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.

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.

The early stages of engineering innovation are typically characterized by high levels of uncertainty, strong dependence on expert experience, and complex coupling among design objectives, manufacturing constraints, and solution maturity. These characteristics make the associated decision-making processes difficult to formalize and reproduce. To address this challenge, an industrial intelligence framework integrating Extenics and the Theory of Inventive Problem Solving (TRIZ) was proposed to support structured reasoning and consistent decision-making in the early phase of engineering innovation. Within the proposed framework, engineering objectives, constraint conditions, and solution maturity are represented as structured industrial knowledge elements, enabling unified processes of conflict identification, rule-based reasoning, and multi-criteria evaluation. Extenics is employed to construct formal representations of problem elements and their interrelationships, while TRIZ is utilized to support systematic principle-based resolution of contradictions. Through this integration, engineering decision-making is shifted from reliance on implicit experiential knowledge toward an explicit, knowledge-driven paradigm. The applicability and effectiveness of the framework were demonstrated through a conceptual design case study of a household product. The results indicate that the proposed approach enhances the transparency and consistency of early-stage engineering decisions, reduces dependence on individual expertise, and provides an interpretable industrial intelligence solution for supporting knowledge-intensive engineering innovation.

In the highly competitive landscape of modern manufacturing, the efficient and timely operation of production processes is paramount for sustaining productivity and ensuring customer satisfaction. Delays and latent bottlenecks, however, often hinder optimal performance. A data-driven methodology for identifying these inefficiencies is presented, employing process mining techniques. By analyzing event logs from Manufacturing Execution Systems (MES) and Enterprise Resource Planning (ERP) systems, the actual execution flow of production processes is reconstructed and compared against the designed process models. Through process discovery, conformance checking, and performance analysis, the underlying causes of delays and capacity bottlenecks are pinpointed. A case study from a manufacturing facility is used to demonstrate the effectiveness of process mining in uncovering critical areas for process improvement. The findings indicate that process mining not only enhances transparency but also provides actionable insights for optimizing resource planning, reducing cycle times, and maximizing overall operational effectiveness. The approach is demonstrated to facilitate the identification of inefficiencies, leading to targeted interventions that significantly improve process performance and business outcomes.

Equipment failure in paper mills represents a critical barrier to operational efficiency and the adoption of Industry 4.0 principles. To address this, a systematic literature review was conducted to identify the multifactorial determinants of such failures. A novel hybrid methodology was proposed, integrating the Functional Analysis Systems Technique (FAST), enhanced by Lean 5S (Sort “Seiri”, Set in Order “Seiton”, Shine “Seiso”, Standardize “Seiketsu”, Sustain “Shitsuke”) principles, to structure the qualitative data collection. The analysis was performed using a Pugh matrix, followed by a Principal Component Analysis (PCA) to extract knowledge systematically. This approach facilitated the development of a conceptual model for downtime causation. The PCA results indicate that two principal components collectively explain 58.5% of the observed variance in failure data. The f irst component was strongly correlated with maintenance practices and operational errors, while the second was associated with intrinsic equipment characteristics and their operating conditions. This data-driven modeling elucidates underlying correlations between disparate factors, providing a robust foundation for prioritizing targeted maintenance optimization actions. This research contributes to the field of industrial intelligence by demonstrating an original methodology for transforming qualitative systematic review data into a quantifiable analytical framework. The application of PCA to this corpus enables the identification of multidimensional interactions that are frequently overlooked in conventional analyses, thereby enriching root-cause failure analysis and informing strategic decision making for predictive maintenance. The identified factors underscore the imperative of a balanced integration between technical data and human factors for the successful digital transformation of production systems.

Material extrusion additive manufacturing (MEX-AM) has emerged as a transformative technology with the potential to redefine industrial production; however, persistent challenges remain regarding variability in part quality, the absence of robust in-process defect detection, and limited capacity for process optimization. To address these limitations, an integrated multi-sensor and machine learning (ML) framework was developed to enhance real-time monitoring and defect detection during MEX-AM. Data were acquired from thermocouples, accelerometers, and high-resolution cameras, and subsequently processed through a multi-sensor data fusion pipeline to ensure robustness against noise and variability. A Multi-Criteria Decision Analysis (MCDA) framework was employed to evaluate candidate ML algorithms based on accuracy, computational cost, and interpretability. Random Forest (RF) and Artificial Neural Network (ANN) models were identified as the most suitable approaches for MEX-AM applications. Validation experiments demonstrated a 92% success rate in corrective interventions, with a reduction of defective components by 38% compared with conventional monitoring methods. The integration of sensor fusion with advanced learning models provided improved predictive capability, enhanced process stability, and significant progress toward intelligent, self-optimizing manufacturing systems. The proposed methodology advances statistical quality control and reduces material waste while aligning with the objectives of Industry 4.0 and smart manufacturing. By demonstrating the efficacy of multi-sensor fusion and ML in real-world AM environments, this study highlights a pathway toward scalable, autonomous, and sustainable industrial production.

The proliferation of the Industrial Internet of Things (IIoT) has transformed energy distribution infrastructures through the deployment of smart metering networks, enhancing operational efficiency while concurrently expanding the attack surface for sophisticated cyber threats. In response, a wide range of Machine Learning (ML)–based Intrusion Detection and Prevention Systems (IDPS) have been proposed to safeguard these networks. In this study, a systematic review and comparative analysis were conducted across seven representative implementations targeting the Internet of Things (IoT), IIoT, fog computing, and smart metering contexts. Detection accuracies reported in these studies range from 90.00% to 99.95%, with models spanning clustering algorithms, Support Vector Machine (SVM), and Deep Neural Network (DNN) architectures. It was observed that hybrid Deep Learning (DL) models, particularly those combining the Convolutional Neural Network and the Long Short-Term Memory (CNN-LSTM), achieved the highest detection accuracy (99.95%), whereas unsupervised approaches such as K-means clustering yielded comparatively lower performance (93.33%). Datasets utilized included NSL-KDD, CICIDS2017, and proprietary smart metering traces. Despite notable classification accuracy, critical evaluation metrics—such as False Positive Rate (FPR), inference latency, and computational resource consumption—were frequently underreported or omitted, thereby impeding real-world applicability, especially in edge computing environments with limited resources. To address this deficiency, a unified benchmarking framework was proposed, incorporating precision-recall analysis, latency profiling, and memory usage evaluation. Furthermore, strategic directions for future research were outlined, including the integration of federated learning to preserve data privacy and the development of lightweight hybrid models tailored for edge deployment. This review provides a data-driven foundation for the design of scalable, resource-efficient, and privacy-preserving IDPS solutions within next-generation IIoT smart metering systems.