Data Acquisition and Data Visualization Framework for Smart Manufacturing

The rapid evolution of manufacturing technologies under the Industry 4.0 paradigm has driven the need for efficient data management and real-time decision-making. The client-server architecture is Central to this transformation, enabling seamless communication, data sharing, and process automation across smart manufacturing environments. This literature review explores the pivotal role of client-server architecture and key communication protocols in enabling Smart Manufacturing and Industry 4.0 initiatives. The study examines the fundamental components of client-server architecture, including clients, servers, and communication protocols, and their interactions in industrial settings. It focuses on three primary protocols, MQTT, HTTP, and OPC UA, analyzing their mechanisms, benefits, and applications in smart manufacturing environments. The research highlights how MQTT’s lightweight, publish-subscribe model facilitates efficient real-time data exchange in IoT-heavy industrial networks. HTTP’s widespread adoption and versatility in handling diverse data types are discussed, emphasizing its role in web-based dashboards and cloud integration. The review also delves into OPC UA’s platform-independent nature and advanced features, which promote interoperability and future-proofing of manufacturing systems. This review synthesizes current knowledge of these technologies and underscores their collective importance in addressing critical requirements of smart manufacturing, including scalability, security, real-time capabilities, and cross-system compatibility. The findings suggest that integrating these communication protocols and architectures is crucial for realizing the full potential of Industry 4.0, paving the way for more agile, efficient, and data-driven production ecosystems.

Recently, along with several technological advancements in cyber-physical systems, the revolution of Industry 4.0 has brought in an emerging concept named digital twin (DT), which shows its potential to break the barrier between the physical and cyber space in smart manufacturing. However, it is still difficult to analyze and estimate the real-time structural and environmental parameters in terms of their dynamic changes in digital twinning, especially when facing detection tasks of multiple small objects from a large-scale scene with complex contexts in modern manufacturing environments. In this article, we focus on a small object detection model for DT, aiming to realize the dynamic synchronization between a physical manufacturing system and its virtual representation. Three significant elements, including equipment, product, and operator, are considered as the basic environmental parameters to represent and estimate the dynamic characteristics and real-time changes in building a generic DT system of smart manufacturing workshop. A hybrid deep neural network model, based on the integration of MobileNetv2, YOLOv4, and Openpose, is constructed to identify the real-time status from physical manufacturing environment to virtual space. A learning algorithm is then developed to realize the efficient multitype small object detection based on the feature integration and fusion from both shallow and deep layers, in order to facilitate the modeling, monitoring, and optimizing of the whole manufacturing process in the DT system. Experiments and evaluations conducted in three different use cases demonstrate the effectiveness and usefulness of our proposed method, which can achieve a higher detection accuracy for DT in smart manufacturing.

Justified in the article is the demand for automation of technological processes at industrial enterprises in Russia. This article identifies the possibilities of using artificial intelligence and implementing smart manufacturing in the industry. Successful practices of IoT technologies at foreign enterprises are analyzed. The priorities of the leading Russian industrial companies in the field of digitalization have been revealed: expanding the use of cloud technologies; predictive analysis; IaaS services (virtual data centers and storage systems); supervisory control and data acquisition (SCADA) etc.Targets for priority manufacturing technologies within the federal project Digital Technologies have been identified: mathematical modelling, Smart design and PLM (product lifecycle management), Smart Manufacturing, manipulators and manipulation technologies; the technology of building a comprehensive architecture and strategy for the use of smart manufacturing has been specified. Smart Design provides the implementation of the concept of digital smart manufacturing through Digital Twin technology based on mathematical models of different levels of complexity and adequacy. Smart Manufacturing involves the preparation and implementation of a production process with minimal human participation based on PLM-system data, operational management of technological processes, production, and enterprise. Manipulation technologies include mathematical modelling techniques for robotic systems and the development of appropriate software.The main possibilities of forming the smart potential of industrial production have been revealed: productivity growth and improved product quality; ensuring the reliability and quality of decision-making; infrastructure security guarantees.Results: 1. The practical need to use artificial intelligence and to implement actively smart manufacturing at russian enterprises is justified. 2. The targets for the period up to 2024 on the main technological trends of the development of smart productions in Russia are specified. 3. Opportunities for the formation of smart potential and its strategic use for the development of industrial production are identified.KeywordsAutomationIndustrial enterpriseArtificial intelligenceIoT technologyDigital technologiesDgital designMathematical modellingSmart manufacturingManipulatorsManipulation technologiesArchitectureJEL CodeO10O20O30

Energy efficiency is one of the key drivers for improving the efficiency of resource utilization in smart manufacturing. Within the manufacturing environments, the importance of energy efficiency has grown. However, in most cases the energy consumption of the manufacturing systems are considered using average energy consumption models or simple scheduling models for the needs of discrete event simulation. This paper researches on energy-efficiency scheduling based on the scenario of multi-tasks and multi-resources in smart manufacturing. The scheduling problem is analyzed and modeled as a mixed integer linear programming (MILP) problem with constraints, such as the feasibility of time domain or space, qualification and initial state. A solution of the mixed integer linear programming which optimized by a fast algorithm based on score ranking is designed under time of use (TOU) electricity pricing scheme in smart grid, in which the electricity price varies throughout a day. The feasibility of the method is verified by simulation case. Under the premise of ensuring the reasonable allocation of manufacturing tasks, the goal of smart manufacturing process and long-term energy optimization would be achieved.

Enhanced time-constraint aware tasks scheduling mechanism based on predictive optimization for efficient load balancing in smart manufacturing

Today, data scientists in the manufacturing domain are confronted with various communication standards, protocols and technologies to save and transfer various kinds of data. These circumstances makes it hard to understand, find, access and extract data needed for use case depended applications. One solution could be a data pipelining approach enforced by a semantic model which describes smart manufacturing assets itself and the access to their data along their life-cycle. Many research contributions in smart manufacturing already came out with with reference architectures like the RAMI 4.0 or standards for meta data description or asset classification. Our research builds upon these outcomes and introduces a semantic model based DIN Spec 91345 (RAMI 4.0) compliant data pipelining approach with the smart manufacturing domain as exemplary use case. This paper has a focus on the developed semantic model used to enable an easy data exploration, finding, access and extraction of data, compatible with various used communication standards, protocols and technologies used to save and transfer data.

Industry 4.0, a new industry initiative in Germany, is impacting strongly both on industry and on society. Many newspapers and technical magazines are publishing the state of the art articles on topics such as smart manufacturing based on IoT (Internet of Things), CPPS (Cyber Physical Production System), and cloud-based systems. Other parts of the world have started initiatives such as the IIC (Industrial Internet Consortium) in the US and the IVI (Industrial Value Chain Initiative) in Japan. Smart manufacturing is the key concept underlying these new initiatives. This special issue addresses the most advanced research on smart manufacturing. Subjects cover cyber-physical product-service systems, machinery production lines, manufacturing system simulation, lot-size energy-consumption dependence per production throughput unit, additive manufacturing processes, sensor network technology, production management technology, supply chain management technology, and smart manufacturing reviews. We thank the authors for their careful work and the reviewers for their incisive efforts without which this special issue would not have been possible. We hope that this special issue will trigger further research on smart manufacturing and its advances.

A digital twin to train deep reinforcement learning agent for smart manufacturing plants: Environment, interfaces and intelligence

Artificial intelligence (AI), imitation learning, big data, cloud and distributed computing, robotics cells, and information communication technology, are some of the key tools and elements of the future digital and smart manufacturing facility. There are a number of challenges that digital and smart manufacturing is facing, especially with the complication of AI (i.e., machine, deep and cognitive learning) algorithms, great amount of data to process, and essential complex coding required, which makes immediate changes needed in manufacturing facilities not straightforward. This is notable in small manufacturing cells which is an integrated part of future smart factories such manufacturing facilities are usually needed some annual and regular updates to meet the update in the design specifications of next generation of products. Imitation learning is offering a great opportunity to overcome these challenges and simplify such complications, where human skills, ability to perform specific tasks, knowledge, and talent could be transferred. This is conveying the knowledge, and skills transfer using imitation learning. However, smart manufacturing and industrial revolution needs robotics cells that has skills beyond this, especially when it comes to process optimisation. Therefore, deep imitation learning could come in to help in the development of self-learning robotic systems and cells. Of course with the powerful tools such as distributed computing, blockchain, cloud computing, edge computing, and 5G the collaboration between such self-learning robotic cells will be possible. This will certainly not eliminate human existence but will enhance the manufacturing environment. This paper is focused on presenting the outcomes of CAD simulation and modelling phase of the ongoing research programme that focused on developing a self-learning robotic system using imitation learning. CAD tools have been used and some initial results is presented. Further work is still undertaken, and this will focus on learning from more than one expert, optimisation, impact of dynamic manufacturing environment.

Modern manufacturing systems are complex and heterogeneous distributed information systems that consist of manufacturing hardware and networking software. Unlike traditional manufacturing environments, modern ones generate large amounts of data in real-time that take the form of sensor measurements from the shopfloor, images from the product inspection, technical text from maintenance records, and many more. The issue of collecting, processing, and analyzing this data is at the core of the Industry 4.0 paradigm's success. What is more important, challenging, and rewarding is that such valuable real-time streaming data can be used for diagnostics, prognostics, and health monitoring of systems and products. This dissertation explores the area of health monitoring of modern manufacturing environments to support the state-of-the-art Industry 4.0 paradigms. We deploy Machine Learning (ML), Deep Learning (DL), and Natural Language Processing (NLP) based approaches to assess the performance of systems and products by identifying anomalies and defects, estimating the Remaining Useful Life (RUL), and determining what sort of corrective maintenance actions may be necessary. In the context of the foregoing, advanced manufacturing environments are very intricate and have a vast array of complex problems that can be researched. This dissertation focuses on four important questions: (1) what are the requirements to achieve interoperability in smart manufacturing? (2) how can the next hardware failure, be it of a machine tool or a cutting tool, be predicted before it occurs? (3) how can we identify the products that fail inspection, and how can we improve the overall quality inspection process by considering the factors affecting it? and (4) how can we leverage the resource-rich maintenance logs from manufacturing systems to find the most frequently occurring problems, and more importantly, how can we predict corrective actions for future maintenance problems? This dissertation addresses the above problems by making the following research contributions. In the first work, we discuss interoperability in manufacturing by reviewing semantic, syntactic, factory, and cloud-based interoperability. We then identify key research challenges that impact the implementation of interoperable manufacturing, and review frameworks for modern manufacturing such as Reference Architecture Model for Industry 4.0 (RAMI 4.0), Industrial Internet Reference Architecture (IIRA), and several others. The second part of the research proposes a framework for Smart Prognostics and Health Management (SPHM) to support manufacturing system health. We apply it to a milling machine operation and assess the condition of the milling machine's tool tip by identifying whether it is safe or degraded and estimate its RUL using ML models. To address the third question, we introduce a methodology for Smart Quality Inspection (SQI) of manufacturing products by using a custom Convolutional Neural Network (CNN) architecture. A shopfloor computer application is developed that helps minimize the effects of the factors affecting the visual inspection process. The fourth component of this dissertation introduces a Technical Language Processing (TLP) framework to deal with the unstructured and complex nature of technical text data from maintenance work orders. We demonstrate its capabilities by applying it to maintenance logs of aircraft to predict corrective actions, and link maintenance problems to standardized maintenance codes as prescribed by the industry. The rapid growth of Artificial Intelligence (AI) suggests that its potential to transform health monitoring capabilities is going to improve in the coming years. Future research in health monitoring of manufacturing systems and products can be aimed towards the following areas: development of standardized communication protocols for interoperability, further automation of the inspection process by integrating localized fault detection, and incorporation of TLP with predictive maintenance approaches and quality control to form a prescriptive approach, to suggest a few.--Author's abstract

Industrial Internet of Things (IIoT) is based on data acquisition and data analytics technologies. A variety of and a large amount of data is collected at management nodes with computing and storage capacities. Recently, the computing ability, denoted by the edge, has been located closer to the service fields to achieve faster and more reliable data-driven service provisioning. Edge computing is a useful resource to facilitate smart manufacturing based on IIoT. Since the current Industrial Wireless Sensor Networks (IWSNs) technologies for IIoT do not perfectly cover all the demands of industries with smart manufacturing such as agile flexibility with asset movement. A major future demand for IWSNs should be to support the mobility of assets in a wireless environment. This paper investigates the shortages of current technologies such as WirelessHART and proposes a novel wireless networking scheme based on edge-based cognitive computing in order to support reliable and low latency communication of mobile assets, which is involved in the smart manufacturing processes. We devise a two-tier geographical networking (TTGN) system that supports position-based mobility detection and networking. Also, resource allocation for the reliable and real-time has been obtained based on the game theory. Evaluation results demonstrate that TTGN can guarantee a high data transmission success ratio, as well as a fast delivery ratio for link path establishment.

Carcinogen risk exposure within manufacturing environments is a major occupational health risk which could have long-term implications for workers. The basic methods of monitoring carcinogen exposure include periodic inspections and periodic sampling, which cannot ensure real-time detection and prevention risk. This article presents a novel carcinogen risk assessment framework to combine IoT-enabled AI sensors with predictive technology to potentially address this issue. The framework proposed here would continuously monitor environmentally hazardous conditions in manufacturing settings where carcinogens exist. It would reliably detect whether harmful chemicals, particulate matter, or gases which could pose carcinogenic risks are present. The hazard prediction-program would use best practice machine learning (ML) models to predict exposure levels before they reach potentially hazardous levels, enabling workers to take preventative action in order to avoid the hazard altogether. Subscriber data-stream from IoT sensors that can detect hazardous gases or particulate matter will be accessed via edge AI modules for primary analysis, while cloud-based predictive modeling (i.e., long short-term memory (LSTM) networks and randomized forests) will advise them about the nature of future levels of exposure and exposures. This framework has empirical evidence demonstrating reasonable accuracy in detecting hazardous conditions and only takes a matter of seconds to deliver an alert. Upon notification of hazardous exposure, immediate real-time action is deployed, including cutting production, ventilation control, etc. Findings from simulations demonstrate the system’s ability to detect risk at over 96% accuracy and predict exposure events 2 hours in advance. This gives managers not only increased worker safety, but environmental health compliance as well. In the future, utilizing predictive modeling with IoT technology will represent a real leap forward in smart manufacturing, creating a scalable solution for managing carcinogen exposure risks and protecting worker health in industrial settings.

The Industrial robots are widely used in smart manufacturing environments to improve productivity and automation, but their increasing energy consumption has become a critical issue. Traditional time-based control methods update control signals at fixed sampling intervals, which often leads to actuator redundancy and energy inefficiency, especially in steady-state or low-dynamic operating modes. To address this issue, this paper proposes an energy-efficient event-based control strategy for energy-efficient operation of industrial robots in smart manufacturing systems. The proposed approach integrates an event-based mechanism with nonlinear robot control so that control adjustments are made only when tracking accuracy or energy-related metrics exceed specified thresholds. This design significantly reduces unnecessary control signals while ensuring satisfactory trajectory tracking accuracy. A Lyapunov function-based analysis is performed to demonstrate the stability of the closed-loop control system using the proposed event-based method. Furthermore, the triggering mechanism guarantees a strictly positive lower bound for the intervals between events, thereby eliminating the Zeno effect and ensuring practical feasibility. The energy efficiency of the proposed strategy is further investigated by analyzing the behavior of the actuators and the overall energy consumption. Experimental results show that the proposed approach achieves significant energy savings compared to traditional time-based control methods without compromising control performance. This highlights its potential for energy-efficient operation of industrial robots in smart manufacturing applications.

With the increasing deployment of 5G private networks in smart manufacturing, transportation, and IoT environments, the high performance of these networks has also introduced new cybersecurity challenges. The convergence of Information Technology (IT), Operational Technology (OT), and Communication Technology (CT) significantly expands the attack surface and exposes systems to cross-domain threats such as rogue base station infiltration, distributed denial-of-service (DDoS) attacks, and lateral movement. These attacks threaten data integrity, service continuity, and user privacy. To address these issues, this study applies the MITRE FiGHT threat model to depict representative attack stages—including reconnaissance, intrusion, lateral propagation, and service disruption—and integrates CVSS 3.1 scoring with corresponding mitigation strategies. Based on these components, we construct an analytical workflow capable of quantifying risks, identifying governance deficiencies, and guiding targeted improvements. A prototype Cybersecurity Governance Maturity Model (CSGMM) tailored for 5G private networks is subsequently proposed. The proposed framework incorporates six governance domains—policy, asset management, risk defense, incident response, control practices, and supply chain management—and formalizes a three-tier structure covering strategic, tactical, and operational layers. It also defines twenty-five practice objectives aligned with international standards such as ISO/IEC 27002 and NIST CSF 2.0, improving both applicability and interoperability. Experimental validation was conducted using a Free5GC and UERANSIM testbed to simulate practical attack scenarios, including traffic-based DDoS and endpoint-level exploitation. The results show that the implementation of the governance framework, together with FiGHT-based mitigation strategies, leads to a measurable reduction in CVSS risk scores and attack success probability. These findings demonstrate that integrating governance mechanisms with technical defense measures enhances incident response, strengthens network resilience, and supports continuous security management throughout the lifecycle of 5G private network deployment. The proposed model provides a structured reference for both industry and government seeking to advance 5G cybersecurity strategies. It offers practical value for smart manufacturing and other mission-critical applications, supporting organizations in addressing increasingly complex cyber threats in the era of digital transformation.

In recent years, the emergence of Cyber-Physical systems (CPS), big data, cloud computing, and industrial wireless networks have promoted the development of Industry 4.0. Intelligent manufacturing systems are the leaders of this change. In smart manufacturing, the emphasis will be placed on using technologies such as big data analytics, cloud computing, edge computing, and artificial intelligence (AI). The advantages of new technologies are significantly reflected in the comparison of preventive maintenance and traditional maintenance based on intelligent technology: preventive maintenance significantly reduces the cost of prevention and, at the same time, dramatically improves the accuracy and timeliness of maintenance, bringing inestimable value to intelligent manufacturing. This paper will analyze new technologies’ innovative advantages and technical prospects through big data analysis and specific application scenarios of AI technology in preventive maintenance. Firstly, in data acquisition, this paper examines the architecture of preventive maintenance systems based on big data. Then, it analyzes the composition of the data pipeline: in the realization of data collection and processing in the field of cloud computing and edge computing, and the aspect of data processing, it discusses the participation of AI technology in data understanding and processing and the unprecedented changes it brings.