Search Results (56)

This study addresses the challenges of manual measurement bias and low robustness in detecting small, occluded targets in complex marine environments during real-time onboard sorting of Portunus trituberculatus. We propose YOLO-DFAM, an enhanced YOLOv11n-based model that replaces the global average pooling in the Focal Modulation module with a spatial–channel dual-attention mechanism and incorporates the ASF-YOLO cross-scale fusion strategy to improve feature representation across varying target sizes. These enhancements significantly boost detection, achieving an mAP@50 of 98.0% and precision of 94.6%, outperforming RetinaNet-CSL and Rotated Faster R-CNN by up to 6.3% while maintaining real-time inference at 180.3 FPS with only 7.2 GFLOPs. Unlike prior static-scene approaches, our unified framework integrates attention-guided detection, scale-adaptive tracking, and lightweight weight estimation for dynamic marine conditions. A ByteTrack-based tracking module with dynamic scale calibration, EMA filtering, and optical flow compensation ensures stable multi-frame tracking. Additionally, a region-specific allometric weight estimation model (R² = 0.9856) reduces dimensional errors by 85.7% and maintains prediction errors below 4.7% using only 12 spline-interpolated calibration sets. YOLO-DFAM provides an accurate, efficient solution for intelligent onboard fishery monitoring. Full article

The rapid growth of the machining market and advancements in additive manufacturing (AM) present new opportunities for innovative tool designs. This preliminary study proposes a design for additive manufacturing (DfAM) approach to redesign a milling cutter head in 17-4 PH stainless steel by integrating topology optimization (TO) and internal coolant channel optimization, enabled by laser powder bed fusion (LPBF). An industrial eight-insert milling cutting tool was reimagined with conformal cooling channels and a lightweight topology-optimized structure. The design process considered LPBF constraints and was iteratively refined using computational fluid dynamics (CFD) and finite element analysis (FEA) to validate fluid flow and structural performance. The optimized milling head achieved approximately 10% weight reduction while improving stiffness (reducing maximum deformation under load from 160 μm to 151 μm) and providing enhanced coolant distribution to the cutting inserts. The results demonstrate that combining TO with internal channel design can yield a high-performance and lightweight milling tool that leverages the freedom of additive manufacturing. As proof of concept, this integrated CFD–FEA validation approach under DfAM guidelines highlights a promising pathway toward superior cutting tool designs for industrial applications. Full article

This study investigates the potential of low-density polymeric materials to enhance the deformation energy absorption of drone fuselage components manufactured using fused filament fabrication (FFF). Two materials—PLA (polylactic acid) and LW-PLA (lightweight polylactic acid)—were selected based on their accessibility, printability, and prior mechanical characterizations. While PLA is widely used in additive manufacturing, its brittleness limits its suitability for components subjected to accidental or impact loads. In contrast, LW-PLA exhibits greater ductility and energy absorption, making it a promising alternative where weight reduction is critical and structural redundancy is available. To evaluate the structural efficiency, a simplified analysis scenario was developed using a theoretical 300 J collision energy, not as a design condition, but as a comparative benchmark for assessing the performance of various metastructural configurations. The experimental results demonstrate that a stiffening core of the LW-PLA metastructure can reduce the component weight by over 60% while maintaining or improving the deformation energy absorption. Modified prototypes with hybrid internal structures demonstrated stable performances under repeated loading; however, the tests also revealed a buckling-like failure of the internal core in specific configurations, highlighting the need for core stabilization within metastructures to ensure reliable energy dissipation. Full article

Stained glass has played important roles in heritage building construction, however, conventional fabrication techniques have become economically prohibitive due to both capital costs and energy inefficiency, as well as high-level artistic and craft skills. To overcome these challenges, this study provides a new design methodology for customized 3D-printed polycarbonate (PC)-based stained-glass window alternatives using a fully open-source toolchain and methodology based on digital fabrication and hybrid crafts. Based on design thinking and open design principles, this procedure involves fabricating an additional insert made of (i) a PC substrate and (ii) custom geometries directly 3D printed on the substrate with PC-based 3D printing feedstock (iii) to be painted after the 3D printing process. This alternative is intended for customizable stained-glass design patterns to be used instead of traditional stained glass or in addition to conventional windows, making stained glass accessible and customizable according to users’ needs. Three approaches are developed and demonstrated to generate customized painted stained-glass geometries according to the different users’ skills and needs using (i) online-retrieved 3D and 2D patterns; (ii) custom patterns, i.e., hand-drawn and digital-drawn images; and (iii) AI-generated patterns. The proposed methodology shows potential for distributed applications in the building and heritage sectors, demonstrating its practical feasibility. Its use makes stained-glass-based products accessible to a broader range of end-users, especially for repairing and replicating existing conventional stained glass and designing new customizable products. The developed custom patterns are 50 times less expensive than traditional stained glass and can potentially improve thermal insulation, paving the way to energy efficiency and economic savings. Full article

Additive manufacturing processes such as the material extrusion of metals (MEX/M) enable the production of complex and functional parts that are not feasible to create through traditional manufacturing methods. However, achieving high-quality MEX/M parts requires significant experimental and financial investments for suitable parameter development. In response, this study explores the application of machine learning (ML) to predict the surface roughness and density in MEX/M components. The various models are trained with experimental data using input parameters such as layer thickness, print velocity, infill, overhang angle, and sinter profile enabling precise predictions of surface roughness and density. The various ML models demonstrate an accuracy of up to 97% after training. In conclusion, this research showcases the potential of ML in enhancing the efficiency in control over component quality during the design phase, addressing challenges in metallic additive manufacturing, and facilitating exact control and optimization of the MEX/M process, especially for complex geometrical structures. Full article

Additive manufacturing (AM) technologies have witnessed remarkable advancements, offering opportunities to produce complex components across various industries. This paper explores the potential of AM for fabricating bipolar plates (BPPs) in fuel cell or electrolysis cell applications. BPPs play a critical role in the performance and efficiency of such cells, and conventional manufacturing methods often face limitations, particularly concerning the complexity and customization of geometries. The focus here lies in two specific AM methods: the laser powder bed fusion of metals (PBF-LB/M) and material extrusion of metals (MEX/M). PBF-LB/M, tailored for high-performance applications, enables the creation of highly complex geometries, albeit at increased costs. On the other hand, MEX/M excels in rapid prototyping, facilitating the swift production of diverse geometries for real-world testing. This approach can facilitate the evaluation of geometries suitable for mass production via sinter-based manufacturing processes. The geometric deviations of different BPPs were identified by evaluating 3D scans. The PBF-LB/M method is more suitable for small features, while the MEX/M method has lower deviations for geometrically less complex BPPs. Through this investigation, the limits of the capabilities of these AM methods became clear, knowledge that can potentially enhance the design and production of BPPs, revolutionizing the energy conversion and storage landscape and contributing to the design of additive manufacturing technologies. Full article

Design for Additive Manufacturing (DfAM) encompasses two primary strategies: adapting traditional designs for 3D printing and developing designs specifically optimized for additive manufacturing. The latter emphasizes consolidating assemblies and reducing weight, leveraging complex geometries and negative space through advanced techniques such as generative design and topology optimization. Critical considerations in the design phase include printing methods, material selection, support structures, and post-processing requirements. DfAM offers significant advantages over conventional subtractive manufacturing, including enhanced complexity, customization, and optimization, with transformative applications in aerospace, medical devices, and automotive industries. This review focuses on the automotive sector, systematically examining DfAM’s potential to redefine vehicle design, production processes, and industry standards. By conducting a comprehensive analysis of the existing literature and case studies, this research identifies gaps in the integration of additive manufacturing into broader manufacturing frameworks. The study contributes to the literature by providing insights into how 3D printing is currently reshaping automotive production by offering a forward-looking perspective on its future implications for the industry. Full article

The amount of non-revenue water, mostly due to leakage, is around 126 billion cubic meters annually worldwide. A more efficient wastewater management strategy would use a parametric design for on-demand, customized pipe fittings, following the principles of distributed manufacturing. To fulfill this need, this study introduces an open-source parametric design of a 3D-printable easy-connect pipe fitting that offers compatibility with different dimensions and materials of pipes available on the market. Custom pipe fittings were 3D printed using a RepRap-class fused filament 3D printer, with polylactic acid (PLA), polyethylene terephthalate glycol (PETG), acrylonitrile styrene acrylate (ASA), and thermoplastic elastomer (TPE) as filament feedstocks for validation. The 3D-printed connectors underwent hydrostatic water pressure tests to ensure that they met the standards for residential, agricultural, and renewable energy production applications. All the printed parts passed numerous hydrostatic pressure tests. PETG couplings can tolerate up to 4.551 ± 0.138 MPa of hydrostatic pressure, which is eight times greater than the highest standard water pressure in the residential sector. Based on the economic analysis, the cost of 3D printing a pipe coupling is from three to seventeen times lower than purchasing a commercially available pipe fitting of a similar size. The new open-source couplings demonstrate particular potential for use in developing countries and remote areas. Full article

Additive manufacturing (AM) technology is gaining acceptance as a strategic manufacturing technique for allowing new product development. Due to ongoing process improvement, design for AM (DFAM) has become a major issue in harnessing AM’s production and development possibilities to achieve design freedom. The classical design process model does not encompass all the knowledge available to take advantage of design freedom. Therefore, a conceptual and in-depth analysis of design alternatives is necessary to determine the manufacturing process. As a result, this research proposed a design process model for a DFAM to attain design freedom with a unique approach and resource selection steps for fused deposition modeling (FDM) that uses an information model based on evolving knowledge and addressing the challenges. The proposed design process model uses an event knowledge graph (EKG) to outline the product manufacturability from the perspective of DFAM limitations. Event-based knowledge representation provides causality information for knowledge-based reasoning in causality analysis tasks. A relationship-aware mechanism is then used to express events on the graph that are directed from entities to occurrences to efficiently extract the most relevant details. Thus, this implements a step-by-step approach to process and resource specifications during the design stage. Consequently, it offers a comprehensive learning approach for establishing and modeling intrinsic relationships to attain flexibility and design freedom. The efficacy and feasibility of the proposed approach are verified by using an application case study of an intake system based on the airflow sensing rate and controls how much air is fed into the engine. Full article

Conflicts between humans and animals in agricultural and settlement areas have recently increased, resulting in significant resource loss and risks to human and animal lives. This growing issue presents a global challenge. This paper addresses the detection and identification of offending animals, particularly in obscured or blurry nighttime images. This article introduces Multi-Channel Coordinated Attention and Multi-Dimension Feature Aggregation (MDA-DETR). It integrates multi-scale features for enhanced detection accuracy, employing a Multi-Channel Coordinated Attention (MCCA) mechanism to incorporate location, semantic, and long-range dependency information and a Multi-Dimension Feature Aggregation Module (DFAM) for cross-scale feature aggregation. Additionally, the VariFocal Loss function is utilized to assign pixel weights, enhancing detail focus and maintaining accuracy. In the dataset section, this article uses a dataset from the Northeast China Tiger and Leopard National Park, which includes images of six common offending animal species. In the comprehensive experiments on the dataset, the $mA{P}_{50}$ index of MDA-DETR was 1.3%, 0.6%, 0.3%, 3%, 1.1%, and 0.5% higher than RT-DETR-r18, yolov8n, yolov9-C, DETR, Deformable-detr, and DCA-yolov8, respectively, indicating that MDA-DETR is superior to other advanced methods. Full article

Design for additive manufacturing (DfAM) has made significant advancements in recent years, with development focusing on pivotal aspects such as topology optimization (TO), generative design (GD), lattice structures, and AI-based algorithms. This paper puts forth a proposed methodology for the development of customizable bike saddle pads for manufacturing with AM. The approach entails the selection of appropriate AM processes and materials, the evaluation of material properties through compression testing, an initial saddle pressure mapping and bike fitting, the design and AM of bespoke saddle pads based on the initial measurements, and a validation pressure mapping and bike fitting. The investigation yielded clear findings regarding improvements in both pressure distribution and the change in pressure peaks, as well as an improvement in riding comfort. The findings indicate that although the overall process is innovative, improvements are required to streamline the measuring, modeling, and manufacturing workflow. Full article

Additive manufacturing (AM) has revolutionized the design and production of complex geometries by offering unprecedented creative freedom over traditional manufacturing. Despite its growing prominence, AM lacks automated and standardized design rules tailored to specific AM processes, resulting in time-consuming and expert-dependent manual verification. To address these limitations, this research introduces a novel design for additive manufacturing (DfAM) framework consisting of two complementary models designed to automate the design process. The first model, based on a decision tree algorithm, evaluates part compliance with established AM design rules. A modified J48 classifier was implemented to enhance data mining accuracy by achieving a 91.25% classification performance accuracy. This model systematically assesses whether input part characteristics meet AM processing standards, thereby providing a robust tool for verifying design rules. The second model features an AM design rule engine developed with a Python-based graphical user interface (GUI). This engine generates specific recommendations for design adjustments based on part characteristics and machine compatibility, offering a user-friendly approach for identifying potential design issues and ensuring DfAM compliance. By linking part specifications to various AM techniques, this model supports both researchers and engineers in anticipating and mitigating design flaws. Overall, this research establishes a foundation for a comprehensive DfAM expert system. Full article

Additive manufacturing (AM) has become a key topic in the manufacturing industry, challenging conventional techniques. However, AM has its limitations, and understanding its convenience despite established processes remains sometimes difficult, especially in preliminary design phases. This investigation provides a hybrid multi-criteria decision-making method (MCDM) for comparing AM and conventional processes. The MCDM method consists of the Best Worst Method (BWM) for the definition of criteria weights and the Proximity Index Value (PIV) method for the generation of the final ranking. The BWM reduces the number of pairwise comparisons required for the definition of criteria weights, whereas the PIV method minimizes the probability of rank reversal, thereby enhancing the robustness of the results. The methodology was validated through a case study, an aerospace bracket. The candidate processes for the bracket production were CNC machining, high-pressure die casting, and PBF-LB/M. The production of the bracket by AM was found to be the optimal choice for small to medium production batches. Additionally, the study emphasized the significance of material selection, process design guidelines, and production batch in the context of informed process selection, thereby enabling technical professionals without a strong AM background in pursuing conscious decisions. Full article

We present a novel, systematic method for evaluating design printability in 3D food printing using a scoring system based on the Design for Additive Manufacturing (DfAM) guidelines. This study addresses a gap in the current literature by proposing a structured approach to assess and enhance the printability of 3D food designs. Our framework consists of a set of nine critical questions derived from the multi-level DfAM guidelines, focusing on key printability factors including unsupported features, geometric accuracy, and surface finish. The evaluation process converts qualitative assessments into numerical values, resulting in a comprehensive printability score that categorizes designs into high, moderate, or low printability levels. To validate the effectiveness of this method, we conducted a case study involving five different designs. The scoring system successfully explores the design space and maximizes the printability of 3D food products. This method alleviates the challenges in design evaluation compared with traditional trial-and-error approaches. The results demonstrate the practicality and efficiency of our framework’s output. The proposed methodology provides a structured approach to design evaluation, offering practical insights and a valuable tool for improving the success rate of 3D printed food products. This research contributes to the field by offering a systematic framework for assessing and enhancing the printability of 3D food designs, potentially accelerating the adoption and effectiveness of 3D food printing technology in various applications. Full article

Additive manufacturing (AM) has impacted the manufacturing of complex three-dimensional objects in multiple materials for a wide array of applications. However, additive manufacturing, as an upcoming field, lacks automated and specific design rules for different AM processes. Moreover, the selection of specific AM processes for different geometries requires expert knowledge, which is difficult to replicate. An automated and data-driven system is needed that can capture the AM expert knowledge base and apply it to 3D-printed parts to avoid manufacturability issues. This research aims to develop a data-driven system for AM process selection within the design for additive manufacturing (DFAM) framework for Industry 4.0. A Genetic and Evolutionary Feature Weighting technique was optimized using 3D CAD data as an input to identify the optimal AM technique based on several requirements and constraints. A two-stage model was developed wherein the stage 1 model displayed average accuracies of 70% and the stage 2 model showed higher average accuracies of up to 97.33% based on quantitative feature labeling and augmentation of the datasets. The steady-state genetic algorithm (SSGA) was determined to be the most effective algorithm after benchmarking against estimation of distribution algorithm (EDA) and particle swarm optimization (PSO) algorithms, respectively. The output of this system leads to the identification of optimal AM processes for manufacturing 3D objects. This paper presents an automated design for an additive manufacturing system that is accurate and can be extended to other 3D-printing processes. Full article