to dynamic demand patterns.
The interplay between population dynamics, resilience, and commuting patterns forms a critical nexus in regional economic and social development. This study investigates these intersections within the Swedish context, focusing on how spatial connectivity and labor market accessibility shape individual residential mobility and long-term population retention. The study addresses two central research questions: First, to what extent does a municipality’s position within the commuting network influence regional resilience as expressed through population retention? Second, how do spatial accessibility conditions contribute to differences in out-migration risks across municipalities? The objective is to analyze resilience not merely as an outcome but as a dynamic interplay between socioeconomic and spatial variables. Using data from the Swedish PLACE database, the study uses a longitudinal panel of individual-level residential, migratory, social, and commuting information. The analysis applies survival models to examine how individual characteristics and municipal-level spatial structures jointly shape the hazard of residential out-migration. The results indicate that population retention is significantly associated with spatial connectivity and labor market accessibility. Municipalities with stronger network integration and better access to nearby employment opportunities exhibit lower out-migration risks. At the individual level, age, education, family structure, and migration background emerge as important predictors of mobility behavior. Commuting diversity is positively associated with migration hazards, suggesting that more dispersed commuting structures may signal weaker local anchoring. Overall, the findings highlight the importance of spatial embeddedness and functional labor market access in shaping demographic stability and regional resilience.
This paper proposes a profit-maximization model for a multi-echelon supply chain, integrating transportation and inventory decisions within a unified framework. The system is formulated as a Variational Inequality, ensuring equilibrium conditions under convex costs and network constraints. By incorporating the Economic Order Quantity policy, inventory variables are eliminated, reducing model complexity. A projection method is used for computation, and numerical results illustrate the effectiveness of the approach in capturing realistic logistics behaviors.
Intercity freight flows vary within the year, especially for seasonal commodities such as fruit, yet many freight-network studies still rely on annually aggregated static networks. Using fulfillment records for fruit shipments from a leading digital freight platform in China, we construct monthly and seasonal directed weighted origin–destination (OD) flow networks for December 2022–November 2023. The analysis asks whether seasonal variation is associated with changes in hub roles, community structure, and OD-graph connectivity at three structural levels: node, community, and network. The results show a stable consumption-side core and a more variable production-side core. Among monthly top-10
This paper investigates spatial disparities in healthcare network infrastructure across Italian provinces from 2010 to 2021 and examines how these imbalances affect the resilience of the healthcare system to shocks such as the COVID-19 pandemic. Using data on hospital beds by type of care (acute, rehabilitation, and long-term) and applying inequality measures from the Generalized Entropy class, we assess both the level and evolution of territorial disparities. The decomposability of these indices allows us to distinguish inequality arising between provinces from inequality within provinces, providing insights about how hospital capacity is spatially configured across the national territory. As the Italian hospital network is organised along a regulated hub-and-spoke logic, the predominance of the within-province component of this spatial configuration should be read jointly with the regulatory framework. Our findings reveal persistent and substantial inequality, with the within-province component accounting for more than half of total inequality in most years, and disparities intensifying after 2016 and during the pandemic. By capturing the concentration of hospital capacity and its evolution over time, entropy-based measures offer a useful monitoring tool to identify structural vulnerabilities in the healthcare network infrastructure that weaken system-wide adaptability and responsiveness. The results show that resilience in Italy is undermined not only by the well-known North–South divide but also by uneven capacity within provinces, highlighting the need for policies that strengthen hospital networks at multiple territorial levels.
The tourism industry plays a significant role in the global economy, contributing not only to economic growth but also to sociocultural advancement. However, the emergence of a new highly contagious disease, COVID-19, triggered a dual crisis, affecting both public health and financial stability worldwide. In spite of this fact, global tourism has shown remarkable resilience. Following several years of severe decline, the industry has experienced a revival, marked by the resurgence of international travel and the strengthening of key tourism trends. Although much research has examined the impact of the pandemic on tourism, comparatively few studies provide direct comparative analyses. In the article, the authors employ conditional inference trees to compare tourist segmentation based on pre- and post-pandemic travel data. The study examines tourist behaviour in Poland using representative survey data collected in 2019 and 2023. The aim is to identify key factors shaping the travel purpose and type, and how these patterns evolved before and after the pandemic. The analysis of six classification models shows that tourist segmentation in both years was primarily driven by travel-related characteristics such as the means of transport, the distance travelled, the number of overnight stays, and the travel purpose. Demographic factors played a minor role. The findings highlight the growing importance of organisational factors in tourist decision making and suggest a post-pandemic convergence of travel behaviours across demographic groups. These insights may support more inclusive tourism strategies in Poland and assist tourism associations and travel agencies in tailoring their offerings to better reflect changing tourist preferences.
Spillover effects are crucial mechanisms in enhancing national innovation capabilities. This study combines the theories of dual innovation and innovation value chain to construct analytical frameworks from the dual innovation value chain perspective. National innovation capabilities are deconstructed into four sub-dimensions of innovation capabilities: knowledge creation, knowledge absorption and sharing, technology commercialization, and technological agglomeration and radiation. An analytical framework for national innovation capabilities was employed to measure the innovation capabilities of 66 countries globally from 2010 to 2023. Based on this premise, this study constructs an analytical framework for multidimensional spillover effects and uses the spatial Durbin model to explore the vertical, horizontal, and cross-dimensional spillover effects of national innovation capabilities. We find that vertical and horizontal spillover effects occur intra-nationally, but not inter-nationally, while cross-dimensional spillover effects are present both intra-nationally and inter-nationally. This finding provides fresh evidence for research on innovation spillovers; distinguishing between intra- and inter-country spillovers unveils novel paths for future empirical inquiries. Based on these results, we recommend that policymakers prioritize strengthening domestic coordination across innovation stages while strategically engaging in international collaboration to harness cross-dimensional spillover benefits for enhanced national competitiveness.
The study introduces a new theoretical framework for analyzing production systems and their resilience, linking output complexity to specific network architecture and the market form that can support it. In particular, it is pointed out that connectivity shapes the ability to generate complex outputs. This idea is integrated into a model of vulnerability and redundancy (Ninivaggi and Cutrini, 2025), highlighting that (a) High-complexity outputs require markets closer to oligopolies, with centralized and quasi-decomposable networks; (b) Low-complexity outputs can be produced in more competitive and fragmented markets. In this context, the analysis of entropy provides fundamental insights into the informational structure of networks and the capacity to handle and generate relevant knowledge. Effective information management allows for reducing entropy, improving productive efficiency, and ensuring the resilience of the resulting market form. The conclusions provide practical insights for resource management and economic policy design, while also pointing to new research directions on how innovation has the potential to reshape connectivity architectures and market forms.
We develop a game-theoretic framework to analyze public incentive allocation between road freight and coastal shipping (cabotage) in a competitive transport system. The interaction among government and transport operators is modeled as a sequential extensive-form game in which policy choices influence strategic investment decisions, generalized transport costs, and fiscal outcomes.The model captures how public incentives affect the strategic response of operators and allows equilibrium regimes to be computed under alternative policy scenarios. Sensitivity analysis over investment effectiveness parameters reveals regime-switching behavior and threshold effects in modal competition, highlighting conditions under which targeted incentives may induce equilibrium-consistent investment in cabotage.The proposed framework contributes to the literature on dynamic policy games by integrating incentive design, fiscal trade-offs, and equilibrium analysis in freight transport systems. Computational experiments illustrate how alternative incentive allocations can alter equilibrium outcomes and provide insights for policy design under strategic interaction.
The growing adoption of decentralized renewable energy systems highlights the need to assess their resilience to disruptions. Renewable Energy Communities (RECs), composed of consumers and prosumers, rely on rooftop photovoltaic generation and peer-to-peer (P2P) energy sharing to meet local demand. Although previous studies have evaluated RECs from energy and economic perspectives, the influence of P2P network topology on resilience remains largely unexplored. This study proposes a framework that combines energy performance assessment with graph-based network vulnerability analysis to evaluate REC resilience under targeted disruption scenarios. P2P energy exchanges are simulated to construct directed, weighted energy-sharing networks, and graph-theoretic centrality measures are used to identify critical nodes whose sequential removal mimics targeted attacks. Results show that RECs provide substantial environmental and economic benefits, reducing greenhouse gas emissions to 61
In an attempt to mitigate their effect on climate change, a number of economies have already, or are in the process of shifting away from a reliance on coal-based power. This just energy transition means that many existing jobs in the coal industry will be lost in favour of so-called “green jobs”, which aim to contribute positively and sustainably to the environment. South Africa is one such economy that is embarking on the process of a just transition, but given that the coal industry is predominantly represented by young people in the province of Mpumalanga, it is not clear how or if this vulnerable group will transition into newly created green jobs. Making use of occupational relatedness metrics, this research investigates the feasibility of green job opportunities to capture displaced youth in Mpumalanga, depending on their employment history. Results of this desktop study suggest that green jobs are relatively different to the existing experience and task competences of young people, and thus some form of reskilling programme is likely to be necessary for young people to take full advantage of any employment opportunities offered by green jobs.
Project disruptions may occur due to unexpected incidents or intentional acts, leading to schedule extensions. The susceptibility of a project to unforeseen occurrences dictates its vulnerability. The assessment of project vulnerability enables managers to incorporate preventive strategies during the planning phase or, in the recovery phase, prioritize projects in decision-making for implementing appropriate strategies. Nonetheless, prior research has mostly neglected the impact of project topology and network structure on performance and susceptibility to interruptions. Additionally, while machine learning has been used in project management, its integration with project network structural metrics to classify projects based on inherent vulnerability is still limited. This study presents an innovative method using network-based metrics to assess project vulnerability. These measurements function as input characteristics for machine learning algorithms to assess vulnerability during the planning phase. A decision tree is first used to derive splitting criteria. A random forest approach, with hyperparameters adjusted using grid search, is used to enhance prediction accuracy. Empirical project data are used to verify the models, which are then evaluated with three alternative algorithms based on established machine learning performance metrics. The assessment findings indicate that the suggested model attains enhanced prediction performance.
The spatial Durbin model has been widely applied in regional economics, real estate, and environmental policy studies for its ability to capture both direct and indirect effects. However, when the target domain suffers from limited sample size or distributional discrepancies with available source domains, conventional single-domain estimation often encounters sample representativeness limitations and external estimation bias. To address this challenge, this paper introduces the concept of transfer learning into the spatial econometric framework and proposes a transfer learning method for spatial Durbin model, which enables knowledge transfer across multiple domains under spatial dependence. Furthermore, under the scenario where transferable sources are unknown, we develop a transferable source detection mechanism that combines instrumental variable transformation and cross-validation to automatically identify source domains most similar to the target domain. Both simulation and empirical analysis demonstrate that our methods outperform the baseline methods.
Transfer learning has been successfully applied across multiple domains, aiming to uncover intrinsic relationships between data or models and transfer knowledge gained from source domain training to target domains. As a core method for evaluating technical efficiency, stochastic frontier models have consistently attracted significant attention. This paper innovatively combines the two approaches, proposing the Spatial Durbin Stochastic Frontier Model with an integrated transfer learning framework (Trans-SDF-STE). This model retains the Spatial Durbin model’s ability to capture “spatial correlation effects between an entity and its neighboring units” while extending the Stochastic Frontier Analysis’s advantage in quantifying heterogeneity in technical efficiency. Simultaneously, it overcomes modeling limitations in target domains with small samples through transfer learning mechanisms. We employ spatial residual bootstrapping to select source domains exhibiting “spatial structural consistency and efficiency mechanism similarity.” By integrating 2SLS estimation to construct instrumental variables addressing endogeneity, followed by a two-stage transfer learning approach (“transfer + bias removal”), we migrate the source domain to the target domain. This enhances the accuracy of model parameter estimation and technical efficiency measurement. Through simulation experiments and real-world data applications, we validate the effectiveness and practicality of the proposed method.
This paper incorporates digital inclusive finance, agricultural total factor productivity, and rural education-based human capital simultaneously into the analytical framework of rural residents’ relative poverty. Based on the panel data of 31 provincial regions in China from 2011 to 2023 and digital inclusive finance data, the spatial Durbin model (SDM) is adopted to analyze the spatial spillover effect of the digital economy on rural residents’ relative poverty. Meanwhile, the mediation model and the panel threshold model are utilized to explore the path of digital inclusive finance on rural residents’ relative poverty. The research has found that there is a spatial correlation between digital inclusive finance and rural residents’ relative poverty; digital inclusive finance can significantly alleviate the problem of rural residents’ relative poverty, and has a spatial impact on the relative poverty of rural residents in the local area and surrounding areas; agricultural total factor productivity and rural educational human capital not only can alleviate rural residents’ relative poverty themselves, but also agricultural total factor productivity can play an intermediary role, and rural educational human capital can play a threshold effect. Therefore, attention should be paid to promoting the development of digital inclusive finance and fully exerting its regional radiation effect. While enhancing agricultural total factor productivity, emphasis should also be placed on the cultivation of rural educational human capital.
As the economic core where the “Belt and Road Initiative” and the Yangtze River Economic Belt converge, the Chengdu-Chongqing Urban Agglomeration plays a crucial role in building a unified national market. Based on the statistical data of 11 major cities in the Chengdu-Chongqing Urban Agglomeration from 2011 to 2022, this paper explores the impact of market integration on regional economic growth and analyzes the mechanism of action and policy implications. The results show that from 2011 to 2022, the level of market integration in the Chengdu-Chongqing Urban Agglomeration has shown a fluctuating upward trend, and it has a significant positive impact on economic growth. Factors such as human capital, opening up, consumption level, and industrial structure also have a significant promoting effect on economic growth. The effect of market integration on economic growth exhibits a non-linear “inverted U-shaped” characteristic and shows spatial heterogeneity. The market integration effect in developed cities is gradually weakening, while there is still considerable room for improvement in less developed cities.
This paper uses Input-Output data to search for trading communities in the world trade network both in final and intermediate goods and it then uses a structural gravity model to conduct counterfactual analyses to tease out the main drivers behind such communities. The main findings are twofold (i) global trade is divided into communities broadly corresponding to regional (continental) areas which are driven entirely by bilateral characteristics such as geography, trade policy and cultural similarities; (ii) the trade network is significantly less modular than the corresponding random networks and this is driven by individual characteristics such as productivity, comparative advantage or size.
