Search Results (310)

In Industry 4.0 manufacturing, predictive maintenance has emerged as a key focus to ensure the reliability of operations, minimize downtime, and contribute to equipment safety. This study proposes a predictive maintenance framework for additive manufacturing systems (AMSs) using an explainable artificial intelligence (XAI) and machine learning (ML) approach, which is supported by real-time multi-sensor monitoring and is synchronized with a digital twin architecture. A heterogeneous dataset of over 10,000 observations and 13 sensor attributes was gathered from a sensing architecture with ESP32 cameras designed to handle heterogeneous signals from thermal, environmental, mechanical and safety-related sensors. Domain-aware feature engineering was done to gain insights into operation indicators such as temperature instability, vibration degradation, smoke risk, humidity anomalies and aggregated maintenance risk scores. Multiple predictive models, such as the Random Forest model, XGBoost model, Logistic Regression model, K-Nearest Neighbours classifier, Multi-Layer Perceptron model, and ARIMA forecast model, were comparatively assessed under highly imbalanced maintenance conditions. The results showed that ensemble learning methods, especially XGBoost and MLP, had better recall and ROC-AUC for fault detection of maintenance-critical problems. SHAP and LIME analyses then showed that a number of physically meaningful indicators, including thermal instability, vibration anomalies, smoke-related features, and aggregated risk scores, have a significant impact on maintenance predictions and hence provide better operational interpretability and maintenance transparency. Full article

This study investigated the integration of 3D-printed tensegrity launch tubes and compliant mechanism components into small-scale industrial line launcher systems. A multi-variable experimental design (n = 108 replicate shots) evaluated a seven-segment PLA tensegrity tube, the potential for a monolithic compliant launcher, and a pneumatic benchmark across various projectile types, tube configurations, and muzzle rifling geometries. The system exhibited a severe 57.4% launch failure rate, with failures concentrated in extended tube configurations and Rigid or Compliant muzzle attachments. To isolate the dominant operational drivers across the dataset, a composite velocity score was analyzed. Non-parametric variance testing identified tube configuration as the primary factor influencing velocity (Kruskal–Wallis H = 26.73, p < 0.001), followed by muzzle geometry (H = 10.20, p = 0.017). Post-test disassembly identified three distinct failure modes, primarily driven by bore clearance rather than the vibrational compliance of the tensegrity architecture. A Failure Mode and Effects Analysis (FMEA) quantified these risks, identifying tensegrity tube bore constriction as the primary threat to system reliability (Criticality = 336). Process capability analysis against Stage 1 prototyping gate criteria confirmed the system is not yet process-capable. Furthermore, a Total Cost of Ownership (TCO) analysis yielded an estimated US $8.07–11.22 per successful launch, challenging the economic scalability of low-cost additive manufacturing materials. This study establishes quantitative benchmarking and a reliability-centered Design for Additive Manufacturing (DfAM) framework required before scaling toward maritime, emergency, or aerospace applications. Full article

Industrial robots in small-batch manufacturing and human–robot collaborative workstations are increasingly required to perform grasping and sorting tasks on workpieces with natural language instructions. Unlike generic object grasping, industrial workpieces may contain local surface defects such as dents, scratches, and geometric irregularities, which affect both target selection and grasp stability. This paper presents a geometric point-cloud perception and defect-aware grasping framework that leverages 3D geometric cues measured from point clouds to detect surface defects and guide robotic manipulation. Rather than learning an end-to-end visuomotor policy, the proposed framework introduces an explicit geometric reasoning layer between language-level task specification and robotic grasp execution. The language model is used only to convert instructions into structured task attributes, whereas defect perception, target ranking, and grasp-contact evaluation are performed using measurable 3D geometric evidence. These components require no task-specific training on annotated industrial defect or grasping datasets, making the formulation potentially useful for small-batch manufacturing scenarios in which large annotated defect datasets are unavailable. A shared defect-response representation supports both instance-level target selection and contact-region screening. We evaluate the proposed method on a controlled prototype tabletop setup, achieving an average defect instance-level F1-score of 0.89, a target grounding accuracy of 0.89, and a grasp success rate of 84.0%. Full article

This study develops an integrated generative artificial intelligence (GAI) framework for improving business process performance in industrial enterprises. The framework treats GAI not as the isolated use of generative tools, but as a governable information systems capability embedded in recurring workflows, enterprise architectures, documented knowledge, and human decision roles. It integrates four functional subframeworks—manufacturing, marketing and sales, accounting and finance, and human resource management—with a shared orchestration and governance layer. This layer coordinates process architecture, approved data and knowledge sources, reusable GAI capabilities, human-in-the-loop validation, traceability, escalation, and performance measurement. A proof-of-concept maturity-readiness validation is conducted in an electronics company using maturity-readiness logic inspired by the Smart Industry Readiness Index (SIRI). The assessment shows an increase in the overall readiness score from 41.60 in the pre-GAI baseline to 79.08 in the post-GAI implementation scenario. Accordingly, the score increase is interpreted as expert-assessed maturity-readiness evidence rather than as a measured causal effect on operational performance. This study contributes a process-centric reference architecture designed for technical implementability, traceability, auditability, and human-supervised enterprise-scale GAI adoption. Full article

This study examines the impact of green innovation investment on ESG rating divergence and the mediating role of product market competition, based on 11,768 firm-year observations of Chinese A-share listed manufacturing enterprises from 2011 to 2023. We measure green innovation by lagged green patent applications (distinguishing invention and utility model patents), ESG rating divergence by the standard deviation and range of scores from six mainstream international and domestic agencies, and product market competition using the Herfindahl–Hirschman Index. Using mediating effect models with multi-dimensional fixed effects and heterogeneity analysis, we find that green innovation significantly widens ESG rating divergence, with green invention innovation exerting a stronger impact. Product market competition plays a partial mediating role, as green innovation intensifies industry competition and further amplifies rating divergence. Heterogeneity analysis shows no significant effect for heavily polluting firms, while green invention innovation of high-attention firms is more likely to cause rating disputes. This study reveals the reverse spillover effect of green innovation on ESG evaluation, providing new perspectives for understanding rating divergence and references for corporate green strategies and rating system optimization. Full article

Reliable pore detection in in situ monitoring is essential for quality control in additive manufacturing, particularly in safety-critical applications. However, existing automated approaches often struggle with low-quality images, computational inefficiency, and the lack of reliable uncertainty estimates to identify the pores accurately. To address these limitations, this paper proposes U-SAMNet, an Uncertainty-Aware Self-Attention Multi-Task Network. The model jointly performs pore segmentation, image-level classification, and denoising on in situ monitoring images, while deriving epistemic uncertainty via Monte Carlo dropout variance to dynamically suppress unreliable attention features. This study uses electron-optical (ELO) images from the E-PBF process. A novel Uncertainty Guidance Attention (UGA) block is introduced. It suppresses channel attention weights using inverted uncertainty estimates. Unlike prior methods that modulate classification weights, U-SAMNet operates on pixel-wise feature maps. The model has only 1.69 M parameters. A single forward pass takes 17.46 ms per image, while uncertainty estimation is performed through multiple Monte Carlo dropout passes. The model was evaluated on real E-PBF images and achieved 99.42% pixel accuracy, 85.31% F1-score, and 74.38% IoU on a held-out test set of 43 real images. It also generalizes to other industrial defect detection benchmarks, reaching 85.09% F1-score on DAGM 2007 and 82.32% on DeepCrack. GAN-synthesized training data improves the F1-score by 21.34 percentage points over rotation augmentation alone. Ablation studies confirm that each component contributes meaningfully to the final performance. Full article

Artificial intelligence is increasingly used to improve industrial quality control, but its practical value depends on whether models remain accurate under different operating conditions and fault classes. This study evaluates an artificial-intelligence-based workflow for gear quality control using vibration signals measured on a real two-stage reduction gearbox. Two orthogonal vibration channels were analyzed for six health states, three shaft speeds, and two load levels. Because the time-series data were only partly stationary, the dataset was divided chronologically into training and test segments. A 54-feature representation was built from rolling-window statistics and operating variables, and six classifiers were compared: Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBM), random forest, decision tree, multilayer perceptron (MLP), and logistic regression. LGBM achieved the best overall accuracy (0.9728) while maintaining substantially lower training time than several competing nonlinear models. Class-wise precision, recall, and F1-score ranged from 0.95 to 1.00, and the nominal response time for most operating-condition transitions was approximately 0.0998 s. The results show that vibration-based machine learning can support robust, near-real-time fault identification in mechanical manufacturing environments. The study also highlights the importance of chronological validation, feature engineering over multiple time windows, and the trade-off between predictive performance and deployment efficiency. Because the validation dataset originates from one gearbox platform, the results should be interpreted as promising internal evidence rather than as proof of universal industrial robustness. Full article

With the rapid development of intelligent sensing systems, digital monitoring platforms, and multisource data acquisition technologies, accurate identification of operational states and potential risks from heterogeneous sensing signals has become an important research issue in artificial intelligence-driven sensing. Existing studies have primarily focused on either textual information understanding or behavioral data analysis, with limited attention paid to jointly modeling the consistency between textual declarations and executed behaviors. As a result, many potential risks that have not yet manifested as significant anomalies but already involve execution deviations are difficult to detect in a timely manner. To address this issue, a language–behavior consistency sensing framework for multisource sensing signals is proposed. Textual sensing signals and behavioral sensing signals are mapped into a shared state logic space, and intelligent perception and quantitative analysis of deviations between textual states and executed states are achieved through a textual state logic extraction module, an observed behavioral state modeling module, and a language–behavior consistency measurement module. Systematic experiments were conducted on a multisource sensing dataset containing public declaration texts, operation reports, behavioral logs, resource allocation records, and state-response information. The results show that the proposed method achieved the best performance in the baseline comparison experiment, with a language–behavior consistency score ($LCS$) of $0.742$, an $AUC$ of $0.846$, an $F1$-score of $0.811$, a Precision of $0.802$, and an explanation consistency score ($ECS$) of $0.821$, clearly outperforming advanced methods such as FinBERT, LSTM, Multimodal Transformer, and the Contrastive Multimodal Model. These results demonstrate that language–behavior consistency sensing can effectively fuse multisource sensing information and improve complex system state identification, anomaly early warning, and risk perception, providing an interpretable artificial intelligence-driven sensing framework with the potential to support industrial operation and maintenance, intelligent manufacturing, digital infrastructure management, and other intelligent monitoring scenarios. Full article

In response to rapid shifts toward electrification, digitalization, and sustainability in the automotive industry, this study developed a sustainability-oriented, evidence-based competency framework for automotive maintenance technicians. Traditional competency frameworks, often derived from manufacturer manuals or curricula, overlook tacit knowledge from real-world maintenance practices, leading to gaps in diagnostic effectiveness, service quality, and resource efficiency. To address this limitation, 8500 maintenance records from 67 service centers (2022–2025) were subjected to quantitative content analysis to identify preliminary competency indicators across five vehicle systems. A three-round Delphi survey involving 24 senior automotive experts was subsequently conducted to validate and prioritize these indicators on the basis of mean importance scores and coefficients of variation (≤0.20). The final framework comprised 39 competencies, such as diagnostic proficiency, electronic system integration, system-level troubleshooting, and technical documentation application. Beyond traditional mechanical skills, cross-system diagnostic capability and digital tool proficiency have become essential competencies for modern electric vehicles. By transforming tacit maintenance knowledge into measurable indicators, the developed framework can contribute to supporting workforce sustainability, enhancing repair accuracy, reducing unnecessary part replacement, and improving resource efficiency. It can also inform vocational education, industry certification, and human capital development aligned with Sustainable Development Goals 8, 9, and 12. Full article

Industrial sustainability assessment needs a transparent and operational framework that can manage multidimensional indicators, expert uncertainty, weighting complexity, and managerial interpretation. This research presents the Sustainable Industrial Measurement (SIM) Model, an AI-assisted sustainability intelligence architecture for manufacturing organizations. The model comprises literature-based indicator synthesis, Fuzzy Delphi Technique (FDT), Pareto 80/20 screening, Group Analytic Hierarchy Process (Group AHP), Utility Value Analysis and the web-based AI-enabled decision support system (AI-DSS) under the Triple Bottom Line (TBL) framework. In FDT validation by consensus, threshold and fuzzy score criteria, 33 experts accepted 64 indicators. The Pareto screening reduced the set to 50 high-impact indicators, consisting of 10 economic, 22 social and 18 environmental indicators. The priority weights were derived from the group AHP weighting by 21 experts and checked for consistency. The environmental and economic indicators represent the dominant sustainability priorities. The weighted structure was embedded in the web-based AI-DSS to enable automated scoring, visualization, gap diagnosis and AI-based managerial recommendations. Thirty industrial practitioners reported excellent perceived usability of the SIM Model, with a System Usability Scale score of 86.0. However, the evaluation assessed usability only, not the accuracy, effectiveness, or organizational impact of AI-assisted recommendations for manufacturing sustainability decisions and future implementation. Full article

Fault-tolerant control systems deployed in manufacturing and process industries must maintain output regulation under actuator degradation and parametric drift while producing operator-readable intervention records. This paper proposes a three-layer cognitive supervisory control framework: a continuously synchronized nominal plant model serves as a lightweight digital twin, driving residual-based anomaly detection that responds to model-mismatch faults before tracking error accumulates; an event-triggered supervisory layer applies composite anomaly scoring, risk estimation, and dwell-time-constrained mode switching over a bank of pre-verified stable controllers; and an optional bounded LLM reasoning agent forming Layer 3 of the cognitive architecture produces real-time natural-language audit records at each supervisory event. The framework is functionally non-intrusive—the closed-loop response is numerically identical to standalone LQI in three of four benchmark scenarios and on an unstable plant variant, with zero supervisory interventions. Under abrupt 75% actuator loss, a single targeted intervention achieves lower overshoot and better tracking than all baselines, including MPC whose feasibility-degradation under reduced actuator authority is confirmed empirically. On near-undamped plants representative of compressor and pipeline dynamics, integral absolute error improves significantly relative to standalone LQI; on a distributed thermal process representative of laser welding and induction heating, IAE improves substantially under abrupt heater power loss. Parametric sensitivity analysis across fault severity, detection threshold, onset timing, measurement noise, score weights, and dwell-time confirms robustness of the detection architecture. The LLM layer’s contribution is regulatory rather than performative: its value is the causally connected audit record generated at the moment of each decision, not IAE improvement—a form of real-time explainability that after-the-fact attribution methods such as SHAP and LIME cannot provide by construction. Full article

General-purpose bibliometric tools do not address domain-specific needs of occupational safety and health (OSH) evidence mapping, such as quantifying whether the literature connects hazard characterisation with real worker exposure or identifying which articles best bridge science and preventive practice. This paper presents ORISMA (Occupational Risk Integrated Systematic Mapping and Analysis), an open-source R package that introduces five preventive bibliometric indicators: the Worker–Risk Disconnection Index (WRDI), Risk Category Saturation Index (RCS), Material–Gap Profile (MGP), Abstract Sufficiency Score (ASS), and Bridge Article Score (BAS). A complete evidence map is produced through three function calls. To demonstrate the package, we apply it to a corpus of bibliographic records on occupational risks in metal additive manufacturing (metal AM), retrieved from Web of Science, Scopus, ProQuest, and PubMed. After conservative relevance-guard filtering (191 records retained) and three-step deduplication, 76 unique records were analysed. The global WRDI was 0.4474, indicating moderate disconnection between hazard characterisation and worker exposure evidence. Industrial hygiene and emerging technology domains dominated preventive coverage, while safety, ergonomics, psychosocial, and biological risk domains were largely absent. Eighteen strong bridge articles were identified. The material–risk co-occurrence map showed no detected co-occurrence between explosion/fire or asphyxiant-gas risks and any material, reflecting an absence of explicit terminology in the retrieved literature rather than a definitive absence of evidence. ORISMA provides a reproducible, domain-aware framework that requires further validation across independent OSH domains before broad generalization. Full article

The transition towards sustainable aluminum manufacturing is commonly assessed through recycling rates, energy savings, and resource efficiency, but its distributive effects across global value chains remain insufficiently examined. This study evaluates whether recycled aluminum value chains contribute to both circularity and north–south redistribution, or whether they reproduce unequal patterns of value capture, industrial upgrading, employment quality, and trade dependency. The analysis combines UN Comtrade trade data for HS 7601–7616, OECD ICIO 2025 value added indicators, ILOSTAT labor statistics, and UN SDG data for the 2018–2020 three-year average. Eighty economies are classified into four groups: advanced industrial economies, emerging industrial economies, lower-middle-income economies, and low-income economies. A composite indicator linked to SDGs 8, 9, 10, and 12, with SDG 17 incorporated only as a trade dependency context, is constructed from normalized industrial, circular material flow, distributive, and job-quality variables. The results show a clear north–south hierarchy: advanced economies concentrate a larger share of exports in aluminum manufactures, while low-income economies remain more dependent on scrap flows. Group A captures most chain value added, whereas Groups C and D retain only marginal shares. Labor productivity falls sharply from advanced to low-income economies, while working poverty increases substantially. By contrast, circularity scores vary less strongly across groups, suggesting that participation in circular material flows does not necessarily imply equitable industrial upgrading. This study shows that circularity in recycled aluminum value chains does not automatically generate redistribution and provides a replicable framework for distinguishing material circularity from distributive justice. Full article

With the rapid development of intelligent manufacturing, edge computing, and industrial and financial–industrial digital systems, large volumes of multisource hardware sensing signals are continuously generated in complex production environments, including environmental, electrical, vibration, network communication, and device operational signals. Owing to the heterogeneity, asynchrony, noise interference, and disturbance sensitivity of these signals, conventional state prediction methods often fail to sufficiently characterize the dynamic response relationships among different sensing sources and cannot maintain stable prediction performance under non-stationary scenarios such as load surges, network congestion, and device anomalies. To address these challenges, a multisource hardware sensing signal fusion network is proposed for the edge-computing and digital production test scenario of an intelligent equipment manufacturing enterprise in Hebei Province, China, with the aim of achieving robust state prediction and anomaly perception in complex digital systems. In the proposed method, environmental sensing, device power, edge-node operation, vibration monitoring, network communication, and system output states are uniformly modeled as multisource engineering sensing signals, and an end-to-end prediction framework is constructed with cross-source sensing signal alignment to facilitate temporal coherence, disturbance-aware residual correction to substantially mitigate disturbance contamination, and context-adaptive fusion. Experimental results show that the proposed method achieves the best performance in the overall state prediction task, with MAE, RMSE, MAPE, and $R^2$ reaching $0.0968$, $0.1457$, $8.12$%, and $0.9416$, respectively, outperforming baseline methods including ARIMA, XGBoost, LightGBM, LSTM, TCN, Transformer, Attention Fusion, and Multimodal Transformer. In the disturbance robustness experiment, the Event-MAE and Event-RMSE of the proposed method are reduced to $0.1126$ and $0.1694$, respectively, with an Avg. Drop of only $28.98$%, indicating that more stable responses can be achieved under non-stationary disturbance scenarios. In the abnormal-state recognition task, Accuracy, Precision, Recall, and F1-score values of $94.32$%, $93.76$%, $92.85$%, and $93.30$% are achieved, respectively. The results demonstrate that the proposed method can effectively improve the state prediction accuracy, disturbance robustness, and anomaly warning capability of multisource hardware sensing data in complex industrial and financial–industrial digital systems, thereby providing an effective modeling scheme for intelligent monitoring and engineering decision-making in AI-driven industrial and financial sensing scenarios. Full article

The aim of the study was to assess the level of implementation of the Industry 4.0 and Quality 4.0 concepts in manufacturing enterprises in Poland and Slovakia, as well as to identify the main barriers, supporting factors, and organizational effects related to digital and quality transformation. The study was conducted using a qualitative–quantitative approach based on a multiple case study strategy and the matched-pair method. The analysis covered 10 manufacturing enterprises (5 from Poland and 5 from Slovakia) representing various industrial sectors. Research triangulation was applied, including a diagnostic questionnaire, semi-structured interviews, analysis of organizational documents, and direct observation. Two synthetic indicators were developed: the Industry 4.0 Readiness Index (I4RI) and the Quality 4.0 Maturity Index (Q4MI). The results indicate a moderate level of implementation of both Industry 4.0 and Quality 4.0 in the analyzed enterprises, with Polish companies achieving slightly higher scores than Slovak enterprises. The highest level of advancement was observed in the area of technologies and automation, while the lowest concerned data integration and the use of analytics. The analysis revealed a strong positive relationship between the level of Industry 4.0 implementation and Quality 4.0 maturity, as well as the positive impact of both concepts on selected operational performance indicators, such as OEE, delivery timeliness, and defect rates. The obtained results confirm the complementary nature of Industry 4.0 and Quality 4.0 and highlight the need for a more integrated and human-centric approach to the digital transformation of manufacturing enterprises in Central and Eastern Europe. Full article