Garuda - Garba Rujukan Digital

Semantic AI-Orchestrated Cross-Domain Automation Framework for Sustainable Smart Factories Milan Bharatkumar Makwana
The Eastasouth Management and Business Vol. 3 No. 02 (2025): The Eastasouth Management and Business (ESMB)
Publisher : Eastasouth Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/esmb.v3i02.1145

Manufacturing sectors pursuing Industry 5.0 objectives increasingly require automation architectures capable of reasoning across heterogeneous cyber-physical domains while advancing sustainability targets. This paper synthesizes evidence from eighteen references spanning semantic web technologies, cyber-physical systems, digital twins, blockchain-enabled traceability, and multi-agent reinforcement learning to propose a Semantic AI-Orchestrated Cross-Domain Automation Framework for sustainable smart factories. The framework integrates a five-layer architecture, comprising physical sensing, cyber-physical integration, semantic reasoning, cross-domain orchestration, and sustainability decision layers, into a unified reasoning pipeline in which ontology-driven knowledge graphs mediate interoperability among heterogeneous equipment, enterprise systems, and human operators. Comparative synthesis indicates that semantic-based autonomous computing architectures reduce unplanned downtime by as much as 37 percent, while multi-agent reinforcement learning scheduling raises resource utilization to 88 percent relative to 58 percent under conventional rule-based scheduling. Interoperability standards such as OPC-UA demonstrate adoption rates near 78 percent among reviewed implementations, and hybrid semantic-blockchain ledgers achieve transaction throughput exceeding 2,100 transactions per second at latencies below 40 seconds, outperforming public proof-of-work ledgers by more than two orders of magnitude. Digital twin adoption trajectories synthesized from the references rose from approximately 8 percent in 2016 to 66 percent by 2024, correlating with a 24 percent reduction in energy consumption and a 31 percent reduction in material waste across reviewed sustainable manufacturing cases. The findings imply that semantic orchestration, combined with decentralized ledgers and human-centric digital twins, offers a scalable pathway toward resilient, low-carbon, and economically viable smart factory operations, while highlighting persistent challenges in ontology standardization, explainability, and cross-organizational governance.

Carbon-Aware Intelligent Electrical Distribution Network Using Digital Twins and Edge Artificial Intelligence Milan Bharatkumar Makwana
The Eastasouth Journal of Information System and Computer Science Vol. 1 No. 03 (2024): The Eastasouth Journal of Information System and Computer Science (ESISCS)
Publisher : Eastasouth Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/esiscs.v1i03.1147

The escalating integration of distributed energy resources and the urgency of decarbonization targets have exposed the limitations of conventional, centrally managed electrical distribution networks. This paper synthesizes findings from twenty-three studies to examine how digital twin technology and edge artificial intelligence can be combined to form a carbon-aware intelligent distribution network. The problem addressed concerns the inability of legacy supervisory control and data acquisition architectures to deliver the millisecond-scale responsiveness, emission transparency, and adaptive reconfiguration required by high-penetration renewable grids. A layered conceptual framework is proposed, integrating field sensing, edge inference, digital twin synchronisation, and carbon-weighted reinforcement learning control. Evidence drawn from the literature indicates that hybrid edge-cloud architectures reduce control-loop latency from several hundred milliseconds to below 100 milliseconds relative to cloud-only deployment, while safe deep reinforcement learning controllers for Volt-VAR optimisation converge within 300 to 500 training episodes and reduce voltage violations substantially. Carbon emission flow modelling combined with temporally shifted, carbon-aware scheduling is shown to yield emission reductions ranging from approximately 10 per cent to more than 25 per cent when co-optimised with digital twin state estimation. The synthesis further identifies bandwidth reduction of up to 82 per cent and inference accuracy gains of nearly 5 percentage points for hybrid configurations. Implications include improved grid resilience, measurable decarbonization, and a pathway toward regulatory-compliant, self-optimising distribution networks, with future work directed toward standardised digital twin interoperability and federated edge learning.

Edge Intelligence-Enabled Self-Calibrating Smart Instrumentation for Autonomous Process Industries Milan Bharatkumar Makwana
The Eastasouth Journal of Information System and Computer Science Vol. 2 No. 01 (2024): The Eastasouth Journal of Information System and Computer Science (ESISCS)
Publisher : Eastasouth Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/esiscs.v2i01.1149

A dense network of smart instrumentation is essential to autonomous process industries, with the accuracy of these sensors constantly being compromised by sensor drift, fouling, temperature changes and degradation over time. Routine manual recalibration is expensive, disruptive and is not able to cope with realtime degradation. Based on sixteen literature references dealing with edge computing, edge intelligence, TinyML, soft sensing and concept drift management, this paper proposes a combined approach for the development of self-calibrating smart instrumentation for autonomous plants. The synthesis results show that the edge intelligence architectures achieve a latency of ~420ms with cloud-only processing, but a latency of <40ms when quantized models are processed on-device, and a reduction in energy consumption per inference from 2.85 joules to 0.31 joules. Cluster-based statistical drift detectors and optimal-transport transfer learning are demonstrated to maintain calibration accuracy 90 percent or higher for more than a year of use without manual corrections. TinyML quantization drastically reduces the footprint of neural drift estimators to fit microcontroller-class devices with less than 256 kilobytes of static random-access memory. The proposed layered architecture integrates field sensing, edge-resident calibration controllers, and federated retraining in the cloud to support the measurement traceability while reducing the bandwidth requirement by up to 87 percent compared to raw data streaming. Results also show that self-calibrating instrumentation decreases the number of unplanned maintenance visits and total measurement uncertainty. The paper finds that edge intelligence and self-calibration represent a technically mature and economically viable roadmap for achieving full autonomy of process instrumentation, though there are still remaining challenges related to standardization, cybersecurity and long-term model drift.

Federated Learning-Based Adaptive Control Architecture for Autonomous Smart Manufacturing Systems Milan Bharatkumar Makwana
The Eastasouth Journal of Information System and Computer Science Vol. 3 No. 02 (2025): The Eastasouth Journal of Information System and Computer Science (ESISCS)
Publisher : Eastasouth Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/esiscs.v3i02.1150

The machine learning architectures for autonomous, smart manufacturing systems must guarantee data privacy across geographically distributed production cells and be able to continuously adapt to the non-stationary process conditions. In this paper, recent advancements in federated learning, secure aggregation, blockchain-based trust management, digital twin simulation, and reinforcement learning are brought together to design a layered adaptive control framework for autonomous smart manufacturing. The architecture incorporates a novel edge-resident local training method, a drift-aware adaptive aggregation mechanism, a digital-twin-validated reinforcement learning control policy, and a blockchain ledger to trace the provenance of the model for auditability. Both results suggest that drift-aware weighted aggregation achieves about 0.958 global model accuracy after 100 communication rounds, while standard federated averaging achieves about 0.887 global model accuracy in the same number of rounds. When adaptive scheduling and secure aggregation are combined, the estimated reduction in communication overhead is around 60% compared to a default schedule. A comparative assessment along privacy, scalability, latency resilience, robustness, auditability and adaptivity dimensions shows that the proposed architecture is superior compared to centralized control and standard federated learning baselines, especially in terms of auditability and robustness to non-independent and non-identically distributed data. The results indicate that a viable path towards confidential, resilient and continuously adaptive control of autonomous manufacturing equipment could be achieved by combining federated optimization with blockchain-verified digital twin validation.

Title

Found 4 Documents
Search

Abstract

Abstract

Abstract

Abstract

Title Search

Found 4 Documents Search

Abstract

Abstract

Abstract

Abstract

Title

Found 4 Documents
Search