2020· International Journal of Data Engineering and Intelligent Computing· 0 citations
TL;DR
This article delves deep into the confluence of simulation, ML, and statistics, showcasing how they synergize to improve engineering workflows and emphasizes that DCE is not just a technological advancement but a foundational strategy for next-generation engineering solutions.
Abstract
Data-centric engineering (DCE) represents a paradigm shift in modern engineering by merging traditional physics-based simulation approaches with machine learning (ML) and statistical methods. As industries face increasing demands for system complexity, efficiency, and reliability, DCE offers a robust framework to model, predict, and optimize engineering processes and products. This article delves deep into the confluence of simulation, ML, and statistics, showcasing how they synergize to improve engineering workflows. By leveraging high-fidelity simulations, advanced ML algorithms, and statistical inference, DCE enables real-time decision-making, anomaly detection, design optimization, and predictive maintenance. In this paper, we explore the historical evolution of DCE, current applications, and its transformative potential across aerospace, automotive, civil infrastructure, and manufacturing domains. The integration of multi-source data, model uncertainty quantification, and digital twins forms the core of our discussion. We present detailed methodologies for hybrid modeling approaches and data fusion techniques and provide experimental results that validate the efficiency of DCE in improving performance metrics. Case studies demonstrate significant improvements in product life cycles, resource allocation, and safety standards. The findings emphasize that DCE is not just a technological advancement but a foundational strategy for next-generation engineering solutions.
It is argued that first-principles models (FPMs) remain essential for mission- and business-critical workflows in industrial automation, both in process design and operations and hybrid intelligence integrates mechanistic rigor with data-driven insights to deliver smarter design, safer operations, and more sustainable processes.
Hongzhi Zhao, Shu Wang, Salvador I. Pérez-Uresti et al.· Industrial & Engineering...· 0 citations
Data science methodologies are playing an increasingly important role in advancing manufacturing systems, enabling improvements in efficiency, energy usage, cost reduction, product quality, and predictive maintenance capabilities. This raises a fundamental question: to what extent can data models reshape manufacturing processes, and what limitations prevent their full-scale adoption? Recent developments show a growing integration of data models within digital twin and digital shadow architectures, facilitating real-time monitoring and decision-making. Nonetheless, the complexity of industrial processes and the scarcity of high-quality, well-structured datasets pose significant challenges, particularly in terms of model robustness, interpretability, and scalability. Importantly, the effectiveness of such models depends more on data quality and representativeness than on data quantity alone. This review presents a structured analysis of data modeling techniques for manufacturing applications, with emphasis on data generation, sampling, preprocessing, and modeling approaches across diverse operational regimes, including steady, transient, and generative processes.
The use of Machine Learning (ML) is rapidly expanding across diverse scientific and engineering domains. ML offers a powerful advantage over traditional modeling approaches for predictive modeling and analysis of variables of interest. This makes it particularly useful for developing advanced methods in analytical chemistry and residential energy systems, including forecasting (such as predicting hot water demand or chromatographic peak behavior), data quality assessment (such as detecting sensor drift or anomalous consumption patterns), and fault detection (such as identifying heat pump malfunctions or degraded separation performance). While traditional modeling approaches struggle to fully exploit complex, high-dimensional features, existing ML studies in the target domains often rely on limited datasets and lack automated and adaptive frameworks capable of handling the scale, variability, and non-stationary data generated in real operational settings. The availability of large amounts of data in this digital era offers unprecedented opportunities for analysis. However, the successful application of ML depends critically on the quality, preparation, and robustness of ML models and their underlying data. In industrial systems, these requirements are shaped by several interacting factors, particularly feature representation, model robustness, and adaptation under changing conditions. The challenges addressed in this research include data quality management, model selection, robustness assessment, and adaptation under changing real-world conditions. The main contributions of this research are: (i) a semi-automatic data preparation workflow with domain-specific feature engineering for large-scale oligonucleotide chromatography datasets; (ii) an unsupervised quality-centric evaluation framework that automatically clusters input data by quality level without requiring labeled annotations; (iii) the FIUL-Data fault injection framework, which quantifies the resilience boundaries of ML models under controlled data degradation; and (iv) a composite adaptive framework that integrates predictive ML with anomaly detection to enable demand-driven heat pump management in residential energy systems. Together, these contributions demonstrate that reliable industrial ML is achieved not by increasing model complexity, but through systematic data-centric practices including structured data preparation, quality-aware pipelines, robustness testing, and adaptive learning, applied across two complementary industrial domains.
Physics-informed machine learning, digital twins, and additive manufacturing are a new direction for the creation of intelligent, adaptive, and high-performance engineering systems that are being integrated into a smart industrial framework. The strategy combines multimodal sensing, data fusion, physics-based modeling, machine learning, process optimization, and closed-loop control, and addresses the challenges of enhancing manufacturing performance across the product life cycle. In physics-informed machine learning, physics principles are incorporated into the data-driven models, which enhances prediction accuracy, decreases the need for large data sets, and facilitates generalization from model to model for different process conditions. Digital twins are virtual models of AM systems that support real-time monitoring and anomaly detection, predictive analysis, virtual experiments, and adaptive process control. The combination of edge computing and intelligent controllers enhances quick decision-making processes during the fabrication process. Aerospace, defense, biomedical engineering, and advanced composite manufacturing are just a few of the applications that show promise for achieving better dimensional accuracy, defect reduction, lightweight design, energy efficiency, material utilization, and process traceability. Industrial deployment is, however, hindered by the lack of high-quality datasets, class imbalance, limited model transferability, interoperability, high computational requirements, cybersecurity, certification, and lifecycle governance. For scalable implementation, standardized data formats, open architectures, benchmark datasets, federated learning, hybrid modeling, and rigorous validation procedures are all important. In general, physics-based informed intelligence and adaptive AM create a promising basis for reliable, efficient, traceable, and environmentally friendly high-performance engineering.
Fahmina Afrin· Journal of Artificial Intell...· 0 citations
Ten contributions are brought together to demonstrate how the systematic integration of physical knowledge can enhance model robustness, reduce data requirements, and improve generalization across manufacturing applications.
Jiewu Leng, Hui Yang, Min Xia et al.· Journal of Computing and Inf...· 0 citations
Machine learning, digital twins, cyber-physical systems, and smart infrastructure are changing the way additive and hybrid manufacturing goes from static, process-defined to adaptive, data-driven manufacturing. The reviewed integrated architectural approach is able to link manufacturing at the physical level with the sensing, data infrastructure, physics-based modelling, surrogate modelling, artificial intelligence, and closed-loop control levels. Digital threads enable ongoing data connectivity and traceability from design to production, inspection, and maintenance phases, and digital twins maintain a dynamic representation of changing conditions in the processes. In the manufacturing sector, Edge and cloud infrastructure make it possible to capture and process data in real time and manage and analyze it at scale in a variety of factory conditions. Surrogate and physics-informed models complement high-fidelity physics-based simulations for reducing computational demands and enabling rapid prediction and optimization. Layer-to-layer and within-layer control strategies further allow the adjustment of manufacturing parameters in an adaptive way using real-time process information. The framework also introduces the possibility of hybrid manufacturing processes: Additive deposition and subtractive machining, finishing, and inspection processes are linked through continuous digital data exchange. Key needs for safe industrial deployment are identified to include safety, cyber security, regulatory compliance, data governance, and model traceability. Overall, the integrated approach offers a way to more autonomous, responsive, traceable, and efficient manufacturing systems, and puts the emphasis on the remaining need for physics-based knowledge, transparent decision making, human supervision, and validated control architectures.
M. Uddin, Sumi Ghosh· Journal of Computer Science...· 0 citations