Human-robot interaction has become a fundamental component of Industry 5.0. Although recent advances in generative artificial intelligence have substantially improved contextual reasoning and interactive capabilities, existing human-robot interaction decision-making frameworks remain predominantly task-oriented and rarely account for human emotional responses during collaborative decision processes. To address this limitation, a regret-based human-robot interaction framework driven by generative artificial intelligence was proposed for adaptive decision-making in Industry 5.0 environments. A dual-regret mechanism was introduced by jointly modeling human regret $\left(R_h\right)$ and robot task regret $\left(R_{t r}\right)$. Contextual knowledge was dynamically retrieved through a retrievalaugmented generation architecture, enabling interaction strategies to be generated according to task conditions, historical user preferences, and evolving human emotional states. Personalized regret profiles were constructed, while explainable counterfactual action generation was incorporated. Furthermore, real-time reinforcement learning was employed to continuously minimize cumulative regret. Experimental evaluations conducted under multiple interaction scenarios demonstrated that the proposed framework effectively improved decision quality, enhanced user satisfaction, and increased task execution efficiency. Compared with conventional human-robot interaction decision-making approaches, cumulative interaction regret was reduced by approximately 20%, while greater robustness and adaptability were achieved under dynamically changing operational conditions. These findings suggest that the proposed framework provides an effective paradigm for integrating emotional intelligence, generative reasoning, and adaptive optimization into human-centered robotic systems, thereby offering a promising decision-making architecture for next-generation collaborative robotics in Industry 5.0.
Prediction of product life cycle (PLC) at the early stage is important to support the planning of production, inventory control, launch timing, and decisions of product renewal. However, newly introduced batik products often have little historical sales data; therefore, it is difficult to apply conventional forecasting methods. This study introduced the Diffusion Bubble Model for Product Diffusion (DBM-PD) as a low-data diffusion model for predicting cumulative product adoption in batik fashion small and medium enterprises (SMEs). The DBM-PD adapts the bubble-decomposition logic into a product diffusion context, where each adoption bubble represents a possible market response at a specific stage of the PLC. To assess the model performance, this study compared the DBM-PD with three benchmark models: the Bass Diffusion Model, Triangle Model, and Circle Model. Four datasets of PLC from batik fashion products were used as empirical cases. Each dataset was transformed into a cumulative adoption ratio to ensure comparable model evaluation. The results showed that the DBM-PD achieved the lowest prediction error across all four datasets. The Bass Diffusion Model ranked second, indicating that innovation and imitation effects remain relevant in batik product adoption. The Triangle and Circle Models were useful as simple geometric benchmarks, but their assumptions were less able to capture irregular adoption patterns. These findings revealed that the DBM-PD could establish a more flexible prediction structure for estimating PLC at the early stage when only limited data were available. A parametersensitivity check discovered that the number and width of bubbles affected in-sample accuracy: additional bubbles could reduce errors but also increased model complexity, while the seven-bubble, $\sigma=0.08$ configuration offered a transparent balance between accuracy and parsimony. The root mean square error (RMSE) rankings were identical across the four datasets, and a Friedman test indicated a significant overall difference among the models $\left(\chi^2(3)=12.00, p=0.007\right)$. This study contributes to the management of technology-based PLC by offering a practical framework of diffusion modelling for batik fashion SMEs.
With the in-depth implementation of the “Industry 4.0” and “Dual Carbon” strategies, the manufacturing of reducer boxes is accelerating its transformation towards intelligence and greenization. To address the frequent dynamic disturbances such as machine failures and urgent order insertions in actual production, as well as the difficulty for traditional scheduling methods to balance production efficiency and green energy saving, a green dynamic scheduling optimization method for flexible job shop driven by digital twin is proposed. First, a digital twin dynamic scheduling framework comprising a physical workshop, a virtual workshop, and a service system is constructed, and a high-fidelity simulation model of the reducer box flexible production line is built based on AnyLogic. Second, a multi-objective dynamic scheduling mathematical model is established by comprehensively considering the makespan, energy consumption, and rescheduling machine deviation. An Improved Multi-Objective Artificial Bee Colony (IMOABC) algorithm is designed to solve the problem, which enhances the global exploration and local exploitation capabilities by fusing Improved Precedence Operation Crossover (IPOX), uniform crossover strategies, and a variable step-size neighborhood search mechanism. Finally, multi-dimensional comparative validation is conducted based on standard benchmark instances and a reducer box manufacturing case. The results demonstrate that the proposed method can effectively cope with dynamic disturbances and outperforms traditional scheduling strategies in shortening production cycles, reducing equipment energy consumption, and maintaining system stability.
Traditional public health disinfection tasks relying on fixed-area coverage often suffer from resource waste, delayed intervention, and low response efficiency. This study proposes a case-density-driven closed-loop intelligent strategy for air-ground-human collaborative disinfection, establishing an end-to-end framework from case perception to task scheduling. Firstly, a spatiotemporal risk field is constructed based on reported case data and population mobility information, and high-risk areas are adaptively identified and prioritized through dynamic evaluation. Secondly, for coordinated execution by unmanned aerial vehicles (UAVs), ground vehicles, and personnel, a multi-objective coupled optimization model is designed, targeting coverage efficiency, suppression timeliness, path conflicts, and resource cost to generate executable collaborative schedules. Furthermore, a closed-loop execution mechanism is developed, enabling real-time rolling re-planning and adaptive strategy correction in response to task feedback, unexpected disturbances (area lockdown, equipment failure, chemical shortage), and risk field updates. Experimental results demonstrate that the proposed closed-loop approach significantly improves coverage, suppression time, and resource utilization compared with traditional static scheduling and single-entity planning methods across multiple scenarios, and exhibits robustness against environmental uncertainties and resource disturbances. This framework provides a feasible theoretical and methodological foundation for intelligent, precise, and resilient public health disinfection operations.
Unmanned aerial vehicles (UAVs) have gained increasing importance due to their expanding application areas and operational flexibility. Selecting the most suitable UAV, however, represents a complex multi-criteria decision-making (MCDM) problem that involves numerous technical and performance-related factors. This study addresses the UAV selection problem by employing four distinct MCDM approaches: Evidential fuzzy MCDM based on Belief Entropy, Intuitionistic Fuzzy Dempster-Shafer Theory (DST), Spherical Fuzzy Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), and Type-2 Neutrosophic Fuzzy CRITIC-MABAC. Each method incorporates different fuzzy set theories, while a common seven-point linguistic scale is utilized to ensure consistency across models. The evaluation criteria were determined through a comprehensive literature review, and expert opinions were collected from experienced UAV pilots and technical personnel. The analysis identified the most suitable UAV alternative among the considered options. Sensitivity analyses were conducted to assess the robustness of the obtained results. The findings demonstrate that the proposed framework enables a simultaneous comparison of different fuzzy set environments on a unified linguistic scale. Overall, the results are consistent, reliable, and practically applicable, offering valuable insights and methodological contributions to the field of UAV selection and fuzzy MCDM applications.
Efficient inventory control in finite-capacity warehouses requires a quantitative understanding of the stochastic interactions between supply, demand, and storage limitations. An entropy-based stochastic decision-support framework was developed to characterize inventory dynamics and evaluate operational uncertainty under finite-capacity constraints. The warehouse was modeled as a finite-capacity birth-death process in which inventory replenishment and order fulfillment were governed by stochastic supply and demand rates, respectively, while the supply-to-demand ratio is defined as the warehouse utilization factor. The steady-state probability distribution of inventory levels was derived, and Shannon entropy was employed to quantify the uncertainty associated with warehouse occupancy. It was demonstrated that entropy reached its maximum when the warehouse utilization factor equaled unity, indicating that all inventory states become equally probable and that the inventory system operates under the greatest level of stochastic uncertainty. This critical operating point also represents the transition between demand-driven and supply-driven regimes. When the utilization factor was less than one, warehouse performance was primarily constrained by stockout risk, whereas values exceeding one led to increasing blocking probability caused by storage saturation. The trade-off between blocking probability and stockout probability was further investigated, revealing that their combined probability is minimized at the critical utilization condition. Consequently, the complementary probability of the warehouse operating in standard inventory states was maximized. In addition, the stochastic race condition between stockout and blocking events was analyzed to determine which operational constraint is expected to occur first under varying utilization levels. The proposed framework provides both theoretical insight and practical decision support for inventory planning, warehouse capacity management, and procurement policy optimization by establishing a direct relationship between stochastic inventory behavior, information entropy, and operational performance in finite-capacity warehouse systems.
Sustainability requirements and industrial digitalization are reshaping warehouse design and operation, yet the choice between incremental efficiency measures and more advanced intelligent systems remains difficult. This study investigates how conventional, energy-efficient, smart automated, and net-zero warehouses differ in their requirements and transformation priorities. A comparative evaluation framework was developed from the literature and organized around four dimensions: economic feasibility, environmental performance, operational performance, and intelligent technological readiness. The four alternatives were qualitatively assessed in terms of investment, operating costs, energy efficiency, resource use, automation, digital integration, real-time data availability, and implementation complexity. The roles of automated storage and retrieval systems (AS/RS), autonomous mobile robots (AMRs), the Internet of Things (IoT), warehouse management systems (WMS), energy monitoring, and renewable energy technologies were also examined. An illustrative calculation was conducted under identical throughput conditions to compare energy intensity across the four alternatives. The comparison showed that the alternatives did not form a universal progression from an inferior system to a superior one. Energy-efficient warehouses offered a comparatively accessible route to lower energy use, whereas smart automated warehouses provided stronger digital integration and operational coordination but required greater technological and organizational capacity. Net-zero warehouses achieved the lowest illustrative energy intensity, although they involved the highest investment and infrastructure demands. The findings indicate that intelligent and sustainable warehouse transformation depends on the alignment of technology, operational needs, financial capacity, and environmental objectives. The proposed evaluation perspective provides a structured basis for assessing warehouse transformation pathways under different industrial and organizational conditions.
The vibrations are part of people’s everyday lives. In mechanical engineering, vibrations are considered an undesirable phenomenon that can have a detrimental effect on machine structures. Vibrations cause fatigue and wear. They are often responsible for the failure that can occur in a machine. Experimental research was conducted, familiarization with measuring equipment and a model of a rotating machine, and identification of characteristic frequencies of rolling bearings in the vibration signal. Timely identification of localized damage, responsible for over 90% of all bearing defects, is crucial to prevent catastrophic downtime. This paper presents an experimental and analytical vibration study conducted on a laboratory model featuring a 6003 2ZR bearing. The presented methodology is simple, practical, and suitable for implementation in industrial condition monitoring systems, providing a reliable basis for preventive and predictive maintenance of rotating machinery.
This paper addressed the computational challenges of solving large-scale fuzzy 0–1 Integer Linear Programming (ILP) problems by proposing a Fuzzy-Hybrid Quantum-Classical Optimization Algorithm (F-HQCOA). The novel approach employed $\alpha$-cut based transformations to convert fuzzy ILP models into a family of crisp Quadratic Unconstrained Binary Optimization (QUBO) problems, which were then solved using hybrid quantum-classical techniques. Unlike multi-objective formulations, the proposed framework performed a parametric analysis over $\alpha$-levels to generate a spectrum of solutions that reflected trade-offs between objective optimality and feasibility satisfaction under uncertainty. The degree of satisfaction ($\mu$) was adopted as an evaluation metric to assess solution quality rather than as an independent optimization objective. The method was evaluated on fuzzy Multi-Dimensional Knapsack and Set Covering problems. Results showed that the proposed approach could efficiently handle larger instances and provided a diverse set of solutions across $\alpha$-levels. While quantum-assisted methods demonstrated reduced Time-to-Solution (TTS) in certain settings, comparisons were presented with careful consideration of differences in computational paradigms. The proposed framework offers a practical decision-support tool for optimization under uncertainty and establishes a foundation for future extensions toward true multi-objective fuzzy quantum optimization.
Intelligent warehousing has become a key component of Industry 4.0-driven logistics systems, where the coordination of autonomous robots directly affects operational efficiency and system responsiveness. This study addresses the joint optimization of task allocation and path planning for warehouse robots in e-commerce fulfillment environments. A grid-based model is first established to represent the warehouse space, and the scheduling objective is formulated to minimize total travel distance while maintaining balanced workload distribution. An improved genetic algorithm is developed for task allocation, incorporating a multi-layer encoding scheme to represent complex task relationships, along with a simulated annealing mechanism to improve solution quality and prevent premature convergence. For path planning, an enhanced A* algorithm is proposed by introducing a turning cost term into the evaluation function, which effectively reduces unnecessary directional changes and improves path smoothness. Simulation results show that the proposed method significantly outperforms conventional approaches, achieving faster convergence and notable reductions in both travel distance and turning frequency. Specifically, the convergence speed is improved by over 70%, while the total travel distance and the number of turning maneuvers are reduced by approximately 48% and 78%, respectively. The proposed framework enables coordinated decision-making for multi-robot systems and provides a scalable and practically applicable solution for intelligent scheduling in smart logistics environments.
Steel surface defect detection is a critical task in intelligent manufacturing, where high accuracy and real-time performance are required for reliable quality inspection. However, existing deep learning-based approaches often rely on complex architectures, leading to increased computational burden and limited applicability in industrial environments with constrained resources. To address these challenges, a lightweight detection framework is developed to improve feature representation while maintaining computational efficiency. The proposed method integrates adaptive sampling with attention-guided feature refinement to enhance multi-scale feature extraction and contextual representation. In addition, an improved regression strategy is introduced to achieve more stable localization for irregular and low-contrast defects. The network structure is further optimized through lightweight design to reduce redundant parameters and support efficient inference. Experimental results on the Northeastern University surface defect detection (NEU-DET) dataset demonstrate that the proposed approach achieves improved detection accuracy with reduced model size and computational cost compared with baseline models. The results indicate that the method provides a practical solution for real-time industrial inspection, offering a balance between accuracy and efficiency in steel surface defect detection.
For businesses, the effective management of cold supply chains is critical to minimizing food losses and ensuring customer satisfaction. Identifying and prioritizing the obstacles that disrupt these processes is therefore a strategic necessity. However, existing literature largely addresses cold supply chain challenges in a fragmented manner, lacking systematic prioritization frameworks that account for the inherent uncertainty and subjective judgments present in real-world operations. To address this deficiency, this study proposes a structured decision framework based on the q-Rung Orthopair Fuzzy (q-ROF) Subjective Weighting Approach. This method effectively captures uncertainty and integrates expert evaluations to determine the relative importance of key cold chain barriers. Through an empirical application involving logistics managers, the framework ranks the identified obstacles to support operational and strategic decision-making. The findings reveal that Time Constraint is the most critical obstacle, directly impacting operational efficiency and customer satisfaction. In contrast, Temperature-Controlled Vehicle Cost is identified as a lower-priority factor in strategic resource allocation. These results offer a clear prioritization scheme that enables managers to focus resources on the most impactful areas, enhancing resilience and efficiency in cold chain operations. This study contributes a robust, uncertainty-aware methodology for barrier prioritization, providing actionable insights for supply chain practitioners and establishing a foundation for future research in cold chain management.
Reliable lane perception is a core enabling function in industrial intelligent driving systems, providing essential structural constraints for downstream tasks such as lane keeping assistance, trajectory planning, and vehicle control. In real-world deployments, lane detection remains challenging due to complex road geometries, illumination variations, occlusions, and the limited computational resources of on-board platforms. This study presents Attention-Guided Cross-Layer Refinement Network (AG-CLRNet), a real-time lane perception framework designed for industrial intelligent driving applications. Built upon an anchor-based detection paradigm, the framework integrates adaptive multi-scale contextual fusion, channel–spatial attention refinement, and long-range dependency modeling to improve feature discrimination and structural continuity while maintaining computational efficiency. The proposed design strengthens the representation of distant and slender lane markings, suppresses background interference caused by shadows and pavement textures, and enhances global geometric consistency in curved and fragmented scenarios. Extensive experiments conducted on the CULane benchmark demonstrate that AG-CLRNet achieves consistent improvements in precision, recall, and F1 score over representative state-of-the-art methods, while sustaining real-time inference performance suitable for practical deployment. Ablation studies further confirm the complementary contributions of the proposed modules to robustness and structural stability under challenging conditions. Overall, AG-CLRNet provides a practical and deployable lane perception solution for industrial intelligent driving systems, offering a balanced trade-off between accuracy, robustness, and real-time performance in complex road environments.
In critical supply chains like pharmaceuticals, balancing operational cost with service resilience is paramount. While complex adaptive models dominate academic literature on inventory routing, the potential of simpler, managerially intuitive heuristics remains underexplored, creating a gap between theory and practice. This study investigates whether a rigorously optimized, simple time-based heuristic can achieve superior performance and robustness compared to a state-of-the-art, multi-parameter adaptive policy within a stochastic Vendor-Managed Inventory (VMI) system. We formalize a time-to-stockout rule into a novel, single-parameter metaheuristic called the Optimized Urgency Threshold (OUT) policy. Using a simulation-optimization framework powered by a Genetic Algorithm, we benchmarked the OUT policy against a non-optimized heuristic and a complex Dynamic Inertial policy across five problem instances subjected to environmental shocks. The OUT policy demonstrated superior performance, achieving the lowest average total cost (€ 58,595.46) and reducing stockouts by 66.3% compared to the Dynamic Inertial model. Sensitivity analysis confirmed the OUT policy's balanced robustness to demand and capacity shocks, whereas the complex policy exhibited service failures under demand surges. Our findings show that a parsimonious, optimized heuristic can outperform a complex adaptive model, challenging the assumption that parametric complexity is necessary for high performance in stochastic IRPs. The OUT policy provides a transparent, effective, and easily implementable solution for enhancing supply chain resilience and mitigating stockouts.
Submit to JII
Review for JII
Propose a Special Issue 
Open Access
News 
Journal Metrics
Journal Overview
Associated Journals