Container shipping faces a significant imbalance between cargo supply and demand, worsened by uneven cargo distribution, particularly since the COVID-19 pandemic. To address this, we propose a decentralized operational network with split tasks and simultaneous pickup and delivery (SPD) to facilitate collaborative ocean freight consolidation. We further segment liner routes into connectable sections aligned with ships' rolling cycles. A bilevel planning model based on cargo operation sequences is developed. The upper level manages cargo transit planning, while the lower level optimizes container layouts, transport routes, and transfer solutions. A genetic algorithm coordinates both levels to balance strategic and tactical objectives. Experimental results confirm the model's effectiveness in identifying key cargo batches for consolidation and optimizing liner frequencies and routes configurations. The analysis reveals clustered route usage and aggregated transfer nodes, offering actionable insights for route selection. By evaluating cargo batch volumes, the framework enables data-driven service adjustments and planning for direct consolidated cargo shipments. Overall, coordinated bilevel optimization significantly enhances efficiency in imbalanced shipping networks.
In the context of global industrial green transformation, the raw material supply chain of building decoration panel enterprises faces key issues including supply deviations, transportation losses, and inventory backlog. This study establishes an end-to-end optimisation model covering supplier screening, order planning, transportation optimisation, and capacity coordination. The model innovatively integrates raw material heterogeneity with uncertainties in supply and transportation, and combines a multidimensional evaluation system with intelligent algorithms including autoregressive neural networks and particle swarm optimisation to achieve dynamic optimisation. Empirical validation is carried out using 240 weeks of order and supply data from 402 suppliers, together with loss rate data from eight transporters. The results show a 31.28% increase in enterprise production capacity. This study establishes a transferable methodology for resource-intensive manufacturing enterprises, enhances their ability to resist supply chain uncertainties, and achieves cost reduction and efficiency improvement. This approach reduces per-unit product raw material consumption and transportation losses, improves resource utilization efficiency, promotes green and low-carbon development while maintaining stable production and logistics scales, and provides quantifiable support for alignment with SDG 9, 12, and 13.
Overloaded heavy vehicles impose substantial costs on urban infrastructure and challenge effective monitoring. From a systems science perspective, overloaded truck monitoring constitutes a complex adaptive logistics system in which sensing infrastructure, enforcement decisions, and routing behavior interact through feedback and learning. This study proposes a tri-level framework integrating strategic deployment of fixed and mobile weighing sensors with the dynamic routing behavior of logistics service providers (LSPs) who cooperatively share real-time information to evade detection. Initially unaware of mobile patrol positions, LSPs gradually learn and adapt through observed enforcement outcomes. The interactions are modeled as a cooperative congestion game, where municipalities sequentially decide on sensor placement and patrols, while LSPs respond strategically. A nested algorithm jointly optimizes sensor deployment and feedback-based routing, capturing realistic evasive and cooperative behavior. Numerical simulations demonstrate that adaptive routing, combined with information sharing, substantially enhances detection rates and infrastructure protection compared to models that neglect driver cooperation. The findings provide actionable insights for municipal authorities and logistics operators, supporting data-driven enforcement and strategic sensor placement to balance operational efficiency with sustainable infrastructure management.
In the context of accelerating digital transformation and increasingly complex global supply chains, this study develops a multi-layered analytical framework to investigate the technological evolution and future trajectory of digital logistics. Using 107,202 patent records from the IncoPat database covering 1960-2023, the study integrates the entropy weight method, grey relational analysis, Latent Dirichlet Allocation (LDA), and a Hidden Markov Model (HMM) to identify core patents, extract technological topics, and forecast their dynamic evolution through 2028. The findings identify 27 major technological topics and reveal a significant transition from traditional operational optimization toward integrated intelligence and autonomous execution. Technologies related to optimization algorithms, intelligent logistics systems, automated handling, data networks, transport automation, and terminal communication emerge as the dominant drivers of future development. The results further indicate that digital logistics technologies are evolving from isolated functional applications toward deeply integrated intelligent systems characterized by automation, connectivity, and data-driven decision-making. Methodologically, the study contributes a dynamic and interpretable framework that combines topic modelling with probabilistic forecasting, extending existing patent-based technology evolution research beyond static analysis. Practically, the findings provide evidence-based guidance for technology investment, industrial upgrading, and policy formulation in the digital logistics sector.
This study examines the application of extreme value theory (EVT) to the newsvendor problem under censored demand conditions. It investigates whether classical economic order quantity (EOQ) logic can approximate stochastic inventory decisions and identifies when such deterministic reasoning remains reliable. The objective is to minimize expected total cost (holding and stockout costs) by utilizing EVT to more accurately estimate the survival function of the demand tail, thereby refining the optimal order-up-to level (Y* ) in the presence of censored data. Traditional inventory models often fail to capture extreme demand fluctuations, particularly during stockouts. EVT, including the generalized Pareto distribution (GPD) and Fisher-Tippett-Gnedenko convergence, enhances tail-end demand estimation. Results show KM is generally more robust in balancing costs, while a split algorithm combining GPD for backorders and a parametric distribution (e.g. Poisson) for holding costs may outperform KM. EVT's predictive potential for optimal reorder quantities with specific asymptotic distributions (e.g. Weibull, Gumbel and Fr & eacute;chet) offers stability for distinct inventory scenarios. The study offers decision criteria for method selection, positioning EOQ-style logic as a useful approximation under moderate service levels and stable demand and EVT models as more appropriate in high-reliability settings where tail risk and extreme demand realizations dominate.
This paper presents a real-world-inspired case study addressing a picking routing problem in a clothing warehouse with a non-conventional layout. In the company under study, picker performance is evaluated based on the number of batches completed within a working period, which constitutes a key performance indicator (KPI). Accordingly, the objective of this research is to minimize the total completion time-defined as the sum of the completion times of all pickers-since reducing this metric enables more orders to be processed within the same time horizon, thereby improving overall worker efficiency. Completion times are computed assuming a constant travel velocity and a fixed picking rate per unit of product. An A*-based approach is proposed to efficiently compute travel distances, and a mathematical formulation, three constructive heuristics, and an Iterated Greedy (IG) metaheuristic are developed and compared with the warehouse's current system logic. The results are evaluated through a comprehensive statistical analysis, showing that the nearest-bin heuristic and the IG algorithm initialized with it significantly enhance order-picking efficiency. The study provides practical insights for warehouses with non-conventional layouts and contributes to more efficient intralogistics operations.
To minimize the impact of human activities on the greenhouse effect, many countries are promoting renewable energy to reduce carbon emissions, with solar power becoming a key solution because of its clean and low-impact characteristics. Taiwan's subtropical climate with abundant sunshine makes it suitable for renewable energy generation systems. This paper studies the optimal configuration of a renewable energy power system in Taiwan under different equipment settings and evaluates not only its economic performance but also its social value within an integrated energy-community system. Using a micro-grid project in Mudan Township and power consumption data from remote villages, we develop a mixed-integer optimization model to determine the optimal combination of photovoltaic generation, energy storage, and diesel backup under budget, space, and reliability constraints, while also incorporating government subsidy mechanisms. From a systems science perspective, the model captures the interdependence among infrastructure investment, policy incentives, operational reliability, and community energy security. Beyond minimizing construction costs, the framework assesses social outcomes, including disaster resilience, reduced outage risk during extreme weather events, and improved electricity access in isolated areas. Compared with the existing plan, the proposed model achieves a 21.03% cost savings while enhancing system reliability and supporting low-carbon regional development.
Accounting and management of energy consumption and carbon emissions in manufacturing are essential for enterprises aiming to establish low-carbon supply chains and reduce the costs imposed by carbon taxation. However, complex product manufacturing systems, with multi-component structures, long supply chains and dynamic process variability, limit the precision of Life Cycle Assessment-driven energy and carbon management (ECM) frameworks. In this paper, the Supply Chain Manufacturing System (SCMS) is first introduced, and a holistic methodology is proposed to support system-wide ECM for complex product manufacturing. First, the concept of SCMS is formalized and its energy-carbon characteristics are analyzed. Second, the proportion of energy consumption and the fluctuation of carbon emissions are defined to classify supply chain enterprises with respect to long-term costs and short-term fluctuation risk. Based on the classification, the differentiated management paradigms of SCMS are designed, including data-coupled, period-coupled and full-process-coupled management. Finally, the methodology is verified on a typical automobile SCMS with three representative suppliers. The results show per-unit carbon emission reductions of 13.29, 7.27, and 0.79 kg CO2 across the three suppliers, indicating that the methodology achieves effective ECM for complex product manufacturing at relatively low cost.
Public transport operators relying on diesel fleets face increasing pressure to adopt environmentally sustainable solutions. Integrating electric vehicles into existing systems represents a promising but complex transition, requiring system-level planning under multiple, policy-oriented scenarios. This study adopts a systems science perspective to address this challenge, recognising the interdependencies among fleet composition, battery technologies, charging infrastructure and policy constraints. To support integrated decision-making, it introduces a mixed-integer linear optimisation model-the MultOptMixedFleet-that determines the optimal mix of electric and non-electric buses, selects suitable battery capacities and identifies strategic charging locations and types. The model formulates fleet electrification as a multi-objective optimisation problem, capturing trade-offs between cost and environmental performance. Applied to a Portuguese public transport provider, the model demonstrates its practical value as a decision-support tool for managers and policymakers seeking cost-effective and low-emission transition strategies. Results show that progressive electrification delivers major emission reductions: fully replacing diesel buses lowers emission-related costs by approximately 67% compared with fully non-electric fleets. Although electrification increases total system costs, longer battery lifecycles mitigate this effect by reducing annual operating and investment costs by 6% and 33%, respectively. Overall, the study highlights the importance of systems-oriented modelling for sustainable operations and logistics.
Digital transformation is a key driver of competitiveness in road freight transport, yet empirical evidence for medium-sized carriers in Central and Eastern Europe remains limited. This article examines how a Slovenian road freight SME can move from partial digitalisation to a structured transformation programme. It applies a seven-dimensional digital maturity framework in an in-depth case study of Transport Company X, using internal documentation, operational and financial data, and workshops with managers and employees. The assessment reveals low to intermediate maturity: basic digital tools are in place, but systems are fragmented, processes remain manual, governance is weak, and digital skills are uneven. A three-phase roadmap is proposed, progressing from process standardisation and basic KPI reporting to ERP-centred integration, and ultimately, customer-facing digital services and analytics. Scenario-based estimates indicate meaningful improvements in efficiency, costs, service quality and incremental sustainability performance, offering practical guidance for road freight SMEs and policymakers.
When implementing strategies and solutions for carbon emission reduction, companies often face the challenge of precisely determining where and how much in a supply chain to reduce emissions in an efficient way. This paper introduces a novel optimisation model designed for the configuration of an end-to-end supply chain by simultaneously optimising mode choice and safety stock placement decisions, such that the tradeoffs between cost, lead time and carbon emission reduction are addressed and exploited. Different from existing studies, we consider the differences in carbon emissions between the production process and the transportation process in the supply chain, and propose customised calculation methods for carbon emissions in these two processes. A case study of the PC supply chain is conducted to demonstrate the capability of our model and solution. Sensitivity analysis is performed to show the impacts of pivotal factors including carbon emission cap, carbon trading price and customer delivery time, concerning the optimal green supply chain configuration and total supply chain costs performance metrics. The research results show that our solutions can assist supply chain decision-makers in balancing costs and customer commitment time while fully considering carbon emissions, thereby achieving the coordination and unification of economic benefits and environmental friendliness.
This paper analyzes the joint coordination of green investment and inventory decisions in a two-echelon supply chain operating under demand uncertainty. The supply chain consists of a manufacturer with imperfect production and a retailer exposed to stockout risk. The manufacturer invests in Industry 4.0 technologies to enhance product greenness, which drives market demand and generates cost spillovers for the retailer. Using an EOQ-based framework, we analyze centralized and decentralized decision-making under deterministic demand and then extend the model to a stochastic setting by incorporating probabilistic lead-time demand. The results show that decentralized decision-making leads to systematic underinvestment in green technologies and lower supply chain profitability compared to the centralized benchmark. These inefficiencies are amplified under demand uncertainty. To address this coordination failure, the paper proposes a cost-sharing contract that align incentives across supply chain members. Numerical analysis shows that the proposed contract increases green investment and improves service performance. Under varying backorder rates, total supply chain profit increases by an average of 10.72%, achieving Pareto efficiency. This study contributes to the EOQ and stochastic inventory literature by integrating green investment decisions, imperfect production, and digital spillover effects into a unified coordination framework, offering actionable insights for sustainable supply chain management in the context of Industry 4.0.
The increasing demand for parcel delivery in Vietnam requires an efficient transportation strategy to minimize operational costs and travel distances while adhering to postal regulations. Vietnam Post operates a three-echelon network following a structured five-phase parcel transportation process: collection at branches, consolidation at hubs, inter-center transfer, distribution back to hubs, and delivery to branches. Current vehicle routing methods often fall short in addressing this specific structure, resulting in significant operational inefficiencies. This paper proposes a routing strategy, which is implemented using equal-size spectral clustering for balanced workload distribution and dynamic programming with bitmasking for intra-cluster route optimization. Route segmentation and vehicle assignment are incorporated to ensure route feasibility. They ensure compliance with key postal operational constraints, including simultaneous parcel pickup and delivery, capacity and route length limits, service time windows, driver regulations, and cross-phase consistency. The strategy is implemented on real-world and large-scale synthetic datasets, encompassing thousands of post offices and up to one million parcels. Experimental results demonstrate significant reductions in travel distance, operational costs, and vehicle use, along with improved load efficiency. Comparisons with a metaheuristic baseline confirm the scalability and practical effectiveness of the strategy in solving large-scale postal logistics challenges.
This study develops a systematic greenhouse-gas (GHG) inventory and provides a preliminary estimation of carbon-credit potential for a logistics enterprise operating within the cold-chain sector. The inventory methodology is based on the IPCC Guidelines and complies with the emission accounting procedures mandated by Circular No. 38/2023/TT-BCT of Vietnam. Primary data on electricity consumption, fuel use, refrigerants, and fire suppression systems were collected from August 2024 to April 2025 and analyzed to quantify the enterprise's CO2-equivalent (CO(2)eq) emissions. The results demonstrated that electricity consumption constitutes the dominant emission source, with 4,093 tCO(2)eq and 3,489 tCO(2)eq at Lotte Logistic and AJ Cold Storage, respectively. This emission source occupied about more than 60% of total emissions, highlighting critical vulnerabilities in energy-intensive cold storage operations. Building on these findings, the study evaluates emission reduction scenarios that focus on improving energy efficiency, promoting renewable energy, and enhancing forest management, and subsequently estimates the associated carbon-credit potential using the Gold Standard methodology. The results underscore the enterprise's capacity to participate in the voluntary carbon market while contributing to national decarbonization objectives. From a systems science and operations perspective, the study provides quantitative evidence to inform logistics enterprises in designing low-carbon strategies, integrating carbon markets into operational planning, and strengthening the resilience of cold-chain systems in transition toward net-zero pathways.
This paper examines an intermodal transport problem characterised by key operational features, including scheduled maritime and rail transport services at seaport and rail terminals, flexible road transport, and temporary storage options at depots near these terminals to reduce high holding costs prior to scheduled departures. To address this challenge, we introduce an operational freight routing problem within a time-sensitive road-rail-maritime-depot intermodal transport network. We develop a novel mixed-integer linear programming (MILP) model to solve this problem and demonstrate its application through a case study on intermodal route planning for a customer order from Denizli, T & uuml;rkiye, to Duisburg, Germany. The model generates Pareto-optimal solutions that balance time and cost objectives by selecting optimal schedules for liner vessels and trains while integrating road transport and warehousing alternatives. A comprehensive sensitivity analysis is also conducted to assess the robustness of the MILP model solutions with respect to variations in the cost and time parameters of the case problem. The proposed model improves efficiency and effectiveness in evaluating operational intermodal routing plans and offers a structured operational framework for enhancing intermodal transport operations.
In response to the increasing demand for door-to-door services and the complexities of modern last-mile delivery models, this study investigates a two-echelon vehicle routing problem for last-mile delivery based on a multi-fleet framework. This framework includes electric vehicles (EVs) to transport robots and cargo, delivery robots (DRs) for terminal deliveries, and vans to recover robots, which is also referred to as an EV-DR-van system. We propose a mathematical model to address the two-echelon electric vehicle routing problem, considering time window constraints, the utilization of delivery robots, and a partial charging strategy, named 2E-EVRPTWDR-PC. We employ the CPLEX solver to obtain optimal solutions for small-scale instances. For larger instances, we design an adaptive large-neighborhood search algorithm (ALNS) to effectively explore and adapt to the complex solution space, producing high-quality results. The data for this research are generated by modifying Solomon's dataset. Our experimental results confirm the proposed model's validity and the algorithm's effectiveness. Comparative analyses indicate that using robots for final deliveries can reduce overall time costs by approximately 20% compared to traditional delivery methods. This research has significant implications for optimizing last-mile delivery systems and improving delivery efficiency.
The timely and reliable delivery of medical supplies is vital in emergencies, where even minor delays can endanger lives. This study proposes a real-time optimisation framework for healthcare logistics that incorporates an urgency-weighted pheromone decay mechanism into an Ant Colony Optimization (ACO) algorithm. The method dynamically adjusts routing priorities according to medical urgency, enabling adaptive path selection under fluctuating traffic, battery, and weather conditions. A hybrid drone-truck collaboration system is developed, combining the endurance of trucks for long-range transport with the agility of drones for last-mile, time-critical deliveries. Simulation experiments using operational data from 40 healthcare facilities in Delhi demonstrate an approximately 30% reduction in average delivery time, a 15% decrease in fuel consumption, and an approximately 30% improvement in overall energy efficiency compared to conventional static routing models. These findings indicate that the proposed framework offers a scalable, energy-aware, and urgency-responsive solution for dynamic healthcare supply chains, with potential applicability to other critical logistics domains.
