Abstract: Accurate temperature prediction is essential for optimizing the microwave preheating of PET preforms prior to blow molding. A key challenge in this context is the strong dependence of electromagnetic field distributions and thermal responses on preform geometry, which varies substantially across product lines. Conventional neural network models trained on specific geometric configurations typically fail to generalize to unseen preform designs, requiring costly retraining for each new geometry. This work proposes a unified geometry-aware deep learning framework that predicts spatial temperature distributions across multiple preform designs using a single neural network model. The approach reformulates temperature prediction as a coordinate-level regression task conditioned on spatial location, geometric descriptors, process parameters, and structural region labels. A domain-bounded training strategy based on extreme feasible preform geometries is introduced, ensuring that predictions for intermediate designs remain within the interpolation regime of the network. The framework is evaluated on six distinct preform geometries, demonstrating that a single model can generalize reliably to new, unseen preform designs when their geometric parameters fall within the bounds of the training data. This is achieved through a domain-bounded training strategy that constructs datasets from the extreme feasible geometries, thereby converting the prediction of any intermediate design into an interpolation task. Since neural networks are inherently limited in their ability to extrapolate beyond the training domain, this formulation is essential for ensuring stable and accurate predictions across the full range of industrially relevant preform configurations. The proposed methodology provides a foundation for geometry-informed surrogate modeling in thermal process control and can be extended to other manufacturing systems characterized by strong geometric variability.

Abstract: Sustainable manufacturing increasingly requires production schedules that balance environmental responsibility with delivery reliability. In flexible job shop environments, this challenge is especially difficult because machine assignment and sequencing decisions affect both the carbon footprint of production and the risk of missing job due dates. Motivated by this trade-off, this paper studies the Carbon-Aware Flexible Job Shop Scheduling Problem with Tardiness Penalty (CAFJSP-T), a flexible job shop formulation in which total carbon emissions and total tardiness penalty are treated as the two primary objectives, while energy consumption and makespan are retained as supporting performance indicators. To solve this problem, we propose a Policy-based Rough Optimization with Large Neighborhood Search (Pro-LNS) framework that combines Proximal Policy Optimization for fast, policy-guided construction of feasible schedules with an adaptive large neighborhood search procedure for targeted refinement. The two phases are aligned through a normalized scalarized objective that balances carbon emissions and tardiness penalty while preserving all precedence, eligibility, and machine-capacity constraints. Computational experiments on benchmark instances spanning small, medium, and large workcenter categories show that Pro-LNS produces high-quality schedules with strong due-date performance and controlled carbon emissions. Under equal objective weighting, the method achieves a median optimality gap of 6.12% relative to the exact formulation, with all reported instances remaining within 14%, while requiring only 4.08 seconds on average and at most 10.51 seconds. These results indicate that Pro-LNS is an effective and computationally practical approach for carbon-aware, tardiness-sensitive flexible job shop scheduling.

Abstract: Predictive maintenance is becoming essential for modern U.S. manufacturing plants as unplanned machine downtime leads to significant productivity losses, supply delays, and increased operational costs. This research proposes an AI-driven predictive maintenance framework that integrates Industrial Internet of Things (IoT) sensor streams, machine learning failure prediction, and reliability-based maintenance scheduling. The model utilizes vibration, temperature, power consumption, and operational cycle data to detect early-stage degradation patterns in industrial equipment. A hybrid deep learning and survival analysis approach is introduced to estimate Remaining Useful Life (RUL) and predict the probability of failure over time. Additionally, an optimization layer is developed to automatically generate cost-effective maintenance schedules that minimize downtime while balancing labor availability and spare parts constraints. The proposed framework is highly scalable and can be implemented across diverse manufacturing sectors, including automotive, semiconductor, and aerospace production. By improving equipment reliability, reducing emergency repairs, and supporting Industry 4.0 modernization, this work directly contributes to U.S. manufacturing competitiveness and industrial resilience.

Abstract: Overall Equipment Effectiveness (OEE) is the dominant metric for manufacturing productivity, computed as the multiplicative product of Availability (A), Performance (P), and Quality (Q). Despite its widespread adoption, the classical OEE formula embeds a structural limitation: the three components are treated as equally important regardless of operational context, a fixed-weight assumption that systematically distorts maintenance prioritisation in environments with asymmetric operational priorities. No published framework has formally addressed this limitation through a structured, auditable multi-criteria weighting model. This paper proposes Adaptive OEE, a FUCOM-TOPSIS framework that replaces the fixed A×P×Q product with a context-driven weighting model. FUCOM elicits context-specific weights for A, P and Q from expert judgement using only n−1 pairwise comparisons with guaranteed consistency, while TOPSIS ranks equipment assets under the weighted criteria, producing a closeness coefficient comparable across assets and contexts. Three illustrative case studies covering availability-dominant, performance-dominant, and quality-dominant contexts demonstrate that the classical OEE ranking is not preserved under any weight configuration, with Divergence Index values ranging from 0.667 to 1.333. Divergence is most severe when one component carries strongly asymmetric weight, precisely the condition equal weighting cannot accommodate. The principal contributions are the formalisation of the equal-weighting assumption as a measurement-theoretic deficiency, the replacement of multiplicative aggregation with a weighted distance measure preserving the A/P/Q decomposition, and the introduction of the Divergence Index as a quantitative measure of context-insensitive rank displacement.

Abstract: Life Cycle Assessment (LCA) is extensively employed to support sustainability evaluation in waste management and manufacturing systems; however, outcomes are highly sensitive to methodological decisions, particularly end-of-life (EoL) allocation approaches. This study examines how different allocation methods—primarily cut-off and substitution approaches—affect the interpretation of energy performance and decarbonisation potential in plastic waste management and injection moulding systems. The analysis applies cut-off logic to open-loop scenarios to establish a baseline impact, while substitution-based modelling is utilised for semi-closed and fully closed-loop configurations to quantify environmental credits and avoided burdens. A dual framework is adopted: first, a literature review examines methodological sensitivities in EoL modelling, focusing on allocation logic and system boundaries; second, a quantitative case study assesses open-loop, semi-closed, and fully closed-loop injection moulding scenarios for polyethylene (PE) products using LCA and hot-spot analysis. The results demonstrate that allocation choices can significantly influence calculated energy savings and greenhouse gas reduction potentials, sometimes reversing the relative ranking of configurations. Substitution-based approaches tend to report higher decarbonisation benefits by crediting avoided primary production, whereas cut-off approaches provide more conservative estimates. In the case study, increased internal material and water looping lead to measurable reductions in energy demand, although trade-offs across impact categories persist. These findings highlight that circular economy (CE) evaluations are strongly shaped by methodological assumptions, with direct implications for energy policy and decarbonisation pathways. The study emphasises the need for transparent allocation decisions and robust frameworks to ensure reliable decision-making in the transition toward low-carbon and energy-efficient systems.

Abstract: Plastic packaging waste has emerged as a critical environmental challenge due to its persistence, low degradation rates, and increasing accumulation in terrestrial and marine ecosystems. Conventional petroleum-based plastics dominate packaging applications because of their durability and low cost; however, their environmental impacts have prompted urgent demand for sustainable alternatives. Bio-based and compostable packaging materials offer promising solutions by utilizing renewable resources and enabling environmentally benign end-of-life pathways. This paper examines the development of bio-based and compostable packaging alternatives aimed at reducing plastic waste. Through a systematic review of material innovations, processing technologies, and life-cycle considerations, the study evaluates the performance, environmental benefits, and limitations of emerging bio-based packaging solutions. The findings indicate that materials such as polylactic acid, polyhydroxyalkanoates, starch-based composites, and cellulose-derived packaging can significantly reduce fossil resource dependency and plastic pollution when supported by appropriate infrastructure. The paper concludes that while bio-based and compostable packaging presents strong environmental potential, successful large-scale adoption requires integrated design strategies, composting infrastructure, and supportive policy frameworks.

Abstract: Artificial Intelligence (AI) and its subset, Machine Learning (ML), play transformative roles in the manufacturing sector, forming the foundation of the “Industry 4.0 and 5.0” frameworks. This research contributes to that evolution by developing AI-based advisory systems that utilize advanced data models to optimize casting processes. These systems exemplify the principles of smart manufacturing, where machines and processes are interconnected, adaptive, and driven by data. They support key objectives such as automation, seamless connectivity, real-time data exchange, human-centric innovation, operational resilience, and sustainability. The models developed in this work enable manufacturers to fine-tune product quality, minimize waste, and accelerate time-to-market through predictive analytics and dynamic process control. By integrating AI-based advisory systems, hybrid modeling, and reduced-order modeling techniques, the systems facilitate real-time decision-making and continuous improvement—essential for achieving flexible, efficient, and customized production environments. A real-world case study further demonstrates the effectiveness of these AI-based advisory systems in casting applications, detailing the steps involved in database construction, data training, and predictive modeling.

Abstract: Optimising production layouts in manufacturing plants is a time-consuming and often manual process that typically only considers individual performance indicators. This paper presents an end-to-end pipeline that uses variational autoencoders to generate and optimise layouts. The method simultaneously considers multiple KPIs such as throughput time, energy consumption, space utilisation, machine density and material flow complexity. Different scenarios like standard, bottleneck, energy focus are supported. Results show that the proposed method generates valid layouts that outperform existing layouts in terms of efficiency, energy consumption and material flow. The pipeline enables fast, reproducible layout generation and can be directly integrated into production control systems to achieve measurable technical improvements.

Abstract: Automated quality inspection is a central component of modern industrial production processes. Over the past few decades, machine vision has evolved from rule-based, traditional image processing methods to data-driven machine learning and deep learning approaches. In particular, with the advent of powerful neural networks, significant progress has been made in the detection, classification, and localization of defects. At the same time, industrial applications place high demands on robustness, real-time capability, explainability, and the handling of rare or unknown defect patterns. This brief survey provides an overview of machine vision methods for industrial quality inspection. It systematizes classical image processing approaches, supervised, unsupervised, and semi-supervised learning methods, and discusses their strengths and limitations in real-world production environments. Furthermore, it examines multisensory and three-dimensional inspection approaches, aspects of industrial implementation, and current developments in the field of explainable artificial intelligence. Finally, this brief overview identifies outstanding challenges and research gaps and outlines future trends in automated quality inspection.

Abstract: Monitoring the condition of rolling bearings is critical for industrial reliability, yet tradi-tional contact-based accelerometers can be impractical in confined or hazardous envi-ronments. This study investigates the use of microphones as a non-invasive diagnostic alternative, focusing on the impact of sensor distance and spatial placement on fault de-tection sensitivity across various rotational speeds and load conditions. Using an accel-erometer mounted directly on the bearing as a benchmark, acoustic data were acquired on a test bench under different speed and load conditions. The experimental setup evaluated three distinct microphones positions and five distances relative to the source to assess spatial influence. Analysis was conducted comparing scalar indicators, such as Root Mean Square (RMS), Kurtosis and Crest-Factor values, with advanced diagnostic tech-niques, specifically the High-Frequency Resonance Technique (HFRT) for envelope spec-trum extraction. Results indicate that while the signal-to-noise ratio (SNR) predictably decreases with distance, diagnostic performance is significantly compromised by acoustic shielding effects caused by bearing housing. Moreover, while simple statistical factors (RMS, Kurtosis, Crest Factor) show limited reliability across varying distances and noise floors, HFRT-based envelope analysis yields robust fault identification even at the max-imum sensor distance. The study concludes that optimal microphone placement is essen-tial for reliable remote monitoring. Particularly, these findings suggest that a preliminary spatial characterization of the acoustic field can significantly enhance the effectiveness of non-contact diagnostic systems in industrial applications.

Abstract: This study investigates hybrid brake-pad composites made by adding different percentages of silicon carbide (15% and 20% SiC) and zinc oxide (10%, 15%, and 20% ZnO). The goal was to find a composite that improves brake working efficiency. Wear and hardness tests were carried out according to ASTM standards. The experimental results were analyzed using Design of Experiments method to study how wear changes over time under different loads. Time-series trend analysis visualizes how the specific wear rate developed. The results showed that sample A5 had the best wear resistance and certified A5 as the optimum structural stability over time composite sample. The hardest samples were A2 and A5. The best composite was selected for a static structural analysis using ANSYS 2022-R1 to evaluate stress, strain, deformation, and elastic energy. The thermal analysis examined heat distribution, heat generation, and heat flux in the hybrid composite material. The numerical results showed that stress levels are lower at outer surfaces compared to the inner regions. The outer surfaces exhibit a uniform distribution heat flux. Directional heat flux showed a slight increase near the inner radius, the disk protrusions and edges. These findings clarified how the optimal composite behaves under braking conditions.

Abstract: Wire Arc Additive Manufacturing (WAAM) is a cost-effective and scalable technique for producing large metallic components; however, coarse columnar microstructures, strong crystallographic texture, and significant residual stresses limit its widespread adoption. In recent years, hybrid WAAM processes integrating deformation-based techniques have been developed to address these limitations. This review provides a comprehensive analysis of deformation-assisted WAAM, encompassing interlayer rolling, friction stir processing (FSP), hammer peening, laser shock peening, and ultrasonic vibration-assisted approaches. These hybrid techniques introduce additional thermomechanical parameters—strain, strain rate, and applied stress—that significantly influence microstructure evolution. The governing physical metallurgy mechanisms are discussed in detail, including dislocation accumulation, recovery, static and dynamic recrystallization, and severe plastic deformation. Studies from 2022 to 2025 are critically reviewed, highlighting the effectiveness of hybrid WAAM in promoting columnar-to-equiaxed grain transformation, reducing anisotropy, mitigating defects, and improving mechanical properties across aluminum, titanium, steels, and nickel-based alloys. The integration of auxiliary processes such as in-situ machining and heat treatment is also discussed. This review establishes a process-structure-property framework for hybrid WAAM and provides guidance for the development of advanced additive manufacturing systems capable of delivering near-net-shape components with microstructures and properties approaching those of wrought or forged counterparts.

Abstract: Smart textiles represent a transformative convergence of materials science, electronics, and sustainability principles, enabling the creation of fabrics that sense, respond, and adapt to environmental stimuli. This systematic review examines recent advancements in sustainable smart textile technologies, synthesizing findings from peer-reviewed literature published between 2018 and 2024. Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework, 78 studies were identified and reviewed. Key innovations include biodegradable electronic fibers, solar-energy-harvesting textiles, thermoelectric wearable systems, self-healing polymer matrices, and bio-based conductive inks. The review highlights a paradigm shift toward circular economy models in textile manufacturing, driven by regulatory pressure and consumer demand for eco-conscious products. Findings reveal that sustainable smart textiles can achieve performance parity with conventional counterparts while reducing environmental impact by up to 60%. The article concludes with targeted recommendations for researchers, industry practitioners, and policymakers seeking to accelerate the responsible commercialization of sustainable smart textile technologies.

Abstract: Petroleum products have led to challenges for the environment on a global scale, such as pollution from microplastics, persistent waste in the environment that is not biodegradable, and an increasing carbon footprint due to plastic production. In quest of sustainable alternatives, environmentally friendly cellulose nanofiber (CNF) based nanopapers have gained potential as biodegradable materials to replace most of plastic derived industrial products. (Cellulose nanofibers derived from renewable biomass that possesses small dimensions which exhibit excellent mechanical properties, low density, high optical transparency, and very good barrier properties against gases and oils.) Advantages like these render CNF nanopapers promising for flexible electronics, packaging materials, coatings, filtration membranes and sustainable structural materials. This study aims to provide a conceptual framework for the development of cellulose nanofiber (CNF) based nanopapers as substitutes for common plastic materials by investigating their performance in industrial applications. This framework combines biomass extraction techniques, nanofibrillation processing, fabrication of nanopaper and material performance validation. Numerical simulations reveal that CNF nanopapers offer mechanical and barrier performance approaching synthetic polymer films with substantial decrease in environmental footprint. The findings demonstrate that cellulose based nanopapers provide a scalable route for sustainable material innovation in industry. Bio nanomaterials hold great promise for contributing to the sustainability of circular manufacturing systems as they can address environmental sustainability challenges linked with plastic based materials.

Abstract: Repeated machining passes (i.e., continuous toolpaths) are common in CNC manufacturing, including multi-level machining of prismatic parts and iso-contour passes in contour machining. They present an opportunity to exploit pass-to-pass learning to improve productivity without sacrificing quality through feedrate optimization. Traditional iterative learning methods provide a means to exploit pass-to-pass learning for quality improvements, but they are not well-suited to feedrate optimization because the reference trajectories change as the feedrate increases. In the authors’ prior work, learning-based feedrate optimization was demonstrated for repeated machining along identical toolpaths. This paper extends that concept to the more challenging case of similar but non-identical cutting paths, as encountered in contour machining. A pass-to-pass learning strategy is proposed in which corresponding sections of non-identical iso-contour passes are identified using a contour-matching method based on geometric similarity. Bayesian linear regression models are then used to learn and predict contour error and spindle torque across passes, with uncertainty explicitly quantified through credible intervals. These predictions are embedded in a window-based feedrate optimization framework solved via sequential linear programming, enabling feedrate maximization subject to kinematic, contour-error, and spindle-torque constraints. The proposed approach is experimentally validated on a 3-axis desktop CNC milling machine through multiple 2.5D contour machining case studies. Results show that the method can rapidly approach near-optimal feedrates after only a few passes, culminating in up to 16.4% increase in productivity compared to an equivalent learning-based feedrate optimization approach for identical toolpaths.

Abstract: Laser hard overlaying is an advanced, perspective technology with wide industrial applications, for example dies. The aim is to improve surface properties like wear resistance using special layers of powder sintered or remelted by laser beam. At present, dies are manufactured by machining with following bulk heat treatment, which is an expensive process particularly due to use of expensive high-alloyed tool steels. Repairs performed using arc or plasma welding introduce a big amount of heat to the part, which can cause dimension changes and material degradation, These methods often fail also due to low weldability of the materials. An advantage of laser overlaying is minimization of these difficulties. The paper contains a comprehensive evaluation of several types of hard overlayed powder of H13 tool steel on a S355 structural steel and on H11 tool steel using laser beam. Macro- and microstructure, hardness and fatigue resistance are evaluated including fatige damage mechanisms. Results are completed with basic measurement of residual stresses using destructive strain-gauge methods. Fatigue resistance is sensitive on surface and subsurface defects, which can significantly reduce endurance limit.

Abstract: In this study, we investigated the change of carbon dioxide (CO₂) gas in PET bottles over time using experimental and numerical methods, as this is an important quality criterion in carbonated beverage production. Gas loss was modeled using the finite element method (FEM) on 2.5-liter PET bottles, and the effects of temperature, internal pressure, and packaging wall thickness were theoretically evaluated within the framework of the ideal gas equation and Fick's law. Validation was achieved by comparing model results with experimental data, and ideal production conditions were determined. Analyses revealed that gas loss was concentrated primarily in the top and shoulder regions of the bottle, and increasing the thickness in this region reduced diffusion. Furthermore, lowering the filling temperature and increasing internal pressure significantly reduced the transfer of dissolved CO₂ from the packaging to the external environment. Modeling studies were conducted using a three-dimensional design of the bottle geometry, defining boundary conditions to investigate the effects of different material distributions and thicknesses. Based on the findings, production processes were reorganized, and standardized recipes were created. As a result, the combination of experimental and numerical data has shown that gas losses have been largely controlled, and quality standards can be maintained for longer periods. This study can provide guidance not only for 2.5-liter PET bottles but also for other packaging types. Thus, it was concluded that more planned, higher standard production can be achieved in the carbonated beverage industry, consumer complaints can be reduced, and product performance can be maintained sustainably.

Abstract: (1) Background: Hugging Face is one of the largest platforms for machine-learning datasets, hosting collections of all kinds beyond its core focus on natural language processing. Whether and how these datasets can be leveraged for agricultural informatics is an open question. (2) Methods: A systematic data-space analysis structured by the PRISMA 2020 methodology was conducted. Using the search terms “farming” and “agriculture”, 128 datasets were identified on the platform, of which 126 could be fully analysed. (3) Results: Datasets cover mostly crops (42 %). English dominates (71 %); 13 languages are represented in total. The distribution of dataset sizes is strongly right-skewed (mean 156,346 entries; median 1,000). Parquet is the most common format (43 %); 92 % of datasets appear to contain human–LLM dialogues. (4) Conclusions: The available agricultural datasets on Hugging Face are thematically and qualitatively heterogeneous. Future work should develop prototypes to test if the available datasets are usable as data base for crop-related applications, and to identify potential gaps in the data space.

Abstract: In the underground operation scenario, with the development of intelligence, the underground electronic monitoring and control systems have gradually become an important tool for mining practitioners to obtain operational information. In view of the existence of low-light environment under the mine and the obvious difference from the above-ground natural light, screen-related factors have a significant impact on the visual search task, so it is critical to study the impact of interface layout in the low-light environment under the mine on the visual search efficiency. In this study, the application scenario of the electronic monitoring and control system under the mine is simulated. The information layout of the current electronic monitoring and control system in the mine and different lighting environments are set as experimental variables. The effects of interface layout design features on user search performance, visual behavior and usability satisfaction are discussed. The experimental results show that interface layout and illumination change have significant main effects on task completion time, fixation times, saccade ratio and subjective usability score. Among them, Three-Column layout mode has outstanding performance in the aspects of task completion time, fixation number and subjective usability score, and the search efficiency is higher in 50lx illumination environment.

Abstract: With Industry 4.0, modern manufacturing systems have undergone significant changes, allowing the collection of data in real time, automation through intelligent systems, and interconnection of production environments. At the same time, one of the most popular approaches to continuous improvement is LSS (LSS), which focuses on the eradication of waste, the efficiency of the process, and the enhancement of quality. The combination of LSS and Industry 4.0 is a developing area of research, even though combining these two paradigms is complementary. This article includes a systematic literature review in which the combination of LSS practices and Industry 4.0 technologies, including the Internet of Things (IoT), artificial intelligence (AI), cyber-physical systems (CPS), and big-data analytics, is discussed. The literature review is based on recent publications published between 2018-2025 and relies on significant academic databases such as Scopus and the Web of Science. The results show that Industry 4.0 technologies significantly improve traditional LSS instruments such as value stream mapping, root cause analysis, and statistical process control owing to their ability to monitor reality in real time, predictive maintenance, and decision-making based on statistics. Nevertheless, integration has also brought up several challenges, including the resistance of the organization to digital transformation, the high cost of its implementation, the skill gap among employees, and cybersecurity issues. Through an overall summary of the available literature and industry case studies and analyses, this study suggests a template for integrating LSS approaches into Industry 4.0. The proposed framework provides viable recommendations for organizations planning to shift to intelligent, data-driven, and sustainable manufacturing systems.