Smart Manufacturing Process Research Articles

Predictive models for 3D inkjet material printer using automated image analysis and machine learning algorithms

Additive manufacturing (AM) is a smart manufacturing process to fabricate components with high precision, minimal post-processing, and increased component complexity in a variety of materials. This research focuses on developing automated image analysis and predictive models for a widely used 3D material inkjet printing (IJP) process. The interplay of four input process parameters, which include frequency, voltage, temperature, and meniscus vacuum, on the output metrics of the inkjet printer was evaluated using statistical measures (ANOVA). Droplet types were classified as no drop, satellite drop, and normal drop using four machine learning classifiers, including random forest, support vector classifier, k-nearest neighbor, and decision trees. Hyperparameter tuning was performed for each model for over 486 data points. Regression predictive models were developed for both ink droplet velocity and volume with three linear models (linear, lasso, and ridge regression) and four non-linear models (random forest, decision tree, support vector regression, and k-nearest neighbor). Mean squared error and the coefficient of determination, r-squared value, were used to evaluate the performance of the predictive models. For the drop type classification models, k-fold of 5 yielded the highest accuracy for the RF, KNN, and DT models of around 98%. Similarly, for the regression based predictive models RF, DT and KNN accuracy results ranged from 97 to 99%. All the machine learning models were validated with experimental data with high prediction accuracies accuracy. This research serves as a foundation for developing design guidelines for 3D material inkjet printing with applications in biosensors, flexible electronics, and regenerative tissue engineering.

2D Super-Resolution Metrology Based on Superoscillatory Light.

Progress in the semiconductor industry relies on the development of increasingly compact devices consisting of complex geometries made from diverse materials. Precise, label-free, and real-time metrology is needed for the characterization and quality control of such structures in both scientific research and industry. However, optical metrology of 2D sub-wavelength structures with nanometer resolution remains a major challenge. Here, a single-shot and label-free optical metrology approach that determines 2D features of nanostructures, is introduced. Accurate experimental measurements with a random statistical error of 18nm (λ/27) are demonstrated, while simulations suggest that 6nm (λ/81) may be possible. This is far beyond the diffraction limit that affects conventional metrology. This metrology employs neural network processing of images of the 2D nano-objects interacting with a phase singularity of the incident topologically structured superoscillatory light. A comparison between conventional and topologically structured illuminations shows that the presence of a singularity with a giant phase gradient substantially improves the retrieval of object information in such an optical metrology. This non-invasive nano-metrology opens a range of application opportunities for smart manufacturing processes, quality control, and advanced materials characterization.

Extended reality implementation possibilities in direct energy deposition-arc

The state-of-the-art cleaner smart manufacturing process in the metal industry is the direct energy deposition-arc (DED-arc) process, which has emerged as an energy-efficient method for producing complex geometry metallic constructions. Process flexibility, material-consumption efficiency and high performance have drawn attention amongst both academics and industry, as DED-arc presents an ecologically viable alternative to traditional manufacturing techniques. Concurrently, the parallel emergence of extended reality (XR) technology has unveiled multiple novel possibilities for enhancing the sustainable development of DED-arc processing toward cleaner manufacturing. However, an evident knowledge gap exists concerning the integration of XR into the DED-arc process chain. This research aims to solve this problem by systematically exploring the potential of implementing XR technology within the DED-arc framework. Therefore, this study identifies through a literature review the technological difficulties and prospects associated with merging XR and DED-arc. Subsequently, a series of practical experiments are executed, presenting various applications of XR within the DED-arc process chain. The current research makes several noteworthy contributions to the practical understanding of how XR can be integrated into the DED-arc manufacturing process. Technological challenges are discussed, while the potential benefits of XR adoption in the DED-arc process chain are illuminated in practical applications.

Investigating smart manufacturing process implementation in the Indian manufacturing industries using tecnomatix and response surface methodology

Investigating smart manufacturing process implementation in the Indian manufacturing industries using tecnomatix and response surface methodology

Enhancing Cybersecurity in Smart Manufacturing: A Comprehensive Approach

The industry 4.0 defines a new era of much efficient and intelligent manufacturing processes that works in integration with other advanced technologies like AI, Big data, Block chain, etc. The smart manufacturing industry has remarkably evolved due to integration of advanced technologies like cyber-physical systems (CPS), Internet of Things (IoT), digital twins that makes up a highly connected infrastructure where data flows in real time. However, the involvement of these technologies has also significantly increased the exposure of manufacturing industry to the cyber threats. This article investigates several security challenges and major threats that are associated with the manufacturing industry. It evaluates various security implementations including digital twins, machine learning, and block chain based on their effectiveness, scalability, and feasibility in terms of security. Through a comprehensive literature review, case study analysis of notable cyberattacks and qualitative exploratory methods, this study explores the best line of defence against the cyber threats. Our findings highlight the critical role of robust and adaptive multiple line of defense mechanisms in mitigating cyber risks. The research concludes by emphasizing the need for an effective security strategy that combines advanced technological solutions with continuous employee education and proactive threat management to safeguard smart manufacturing processes against evolving cyber threats.

P‐234: Late‐News Poster: A Data‐Centric Approach to Minimize Defect Leakage in an AI‐based Automated Surface Inspection System for Display Manufacturing Process

This study advances Surface Inspection (SI) in display panel manufacturing using a data‐centric approach, addressing high gray zone rates and data discrepancies. We employed a data flywheel method with dual labeling to improve dataset quality. Results show expanded automated coverage and enhanced classification, reducing defect leakage, demonstrating AI's impact in smart manufacturing processes.

A Smart Manufacturing Process for Textile Industry Automation under Uncertainties

Most textile manufacturing companies in the world heavily rely on manual labor, particularly in the fabric inspection section, especially for cotton fabric. Establishing smart manufacturing systems like industrial automation in the textile industry for cotton fabric inspection is important for error-free inspection. The proposed make-to-order (MTO) inventory model focuses on the strategic development of a supply chain network under fuzzy uncertainty. The distinctiveness of this research lies in integrating a methodology that involves human and machine interaction, along with allocating resources to investment in smart manufacturing. This article presents a case study of the Jagatjit Cotton Textiles (JCT) manufacturing company in Punjab, India, as an example to validate the model and check the performance of SMT in the fabric inspection process in cotton TC mills. This paper contributes by developing four distinct textile supply chain models with industrial automation under triangular and trapezoidal fuzzy demand. A numerical analysis is conducted to verify the effectiveness of installing automated fabric inspection machines in the cotton plant. This article proposes an iterative solution algorithm (KDPMG) to obtain the global optimum for the proposed model. A comparative study of the proposed algorithm, KDPMG, and the genetic algorithm (GA) is presented in this study to verify the credibility of the obtained results. It is observed that KDPMG provides more appropriate solutions to the problem compared to the GA. Moreover, the computational time of KDPMG is significantly less than that of the GA. The rigorous analysis reveals that maximum profit can be achieved under trapezoidal fuzzy demand with fully automated fabric inspection technology. Using a triangular fuzzy demand pattern, the model with fully automated smart manufacturing achieves an 8.62% higher profit compared to a traditional system. Similarly, in the case of a trapezoidal fuzzy demand pattern, the adoption of automation in cotton plants can achieve an 8.69% higher profit. Hence, the implementation of smart manufacturing systems in the mending section of the cotton textile industry proves to be more profitable compared to the traditional inspection process.

Intrinsic and post-hoc XAI approaches for fingerprint identification and response prediction in smart manufacturing processes

AbstractIn quest of improving the productivity and efficiency of manufacturing processes, Artificial Intelligence (AI) is being used extensively for response prediction, model dimensionality reduction, process optimization, and monitoring. Though having superior accuracy, AI predictions are unintelligible to the end users and stakeholders due to their opaqueness. Thus, building interpretable and inclusive machine learning (ML) models is a vital part of the smart manufacturing paradigm to establish traceability and repeatability. The study addresses this fundamental limitation of AI-driven manufacturing processes by introducing a novel Explainable AI (XAI) approach to develop interpretable processes and product fingerprints. Here the explainability is implemented in two stages: by developing interpretable representations for the fingerprints, and by posthoc explanations. Also, for the first time, the concept of process fingerprints is extended to develop an interpretable probabilistic model for bottleneck events during manufacturing processes. The approach is demonstrated using two datasets: nanosecond pulsed laser ablation to produce superhydrophobic surfaces and wire EDM real-time monitoring dataset during the machining of Inconel 718. The fingerprint identification is performed using a global Lipschitz functions optimization tool (MaxLIPO) and a stacked ensemble model is used for response prediction. The proposed interpretable fingerprint approach is robust to change in processes and can responsively handle both continuous and categorical responses alike. Implementation of XAI not only provided useful insights into the process physics but also revealed the decision-making logic for local predictions.

The Role of Ai in Transforming Smes: Opportunities and Challenges

Artificial Intelligence (AI) systems assist in enhancing data analysis with meaningful insights and promote real-time decision construction that implements blockchain technology in data security. This factor guides SMEs in managing big data by usage of predictive analysis and algorithms in reducing operational complexity in the USA. There are 33.2 million SMEs in the US and contribute 50% of the overall GDP in this country as per the 2023 report [1]. The application of AI and automation technology leads to improved work distribution and enhanced overall performance in handling the SME's services. This factor encouraged operational facilities to use the real-time tracking system and promote the level of the smart manufacturing process. The primary research used 101 participants to conduct a survey on SME employees for the transformation of SMEs with AI in analyzing opportunities and challenges. This analysis reflects that most of the employees felt encouraged to adopt AI and blockchain in expanding the overall productivity of USA SMEs.

Toward smart and sustainable cement manufacturing process: Analysis and optimization of cement clinker quality using thermodynamic and data-informed approaches

Cement manufacturing is widely recognized for its harmful impacts on the natural environment. In recent years, efforts have been made to improve the sustainability of cement manufacturing through the use of renewable energy, the capture of CO2 emissions, and partial replacement of cement with supplementary cementitious materials. To further enhance sustainability, optimizing the cement manufacturing process is essential. This can be achieved through the prediction and optimization of clinker phases in relation to chemical compositions of raw materials and manufacturing conditions. Cement clinkers are produced by heating raw materials in kilns, where both raw material compositions and processing conditions dictate the final chemical makeup of the clinkers. This study uses thermodynamic simulations to analyze phase assemblages of alite- and belite-enriched clinkers based on chemical compositions of raw materials and to create a database. The thermodynamic simulations can accurately reproduce clinker phases in comparison with experimental results. Subsequently, the simulated database is employed to train a data-informed model, and the predictions are used to determine the optimal composition domains that produce high quality clinker (C3S>50 %) at different calcination temperatures. Additionally, optimal lime saturation factor and alumina modulus are investigated to achieve target clinker phases. Overall, this study demonstrates the potential of using a data-informed approach to achieve smart and sustainable cement manufacturing process.

Development of a Human Centric Cyber Physical Production System Framework for Enhanced Social Sustainability

Social sustainability focuses on building a sustainable workplace that prioritizes the occupational health, safety, and overall well-being of workers. It is considered one of the three fundamental pillars of sustainability. Human centric cyber physical production system (CPPS) emphasizes the central role of humans in the smart manufacturing process that has become an essential requirement to enhance social sustainability. This paper proposes a human centric CPPS framework for enhanced social sustainability. The social needs/requirements are identified and categorized into various types. Similarly, functional and design requirements are categorized into different elements and sub-elements of CPPS. The findings are used to create a QFD (Quality Function Deployment) matrix that integrates social requirements with the functional and design requirements of modular CPPS elements. The present work will be significant in enhancing the management capabilities and performance of traditional manufacturing systems, while also meeting diverse social needs and requirements.

XAI-driven digital twin for cobot dynamic error compensation

Process and product fingerprints (FP) have been approved as effective parameters to reveal the principal contributing factors towards functionality in smart manufacturing processes. Though AI-driven methods outperform other approaches for fingerprint extraction, the lack of explainability in its black-box style predictions leads to misconceptions and trust issues among stakeholders. In this study, a novel explainable-AI (XAI) approach is proposed to identify mathematical fingerprint expressions by formulating them as graphs using the QLattice algorithm, inspired by path integral formulation. Here, the Qlattice model identifies explainable and human-comprehensible FP expressions for cobot dynamic error based on accelerometer signal features. The discovered symbolic model is subsequently applied to a digital twin which successfully tracked and compensated for dynamic errors autonomously in real time.

On the Verification of Distributed Control for Multi Job Shop Assignment Problem in Smart Manufacturing System

The increasing need of product individualization and rapid demand fluctuation leads industrial manufacturers to leveraging new control solutions for production planning and control. Among these, distributed control approaches for the multi job shop scheduling problem could be a promising solution to make smart manufacturing systems more responsive to the market. In this context, this paper aims at investigating the effectiveness and the benefits of distributed control approaches for solving the multi job shop assignment problem on a real-world smart manufacturing process involving the production of high-vacuum solar panels. The validation is carried out by leveraging Cyber-Physical Systems modelling and a digital model-based procedure which, interacting with the physical plant, provides the proper job assignment in order to optimize the overall makespan and work-in-progress in a fully-distributed fashion.

Exploring the Potential Network Vulnerabilities in the Smart Manufacturing Process of Industry 5.0 via the Use of Machine Learning Methods

Exploring the Potential Network Vulnerabilities in the Smart Manufacturing Process of Industry 5.0 via the Use of Machine Learning Methods

Exploring Data Acquisition and Real-Time Analysis Algorithms in Smart Manufacturing with a Focus on Automation and Inspection Technologies

The conducted research project is based on the process followed in the smart manufacturing method. It has been observed that there are different types of methods that have been used in the smart manufacturing process and among these methods most of the methods are directly associated with different modern technology and artificial intelligence as well. Along with the inclusion of automation systems in the smart manufacturing process, the incorporation of inspection technologies is also observed in the case of performing smart manufacturing systems. In association with these, the usage of an acquisition system and real-time analysis process is also observed as well. Independent steps are present inside the phases of production process of the smart manufacturing methods. The incorporation of automation processes significantly enhances the performance of individual phases within the smart manufacturing process, thereby raising the overall production standards. Automation minimizes the likelihood of human errors, data misplacement, and mishandling. Moreover, it proves highly advantageous in the data acquisition process, ensuring error-free data transfer through automated systems.

Artificial neural network-based prediction assessment of wire electric discharge machining parameters for smart manufacturing

Abstract Artificial intelligence (AI), robotics, cybersecurity, the Industrial Internet of Things, and blockchain are some of the technologies and solutions that are combined to produce “smart manufacturing,” which is used to optimize manufacturing processes by creating and/or accepting data. In manufacturing, spark erosion technique such as wire electric discharge machining (WEDM) is a process that machines different hard-to-cut alloys. It is regarded as the solution for cutting intricate parts and materials that are resistant to conventional machining techniques or are required by design. In the present study, holes of different radii, i.e. 1, 3, and 5 mm, have been cut on Nickelvac-HX. Tapering in WEDM is a delicate process to avoid disadvantages such as wire break, wire bend, wire friction, guide wear, and insufficient flushing. Taper angles viz. 0°, 15°, and 30° were obtained from a unique fixture to get holes at different angles. The study also shows the influence of taper angles on the part geometry and area of the holes. Next, the artificial neural network (ANN) technique is implemented for the parametric result prediction. The findings were in good agreement with the experimental data, supporting the viability of the ANN approach for the evaluation of the manufacturing process. The findings in this research provide as a reference to the potential of AI-based assessment in smart manufacturing processes and as a design tool in many manufacturing-related fields.

An implementation model for digitisation of visual management to develop a smart manufacturing process

PurposeThis paper aims to introduce a model using a digital twin concept in a cold heading manufacturing and develop a digital visual management (VM) system using Lean overall equipment effectiveness (OEE) tool to enhance the process performance and establish Fourth Industrial Revolution (I4.0) platform in small and medium enterprises (SMEs).Design/methodology/approachThis work utilised plan, do, check, act Lean methodology to create a digital twin of each machine in a smart manufacturing facility by taking the Lean tool OEE and digitally transforming it in the context of I4.0. To demonstrate the effectiveness of process digitisation, a case study was carried out at a manufacturing department to provide the data to the model and later validate synergy between Lean and I4.0 platform.FindingsThe OEE parameter can be increased by 10% using a proposed digital twin model with the introduction of a Level 0 into VM platform to clearly define the purpose of each data point gathered further replicate in projects across the value stream.Research limitations/implicationsThe findings suggest that researchers should look beyond conversion of stored data into visualisations and predictive analytics to improve the model connectivity. The development of strong big data analytics capabilities in SMEs can be achieved by shortening the time between data gathering and impact on the model performance.Originality/valueThe novelty of this study is the application of OEE Lean tool in the smart manufacturing sector to allow SME organisations to introduce digitalisation on the back of structured and streamlined principles with well-defined end goals to reach the optimal OEE.

The decay of Six Sigma and the rise of Quality 4.0 in manufacturing innovation

Smart manufacturing (SM) processes exhibit rapidly increasing complexity, nonlinear patterns in hyperdimensional spaces, high volumes of data, transient sources of variations, reduced lifetime, ultrahigh conformance, and non-Gaussian pseudo-chaotic behaviors. Standard quality control techniques and paradigms are not up to handling all these dynamics. Therefore, quality engineers went stagnant, with little innovation to offer to the manufacturing industry. Artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL) have been applied to solve complex engineering problems and drive innovation. This new era where computer science principles are applied to quality control is called Quality 4.0 (Q4.0). However, the Six Sigma five-step problem-solving strategy (define, measure, analyze, improve, and control) does not fit the full ML cycle. The limitations of the Six Sigma techniques and paradigms in driving manufacturing innovation are discussed. A case study where a 3D quality pattern that can be easily detected by an MLA is not detected by traditional process monitoring methods. Early results motivate the development of the new era of Q4.0 without the limitations of Six Sigma and the potential of AI.

To achieve sustainability in supply chain with Digital integration: A TISM approach

Conventional supply chain has been shown to be incapable of meeting the ever-increasing demands of customers as well as the requirements of innovation. Due to various uncertainty volatility, ambiguity, and intricacy, the sustainability of supply chain becomes a major topic for organisations. Now there is need to integrate the digital technologies like cloud computing, internet of things, artificial intelligence, big data analysis etc. which improve the performance of supply chain in efficient and responsiveness manner. Due to this digital integration, the system undergoes various changes at organisational, operational, performance and technological level. This study aims to identify nine major critical factors which are enablers to achieve the sustainability in digitally integrated supply chain. A TISM model is developed to address their interrelationship among them. The factors are classified as dependent and independent factors according to their driving and dependence power through the use of the MICMAC analysis. If is confirmed in MICMAC analysis that the factors Agile Organisational structure, Smart logistics Capabilities, Smart Manufacturing Process and financial planning enhance the sustainability of digitally enabled supply chain. This study provides a comprehensive list of enablers that are necessary to achieve sustainability of digitally integrated supply chains; nevertheless, the list is not exhaustive. This paper was written with the intention of contributing a pool of knowledge on achieving sustainability in digitally integrated supply chain. This study has the potential to make it possible for market specialists and executives to focus on critical elements that lead to tactical decisions and maximise value for companies. It establishes a baseline from which future studies can build.

Design of an Optimal Scheduling Control System for Smart Manufacturing Processes in Tobacco Industry

The whole process of tobacco production is composed of many components, in which their operation and administration are currently independent. It is required to deploy smart manufacturing workflow for the whole production process, in order to realize centralized effective global scheduling. This requires an advanced administration control platform that has strong abilities of multisource data integration and automatic decision support. To bridge such research gap, this paper designs an optimal scheduling control system for smart manufacturing processes of tobacco industry. First of all, this work discusses major characteristics of future-generation production control patterns in intelligent tobacco factories (ITF). Then, a five-layer architecture for optimal scheduling control of ITF is proposed, which contains Internet-of-Things layer, centralized control layer, model layer, platform layer and operation layer. In addition, a production scheduling optimization strategy is also developed for the proposed system to serve as the software algorithm that drives the running of whole smart manufacturing processes. Finally, this paper presents a comparative analysis of the proposed system’s transformation in a cigarette factory. Naturally, the effectiveness of the proposed production optimization scheduling strategy is verified through simulation.