Hasil untuk "Manufacturing industries"

Masoud Deylami, Negar Izadipour, Adel Alaeddini

Manufacturability assessment is a critical step in bridging the persistent gap between design and production. While artificial intelligence (AI) has been widely applied to this task, most existing frameworks rely on geometry-driven methods that require extensive preprocessing, suffer from information loss, and offer limited interpretability. This study proposes a methodology that evaluates manufacturability directly from parametric design features, enabling explicit incorporation of dimensional tolerances without requiring computer-aided design (CAD) processing. The approach employs Kolmogorov-Arnold Networks (KANs) to learn functional relationships between design parameters, tolerances, and manufacturability outcomes. A synthetic dataset of 300,000 labeled designs is generated to evaluate performance across three representative scenarios: hole drilling, pocket milling, and combined drilling-milling, while accounting for machining constraints and design-for-manufacturing (DFM) rules. Benchmarking against fourteen machine learning (ML) and deep learning (DL) models shows that KAN achieves the highest performance in all scenarios, with AUC values of 0.9919 for drilling, 0.9841 for milling, and 0.9406 for the combined case. The proposed framework provides high interpretability through spline-based functional visualizations and latent-space projections, enabling identification of the design and tolerance parameters that most strongly influence manufacturability. An industrial case study further demonstrates how the framework enables iterative, parameter-level design modifications that transform a non-manufacturable component into a manufacturable one.

Fanchao Zhong, Yang Wang, Peng-Shuai Wang et al.

Manufacturability is vital for product design and production, with accessibility being a key element, especially in subtractive manufacturing. Traditional methods for geometric accessibility analysis are time-consuming and struggle with scalability, while existing deep learning approaches in manufacturability analysis often neglect geometric challenges in accessibility and are limited to specific model types. In this paper, we introduce DeepMill, the first neural framework designed to accurately and efficiently predict inaccessible and occlusion regions under varying machining tool parameters, applicable to both CAD and freeform models. To address the challenges posed by cutter collisions and the lack of extensive training datasets, we construct a cutter-aware dual-head octree-based convolutional neural network (O-CNN) and generate an inaccessible and occlusion regions analysis dataset with a variety of cutter sizes for network training. Experiments demonstrate that DeepMill achieves 94.7% accuracy in predicting inaccessible regions and 88.7% accuracy in identifying occlusion regions, with an average processing time of 0.04 seconds for complex geometries. Based on the outcomes, DeepMill implicitly captures both local and global geometric features, as well as the complex interactions between cutters and intricate 3D models.

Bestoun S. Ahmed, Tommaso Azzalin, Andreas Kassler et al.

We explore a Digital Twin-Based Approach for Smart Manufacturing to improve Sustainability, Efficiency, and Cost-Effectiveness for a steel production plant. Our system is based on a micro-service edge-compute platform that ingests real-time sensor data from the process into a digital twin over a converged network infrastructure. We implement agile machine learning-based control loops in the digital twin to optimize induction furnace heating, enhance operational quality, and reduce process waste. Key to our approach is a Deep Reinforcement learning-based agent used in our machine learning operation (MLOps) driven system to autonomously correlate the system state with its digital twin to identify correction actions that aim to optimize power settings for the plant. We present the theoretical basis, architectural details, and practical implications of our approach to reduce manufacturing waste and increase production quality. We design the system for flexibility so that our scalable event-driven architecture can be adapted to various industrial applications. With this research, we propose a pivotal step towards the transformation of traditional processes into intelligent systems, aligning with sustainability goals and emphasizing the role of MLOps in shaping the future of data-driven manufacturing.

Junchun Ding

Under the market background of increasingly personalized product demand and compressed response cycle, the traditional manufacturing model with standardized mass production as the core has been difficult to meet the dual expectations of customers for differentiation and fast delivery. In order to improve the efficiency of resource allocation and market response, automobile manufacturers need to build a production system that takes into account cost and flexibility. Based on the delayed response manufacturing strategy, this study built an order response node configuration model suitable for automotive manufacturing scenarios, focusing on the positioning of order driven intervention points in the production process. The model comprehensively considers the structural cost changes brought by process adjustment, the dynamic characteristics of the changes of unit manufacturing cost and intermediate inventory cost at different stages with the location of nodes, and introduces delivery time constraints to embed time factors into the inventory decision logic to enhance the practicality of the model and the adaptation of realistic constraints. In terms of solution methods, this paper adopts function fitting and simulation analysis methods, combined with mathematical modeling tools, systematically describes the change trend of total cost, and verifies the rationality and effectiveness of the model structure and solution through actual enterprise cases. The research results provide a theoretical basis and decision support for automobile manufacturing enterprises to realize the synergy of flexible production and cost control in the environment of variable demand, and also provide an empirical reference for the implementation path and system optimization of subsequent relevant strategies.

Ananya Ipsita, Ramesh Kaki, Ziyi Liu et al.

To address the shortage of a skilled workforce in the U.S. manufacturing industry, immersive Virtual Reality (VR)-based training solutions hold promising potential. To effectively utilize VR to meet workforce demands, it is important to understand the role of VR in manufacturing education. Therefore, we conduct a scoping review in the field. As a first step, we used a 5W1H (What, Where, Who, When, Why, How) formula as a problem-solving approach to define a comprehensive taxonomy that can consider the role of VR from all relevant possibilities. Our taxonomy categorizes VR applications across three key aspects: (1) Domains, (2) Levels, and (3) Entities. Using a systematic literature search and analysis, we reviewed 108 research articles to find the current state, benefits, challenges, and future opportunities of VR in the field. It was found that VR has been explored in a variety of areas and provides numerous benefits to learners. Despite these benefits, its adoption in manufacturing education is limited. This review discusses the identified barriers and provides actionable insights to address them. These insights can enable the widespread usage of immersive technology to nurture and develop a workforce equipped with the skills required to excel in the evolving landscape of manufacturing.

Yongxue Chen, Tao Liu, Yuming Huang et al.

This paper presents a method for computing interleaved additive and subtractive manufacturing operations to fabricate models of arbitrary shapes. We solve the manufacturing planning problem by searching a sequence of inverse operations that progressively transform a target model into a null shape. Each inverse operation corresponds to either an additive or a subtractive step, ensuring both manufacturability and structural stability of intermediate shapes throughout the process. We theoretically prove that any model can be fabricated exactly using a sequence generated by our approach. To demonstrate the effectiveness of this method, we adopt a voxel-based implementation and develop a scalable algorithm that works on models represented by a large number of voxels. Our approach has been tested across a range of digital models and further validated through physical fabrication on a hybrid manufacturing system with automatic tool switching.

James A. Levinsohn

Fiorella P. Cárdenas-Toro, Jennifer H. Meza-Coaquira, Gabriela K. Nakama-Hokamura et al.

Currently, <i>Bixa orellana</i> L. extracts are used as a color source in the food, pharmaceutical, and cosmetic industries because they are important as a potential source of antioxidant activity. The extraction is carried out by conventional methods, using alkaline solutions or organic solvents. These extraction methods do not take advantage of the lipid fraction of annatto (<i>Bixa orellana</i> L.) seeds, and the process is not friendly to the environment. In this work, the objective was to obtain an extract rich in nutraceuticals (bixin and tocols) of high antioxidant power from Peruvian annatto seeds as a potential source for a functional food or additive in the industry using supercritical fluid extraction (SFE). Experiments related to extraction yield, bixin, tocotrienols, tocopherols, and antioxidant activity were carried out. The SFE was performed at 40 °C, 50 °C, and 60 °C, and 100, 150, and 250 bar with 0.256 kg/h carbon dioxide as the supercritical solvent (solvent-to-feed ratio of 10.2). Supercritical extraction at 60 °C and 250 bar presented the best results in terms of global extraction yield of 1.40 ± 0.01 g/100 g d.b., extract concentration of 0.564 ± 0.005 g bixin/g extract, 307.8 mg α-tocotrienol/g extract, 39.2 mg β-tocotrienol/g extract, 2 mg γ-tocopherol/g extract, and IC₅₀ of 989.96 μg extract/mL. Economical evaluation showed that 60 °C, 250 bar, and 45 min presented the lowest cost of manufacturing (2 × 2000 L, COM of USD 212.39/kg extract). This extract is a potential source for functional food production.

Payam Maleki, Mohammadmehdi Shahzamanian, Wan Jefferey Basirun et al.

Steel/polymer/steel laminates, also known as laminated steels, are composite materials consisting of bonding layers of steel and polymer. The polymer layer acts as a bonding agent between the steel layers, imparting additional properties such as low density, impact resistance, and thermal insulation, while the steel layers provide strength and formability. These laminated steels have found increasing applications in automotive, aerospace, and construction industries to reduce weight and improve fuel efficiency. The bending behavior of this laminates is more complex compared to that of a single layer of metallic sheets. This complexity arises from significant differences in mechanical properties, as well as the thickness ratio between the skin and the core. The flexural properties and behavior of different St14/TPU/St14 laminate sheets that were fabricated using the direct roll bonding (DRB) process were investigated through three-point and hat-shaped bending tests. The direct roll bonding process involves the bonding of steel and semi-melt polymer sheets under the pressure of rollers, ensuring a cohesive and durable composite material. The microscopic analysis of the cross-section of the SPS laminates after the bending processes shows the absence of delamination or slippage between the layers, which indicates the correct selection of materials and the bonding method. The results showed that the springback of three-layer laminates has an inverse relationship with the work-hardening exponent, yield strength, and yield point elongation value, while possessing a direct relationship with normal anisotropy and elastic modulus. Furthermore, the flexural strength and flexural modulus decrease with the increase in the volume fraction of the polymeric core, while the flexural rigidity increases. The findings indicate the DRB technique as a promising method for manufacturing a lightweight metal–polymer laminate with a high formability performance.

Marina S. Reshetnikova, Svetlana S. Tretyakova

The International Federation of Robots claims that the number of robots being produced today is at an all-time high, particularly in sectors like electronics and the automotive industry. Considering this, as artificial intelligence (AI) becomes more and more popular, it is imperative to study the robotics market, particularly in the nations that are major players in it. Robots are now integrated into all industries, especially automotive and electronics. The need for modernization and increasing competition are pushing countries to automate production and improve business processes; the United States and China have achieved the greatest success in this. According to the authors, it is China that will succeed in the industrial robotics market by introducing “smart manufacturing” and “smart factories.” In addition, there is now a growing trend towards creating robots that interact with people: the development of technology, the emergence of artificial intelligence, as well as human acceptance of the robot - all this contributes to the introduction of robots into our lives. Thus, according to the authors, it is the United States that has achieved the greatest success in the field of introducing service robots responsible for interaction with humans. The evolution of the robotics industry in the two major market participants - China and the United States - is analyzed. The study analyzes the industries in both countries to pinpoint development areas and reviews government initiatives that support business growth in both China and the United States.

ZHANG Ying, YAN Dingbo, HU Jinghui, ZHANG Jiaojiao, CHEN Bin, ZHAO Jinsong, HAN Xinglin

In recent years, with the improvement of consumers' living standards, more and more people's preference for Baijiu style has changed from 'spicy' to 'elegant'. As one of the 12 major flavor-style Baijiu, the style characteristics of light-flavor (Qingxiangxing) Baijiu are just in line with the pursuit of consumers, thus promoting the development of light-flavor Baijiu. In this paper, the development of light-flavor Baijiu since the founding of the People's Republic of China was summarized from the aspects of development history, industry overview, style characteristics, production process and research progress, and the future development of light-flavor Baijiu was prospected, in order to provide references for light-flavor Baijiu enterprises and related workers.

Mavis Akolor, Tripti Gujral

This work was conducted to determine how firm size, investment, inflation, and government effectiveness influence dividend policies proxy by dividend payout of quoted firms in Ghana. Twenty-two firms in total were sampled for this study and included trading, manufacturing, construction, and mining companies for the period 2011 to 2020. The firms cut across different industries namely telecommunications, consumer goods, constructions, oil and gas, technological companies, consumer services, and manufacturing companies quoted in Ghana. Panel data for all 22 non-financial firms were collected from the financial statement, ratios were calculated to ascertain the measurement of the variables under study. The Generalized Method of Moment was adopted and a multiple regression analysis was performed to determine the impact of the variables on dividend policy proxy by dividend payout ratio. The results of the study indicated that the log of total assets had a negative relationship with dividend policy significant at all conventional levels. Investment and government effectiveness also had a positive relationship with dividend policy. Unfortunately, inflation had an insignificant effect on the dividend payout of quoted firms in Ghana for the period under review.

Jeongguk Kim, Sungil Seo

In mechanical braking systems, there are hot spots on the surface of a braking disc due to thermal deformation with a high thermal gradient. Controlling such hot spots is important for extending the life of a braking disc. In this study, surface temperatures of railway brake discs were monitored using infrared (IR) thermal imaging technique. A high-speed infrared camera with a maximum speed of 380 Hz was used to monitor surface temperature changes of the braking disc. Braking tests were performed with a full-scale dynamometer. During the braking test, the surface temperature change of the braking disc were monitored using a high-speed infrared camera. Hot spots and thermal damage observed on the surface of railway brake discs during braking tests were quantitatively analyzed using infrared thermographic images. Results revealed that monitoring disc surface temperature using IR thermographic technique can be a new method for predicting surface temperature changes without installing a thermocouple inside the disc.

Xingyu Li, Fei Tao, Wei Ye et al.

In this study, we introduce Generative Manufacturing Systems (GMS) as a novel approach to effectively manage and coordinate autonomous manufacturing assets, thereby enhancing their responsiveness and flexibility to address a wide array of production objectives and human preferences. Deviating from traditional explicit modeling, GMS employs generative AI, including diffusion models and ChatGPT, for implicit learning from envisioned futures, marking a shift from a model-optimum to a training-sampling decision-making. Through the integration of generative AI, GMS enables complex decision-making through interactive dialogue with humans, allowing manufacturing assets to generate multiple high-quality global decisions that can be iteratively refined based on human feedback. Empirical findings showcase GMS's substantial improvement in system resilience and responsiveness to uncertainties, with decision times reduced from seconds to milliseconds. The study underscores the inherent creativity and diversity in the generated solutions, facilitating human-centric decision-making through seamless and continuous human-machine interactions.

Yu-hsuan Chen, Jonathan Cagan, Levent Burak kara

Designing for manufacturing poses significant challenges in part due to the computation bottleneck of Computer-Aided Manufacturing (CAM) simulations. Although deep learning as an alternative offers fast inference, its performance is dependently bounded by the need for abundant training data. Representation learning, particularly through pre-training, offers promise for few-shot learning, aiding in manufacturability tasks where data can be limited. This work introduces VIRL, a Volume-Informed Representation Learning approach to pre-train a 3D geometric encoder. The pretrained model is evaluated across four manufacturability indicators obtained from CAM simulations: subtractive machining (SM) time, additive manufacturing (AM) time, residual von Mises stress, and blade collisions during Laser Power Bed Fusion process. Across all case studies, the model pre-trained by VIRL shows substantial enhancements on demonstrating improved generalizability with limited data and superior performance with larger datasets. Regarding deployment strategy, case-specific phenomenon exists where finetuning VIRL-pretrained models adversely affects AM tasks with limited data but benefits SM time prediction. Moreover, the efficacy of Low-rank adaptation (LoRA), which balances between probing and finetuning, is explored. LoRA shows stable performance akin to probing with limited data, while achieving a higher upper bound than probing as data size increases, without the computational costs of finetuning. Furthermore, static normalization of manufacturing indicators consistently performs well across tasks, while dynamic normalization enhances performance when a reliable task dependent input is available.

Jonghan Lim, Leander Pfeiffer, Felix Ocker et al.

Improving the overall equipment effectiveness (OEE) of machines on the shop floor is crucial to ensure the productivity and efficiency of manufacturing systems. To achieve the goal of increased OEE, there is a need to develop flexible runtime control strategies for the system. Decentralized strategies, such as multi-agent systems, have proven effective in improving system flexibility. However, runtime multi-agent control of complex manufacturing systems can be challenging as the agents require extensive communication and computational efforts to coordinate agent activities. One way to improve communication speed and cooperation capabilities between system agents is by providing a common language between these agents to represent knowledge about system behavior. The integration of ontology into multi-agent systems in manufacturing provides agents with the capability to continuously update and refine their knowledge in a global context. This paper contributes to the design of an ontology for multi-agent systems in manufacturing, introducing an extendable knowledge base and a methodology for continuously updating the production data by agents during runtime. To demonstrate the effectiveness of the proposed framework, a case study is conducted in a simulated environment, which shows improvements in OEE during runtime.

Audrey Chung, Francis Li, Jeremy Ward et al.

As electronics manufacturers continue to face pressure to increase production efficiency amid difficulties with supply chains and labour shortages, many printed circuit board assembly (PCBA) manufacturers have begun to invest in automation and technological innovations to remain competitive. One such method is to leverage artificial intelligence (AI) to greatly augment existing manufacturing processes. In this paper, we present the DarwinAI Visual Quality Inspection (DVQI) system, a hardware-integration artificial intelligence system for the automated inspection of printed circuit board assembly defects in an electronics manufacturing environment. The DVQI system enables multi-task inspection via minimal programming and setup for manufacturing engineers while improving cycle time relative to manual inspection. We also present a case study of the deployed DVQI system's performance and impact for a top electronics manufacturer.

H. Shewan, J. Stokes

Abir Gallala, Atal Anil Kumar, Bassem Hichri et al.

There has been a rapid increase in the use of collaborative robots in manufacturing industries within the context of Industry 4.0 and smart factories. The existing human–robot interactions, simulations, and robot programming methods do not fit into these fast-paced technological advances as they are time-consuming, require engineering expertise, waste a lot of time in programming and the interaction is not trivial for non-expert operators. To tackle these challenges, we propose a digital twin (DT) approach for human–robot interactions (HRIs) in hybrid teams in this paper. We achieved this using Industry 4.0 enabling technologies, such as mixed reality, the Internet of Things, collaborative robots, and artificial intelligence. We present a use case scenario of the proposed method using Microsoft Hololens 2 and KUKA IIWA collaborative robot. The obtained results indicated that it is possible to achieve efficient human–robot interactions using these advanced technologies, even with operators who have not been trained in programming. The proposed method has further benefits, such as real-time simulation in natural environments and flexible system integration to incorporate new devices (e.g., robots or software capabilities).

Haiming Yao, Wenyong Yu, Xue Wang

Recent advances in the industrial inspection of textured surfaces-in the form of visual inspection-have made such inspections possible for efficient, flexible manufacturing systems. We propose an unsupervised feature memory rearrangement network (FMR-Net) to accurately detect various textural defects simultaneously. Consistent with mainstream methods, we adopt the idea of background reconstruction; however, we innovatively utilize artificial synthetic defects to enable the model to recognize anomalies, while traditional wisdom relies only on defect-free samples. First, we employ an encoding module to obtain multiscale features of the textured surface. Subsequently, a contrastive-learning-based memory feature module (CMFM) is proposed to obtain discriminative representations and construct a normal feature memory bank in the latent space, which can be employed as a substitute for defects and fast anomaly scores at the patch level. Next, a novel global feature rearrangement module (GFRM) is proposed to further suppress the reconstruction of residual defects. Finally, a decoding module utilizes the restored features to reconstruct the normal texture background. In addition, to improve inspection performance, a two-phase training strategy is utilized for accurate defect restoration refinement, and we exploit a multimodal inspection method to achieve noise-robust defect localization. We verify our method through extensive experiments and test its practical deployment in collaborative edge--cloud intelligent manufacturing scenarios by means of a multilevel detection method, demonstrating that FMR-Net exhibits state-of-the-art inspection accuracy and shows great potential for use in edge-computing-enabled smart industries.