International Journal of Materials and Product Technology (16 papers in press)

Special Issue on: OA The Application of Advanced Materials for Smart and Sustainable Manufacturing

• Open Access

  Adaptability of welding process parameters and defect control in ship cabin based on low-carbon steel-based materials ( Free Full-text Access )

  by Zhifeng Zhao, Shan Zhou 
  Abstract: Due to the mismatch between current welding process parameters for ship compartments and the metallurgical response of low-carbon steel-based materials, defects such as hot cracking and porosity are easily generated, leading to insufficient joint reliability. To address this issue, this paper first uses Gleeble thermal simulation software to calibrate the phase transformation kinetics and critical strain threshold for hot cracking of the material. Second, the aforementioned metallurgical response characteristics are embedded as constraints into a thermo-mechanical coupled finite element model to extract local thermal cycles and strain rate fields. Then, a Gaussian process surrogate model is constructed with the dual objectives of minimising the HTI (hot cracking tendency index) and maximising the equivalent strength. Finally, the optimal combination of process parameters is searched sequentially using the expected improvement criteria in Bayesian optimisation. Experimental verification shows that this method can achieve a median HTI of 0.18 for the weld, a tensile strength of 512 MPa at the weld centre, and effectively suppress inherent defects.
  Keywords: low-carbon steel; ship cabin welding; adaptive parameter optimisation; thermo-metallurgical-mechanical coupling; welding defect control.
  DOI: 10.1504/IJMPT.2026.10079471
   
  

• Open Access

  Intelligent prediction of geometric morphology of missing parts in Clarke ceramic tile based on 3D-GAE image autoencoder and precise completion in 3D printing ( Free Full-text Access )

  by Yibing Yang, Qiang Zhao 
  Abstract: In the restoration of Clark porcelain, in order to solve the problem of insufficient completion accuracy caused by incomplete geometric information and highly complex fragment shapes, this paper adopts an intelligent prediction method based on 3D graphic autoencoder (3D-GAE). This paper first constructs a point cloud structure of Clark porcelain fragments, integrating local adjacency and global symmetry constraints; then, designs a three-layer graph convolutional encoder to extract 128 dimensional latent vectors; and adopts a mask guided training strategy, combined with L2 loss and chamfer distance, to optimise the reconstruction of missing areas; finally, generates a watertight mesh model suitable for FDM printing through Poisson reconstruction, curvature smoothing, and wall thickness detection. The experiment showed that the proposed method achieved a CD of only 0.62 millimetres and an HD of only 2.87 millimetres in areas with high missed detection rates (70%), demonstrating high geometric fidelity.
  Keywords: Clark ceramics; 3D graph autoencoder; 3D-GAE; intelligent prediction; geometric completion; 3D printing; symmetry constraints.
  DOI: 10.1504/IJMPT.2026.10079715
   
  

• Open Access

  Visual communication morphology generation and algorithm control of photosensitive resins in UV curing process ( Free Full-text Access )

  by Boying Qie 
  Abstract: This paper proposes a temporal UV exposure path generation algorithm driven by visual semantic parameterised mapping for photosensitive resin curing. Structural rhythm, spatial hierarchy, and morphological tension are encoded into a continuous semantic parameter field and jointly modelled with UV exposure intensity, density coefficients, and priority factors, then converted into machine-executable exposure trajectories, establishing a direct pathway from visual semantics to physical curing instructions. Experimental results show that the proposed method achieves significantly higher morphological tension intensity (0.87) and rhythmic change rate (0.83) than static exposure strategies (0.42 and 0.35). Curing analysis further verifies effective characterisation of nonlinear resin curing behaviour, demonstrating a controllable and expressive morphology generation framework for digital cultural and visual communication applications.
  Keywords: visual semantic parameterisation; time-series UV exposure path generation; photosensitive resin curing; visual communication; closed-loop coupling mechanism.
  DOI: 10.1504/IJMPT.2026.10079716
   
  

• Open Access

  An algorithm for ranking environmentally friendly material floor plan design schemes integrating life cycle assessment ( Free Full-text Access )

  by Shuqin Zhang, Hongying Yan, Yanchi Fan, Yangyang Ma, Xizhan Tu 
  Abstract: Current evaluation of environmentally friendly material floor plan designs in exhibition design lacks a life cycle perspective considering design contexts such as service life and disassembly frequency, and subjective design preferences are difficult to effectively integrate with objective environmental data. To address this, this paper constructs a ranking algorithm framework for environmentally friendly material floor plan design schemes that integrates life cycle assessment (LCA) within the exhibition design context. This framework first constructs an LCA model covering the entire process of raw material acquisition, manufacturing, transportation, installation and dismantling, use and maintenance, and decommissioning, using each square metre/single exhibition period as a functional unit. The model explicitly incorporates assumptions about material lifespan and dismantling frequency to address the issue of traditional evaluations neglecting differences in usage stages. This method can provide a quantifiable, interpretable, and context- and uncertainty-tested decision support tool for selecting environmentally friendly materials in exhibition design.
  Keywords: environmentally friendly materials; scheme ranking; life cycle assessment; LCA; relative proximity; exhibition design; multi-criteria decision making; service life.
  DOI: 10.1504/IJMPT.2026.10079717
   
  

• Open Access

  Optimisation design of clothing 3D printing materials by combining machine learning and computer graphics ( Free Full-text Access )

  by Pei Wen, Yingqi Fu 
  Abstract: This paper addresses the challenges of poor forming accuracy, low design efficiency, and comfort in 3D clothing printing. It proposes an intelligent optimisation framework integrating computer graphics and machine learning, shifting from experience-driven to closed-loop optimisation. Implicit surface modelling and finite element simulation are used to extract geometric and physical features, while a deep neural network surrogate model enables rapid prediction of forming performance. The multi-objective Bayesian optimisation (MOBO) algorithm optimises lattice parameters and printing paths for comfort and material efficiency. A parametric-machine learning collaborative design framework supports large-scale parameter optimisation. Experimental results show high geometric forming accuracy, with average root mean square error (RMSE) values of 0.081 mm (shoulder), 0.092 mm (armhole), 0.073 mm (waistline), and 0.087 mm (hem). Average printing time is reduced to 85.3 minutes, and average material consumption is only 100.4 grams. The method successfully achieves multi-objective synergistic optimisation of forming quality, mechanical properties and wearing comfort.
  Keywords: machine learning; computer graphics; 3D printing of garments; implicit surface modelling; multi-objective Bayesian optimisation; MOBO; material forming accuracy.
  DOI: 10.1504/IJMPT.2026.10079855
   
  

• Open Access

  Zonal material design and performance evaluation of cooling protective clothing based on human body thermal moisture distribution map ( Free Full-text Access )

  by Xiaoxi Zhao, Feifei Song 
  Abstract: In the design of protective clothing for high-temperature operations, this paper addresses the issues of insufficient overall thermal comfort and redundant material functions caused by differences in local thermal and moisture loads on the human body. A zonal material design method based on measured thermal and moisture distribution maps is adopted. The core of this method is to map the measured human body thermal and moisture load distribution into a spatial demand map, and accordingly divide it into high, medium and low functional zones. For each zone, thermal conductivity and moisture permeability parameters are precisely matched to construct a differentiated material library. Zonal cutting is performed based on a three-dimensional human body model, and a seamless hot-pressing process is used to achieve multi-material integration. Furthermore, gradient transition rules between materials and a double-layer composite structure are defined to ensure interface thermal and moisture compatibility and dynamic wearing stability, forming a closed-loop design system from physiological data to product structure.
  Keywords: high-temperature work; cooling protective clothing; thermal and humidity distribution map; zoned materials; 3D human body model.
  DOI: 10.1504/IJMPT.2026.10079856
   
  

• Open Access

  Microstructure design of bionic clothing materials based on generative adversarial networks ( Free Full-text Access )

  by Fan Yang, Rui Zhang, Shangkui Yang 
  Abstract: This paper proposes a hierarchical bio-inspired conditional generative adversarial network (HB-cGAN) microstructure design method. By constructing a multi-scale structural database composed of natural biomimetic structures and clothing material microstructures, porosity, fibre orientation angle, and structural connectivity are parameterised and encoded. The method embeds structural parameter vectors and hierarchical structural encoding modules into the generator latent space, and utilises residual convolutional units and cross-scale feature fusion mechanisms to establish hierarchical relationships between nanoporous structures and microfibre networks. Simultaneously, a structural statistical consistency constraint layer is introduced into the discriminator to constrain pore distribution and topological connectivity features. The structure generation process is coupled with a material performance prediction network. By constructing a structure-property mapping loss function and adversarial loss for air permeability, thermal resistance, and tensile modulus, the latent space search process and target performance constraints are synchronously embedded in the network training.
  Keywords: hierarchical bio-constrained conditional generative adversarial network; bionic clothing materials; microstructure design; structure-property mapping; performance constraint.
  DOI: 10.1504/IJMPT.2026.10081475
   
  

• Open Access

  Optimisation model of 3D product interface design based on reinforcement learning algorithm ( Free Full-text Access )

  by Junjun Li, Jiajie Li, Pengcheng Ju, Binni Lin 
  Abstract: Current 3-D product interface design lacks dynamic adaptive optimisation for coordinating multiple experience objectives under high-dimensional continuous parameters and complex human-computer interaction feedback. This study develops a 3-D interface optimisation model based on a Markov decision process (MDP), vectorising spatial layout, hierarchy, and interaction controls as states and defining continuous actions for flexible interface adjustment. A multi-objective reward function integrates operational efficiency, misoperation penalties, and cognitive-load weighting. Deep reinforcement learning (DRL) updates the policy network through policy gradients, transforming interface design into dynamic, learnable decision-making. The model converged within 187 training rounds, with a reward standard deviation of 0.018 and an average cumulative reward of 0.91. User testing achieved 12.6 s average task time, 0.27 error rate, and 4.22/5 satisfaction, demonstrating algorithm stability, structural optimisation quality, and consistent user experience.
  Keywords: deep reinforcement learning; DRL; three-dimensional interface design; multi-objective optimisation; Markov Decision process modelling; human-computer interaction evaluation.
  DOI: 10.1504/IJMPT.2026.10081597
   
  

• Open Access

  Digital design and manufacturing of complex jewellery based on generative algorithms ( Free Full-text Access )

  by Xianting Qiao 
  Abstract: Traditional manual jewellery design struggles to efficiently generate complex topologies balancing aesthetics and mechanics. This paper presents an end-to-end generative framework to address this. A CVAE learns an organic latent space to generate surfaces from sketches or semantics, while a GAN with embedded mechanical evaluation enforces stress constraints to avoid high concentration. Progressive topology optimisation then performs lightweight hollowing to create a robust biomimetic skeleton. Finally, manufacturing semantic adaptation tailors the structure for DLP and LPBF printing. The proposed method significantly outperforms comparative approaches: volume error is < 2%, maximum stress remains below the materials allowable limit, minimum wall thickness compliance reaches 95.3%, and overall constraint satisfaction is 92.5%. Moreover, it produces highly hollowed organic morphologies with 40.6 connected domains and a complexity index of 9.8, verifying its comprehensive effectiveness in co-optimising complex generation, mechanical performance, and manufacturability.
  Keywords: generative algorithm; complex structural jewellery design; conditional variational autoencoder; topology optimisation; additive manufacturing adaptation.
  DOI: 10.1504/IJMPT.2026.10081617
   
  

• Open Access

  Material sensing and functional integration design of smart age-friendly home products ( Free Full-text Access )

  by Tianhui Liang, Lan Zou, Wenjie Yang 
  Abstract: This paper proposes an integrated design framework for age-friendly home products based on multimodal material perception to address declining tactile recognition and delayed interaction feedback in elderly users. A tactile material library is established, mapping physical properties to three bipolar labels (soft/hard, warm/cool, smooth/rough) for non-visual navigation. Event-driven vibration encoding converts home events into distinguishable tactile patterns embedded in high-contact areas such as handrails. A thin-film pressure sensor array combined with a sliding-window algorithm enables real-time centre-of-gravity tracking and pre-fall warning. A self-responsive emotional loop using shape-memory alloys and physiological sensors simulates companionship through slow deformation and colour changes. Experiments show 92.1% recognition accuracy for silicone-based composites, a fall false-negative rate of 0.5% (AUC 99%), and nighttime active contact of 17.2 min/h (vs. 10.8 min/h average); removing the emotional module reduces interaction time by ~30%, confirming its key role in companionship and user engagement.
  Keywords: intelligent age-friendly home products; multimodal material perception; tactile feedback encoding; pressure distribution perception; emotional interaction design.
  DOI: 10.1504/IJMPT.2026.10081618
   
  

• Open Access

  Transmittance enhancement technology and durability analysis of ultra-clear patterned glass for photovoltaic module encapsulation ( Free Full-text Access )

  by Ming Sun 
  Abstract: To improve the durability and transmittance stability of ultra-clear patterned glass for photovoltaic module encapsulation, this study develops a gradient corrosion-resistant antireflective layermicrostructure synergistic system. A SiO2/TiO2 gradient layer was fabricated via sol-gel deposition and annealed at 350, followed by PECVD grafting of a fluorosilane hydrophobic film and RIE-based regulation of microgroove aspect ratios (1.21.5). The microstructural evolution and optical degradation were investigated under damp heat, thermal cycling, and wind loading using FIB-SEM and in situ spectral characterisation. Results demonstrate that the gradient structure effectively mitigates thermal mismatch stress, while the hydrophobic layer enhances moisture resistance. The optimised sample maintains transmittance attenuation deviation within 0.17%, outperforming the control group. This work reveals the cross-scale failure mechanism linking interface chemistry, mesoscopic damage, and optical performance, providing a reliable strategy for durable photovoltaic packaging.
  Keywords: ultra-clear patterned glass; gradient antireflective coating; multi-stress coupled aging; light transmittance retention; interface failure.
  DOI: 10.1504/IJMPT.2026.10081703
   
  

• Open Access

  Thermal deformation and mechanical vibration response characteristics of high-pressure GIS cylinder materials ( Free Full-text Access )

  by Fei Gao, Gui Fu, Yan Lu, Juan Du, Aisheng Yang, Junkai Zhao, Xiaojun Guo 
  Abstract: A thermo-vibrational joint loading platform is established to investigate the thermally induced deformation and vibration response of high-pressure GIS cylinders under coupled environments. A sequential decoupling procedure quantifies the contributions of elastic modulus softening, thermal pre-stressing, and geometric stiffness changes to frequency drift. Eddy current displacement sensors extract axial constraint coefficients, and frequency response function analysis determines the first two natural frequencies and half-power bandwidths. At 60?C temperature rise, the net axial elongations of 5083 Al alloy, 6063-T6 Al alloy, and 304 SS are 1.213 mm, 1.218 mm, and 1.158 mm, with constraint coefficients of 0.72, 0.73, and 0.93, and damping ratios increasing from 0.0045, 0.0047, 0.0030 to 0.0060, 0.0062, and 0.0055, respectively. Thermal pre-stressing contributes 51% and 52% to the frequency drift of the two aluminium alloys, while modulus softening contributes 65% for 304 stainless steel, revealing distinct material-specific thermal-vibration response mechanisms.
  Keywords: high-pressure GIS cylinder; thermally induced deformation; mechanical vibration; frequency drift; softening of elastic modulus.
  DOI: 10.1504/IJMPT.2026.10081769
   
  

• Open Access

  Optimising financial dynamics in the supply chain of conjugated polymer thin film materials: a simulation model and algorithm approach ( Free Full-text Access )

  by Yiming Liu, Qingwei Li 
  Abstract: Conjugated polymer thin film materials are essential for photovoltaics, organic electronics, and sensors, yet the supply chain demands robust financial management. This study presents a simulation model and algorithm (SMA) to optimise financial dynamics throughout material procurement, distribution, manufacturing, and end-user stages. The model assesses transportation overheads, production expenses, procurement strategies, and pricing at each phase, integrating optimisation to guide decisions on lead times, demand variability, price trends, and production capacity. These elements enable adaptive financial strategies that enhance overall efficiency. Case study evaluations confirm that the SMA improves revenue generation, cost control, and financial resilience. By applying these tools, supply chain stakeholders can strengthen financial security and sustain the reliable availability of conjugated polymer thin film materials.
  Keywords: conjugated polymer thin film material; supply chain; financial dynamic systems; financial resilience; simulation model and algorithm; SMA; decision making.
  DOI: 10.1504/IJMPT.2026.10081881
   
  

• Open Access

  An end-to-end generation method from transformer-based clothing design sketches to 3D virtual garments ( Free Full-text Access )

  by Liwan Shangguan 
  Abstract: Traditional virtual garment reconstruction from 2D sketches suffers from fixed template constraints and unrealistic physical details. We propose an end-to-end transformer-based framework directly mapping 2D planar semantics to 3D spatial geometry. Our method utilises multi-scale spatial attention for global intent extraction, a cross-modal network, and an implicit signed distance field (SDF) to decode free-topology 3D geometries without pre-defined templates. Furthermore, we integrate explicit physics priors into a joint loss function to refine the mesh surface. Experiments demonstrate our method significantly outperforms mainstream models: it reduces chamfer distance (CD) by 23.5%, optimises Earth movers distance (EMD) by 18.2%, and stably controls mesh edge length error within 1.5 mm. This framework offers an efficient, template-free solution for digital apparel research and intelligent manufacturing.
  Keywords: cross-modal generation; transformer architecture; end-to-end generation; physics-informed prior; 3D garment reconstruction.
  DOI: 10.1504/IJMPT.2026.10081975
   
  

• Open Access

  Research on biosensing and feedback of innovative conjugated materials in sports training ( Free Full-text Access )

  by Yutong Yang, Xiaoming Liu 
  Abstract: This paper presents a biosensing framework that integrates conjugated materials, real-time biometric monitoring, and smart exercise equipment to enhance sports training. An innovative conjugated biosensing technology (ICBT) is proposed to build the BioFitTrainer (BFT) system. An electrochemical biosensing platform was constructed using conjugated material-modified electrodes, and physiological data including lactate, glucose, uric acid, and heart rate variability were collected from athletes during treadmill exercise. Sensor outputs were validated against standard laboratory assays using concurrent sweat and blood samples. The system provides individualised training plans and real-time feedback by continuously assessing metabolic stress, biomechanical load, and fatigue biomarkers. This approach supports adaptive exercise management, reduces the risk of overexertion, and enables safer, more efficient training regimens. The combination of high-conductivity conjugated materials and biosensing technology offers a promising direction for next-generation wearable sports-monitoring devices.
  Keywords: biosensing technology; conjugated materials; BioFitTrainer; BFT; innovative conjugated-biosensing technology; sports training.
  DOI: 10.1504/IJMPT.2026.10082127
   
  

Regular Issues

• Improving zinc-nickel coated bolt performance by optimising electroplating parameters to reduce hydrogen embrittlement  
  by Ebru Saraloğlu Güler, Mehmet Can Altıntaş 
  Abstract: Industries such as automotive and defence engage commonly employ surface enhancement processes to protect components from harsh environmental conditions. Among these, zinc-nickel electrodeposition is widely used due to its superior corrosion resistance. However, the long-term performance of such coatings can be compromised by hydrogen embrittlement, a critical issue that may lead to premature failure of structural components. This study investigates hydrogen embrittlement in zinc-nickel deposits by optimising the electroplating parameters including current density, temperature of the bath and immerison time in acid prior to deposition. Steel samples were electroplated under varying conditions and X-ray fluorescence (XRF) measurements were performed to determine the nickel and zinc content in the deposit. In addition, bolts that are commonly used fasteners in armoured vehicles are used as substrates. Thus, a coating set-up was established for bolt deposition using two anodes. After that the bolts were tested in accordance with EN ISO 15330 standards. Microstructural examinations were carried out after torque testing to assess coating integrity. Results revealed that a current density of 2.5 although coating temperature had no significant impact on hydrogen embrittlement, its influence became apparent with the change in coating thickness.
  Keywords: electroplating; zinc-nickel; hydrogen embrittlement; bolt; torque.
  DOI: 10.1504/IJMPT.2026.10080553