Search Results (553)

This study presents a gradient-based topology optimisation framework for heat sinks embedded with phase-change material (PCM) that targets the mitigation of temperature oscillations under cyclic thermal loads. The approach couples transient thermal diffusion modelling in FEniCS with automatic adjoint sensitivities and GCMMA, and uses a simple analytical homogenisation to parametrise a composite of PCM and conductive material. With latent-heat buffering using PCM, the optimised layouts reduce the temperature variance by 41% when the full time history is used and by 32% when only the quasi-steady-state cycle is used. To improve physical manufacturability, explicit penalisation yields near-discrete designs with only ∼10% performance loss, preserving most oscillation reduction benefits. The results demonstrate that adjoint-driven PCM topology optimisation can systematically suppress thermal oscillations. Full article

Composite laminates designed for additive manufacturing require efficient material distribution to minimize weight while maintaining structural integrity. Traditional topology optimization methods, however, produce continuous density fields incompatible with layer-based fabrication. This work presents Selective Reinforcement Optimization (SRO), a stress-driven methodology that converts uniformly loaded laminate layers into localized reinforcement regions, or “patches”, at critical stress concentrations. The approach employs layer-wise statistical analysis of Tsai–Wu failure indices to identify high-variance layers; applies DBSCAN clustering to extract spatially coherent stress regions while rejecting artificial concentrators; and generates CAD-compatible and manufacturing-ready boundary geometries through a custom concave hull algorithm. The method operates iteratively in dual modes: lightweighting progressively removes full layers and replaces them with localized regions when the structure is safe, while strengthening adds reinforcement without layer removal when failure criteria are approached. Case studies demonstrate weight reductions of 10–30% while maintaining failure indices below unity, with typical convergence achieved within 100 iterations. Unlike classical topology optimization, which requires extensive post-processing, SRO directly outputs discrete patch geometries compatible with composite additive manufacturing, offering a computationally efficient and production-oriented framework for the automated design of layered composite structures. Full article

Large-Format Additive Manufacturing (LFAM) offers the ability to 3D print composites at multi-meter scale and high throughput by utilizing a screw-based extrusion system that is compatible with pelletized feedstock. As such, LFAM systems like the Big Area Additive Manufacturing (BAAM) system provide a pathway for incorporating AM techniques into industry-scale production. Despite significant growth in LFAM techniques and usage in recent years, typical Multi-Material (MM) techniques induce weak points at discrete material boundaries and encounter a higher frequency of delamination failures. A novel dual-hopper configuration was developed for the BAAM platform to enable in situ switching between material feedstocks that creates a graded transition region in the printed part. This research studied the influence of extrusion screw speed, component design, transition direction, and material viscosity on the transition behavior. Material transitions were monitored using compositional analysis as a function of extruded volume and modeled using a standard Weibull cumulative distribution function (CDF). Screw speed had a negligible influence on transition behavior, but averaging the Weibull CDF parameters of transitions printed using the same configurations demonstrated that designs intended to improve mixing increased the size of the blended material region. Further investigation showed that the relative difference and change in complex viscosity influenced the size of the blended region. These results indicate that tunable properties and material transitions can be achieved through selection and modification of composite feedstocks and their complex viscosities. Full article

The use of vibrational transport for granular materials has significantly increased in the technological industry due to its reliability, operational efficiency, cost-effectiveness, and relatively uncomplicated technological setup. These transportation methods typically utilize various forms of asymmetry, such as kinematic, temporal (time), wave, and power asymmetry, to induce controlled motion on oscillating surfaces. This study analyses the motion of the granular materials on an inclined plane, where the central innovation lies in the creation of an additional system asymmetry of frictional conditions that enables the granular materials to move upward. This asymmetry is created by introducing dry friction dynamic manipulations. A mathematical model has been developed to describe the motion of particles under these conditions. The modelling results proved that in an inclined transportation system, the asymmetry of frictional conditions during the oscillation cycle—created through dynamic dry friction manipulations—generates a net frictional force exceeding the gravitational force, thereby enabling the upward movement of granular particles. Additionally, the findings highlighted the key control parameters governing the motion of granular particles. λ, which represents the segment of the sinusoidal period over which the friction is dynamically louvered, serves as a parameter that controls the velocity of a moving particle on an inclined surface. The phase shift ϕ serves as a parameter that controls the direction of the particle’s motion at various inclination angles. Experimental investigations were conducted to assess the practicality of this method. The experimental results confirmed that the granular particles can be transported upward along the inclined surface with an inclination angle of up to 6 degrees, as well as provided both qualitative and quantitative validation of the model by illustrating that motion parameters exhibit comparable responses to the control parameters, and strongly agree with the theoretical findings. The primary advantage of the proposed vibrational transport method is the capacity for precise control of both the direction and velocity of granular particle transport using relatively simple mechanical setups. This method offers mechanical simplicity, low cost, and high reliability. It is well-suited to assembly line and manufacturing environments, as well as to industries involved in the processing and handling of granular materials, where controlled transport, repositioning, or recirculation of granular materials or small discrete components is required. Full article

Improving responsiveness and efficiency in production systems requires an understanding of how manufacturing flexibility and inventory management interact under conditions of uncertainty. This study examines the combined effect of four types of flexibility, machine, labor, routing, and volume, together with the use of buffers, on the total flow time of production batches. A total of 84 experimental configurations were simulated, of which 35 were feasible and statistically valid, using a discrete-event simulation model developed in Arena and validated with industrial data. The results show that combining high machine and labor flexibility reduces total flow time from 5450 to 3050 min (a 44% decrease), whereas routing and volume flexibility exhibit minor effects. Moreover, the inclusion of buffers further improves performance, reducing times by approximately 1000 min in low-flexibility configurations. These findings provide robust quantitative evidence to guide the design of adaptive production systems by jointly evaluating the flexibility and inventory management dimensions that are typically studied in isolation. Full article

This article proposes a framework for the development of process digital twins (DTs) in the shipbuilding sector, based on the ISO 23247 standard and structured around the achievement of three levels of digital maturity. The framework is demonstrated through a real pilot cell developed at the Innovation and Robotics Center of NAVANTIA—Ferrol shipyard, incorporating various cutting-edge technologies such as robotics, artificial intelligence, automated welding, computer vision, visual inspection, and autonomous vehicles for the manufacturing of minor pre-assembly components. Additionally, the study highlights the crucial role of discrete event simulation (DES) in adapting traditional methodologies to meet the requirements of Process digital twins. By addressing these challenges, the research contributes to bridging the gap in the current state of the art regarding the development and implementation of Process digital twins in the naval sector. Full article

Raw material price volatility, freight rates, and foreign exchange all pose significant uncertainty for lithium-ion battery manufacturers, jeopardising procurement planning and financial stability. In this paper, we formulate a discrete Conditional Value-at-Risk (CVaR) optimisation model to design implementable robust hedging strategies for multi-factor cost exposure. Unlike conventional continuous hedge models, which are often severely parameter-sensitive and require frequent rebalancing, the discrete approach takes hedge ratios to be fixed at a finite implementable grid (0%, 50%, 100%) and simultaneously minimises the expected cost and tail risk. We conduct two case studies: the first evaluates the model behaviour under stochastic price shocks using a multi-market simulation data set, and the second subjects the model to stress testing on correlation drift and tail amplification in order to examine systemic robustness. Our results show that, compared with an OLS-based hedge or a fully hedged benchmark, the discrete CVaR framework yields smoother hedge patterns, lower tail losses, and improved liquidity stability; in addition, our results indicate that, when combined with tail-risk penalisation, decision discretisation can endogenously confer robustness to the industrial procurement horizon. This work contributes to the stochastic optimisation literature and provides a practical tool for mitigating volatility in the global lithium supply chain. Full article

Equipment failure is the leading cause of industrial operational disruption, with unplanned downtime accounting for up to 11% of manufacturing revenue, highlighting the need for effective proactive maintenance strategies, such as protective sensors that can detect potential failures in critical equipment before a functional failure occurs. However, sensors are also subject to hidden failures themselves, requiring periodic failure-finding inspections. This study proposes a novel integrated multimethodological approach combining discrete event simulation, Monte Carlo, optimization, risk analysis, and multicriteria decision analysis methods to determine the optimal inspection period for protective sensors subject to hidden failures. Unlike traditional single-objective models, this approach evaluates alternative inspection periods based on their risk-informed overall values, considering multiple conflicting key performance indicators, such as maintenance costs and equipment availability. The optimal inspection period is then selected considering uncertainties and the intertemporal, intra-criterion, and inter-criteria preferences of the organization. The approach is demonstrated through a case study at the leading Portuguese electric utility, replacing previous empirical inspection standards that did not consider economic costs and uncertainties, supported by an open, transparent, auditable, and user-friendly decision support system implemented in Microsoft Excel using only built-in functions and modeled based on the principles of probability management. The results identified an optimal inspection period of 90 h, representing a risk-informed compromise distinct from the 120 h interval suggested by cost minimization alone, highlighting the importance of integrating organizational preferences into the decision process. A sensitivity analysis confirmed the robustness of this solution, maintaining validity even as the organizational weight for equipment availability ranged between 35% and 82%. The case study shows that the proposed approach enables the identification of inspection intervals that lead to quantitatively better maintenance cost and availability outcomes compared to empirical inspection standards. Full article

Population initialization significantly influences metaheuristic algorithm performance, yet random initialization dominates despite problem-specific needs. This study investigates initialization strategies for the Job Shop Scheduling Problem (JSSP), an NP-hard combinatorial optimization challenge in manufacturing systems, addressing the gap in understanding how different initialization approaches affect solution quality and reliability in constrained discrete problems. The research employs a two-phase experimental design using Particle Swarm Optimization (PSO) on Taillard benchmark instances. Phase 1 evaluates seventeen initialization methods across four categories: random-based, problem-specific heuristics, hybrid methods, and adaptive strategies. Each method is tested through 30 independent runs on six problem instances. Phase 2 develops machine learning-enhanced initialization using variational autoencoders (VAEs) trained on high-quality solutions from successful traditional methods, comparing three VAE variants against conventional approaches. Results show all random-based methods failed completely, while only First-In-First-Out and Most Work Remaining heuristics succeeded consistently among traditional approaches. VAE-based methods achieved 100% solution validity (540/540) versus 97% for traditional methods (349/360), with statistical significance (χ² = 14.27, p < 0.001). The Friedman test confirmed performance differences (χ² = 19.87, p < 0.001, Kendall’s W = 0.828), with VAE methods achieving lower mean ranks and makespan reductions. Despite starting with lower initial diversity, VAE methods exhibited larger diversity increases during optimization, suggesting structured initialization enables more effective exploration than random dispersion. Full article

Welding deformation in large thick-plate structures can severely compromise manufacturing quality, making its prediction and control a critical engineering challenge. This study focuses on an engineering vehicle frame, for which a finite element model was developed to investigate the effects of welding sequence and direction on residual deformation, using a local-global strain mapping approach. Thermo-elasto-plastic simulations were first performed on T-joints and fillet joints to extract local plastic strains, which were subsequently mapped onto the global elastic model to compute overall structural deformation. The simulation results showed good agreement with experimental measurements, with a deviation of approximately 5.6%, confirming the reliability of the proposed method for predicting welding-induced deformation and stress in complex assemblies. To further optimize the welding strategy, a surrogate model was constructed based on Design of Experiments (DOE), and a Discrete Particle Swarm Optimization (DPSO) algorithm was employed. The optimized welding sequence and direction reduced the maximum deformation by 43%, while significantly lowering computational cost without sacrificing accuracy. This integrated approach offers valuable guidance for welding process design in engineering vehicle frames and other large welded structures. Full article

This study introduces a model-free reinforcement learning framework based on Q-Learning (QLA) for the multi-objective optimization of Selective Laser Melting (SLM) process parameters for Inconel 718. To efficiently handle the limited experimental dataset, a tabular Q-Learning approach was implemented, in which each parameter combination was treated as a discrete state and every possible transition as an action. Four key process variables laser power (P), scan speed (S), layer thickness (T), and hatch spacing (H) were optimized for two output responses: relative density (RD) and Vickers hardness (VH). The Q-Learning agent iteratively explored various parameter combinations, observed the resulting material properties, and continuously updated its policy to converge toward optimal conditions. The optimal parameter set identified by the framework was P = 270 W, S = 800 mm/s, H = 0.1 mm, and T = 0.08 mm. Despite relying on only 16 experimental trials, the model achieved exceptionally low prediction errors of 0.0503% for RD and 0.0857% for VH, demonstrating substantial reductions in both experimental effort and material consumption. The results confirm that reinforcement learning can autonomously and effectively identify optimal SLM parameter settings, highlighting its strong potential to enhance precision, efficiency, and overall quality in the additive manufacturing of metallic components. Full article

This work presents a highly compact and scalable 1.2-kV SiC MOSFET half-bridge building-block module enabled by a die-integrated 3D PCB packaging technology. Compared with conventional DBC-based or TO-247-based SiC half-bridge modules, the proposed design reduces the physical volume and weight by more than 90% while maintaining full compatibility with standard PCB manufacturing processes. The vertically laminated DC+/DC− conductors and symmetric PCB–die–PCB stack establish a tightly confined commutation loop, resulting in a measured power-loop inductance of 2.2 nH and a 3.8 nH gate-loop inductance—representing up to 94% and 89% reduction relative to discrete device implementations. Because the parasitic parameters are intrinsically well-balanced across replicated units and the mutual inductance between adjacent modules remains extremely small, the structure naturally supports current sharing during parallel operation. Thermal and insulation evaluations further confirm the suitability of copper filling via high-Tg laminated PCB substrates for high-power SiC applications, achieving withstand voltages exceeding twice the rated bus voltage. The proposed module is experimentally validated through finite-element parasitic extraction and 950 V double-pulse testing, demonstrating controlled dv/dt behavior and robust switching performance. This work establishes a manufacturable and parallel-friendly packaging approach for high-density SiC power conversion systems. Full article

The actual problem in manufacturing functionally graded materials (FGMs) produced in the laser powder bed fusion (LPBF) process remains the controllability of the materials gradient and the properties gradient of the final product. The key element in gradient formation is the delivery system in conjunction with the properties of the powder materials. This paper presents the first preliminary stage of the study, an application of a model based on the discrete element method to simulate several powder delivery systems and the analysis of the results obtained. Two designs of LPBF machine constructions with one and two movable platforms are simulated with and without separation walls. The variants of initial powder material separation were modeled along the longitudinal axis, inclined, and periodic lines. The powder material of the same or different densities and particle sizes was analyzed. The mean diameters of the powder particles in simulations are 0.78 and 0.6 mm, and the ratio of the material densities is 1.0 or 1.5. The 15 multi-stage delivery processes were simulated. The influence of various constructive and material parameters on the segregation (percolation) process and final distribution of powder materials was analyzed. It is shown that constructive elements can be more significant than initial material distribution in controlling the final distribution; limiting percolation in the transverse direction remains a major challenge for the distribution system in gradient control. The results demonstrate the usefulness and suitability of applying simulations with the developed model to the design of the powder delivery system and define a direction for further theoretical and experimental research. Full article

Multi-objective Bayesian Optimization (MOBO) has become a promising strategy for accelerating process development in Laser Powder Bed Fusion of Metals (PBF-LB/M), where experimental evaluations are costly, and design spaces are high-dimensional. This study investigates how different initialization strategies affect MOBO performance in a discrete, machine-limited parameter space during the fabrication of AZ31 magnesium alloy. Three approaches to constructing the initial experimental dataset—Latin Hypercube Sampling (V1), balanced-marginal selection (V2), and prior fractional-factorial sampling (V3)—were compared using two state-of-the-art MOBO algorithms, DGEMO and TSEMO, implemented within the AutoOED platform. A total of 180 samples were produced and evaluated with respect to two conflicting objectives: maximizing relative density and build rate. The evolution of the Pareto front and hypervolume metrics shows that the structure of the initial dataset strongly governs subsequent optimization efficiency. Variant V3 yielded the highest hypervolumes for both algorithms, whereas Variant V2 produced the most uniform Pareto approximation, highlighting a trade-off between global coverage and structural distribution. TSEMO demonstrated faster early convergence, whereas DGEMO maintained broader exploration of the design space. Analysis of duplicate experimental points revealed that discretization and batch selection can considerably limit the effective search diversity, contributing to an early saturation of hypervolume gains. The results indicate that, in constrained PBF-LB/M design spaces, MOBO primarily serves to validate and refine a well-designed initial dataset rather than to discover dramatically new optima. The presented workflow highlights how initialization, parameter discretization, and sampling diversity shape the practical efficiency of MOBO for additive manufacturing process optimization. Full article

In the context of intelligent manufacturing, the integrated scheduling problem of dual rail-guided vehicles (RGVs) and multiple parallel processing equipment in flexible manufacturing systems has gained increasing importance. This problem exhibits spatiotemporal coupling and dynamic constraint characteristics, making traditional optimization methods ineffective at finding optimal solutions. At the problem formulation level, the dual resource scheduling task is modeled as a mixed-integer optimization problem. An intelligent scheduling framework based on action mask-constrained Proximal Policy Optimization (PPO) deep reinforcement learning is proposed to achieve integrated decision-making for production equipment allocation and RGV path planning. The approach models the scheduling problem as a Markov Decision Process, designing a high-dimensional state space, along with a multi-discrete action space that integrates machine selection and RGV motion control. The framework employs a shared feature extraction layer and dual-head Actor-Critic network architecture, combined with parallel experience collection and synchronous parameter update mechanisms. In computational experiments across different scales, the proposed method achieves an average makespan reduction of 15–20% compared with numerical methods, while exhibiting excellent robustness under uncertain conditions including processing time fluctuations. Full article