Figure - available from: The International Journal of Advanced Manufacturing Technology
This content is subject to copyright. Terms and conditions apply.
Source publication
This study explores the application of machine learning algorithms for supporting complex product manufacturing quality through a focus on quality assurance and control. We aim to take advantage of ML technics to solve one of the complex manufacturing problems of the tempered glass manufacturing industry as a first attempt to automate product quali...
Citations
... A deep reinforcement learning framework was proposed using three methods as deep Q-network, proximal policy optimization and advantage actor-critic algorithms for ensuring the product quality and minimizing the overall energy consumption of an industrial glass manufacturing process [31]. In the area of quality forecasting, 14 algorithms were compared (ridge regression, linear regression, light gradient boosting machine, lasso regression, random forest regressor, artificial neural networks, gradient boosting regressor, extra trees regressor, elastic net, Bayesian ridge, K neighbors regressor, AdaBoost regressor, least angle regression and orthogonal matching pursuit); the ridge regression algorithm presented the best overall predictive performance for the test examples [32]. ...
Nowadays, one of the important and indispensable conditions for the effectiveness and competitiveness of industrial companies is the high efficiency of manufacturing and assembly. These enterprises based on different methods and tools systematically monitor their efficiency metrics with Key Performance Indicators (KPIs). One of these most frequently used metrics is Overall Equipment Effectiveness (OEE), the product of availability, performance and quality. In addition to monitoring, it is also necessary to predict efficiency, which can be implemented with the support of machine learning techniques. This paper presents and compares several supervised machine learning techniques amongst other polynomial regression, lasso regression, ridge regression and gradient boost regression. The aim of this article is to determine the best estimation method for semiautomatic assembly line and large batch size. The case study presented with a real industrial example gives the answer as to which of the cumulative or rolling horizon prediction methods is more accurate.
The synthesis of nanoscale particles and particle aggregates from liquid or gaseous precursors is affected by a variety of trade‐off relations, for example, in terms of product composition, yield, or energy efficiency. Machine‐supported process evaluation and learning (ML) of these relations enables optimization strategies for advanced material processing. Such a workflow is demonstrated on the example of plasma‐assisted aerosol deposition (PAAD) of alumina powders. Depending on processing conditions, these powders comprise of hetero‐aggregate mixtures of crystalline and amorphous polymorphs. Process optimization toward a specific target composition calls for ML approaches. For this, a sufficiently large and consistent dataset of PAAD input (processing) and output (product) parameters is initially generated by real‐world processing, and subsequently extrapolated into a cloud of ≈10⁶ input‐output parameter matrices using Gaussian process regression with multivariate output and input‐output feature analysis. It is subsequently demonstrated how not only the phase composition of the obtained alumina powders, but also product resilience to variations in specific processing parameters, or – as a perspective – the energy efficiency of material processing can be predicted.
This research addresses the need for an advanced and efficient approach to ergonomic assessment by leveraging the METEO (Work and Organization Assessment Tool) method in conjunction with Artificial Intelligence (AI) and computer vision (CV). The primary objective is to identify and quantify the risk factors within work environments that contribute to musculoskeletal disorders (MSD) and injuries, enabling measurable improvements in workplace conditions. To achieve this, a dataset was collected from the mechanical department of “The National School of Arts and Crafts Meknes” for pose estimation using computer vision techniques. The dataset encompassed a diverse range of participants, ensuring representative data for analysis. The developed approach automates the assessment process, effectively eliminating human error variability inherent in traditional observation-based methods. By employing AI and computer vision algorithms, real-time evaluations can be conducted, drastically reducing assessment time from hours to minutes or even seconds. To validate the accuracy and effectiveness of the proposed approach, comprehensive evaluation and validation processes were employed. Performance metrics were established, and comparisons with existing approaches were made to demonstrate the superiority of the proposed method. The research outcomes offer significant implications for workplace safety and employee well-being. By providing an accessible web application, built using appropriate technologies, organizations across various industries can benefit from the automated ergonomic assessment, thereby optimizing work environments and reducing the risk of MSDs. Future implications of this research include expanding the applicability of the developed approach to different industry sectors and exploring potential enhancements in assessing other ergonomic factors beyond posture estimation. This research presents a novel and efficient approach to ergonomic assessment, utilizing the METEO method augmented by AI and computer vision. The integration of these technologies enhances accuracy, reduces assessment time, and holds promise for improving workplace conditions and mitigating MSD risks.
In the manufacturing sector, the ability to predict key performance indicators (KPIs) such as Overall Equipment Efficiency (OEE) is crucial for process optimization and informed decision-making. Although several machine learning methods have been employed to forecast KPIs, their real-time predictive capabilities remain underutilized. This paper presents a novel framework for real-time monitoring and prediction of manufacturing system KPIs, focusing on production and maintenance metrics such as OEE, its constituent factors (performance rate, quality rate, availability), macro/micro stoppages, MTTR (Mean Time To Repair), MTBF (Mean Time Between Failures), and maintenance team availability. Leveraging One-Dimensional Convolutional Neural Networks, the proposed approach offers accurate time series forecasting for KPIs with self-adaptive models that evolve in response to data changes. We implemented this framework in an advanced automotive industry setting, where the results demonstrated high predictive accuracy, enabling better data utilization from the production floor and improved understanding of the plant's status.
Work measurement is a critical aspect of industrial engineering, where various methods are employed to analyze work and optimize productivity. One of the primary tools used by industrial engineers is cycle time, which determines the time taken to complete tasks within a company. However, traditional human analysis of work can be challenging and time-consuming, requiring significant effort to measure and analyze work through direct observations. This paper proposes a new methodology that uses artificial intelligence algorithms, combining ConvLSTM and Faster R-CNN, to identify and track the actions and items manipulated by operators while computing important metrics like cycle time and total wastes. Our approach enables the analysis of human-object interaction from an egocentric perspective, providing an alternative to traditional methods that involve human analysts. We tested our method on the Meccano Dataset and our own dataset, and the results are promising. The proposed methodology is robust, cost-effective, and useful for measuring standard time and identifying waste in various industrial environments. This research demonstrates the potential of artificial intelligence in work measurement and analysis and can lead to significant improvements in productivity and efficiency.
