Aug 2026· Journal of Engineering Research and Reports· 0 citations
TL;DR
The transition from localised control to an AI-driven autonomous framework, supported by Digital Twins and Explainable AI, provides a transparent approach to modernising glass manufacturing and may reduce operational risks and environmental impacts, thereby supporting intelligent and sustainable industrial automation.
Abstract
Aims: This research aims to advance industrial glass manufacturing by implementing a next-generation autonomous control architecture. The study focuses on maximising energy efficiency and product quality by integrating Digital Twin technology, Deep Reinforcement Learning (DRL), and Explainable AI (XAI) to overcome the limitations of legacy control systems.
Study Design: This is an analytical and simulation-based research study focused on the cognitive optimisation of industrial float glass production processes.
Place and Duration of Study: The study was conducted in the Department of "Instrumentation Engineering" (Intellectual Measurement and Control Systems), Azerbaijan State Oil and Industry University (ASOIU), between September 2025 and June 2027.
Methodology: A high-fidelity Digital Twin of a float glass furnace was developed to simulate production dynamics. A DRL-based agent was implemented for real-time furnace regulation, allowing for continuous self-optimisation of thermal zones. The system integrated a Convolutional Neural Network (CNN)-based visual inspection layer for defect detection, complemented by an XAI module that provides transparent, logic-based justifications for autonomous operational adjustments. Theoretical evaluations of hydrogen-natural-gas blending were also conducted to assess sustainability impacts.
Results: The proposed DRL-driven architecture achieved improved thermal regulation, maintaining stability within ±0.5°C compared with traditional Fuzzy-PID models. The combustion strategy, optimised via reinforcement learning, produced a reported reduction in fuel consumption of more than 6% (P < 0.05). The XAI module provided real-time interpretability of system decisions, while the integration of hydrogen-enriched combustion pathways indicated a potential decrease in carbon emissions of approximately 8-10% and maintained high combustion efficiency compared with standard natural gas use.
Conclusion: The transition from localised control to an AI-driven autonomous framework, supported by Digital Twins and Explainable AI, provides a transparent approach to modernising glass manufacturing. These systems may reduce operational risks and environmental impacts, thereby supporting intelligent and sustainable industrial automation.
: The rapid adoption of artificial intelligence (AI) in the production sector has triggered a revolutionary change in the design, manufacturing, monitoring, and optimization. The paper is a critical analysis of AI-based additive manufacturing (AM) and smart manufacturing systems with an emphasis on design intelligence and process optimization, real-time quality control, and sustainable production. Based on recent literature, this paper considers the major AI methods, such as machine learning (ML), deep learning (DL), reinforcement learning (RL), and physics-informed neural networks (PINNs) throughout the manufacturing lifecycle. The results show that AI can greatly decrease the time spent on design iterations, increase the accuracy of predictions of process parameters, and allow in-situ defect detection with high accuracy. Moreover, AI
A new hybrid artificial intelligence system of smart manufacturing is suggested, combining Long Short-Term Memory networks, Convolutional Neural Networks, Convolutional Neural Networks, ensemble tree-based classifiers, and a Proximal Policy Optimization-based Reinforcement Learning agent in a four-layer system that includes data acquisition, AI processing, decision control, and feedback actuation.
C. T· Materials Research Proceedin...· 0 citations
Abstract. The intersection of artificial intelligence (AI) and sustainable manufacturing offers an innovative possibility to decrease industrial waste and, at the same time, improve the production yield. The current paper suggests a new multi-layer AI architecture including a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) neural network and a Multi-Objective Genetic Algorithm (MOGA) optimizer to control real-time process in green manufacturing setting. The architecture acts on the heterogeneous sensor streams, energy metering information, and process logs in a five-stage pipeline including data acquisition, preprocessing, AI inference, decision support and closed-loop feedback. Two industrial datasets (n = 18,400 samples and n = 14,200 samples) of semiconductor fabrication and automotive body stamping data (20192023) are experimentally validated. The proposed system attains a waste reduction of 34.7, a yield improvement of 17.1 and prediction RMSE of 1.72 as compared to the traditional machine learning baselines. The explainability analysis using SHAP reveals that the dominant process variables are the coolant temperature, spindle speed and feed rate. The findings affirm that the proposed hybrid AI architecture is more effective than the current approaches and provides a deployable and scalable system towards smart green manufacturing.
Irshadullah Asim Mohammed· Materials Research Proceedin...· 0 citations
The development of AI-enabled advisory systems for casting processes are presented, integrating singular value decomposition (SVD)-based reduced-order models with a Variational Autoencoder with Arbitrary Conditioning (AC-VAE) and hybrid simulation frameworks to support real-time process prediction and optimization.
Sofija Milicic, A. Horr, S. Elgeti et al.· Processes· 0 citations
Quality control is a fundamental function of manufacturing because product conformity, process stability, customer satisfaction and operational efficiency depend on the ability of manufacturers to detect and prevent deviations from specified requirements. Conventional quality-control practices, although effective in many applications, frequently depend on manual inspection, statistical process control and rule-based decision-making, which may be inadequate for highly customised, high-speed and data-intensive production environments. The emergence of artificial intelligence (AI), machine learning (ML), deep learning (DL), computer vision, industrial Internet of Things (IIoT), edge computing and digital twins has created new opportunities for transforming quality control from a predominantly reactive activity into a predictive and autonomous manufacturing function. This paper examines the application of AI-driven quality control in modern manufacturing systems through a critical review of recent literature. The paper discusses AI-enabled visual inspection, predictive quality, process monitoring, anomaly detection, defect classification, intelligent metrology, predictive maintenance and closed-loop quality control. Particular attention is given to the integration of AI with Industry 4.0 technologies and the implications for production efficiency, waste reduction, process capability and decision-making. The paper also identifies major challenges, including data scarcity, model explainability, algorithmic generalisation, cybersecurity, computational requirements, system integration and workforce readiness. A conceptual framework for AI-driven quality control is proposed, linking data acquisition, intelligent analytics, quality prediction, decision support and corrective action. The paper concludes that AI should not be viewed simply as a replacement for human inspection but as an intelligent layer that augments engineering knowledge and enables manufacturing systems to anticipate, diagnose and correct quality problems. Future research should emphasise explainable, robust, resource-efficient and human-centred AI systems capable of operating reliably in real industrial environments.
Olusegun Olukayode Olaleye, E.J. Wilson· International Journal of Afr...· 0 citations
Smart manufacturing is undergoing a transformation due to the development of Generative Artificial Intelligence (GenAI), which introduces a new level of autonomy and data-driven decision-making in industrial settings. Unlike conventional prediction and classification type AI, GenAI can also be applied to develop optimized production plans, adjust production plans to changes, generate synthetic engineering data, intelligent design options, and real-time production recommendations. The review provides an in-depth overview of the use of GenAI in contemporary manufacturing systems such as process optimization, predictive maintenance, quality assurance, digital twins, intelligent robotics, supply chain resilience, and autonomous production control. It covers the most recent developments in generative models, including Large Language Models (LLMs), Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs) and diffusion models, as well as how these models are used in manufacturing process planning, defect detection, scheduling optimization and human-machine collaboration. This also encompasses the relationship between GenAI and Industrial Internet of Things (IIoT), cyber-physical systems, cloud-edge computing, and Industry 5.0, which will be used to create a self-adaptive manufacturing environment. Technical, ethical, and organizational issues, including data quality, model interpretable, cybersecurity, computational burden, workforce adaptation, and regulatory compliance issues, are explored. Lastly, the review spills out research gaps in this rapidly evolving technology field and provides directions for developing trustworthy, explainable and sustainable GenAI-enabled manufacturing systems.
Fischer Erik Krisztián, Viktor Gulyás-Oldal, Imre Gálóczi et al.· Veredas do Direito· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
