Rolling-element bearings remain a leading cause of unplanned downtime in industrial machinery, while vibration-based condition monitoring has matured, the post-2018 literature has converged on machine-learning classifiers whose interpretability layer remains restricted to expert analysts. This paper presents an operator-centred visualization framework supported by two time-frequency methods: the Zhao–Atlas–Marks Distribution (ZAMD), a Cohen’s-class representation with a cross-term-suppressing cone kernel, and Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), evaluated through its Hilbert spectral analysis output. Both methods produce two-dimensional time-frequency artefacts with a similar visual structure—impact-related energy bursts that recur at the characteristic fault frequencies—and are presented in side-by-side form for each fault class. A four-stage framework wraps either method with the characteristic fault frequencies (supplied as a comparison reference) and colour-coded, healthy baseline-referenced scaling. The framework is demonstrated on a laboratory bearing rig (KOYO 6302, 600 RPM) across inner-race, outer-race, and ball-spin fault classes. A preliminary readability assessment of annotated ZAMD-generated artefacts, with twelve non-specialist participants from a brewing and packaging industrial context, recorded 89.8% aggregate classification accuracy (194 of 216 trials) at a mean response time of 15.4 s. Because no label-free or alternative-format control conditions were included, this result characterises the annotated artefact as a whole and does not isolate the contribution of the time-frequency representation from that of the annotation layer; it is established for the ZAMD engine only. The two methods are compared as visualization engines—qualitatively, through the structure of their side-by-side time-frequency artefacts, and quantitatively, through computational cost—whereas the non-specialist readability assessment characterises the ZAMD-based framework specifically. CEEMDAN is positioned as a candidate alternative engine whose time-frequency output is shown to be structurally similar but whose operator readability has not been tested with human participants and is identified as future work.

The increasing demand for high-performance thermoplastics in functional and engineering applications has led to a rise in the research of their processability through additive manufacturing (AM) methods. Polysulfone (PSU) has emerged as an interesting candidate for use in advanced 3D printed components for its high thermal stability, excellent chemical resistance, and good mechanical strength. However, polysulfone has challenges in material extrusion (MEX) printing due to high processing temperature and melt viscosity. This work aims to improve quality metrics (surface quality, porosity, and dimensional accuracy) of PSU parts through processing the MEX process parameters. The experimental manufacturing and evaluation procedures were conducted based on the Taguchi L16 Robust design experimental model. The toolpath orientation angle (TA), nozzle heat level (NHL), deposition speed (DS), structure density (DS), and line width (LW) were selected as control variables. Their effect was evaluated against the average roughness - Ra, Root Mean Square roughness - Rq, Actual to Nominal Dimensional deviation at 95 % - A2N95, and Computed Tomography Scan porosity- PCT quality responses. TA had a remarkable effect on roughness responses, SD on dimensional deviation, and NHL on porosity. Data processing was implemented with the reduced quadratic regression model (RQRM) and the quadratic regression model (QRM), with RQRM being selected due to the higher F-values (R2 values were sufficient for three metrics, ∼65%, while in porosity they were high, ∼82%). The validity of the prediction functions was confirmed by two additional runs (>10% error), advancing the process understanding of HPPS in MEX AM.

The techniques used to produce material culture objects represent fundamental aspects of intangible cultural heritage. While they rely on tangible tools and materials, their execution—how these elements intertwine with the artisan’s craft to create the final product—is inherently intangible. Traditionally, such techniques are passed down through generations via oral instruction and long-term practical apprenticeship. However, they leave little physical evidence in the resulting objects. The study of construction techniques in handmade (non-wheel-thrown) pottery is a key focus of archaeological and ethnographic research. Handmade pottery retains a wealth of manufacturing traces within its vessel walls, many of which can be effectively revealed using modern, non-invasive imaging techniques such as X-ray μ-CT. By enabling direct visual analysis of 2D virtual thin sections and 3D representations, μ-CT reveals the precise morphology of construction units (e.g., coils, slabs) and their joins, as well as the shape, size, orientation, and distribution of pores and non-plastic inclusions within the ceramic fabric. This paper presents the results of μ-CT scanning on a pottery assemblage from Middle Neolithic Sesklo (6th millennium BC), addressing three key aspects: 1. The potential of visual analysis, 2. The methodological implications of μ-CT for studying ancient ceramics, and 3. The contribution of μ-CT analysis to documenting intangible cultural heritage through material remains.

Rolling-element bearings are a leading cause of unplanned downtime in continuous-process manufacturing, and the migration toward Industry 4.0 condition-based maintenance (CBM) has intensified the need for diagnostic methods evaluated on real in-service assets. This paper reports an exploratory, longitudinal single-asset industrial field demonstration of Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN)-based Hilbert spectral analysis for rolling-element bearing condition monitoring. An in-service bearing on a critical production machine was monitored over eight measurements spanning approximately four months and analyzed with the Hilbert–Huang Transform, using CEEMDAN in place of the classical Empirical Mode Decomposition to suppress mode mixing. The Hilbert spectra tracked the evolution of the bearing’s vibration signature as a growing concentration of vibration amplitude in a stable band of the 0–400 Hz analysis window (approximately 280–380 Hz); because the analysis characterizes the distribution of amplitude within the band rather than resolving discrete defect lines, this is reported as band-amplitude trending, and the band is treated as compatible with bearing-related excitation rather than attributed to a specific kinematic fault frequency. At the final pre-replacement measurement (M8), a bounded consistency cross-check against a conventional single-sided Fast Fourier Transform (FFT) amplitude spectrum computed from the same exported waveform recovered co-located dominant content. This establishes cross-method consistency for the measurement examined, but does not independently validate diagnostic correctness or the full campaign trend. Interpretability of the representation is argued qualitatively and remains to be tested. The contribution is therefore the documented field application and its explicit consistency cross-check procedure; applicability beyond this asset and its operating conditions requires multi-asset evaluation.

In three-dimensional (3D) printing, the produced components quality in terms of their surface roughness, porosity, and dimensional inaccuracy is a weak point, affecting also their performance. This is critical for Polyvinylidene fluoride (PVDF), a high-performance thermoplastic, which is capable of functional parts production, due to its high chemical and thermal resistance. This study investigates how six key process parameters, i.e., raster deposition angle, nozzle temperature, bed temperature, infill density, print speed, and layer height, affect the quality of PVDF MEX 3D‑printed components. A Taguchi L25 design of experiments combined with a regression model (reduced quadratic) was employed for the analysis of the experimental data. Through the optimization process followed, surface roughness and dimensional accuracy improved by roughly 30

This work takes a step toward Automated Machine Learning (AutoML) by moving beyond the traditional black-box approach often seen in engineering applications of neural networks. The primary objective is to advance the understanding of Neural Architecture Search (NAS), with a specific emphasis on using Bayesian Optimization (BO) with conditional constraints to fine-tune neural archi-tectures and hyperparameters. To the best of our knowledge, NAS frameworks remain underexplored in engineering contexts, where neural architectures are typically developed through ad hoc, manually crafted designs. In this context, Long Short-Term Memory (LSTM) recurrent neural architectures for predicting highly nonlinear delamination growth patterns in composite laminates are automatically designed using NAS. The sequential data, such as nonlinear degradation with increasing load, are generated through finite element analysis (FEA). The automated selection process produces neural network models specifically tailored to capture the complex, history-dependent nature of delamination growth. The study investigates both shallow and progressively deeper architectures through extensive systematic experiments, statistical analysis and dimensionality reduction techniques, offering insights into the inner workings of BO in optimizing neural architectures. NAS results reveal that the objective function in BO can be high irregular with no clearly defined local minima, and that deeper neural networks are not universally superior to shallower ones. Additionally, random search is used as a natural baseline to judge BO performance. The findings underscore the effectiveness of LSTM in modeling sequence-to-sequence relationships in composite damage evolution, and exhibit strong potential as surrogate models, delivering near real-time predictions, a capability challenging to achieve with traditional FEA.

In structural and biomedical applications, where high-speed loading is a major concern for mechanical reliability, high-performance polymers (HPP) are increasingly required. Still, their thermomechanical response produced by additive manufacturing methods has not been fully characterized. In this study, the strain-rate-dependent compressive behavior of specimens, 3D printed by the material extrusion (MEX) method with the polyether ether ketone (PEEK) biopolymer, has been investigated. Simultaneously, the evolution of specimen temperature has been monitored using an infrared camera. Compression testing was performed over a range of test speeds up to 200 mm/min. The aim was to quantify the mechanical performance under compression loads and concurrently thermal self-heating phenomena as a function of applied strain rate. Results show that the compressive strength of the MEX-processed PEEK had a positive strain-rate sensitivity of 14.1% at higher strain rates. The strain-rate sensitivity index had higher values at lower test velocities, suggesting that viscoelastic effects play a larger role in the deformation mechanism. At the same time, the maximum specimen temperature increased by 33% as the strain rate increased (63 to 85 °C) (thermomechanical self-heating), which can affect material response at elevated strain rates. This work offers fundamental insights into the mechanical response of MEX-fabricated PEEK biopolymer under loads applied at various speeds and has direct merit for the design of additively manufactured biomedical components which are often subjected to such loading conditions.

In Fused Filament Fabrication (FFF) 3D printing, surface quality, dimensional accuracy, and porosity are often reasons for negative criticism affecting the quality of 3D printed parts. These parameters affect the mechanical performance of the parts, too. This is more intense for high-performance polymers (HPPs), such as polyimide (PI), owing to their high cost and demanding operating conditions, in addition to their printing challenges. In this study, a Taguchi L16 orthogonal array was formed to evaluate the influence of nozzle temperature, printing speed, Raster Orientation (RO), infill density, and Strand Width (SW) on these quality responses. The aim of this study was to determine the optimal set of parameters. All metrics were notably improved (porosity was reduced by approximately 60

The use of 3D printed components often requires materials with good resistance to weather conditions, which is usually a weak point of thermoplastics. Acrylonitrile styrene acrylate (ASA) is a thermoplastic capable of withstanding extreme environmental conditions. Thus, its investigation is of great importance, especially when it comes to being processed by additive manufacturing (AM). Herein, novel ceramic nanocomposites loaded with silicon carbide (SiC) in five (5) different filler quantities (2.0–10.0 wt%) were prepared and characterized, aiming to be used in lightweight automotive interior and exterior parts, customized aerospace, electrical and electronic housing, or tooling and fixtures. Matrix-filler mixtures were extruded into filaments to 3D print a series of specimens. The samples underwent tests for their performance during thermal, rheological, (thermo-) mechanical, structural, and morphological examinations. 6.0 wt% was distinguished overall for its performance over ASA pure, showing improvement in tensile strength (15.1%), modulus of elasticity (14.1%), flexural strength (16.8%), and modulus of elasticity (19.1%). 8.0 wt% had the highest tensile toughness (13.8%) and flexural toughness (10.6%). Porosity was reduced by the addition of the SiC filler (10.0 wt% loading improved porosity by 36.5%). In total, the ASA/ SiC nanocomposites have great potential to be utilized for outdoor applications requiring 3D printed components with reinforced mechanical performance.

Dual-material additive manufacturing enables the design of cellular structures with a tailored mechanical response through controlled material distribution and interfacial architecture. In this research, honeycomb structures fabricated by Fused Filament Fabrication (FFF) using dual-material TPU/PLA configurations have been systematically investigated. Particular emphasis is placed on interlocking TPU/PLA joint designs, implemented through tau-shaped and teeth-based geometries, to evaluate their role in load transfer and structural performance. An experimental-analytical model has been developed to characterize the compressive force-displacement response of dual-material honeycombs, capturing the three characteristic deformation regimes-initial stiffness, progressive collapse, and densification-while linking the effective stiffness to the underlying beam-lattice mechanics. The relative contributions of axial and bending deformation mechanisms are quantified through a comparative beam element approach, introducing dimensionless coefficients that reflect the governing deformation mode. The results reveal that the mechanical response is bending-dominated for the examined configurations. The configuration with PLA at the nodes and TPU at the struts exhibits a higher load-carrying capacity and a more stable collapse regime due to a more balanced axial-bending interaction. In contrast, alternative material distributions lead to earlier instability and reduced structural efficiency. The proposed analytical model demonstrates excellent agreement with the experimental data across all configurations. The results demonstrate that properly designed dual-material interlocks can enhance load transfer, decrease stress concentrations, and refine the overall mechanical performance of lightweight cellular structures.

Polycarbonate (PC) is a widely used thermoplastic. Therefore, the amount of waste produced is notable. The exploitation of such waste is of great interest nowadays in the industry and academic society, due to its contribution to environmental pollution and other negative consequences. Herein, the possibility of using PC scrap as a raw material in 3D printing (material extrusion - MEX) is reported. The efficacy of the PC polymer after six thermomechanical courses was evaluated. The effect on rheology, mechanical performance, and thermal behavior is reported. The morphological characteristics were also assessed through scanning electron microscopy, while two quality metrics, i.e., geometrical accuracy and 3D printing structure porosity of the parts, were investigated through micro-computed tomography. The findings were correlated to report the impact of thermomechanical processing on the PC polymer properties. A 9% tensile strength increase compared to the virgin polymer is reported (third round), while the flexural strength was improved by 14% (second round). Then the strength declined. It was lower than that of the virgin material on the sixth thermomechanical repetition. The findings showed that the life of PC can be extended through thermomechanical recycling for 3D printing applications.

To promote environmental sustainability, this research investigated the potential of utilizing recycled polymethylmethacrylate (PMMA) as raw material in material extrusion (MEX) additive manufacturing (AM). To enhance its mechanical response, carbon black (CB) was employed as the filler in nanocomposite formation. Filament extrusion of the mixture at different concentrations produced printable feedstocks for MEX AM. Rheological analysis (viscosity and material flow rate) showed that the CB introduction to the matrix was beneficial for consistent layer deposition, while differential scanning calorimetry and thermogravimetric analyses verified the thermal stability of the nanocomposites during processing. Mechanical properties were optimized, with increases in modulus (27.8% and 25.8%, respectively, in tensile and bending loadings) and tensile strength at optimal CB loadings. Dynamic mechanical analysis revealed the viscoelastic response of the nanocomposites. Raman and energy dispersive spectroscopy provided element-related insights. Surface morphology and parts structure were observed employing scanning electron microscopy and micro-computed tomography, respectively, revealing a positive impact on the AM parts due to the CB presence in the nanocomposites. The 4 wt.% in CB content nanocomposite was the optimum one. This research pioneers the development of new sustainable nanocomposite filaments and highlights the potential of next-generation MEX-based AM.

Recyclability has emerged as a critical aspect in the pursuit of sustainability, waste reduction, and the circular economy, resulting in its extensive adoption in both research and industry. Material extrusion (MEX) additive manufacturing (AM) technologies present considerable potential for capitalizing on this trend by employing recycled polymers, such as polymethylmethacrylate (PMMA), as evidenced herein. The integration of ceramic fillers, such as titanium carbide (TiC), utilized in this research, provides a method to enhance the mechanical properties of recycled PMMA and expand its use. Three-dimensional printed samples were fabricated in appropriate forms to evaluate their mechanical strength (tensile, flexural, impact, Dynamic Mechanical Analysis, microhardness), structure, and morphology (computed tomography and scanning electron microscopy). Thermal and rheological properties were assessed to evaluate the TiC's impact on the properties of PMMA. Filler concentrations varied from 0.0 - 8.0 wt. %. The 2.0 wt. % composite demonstrated the most favorable impact on strength (28.5% in tensile and 27.3% in flexural loadings), whereas the 8.0 wt. % composite exhibited the greatest enhancement in stiffness (47.2% Young’s modulus increase). New nanocomposites for MEX AM are presented, promoting sustainability and enhancing mechanical performance at the same time.

The integration of additive manufacturing (AM) with advanced joining techniques presents potential for high-performance polymeric structures. This research investigates the performance and feasibility of friction stir welding (FSW) applied to three-dimensional (3D)-printed acrylonitrile styrene acrylate (ASA) sheets, a thermoplastic renowned for its enhanced weatherability and mechanical stability. A two-stage process was followed for maximum optimization performance, featuring initially nine screening runs (Taguchi L9) and then a full factorial (FF) analysis (six runs), with adjusted values for travel speed (TS), rotation speed (RS), shoulder diameter (SD) and pin diameter (PD). The response metrics were tensile and bending strength and stiffness, force on the three axes, welding temperature, and mu (Fy/Fz) index. The welding efficiency exceeded 100%. The tensile and the flexural strength were 44.1 MPa and 86.5 MPa, respectively (improved similar to 35% after optimization). The highest strength (FF analysis) was achieved with TS = 3 mm/min and RS = 1800 rpm, with a welding tool with SD = 10 mm and PD = 4 mm. The reduced quadratic regression model (RQRM) proved more accurate than the linear regression model (LRM). The prediction formulas were evaluated with a confirmation run. The derived results underscore FSW as an effective process for constructing advanced ASA-based architectures, offering new possibilities in various engineering fields.

Polyethylene terephthalate glycol (PETG) is an amorphous polymer that has been widely used in numerous applications, from everyday life to medical and even defense-related applications. The latter constitute very demanding environments in which, in many cases, specific multifunctionalities are required. Herein, we aim for specific functionalities to appear simultaneously, thus creating novel materials that can provide important solutions to applications. Therefore, inducing antibacterial properties along with enhanced mechanical properties for use in the defense and security domains constitutes an additional asset when disease spread becomes very important. To address this challenge, we mixed pure PETG with an antibacterial nanopowder to investigate these novel multifunctionalities in detail. Concomitantly, the enhancement of the mechanical properties of the 3D printed PETG/antibacterial nanocomposites was thoroughly examined. Several PETG nanocomposites were manufactured with different nanopowder loadings and turned into filaments for use in the AM method of material extrusion (MEX). The several 3D printed PETG/antibacterial nanocomposites were thoroughly investigated for their mechanical and rheological properties, thermal stability, and morphological, structural, and chemical characteristics, combined with antibacterial performance, against two common pathogens, s. aureus and e. coli, using the agar well diffusion method. The outcome of the nanopowder introduction to the quality metrics of the 3D printed PETG, namely the geometrical accuracy and pores of the 3D printed structure was also investigated through high-resolution micro-computed tomography. The PETG/antibacterial nanocomposites exhibited improved mechanical properties. A 13.6 % tensile strength increase was achieved with 8 wt% content. 10 wt % achieved 17 % Young's modulus increase, 19 % flexural strength and 18.2 % flexural modulus improvement and can be considered the optimum loading of the research. Nanocompounds also showed strong antibacterial activity against s. aureus and E. coli. These induced multifunctionalities can constitute a new class of materials where the desired properties can have significant applications in two or more different fields for functional, durable, and infection-resistant materials, such as in the demanding defense and security sector, the medical field, or both.

Polyhydroxyalkanoate (PHA) is a biopolymer that can be 3D printed using the material extrusion method. Nevertheless, their mechanical properties are inferior to those of petroleum-derived polymers, which restricts their broader application. Herein, nanobiocomposites comprising naturally sourced PHA and cellulose nanocrystals (CNC) as fillers were successfully synthesized. These nanobiocomposites were prepared with filler concentrations ranging from 0.5 to 2.5% by weight, in increments of 0.5 wt %. Filaments were produced from binary PHA/CNC mixtures and subsequently employed for 3D printing of the respective nanobiocomposite samples. They were subjected to mechanical, rheological, thermal, and structural analyses using analytical techniques and the respective standards. The integration of CNC into the pure PHA polymer matrix has been reported to enhance PHA's mechanical properties of PHA, with an increase in flexural strength by 23.3%, flexural modulus by 20.8%, and Young's modulus by 47.3%, although there was a reduction in impact strength and microhardness. Morphological characterization confirmed the homogeneous dispersion of CNC, whereas the thermal and rheological properties remained almost unchanged. The porosity and geometric accuracy of the 3D-printed samples, evaluated using micro-CT, were improved by incorporating CNC into the PHA matrix. The 0.5 wt % CNC concentration was the optimum one, improving mechanical and quality metrics. These findings highlight the potential of PHA/CNC nanocomposites as innovative high-performance biodegradable materials for 3D printing for biomedical, packaging, and structural engineering applications. Such nanobiocomposites can contribute to reducing the environmental impact of petroleum polymers through a cost-effective additive manufacturing method.

The production of polymeric welds remains a challenging task. This study aimed to investigate the feasibility and performance of joining two popular thermoplastics, acrylonitrile styrene acrylate (ASA) and acrylonitrile butadiene styrene (ABS), with differing compositions and thermal behaviors, produced via additive manufacturing, which adds an additional challenge to the process. In the first optimization stage, a Taguchi L9 analysis for screening was performed, while the second utilized a Full Factorial design. The response metrics were the tensile and flexural strengths and stiffness, forces along the three axes, welding temperature, and friction index. The initial phase involved four control parameters: travel (TS) and rotation (Rs) speeds, and the shoulder and pin diameter of the welding tool (nine experimental runs). In the second phase, the optimal welding tool was employed, and Ts and Rs were tested (six experimental runs). The most notable performance was observed with a 3 mm/min Ts, 1800 rpm Rs, and a welding tool with a 10 mm/4 mm diameter geometry. The regression models assessed included the Linear Regression Model and the Reduced Quadratic Regression Model, achieving R2 values higher than 99 %. A welding efficiency (ratio of the welded and unwelded parts tensile strength) of greater than 100 % was achieved. The welding conditions sensitivity was revealed as more than 200 % improvement in the mechanical properties was reported (33MPa/14 MPa tensile, and 76.1MPa/37.6 MPa flexural strength, respectively). These findings provide valuable insights for scientific and industrial communities by offering intriguing and applicable information for future applications in challenging environments.

In light of the extensive application of polymeric materials in additive manufacturing (AM) via three-dimensional (3D) printing technology, there is an increasing need to explore and employ materials with diminished environmental impacts. Polyhydroxyalkanoates (PHAs) represent such materials; however, despite their promising potential, they remain underutilized and insufficiently examined within the AM industry. Furthermore, the quality of AM parts is always a weak point, although it has been proven to affect their performance. This study examines 3D-printed coupons fabricated from PHAs and their optimization of quality metrics through the Taguchi L9 experimental design. The selected printing control parameters included feed rate, layer thickness, nozzle heater temperature, and strand width. The quality-related response metrics comprised geometrical accuracy, porosity (through micro-computed tomography), and mean and maximum surface roughness. The thermal and rheological behavior of the material were assessed along with optical microscopy. Nozzle heater temperature was identified as exerting the most significant influence on the majority of the responses. Optimization improved geometrical accuracy by 23

Additive Manufacturing (AM) can provide customized parts that conventional techniques fail to deliver. One important parameter in AM is the quality of the parts, as a result of the material extrusion 3D printing (3D-P) procedure. This can be very important in defense-related applications, where optimum performance needs to be guaranteed. The quality of the Polyetherimide 3D-P specimens was examined by considering six control parameters, namely, infill percentage, layer height, deposition angle, travel speed, nozzle, and bed temperature. The quality indicators were the root mean square (Rq) and average (Ra) roughness, porosity, and the actual to nominal dimensional deviation. The examination was performed with optical profilometry, optical microscopy, and micro-computed tomography scanning. The Taguchi design of experiments was applied, with twenty-five runs, five levels for each control parameter, on five replicas. Two additional confirmation runs were conducted, to ensure reliability. Prediction equations were constructed to express the quality indicators in terms of the control parameters. Three modeling approaches were applied to the experimental data, to compare their efficiency, i.e., Linear Regression Model (LRM), Reduced Quadratic Regression Model, and Quadratic Regression Model (QRM). QRM was the most accurate one, still the differences were not high even considering the simpler LRM model.

The extensive utilization of materials in daily life can result in environmental pollution, resource depletion, and numerous other consequences. One approach to mitigate these issues could be adopting recycling techniques for high-consumption materials. In this investigation, Acrylic Styrene Acrylonitrile (ASA) was selected for examination under six successive recycling cycles to assess its performance and suitability for material extrusion (MEX) 3D printing. This material was used for filament fabrication and three-dimensional (3D) specimen printing. Thermomechanical processes are expected to influence the behavior of the ASA samples. Furthermore, mechanical, thermal, and rheological tests were conducted, and the morphology and structure of the parts were investigated. The mechanical properties of the filaments were also assessed. Morphological and structural analyses were performed by micro-computed tomography and scanning electron microscopy. The second recycling cycle samples were notable for their performance relative to the first cycle (20% higher tensile strength), whereas all five cycles demonstrated higher strength than the first cycle. These results indicate the promising potential of using recycled ASA in applications for various purposes.