Search

Self-supervised learning has demonstrated state-of-the-art performance on various anomaly detection tasks. Learning effective representations by solving a supervised pretext task with pseudo-labels generated from unlabeled data provides a promising concept for industrial downstream tasks such as process monitoring. In this paper, we present SSMSPC a novel approach for multivariate statistical in-process control (MSPC) based on self-supervised learning. Our motivation for SSMSPC is to leverage the potential of unsupervised representation learning by incorporating self-supervised learning into the general statistical process control (SPC) framework to develop a holistic approach for the detection and localization of anomalous process behavior in discrete manufacturing processes. We propose a pretext task called Location + Transformation prediction, where the objective is to classify both, the type and the location of a randomly applied augmentation on a given time series input. In the downstream task, we follow the one-class classification setting and apply the Hotelling's T 2 statistic on the learned representations. We further propose an extension to the control chart view that combines metadata with the learned representations to visualize the anomalous time steps in the process data which supports a machine operator in the root cause analysis. We evaluate the effectiveness of SSMSPC with two real-world CNC-milling datasets and show that it outperforms state-of-the-art anomaly detection approaches, achieving 100 % and 99.6 % AUROC, respectively. Lastly, we deploy SSMSPC at a CNC-milling machine to demonstrate its practical applicability when used as a process monitoring tool in a running process. [ABSTRACT FROM AUTHOR]

Gear factories are most typical discrete manufacturing factories. Many gear factories are striving to explore how to achieve intelligent manufacturing in order to improve efficiency and reduce costs. Digital twin technology is currently one of the most reliable ways to achieve intelligent manufacturing. This article aims to establish a real-time control system in order to promote intelligent manufacturing for discrete manufacturing factories. Firstly, a model for a digital twin gear factory is put forward based on the characteristics of gear factories, and the composition of a real-time control system for gear factories is clarified. Then, a human–computer interaction architecture for the real-time control system is proposed. The real-time control system consists of three parts as follows: a monitoring module, a virtualizing module, and a controlling module. At work, it appears as a kind of human–machine interaction form with the three following interfaces: a monitoring window, a virtualizing window, and a controlling window. Finally, a gear factory, which is specialized in manufacturing the intermediate shaft dual gear of a new energy vehicle gearbox, develops a set of software for the real-time control system. The prototype software is obtained through some development activities such as 3D MAX and WebGL virtualization modeling and OPC UA and REST communication design. [ABSTRACT FROM AUTHOR]

Along with the development of new-generation information technology, digital twins (DTs) have become the most promising enabling technology for smart manufacturing. This article presents a statistical analysis of the literature related to the applications of DTs for discrete manufacturing lines, researches their development status in the areas of the design and improvement of manufacturing lines, the scheduling and control of manufacturing line, and predicting faults in critical equipment. The deployment frameworks of DTs in different applications are summarized. In addition, this article discusses the three key technologies of high-fidelity modeling, real-time information interaction methods, and iterative optimization algorithms. The current issues, such as fine-grained sculpting of twin models, the adaptivity of the models, delay issues, and the development of efficient modeling tools are raised. This study provides a reference for the design, modification, and optimization of discrete manufacturing lines. [ABSTRACT FROM AUTHOR]

The Confirmatory Factor Analysis of Safety Culture in the Discrete Manufacturing Industry for Thailand is important for developing standards to assess safety culture. This study explores the confirmatory factor analysis of safety culture within Thailand's discrete manufacturing sector, utilizing structural equation modeling to assess six key elements of safety culture. It innovatively validates various sub-elements, offering critical insights for enhancing safety standards. This research aimed to analyze the confirmatory factor analysis of safety culture in discrete manufacturing industry for Thailand. The research method involved Confirmatory Factor Analysis (CFA) with a sample of observations. By examining the consistency of the six elements and indicators, along with their respective factor loading, the process initiates with the specification of model-specific data. This was accomplished through a confirmatory measurement model, starting with the assignment of data to the model based on the confirmatory ranking of the first and second orders, sequentially, considering chi-square (χ²), relative chi-square (χ² /df), CFI, TLI, RMSEA, and SRMR. These indices were used to evaluate the adequacy of the measurement model. The study identified six core elements of safety culture, each with specific subelements and observed variables meeting established criteria: Input, encompassing Man, Method, Material, and Machine; Processing, including Workforce and Work Systems; Output, with Accident Rate and Enterprise Damage; Working Environment, comprising Climate, Facilities, and Cultural Differences; Ergonomics, involving Job and Personal Characteristics; and Safety Experience, consisting of Duration, Involvement, Training, and Leadership. This structure offers a comprehensive framework for understanding and enhancing safety culture in the workplace. The research offers an in-depth analysis of safety culture in Thailand's discrete manufacturing sector, utilizing Confirmatory Factor Analysis to delineate various elements and sub-elements. This study significantly advances the understanding and assessment of safety culture, underlining the crucial role such evaluations play in enhancing workplace safety across the manufacturing industry. [ABSTRACT FROM AUTHOR]

To solve the problems of strong uncertainty, dynamics, and high complexity during the operation process in a discrete manufacturing shop floor that produces a variety of variable-volume products, a data-driven smart management and control framework of a digital twin shop floor (DTS) is proposed. Its implementation process is analyzed. Five key tasks are illustrated in detail: (1) the construction of a shop floor digital twin (DT) model from the multi-dimensional multi-scale perspective; (2) data acquisition and management technology in a DTS; (3) the real-time data-driven synchronous modeling of the shop floor operating status; (4) the model- and data-driven online prediction of the shop floor operation status; and (5) the multi-agent-based operation decision of a DTS. In addition, for products that are complex to produce on the assembly shop floor, a DT-based smart management and control system named the DT-VPPC is developed. The effectiveness of the proposed method is verified by a specific application example on an assembly shop floor. [ABSTRACT FROM AUTHOR]

Gear factories are most typical discrete manufacturing factories. Many gear factories are striving to explore how to achieve intelligent manufacturing in order to improve efficiency and reduce costs. Digital twin technology is currently one of the most reliable ways to achieve intelligent manufacturing. This article aims to establish a real-time control system in order to promote intelligent manufacturing for discrete manufacturing factories. Firstly, a model for a digital twin gear factory is put forward based on the characteristics of gear factories, and the composition of a real-time control system for gear factories is clarified. Then, a human–computer interaction architecture for the real-time control system is proposed. The real-time control system consists of three parts as follows: a monitoring module, a virtualizing module, and a controlling module. At work, it appears as a kind of human–machine interaction form with the three following interfaces: a monitoring window, a virtualizing window, and a controlling window. Finally, a gear factory, which is specialized in manufacturing the intermediate shaft dual gear of a new energy vehicle gearbox, develops a set of software for the real-time control system. The prototype software is obtained through some development activities such as 3D MAX and WebGL virtualization modeling and OPC UA and REST communication design.

With the development of sensing and communications technology, some new features have emerged in manufacturing processes, such as highly correlated, deeply integrated, dynamically integrated, and a huge volume of data. There is a strong need to deeply excavate information from manufacturing Big Data, especially the energy consumption data, for energy-efficient manufacturing operations management and analysis. However, relevant data reduction and association analysis to support energy-efficient manufacturing are still ineffective and error-prone, especially for discrete manufacturing workshops. In this paper, an energy-aware Cyber Physical System (E-CPS) is proposed for energy Big Data analysis and recessive production anomalies detection. Firstly, E-CPS is introduced to acquire manufacturing Big Data. Then, a Big Data analysis method, including data reduction and data association analysis, is proposed to analyse the manufacturing data in the E-CPS. Considering the complexity and dynamics of manufacturing processes, an energy Big Data-driven recessive production anomalies analysis method is proposed based on deep belief networks. The proposed method in this paper realises the integrated utilisation of production Big Data and energy Big Data in the E-CPS. Further, the efficiency evaluation and recessive anomalies detection methods can be used in existing production information systems. [ABSTRACT FROM AUTHOR]

The deployment of a manufacturing execution system (MES) holds promising potential in facilitating the accumulation of a substantial amount of inspection data. Low quality levels in discrete manufacturing environments are the result of multi-factor coupling and the failure to detect quality issues promptly in accordance with manufacturing settings, which may trigger the propagation of downstream defects. Currently, most inspection quality methods consist of direct measurements followed by manual judgment. The integration of deep learning methods provides a feasible way in which to identify defects on time, thus improving the acceptance rate of factories. This paper focuses on the design of a data-driven quality prediction and control model, built around discrete manufacturing characteristics, and uses fuzzy theory to evaluate the quality levels of production stages. Furthermore, a multivariate long- and short-term memory sequence model is proposed in order to explore the qualitative information from time domain features. The data regarding the produced water dispensers are validated using three evaluation indices, namely, RMSE, MAE, and MAPE. The results indicate that the multivariate long- and short-term memory model exhibits stronger prediction performance. [ABSTRACT FROM AUTHOR]

With the development of new-generation information technologies, such as big data and artificial intelligence, digital twins have become a key technology in intelligent manufacturing. The introduction of digital twin technology has addressed many problems in discrete manufacturing lines, including low visualization and difficult cyber–physical integration. However, the application of digital twin technology to discrete manufacturing lines still faces problems of low modeling accuracy, response delay, and insufficient production line control accuracy. Therefore, this paper proposes a digital twin four-dimension fusion modeling method to solve the above problems. First, a digital twin system architecture for a discrete manufacturing production line is designed. Then, the information control dimension is integrated into traditional digital twin modeling methods. Further, a digital twin geometry–physics–behavior–information control four-dimension fusion modeling method is proposed. This method can describe the geometric and physical characteristics of a physical entity and map its behavior mechanism. More importantly, it reveals the control logic and virtual–real mapping rules, which provides important support for the virtual–real intelligent mutual control. Finally, the feasibility and effectiveness of the proposed method are verified by experiments on a fidget spinner discrete manufacturing line, and a digital twin operation and maintenance management system is developed. The results presented in this study could provide ideas for the digital transformation of discrete manufacturing enterprises. [ABSTRACT FROM AUTHOR]

Along with the development of new-generation information technology, digital twins (DTs) have become the most promising enabling technology for smart manufacturing. This article presents a statistical analysis of the literature related to the applications of DTs for discrete manufacturing lines, researches their development status in the areas of the design and improvement of manufacturing lines, the scheduling and control of manufacturing line, and predicting faults in critical equipment. The deployment frameworks of DTs in different applications are summarized. In addition, this article discusses the three key technologies of high-fidelity modeling, real-time information interaction methods, and iterative optimization algorithms. The current issues, such as fine-grained sculpting of twin models, the adaptivity of the models, delay issues, and the development of efficient modeling tools are raised. This study provides a reference for the design, modification, and optimization of discrete manufacturing lines.

At present, most risk management work mainly relies on manpower, and manpower relies on the professional knowledge of relevant skilled workers to discover hidden safety risks in production activities. This article combines relevant big data theories and 4V characteristics to analyze and investigate safety production and big data, study data structure, data source, and data type. Using 5W1H scientific big data and applications, this analysis method analyzes the theoretical basis, applications, and beneficiaries of big data related to safety production, some of which are application links and important theoretical issues. Secondly, it studies the security risk management model based on big data, proposes a risk management model based on big data, the technical basis of big data, and the idea of a three-dimensional model, and applies the systematic space method, which is reflected in three aspects of risk management. In the end, a risk identification model based on big data, a risk assessment classification model, and a risk early warning and pre-control model are defined. [ABSTRACT FROM AUTHOR]

Due to the complexity of the production shop in discrete manufacturing industry, the traditional genetic algorithm (GA) cannot solve the production scheduling problem well. In order to enhance the GA-based method to solve the production scheduling problem effectively, the simulated annealing algorithm (SAA) is used to develop an improved hybrid genetic algorithm. Firstly, the crossover probability and mutation probability of the genetic operation are adjusted, and the elite replacement operation is adopted for simulated annealing operator. Then, a mutation method is used for the comparison and replacement of the genetic operations to obtain the optimal value of the current state. Lastly, the proposed hybrid genetic algorithm is compared with several scheduling algorithms, and the superiority and efficiency of the proposed method are verified in solving the production scheduling. [ABSTRACT FROM AUTHOR]

Blockchain technology ensures the security of cross-organizational data sharing in the process of collaborative innovation. It drives the development of collaborative innovation in discrete manufacturing to intelligent innovation. However, collaborative innovation is a multi-role, networked, and open resource-sharing process. Therefore, it is easy to form information barriers and increase the risk of cooperation between organizations. In this paper, we firstly analyze the blockchain- based information management models in the traditional discrete manufacturing collaborative innovation process. Then, we found that in the process of industry-university-research (IUR) collaborative innovation, consensus servers maintain too many connections due to the high latency between them, which leads to lower consensus performance and efficiency. To solve this problem, we proposed the dependency analysis (DA) model, which separates and assembles transactional data and non-transactional data into different blocks by analyzing read_set and write_set of transactions. Besides, we propose the out-of-order Raft (OORaft) model to allow non-transactional blocks to be copied in parallel, which can also prioritize transactional blocks. Finally, we implement the blockchain model in the discrete manufacturing scenario based on Hyperledger Fabric. The experimental results show that our model improves the transactions per second (TPS) performance of 1.9X-3.7X and reduces transactional data committing latency by more than 40% in the target scenario. [ABSTRACT FROM AUTHOR]

The deployment of a manufacturing execution system (MES) holds promising potential in facilitating the accumulation of a substantial amount of inspection data. Low quality levels in discrete manufacturing environments are the result of multi-factor coupling and the failure to detect quality issues promptly in accordance with manufacturing settings, which may trigger the propagation of downstream defects. Currently, most inspection quality methods consist of direct measurements followed by manual judgment. The integration of deep learning methods provides a feasible way in which to identify defects on time, thus improving the acceptance rate of factories. This paper focuses on the design of a data-driven quality prediction and control model, built around discrete manufacturing characteristics, and uses fuzzy theory to evaluate the quality levels of production stages. Furthermore, a multivariate long- and short-term memory sequence model is proposed in order to explore the qualitative information from time domain features. The data regarding the produced water dispensers are validated using three evaluation indices, namely, RMSE, MAE, and MAPE. The results indicate that the multivariate long- and short-term memory model exhibits stronger prediction performance.

With the development of new-generation information technologies, such as big data and artificial intelligence, digital twins have become a key technology in intelligent manufacturing. The introduction of digital twin technology has addressed many problems in discrete manufacturing lines, including low visualization and difficult cyber–physical integration. However, the application of digital twin technology to discrete manufacturing lines still faces problems of low modeling accuracy, response delay, and insufficient production line control accuracy. Therefore, this paper proposes a digital twin four-dimension fusion modeling method to solve the above problems. First, a digital twin system architecture for a discrete manufacturing production line is designed. Then, the information control dimension is integrated into traditional digital twin modeling methods. Further, a digital twin geometry–physics–behavior–information control four-dimension fusion modeling method is proposed. This method can describe the geometric and physical characteristics of a physical entity and map its behavior mechanism. More importantly, it reveals the control logic and virtual–real mapping rules, which provides important support for the virtual–real intelligent mutual control. Finally, the feasibility and effectiveness of the proposed method are verified by experiments on a fidget spinner discrete manufacturing line, and a digital twin operation and maintenance management system is developed. The results presented in this study could provide ideas for the digital transformation of discrete manufacturing enterprises.

In recent years, the use of lithium-ion batteries has greatly expanded into products from many industrial sectors, e.g. cars, power tools or medical devices. An early prediction and robust understanding of battery faults could therefore greatly increase product quality in those fields. While current approaches for data-driven fault prediction provide good results on the exact processes they were trained on, they often lack the ability to flexibly adapt to changes, e.g. in operational or environmental parameters. Continual learning promises such flexibility, allowing for an automatic adaption of previously learnt knowledge to new tasks. Therefore, this article discusses different continual learning approaches from the group of regularization strategies, which are implemented, evaluated and compared based on a real battery wear dataset. Online elastic weight consolidation delivers the best results, but, as with all examined approaches, its performance appears to be strongly dependent on task characteristics and task sequence. [ABSTRACT FROM AUTHOR]

Detecting abnormal conditions in manufacturing processes is a crucial task to avoid unplanned downtimes and prevent quality issues. The increasing amount of available high-frequency process data combined with advances in the field of deep autoencoder-based monitoring offers huge potential in enhancing the performance of existing Multivariate Statistical Process Control approaches. We investigate the application of deep auto encoder-based monitoring approaches and experiment with the reconstruction error and the latent representation of the input data to compute Hotelling's T2 and Squared Prediction Error monitoring statistics. The investigated approaches are validated using a real-world sheet metal forming process and show promising results. [ABSTRACT FROM AUTHOR]

This study proposes a microcontroller-based closed-loop water level control system that utilizes a sonar ultrasonic sensor to monitor and regulate water levels in a tank. The system operates by continuously checking the water level in the tank and adjusting the valve’s opening and closure times to maintain the desired water level. The closed-loop control approach employed by the system offers a precise water level control solution that can significantly reduce water waste in water pumping systems. The paper also assesses the theoretical and experimental response of an open-loop water level control system. The findings demonstrate that the closed-loop system outperforms the open-loop system, indicating the effectiveness of the proposed approach in maintaining water levels in a tank. The proposed system is not only applicable to water pumping systems but can also be used to manage the flow of hot and cold water from home water heaters. Additionally, the system can be utilized in various industries that require level or flow control, with particular emphasis on the production of distilled water, juice, and milk.

WIA-FA and IEEE 802.11 are two most widely adopted industrial wireless standards in discrete manufacturing. However, comprehensive performance comparisons between WIA-FA and IEEE 802.11 are still missing and industrial applications urgently need experimental methods to guide the selection of appropriate wireless technologies. To this end, this article performs extensive experiments between WIA-FA and IEEE 802.11 in two practical industrial scenarios, with one ordered scenario defining the transmission order of devices and the other order-free scenario imposing no order constraints to the transmission order of devices. Network performance indices of the WIA-FA and IEEE 802.11 networks, including reliability, delay, jitter, and disorder rate, are compared for different network sizes and data generation periods. Experimental results show that the WIA-FA protocol provides stable network performance, while the network performance of the IEEE 802.11 protocol is random and uncontrollable. Additionally, we perform preliminary comparisons of WIA-FA with IEEE 802.11ax and 5G New Radio. [ABSTRACT FROM AUTHOR]

In discrete manufacturing the variation in process parameters and duration is often large. Common data storage and analytics systems primarily store data in univariate time series, and when analysing machine components of strongly varying lifetime and behaviour this causes a challenge. This paper presents a data structure and an analysis method for outlier detection which intends to deal with this challenge, as an alternative to predictive maintenance which often requires more data with higher quality than what is available. A case study in aluminium extrusion billet manufacturing is used to demonstrate the approach, predominantly detecting anomalies at the end of a critical component's lifetime. [ABSTRACT FROM AUTHOR]

The early and robust detection of anomalies occurring in discrete manufacturing processes allows operators to prevent harm, e.g. defects in production machinery or products. While current approaches for data-driven anomaly detection provide good results on the exact processes they were trained on, they often lack the ability to flexibly adapt to changes, e.g. in products. Continual learning promises such flexibility, allowing for an automatic adaption of previously learnt knowledge to new tasks. Therefore, this article discusses different continual learning approaches from the group of regularization strategies, which are implemented, evaluated, and compared based on a real industrial metal forming dataset. [ABSTRACT FROM AUTHOR]

The neural network (NN) has an advantage in handling the massive real-time monitoring data on discrete manufacturing. Therefore, this paper proposes a production management and control method for discrete manufacturing job-shops based on ant colony optimization (ACO). Firstly, the production management and control problem for discrete manufacturing job-shops was described through the functional analysis on the management and control system, followed by establishing the corresponding mathematical model. After that, the ACO was improved to solve the static multi-objective production management and control problem. Then, the authors set up an NN-based production management and control model for dynamic discrete manufacturing job-shop, and detailed the way to select and transform the judgement result on production state and to set up the training set. Finally, the effectiveness of our algorithm was verified through experiments. [ABSTRACT FROM AUTHOR]

Distributed production paradigms have grown in discrete manufacturing as discrete products are increasingly made by global, distributed networks. Challenges faced by discrete manufacturing, such as increased globalization, market volatility, workforce shortages, and mass personalization have necessitated scalable solutions that improve the agility of production systems. These challenges have driven the need for better collaboration and coordination in production via improved integration of production systems across the product life cycle. This paper describes key industry use cases to motivate the research and development needed for distributed production in discrete manufacturing. The technological challenges that have hindered distributed production in discrete manufacturing are presented as a state-of-the-art review of the standards and technologies that have been developed to overcome these challenges. Based on this review, future research directions are described to address the needs of industry and achieve the goals of distributed production in discrete manufacturing. [ABSTRACT FROM AUTHOR]

Copyright of Ordnance Industry Automation is the property of Editorial Board for Ordnance Industry Automation and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this paper, we explore the use of network routing for production planning and control in manufacturing of complex industrial products. Such a product is the result of a joint effort of many manufacturing companies; each considered a collection of work centre nodes connected by transportation links forming a company-wide manufacturing network. Each company is, in turn considered a manufacturing node connected to other manufacturers by transportation links forming a distributed manufacturing network that produces the final product. We model the manufacturing network as a distribution network where the incoming and outgoing inventories of products are distinct nodes in addition to the work centre nodes. Production planning and control are done by minimizing the cost of handling all products in all work centre nodes. This formulation provides a scalable and flexible production planning and control scheme adhering to the networked structure of manufacturing of complex products. We apply the model to a company-wide manufacturing network as a first approach. A case study using the model demonstrates production planning using network routing at a manufacturer of ship propulsion engines.

In order to obtain a decision model with universality, the manufacturing unit was regarded as the most basic carrier for the functional objectives of the manufacturing system. This paper has established the functional objective decision model of discrete manufacturing system by characterizing the manufacturing objectives of cost, efficiency, quality, time, agility and greenness, and has introduced the concept of coordination degree between manufacturing units. In weight calculation, the model could balance the importance of the functional objectives required by the customer and the producer. Moreover, according to the NP-hard characteristics of the model, ant colony algorithm and particle swarm optimization (ACO-PSO) algorithm was designed to solve the problem. The feasibility and validity of the algorithm were verified by simulation examples, which could promise the experimental results more satisfactory than the traditional genetic algorithm. In addition, the model can provide more choices for decision-making of functional objectives in discrete manufacturing systems by adjusting the fitness value. [ABSTRACT FROM AUTHOR]

Automated manufacturing machines in the discrete manufacturing domain frequently face changes in requirements, such as volatile customer demands or changes in product variants. Due to this, machines need to become more flexible to cope with these changing conditions. Therefore, manufacturing machines have to undergo adaptation processes during their operational phase. The adaptation processes might include mechanical, electrical, and software changes. In industrial practice, experts individually perform these adaptation processes without methodological support, which is time-consuming and highly error-prone. This article proposes a systematic approach for supporting the different phases of the adaptation process. The producibility check of a production request based on a suitable skill model of the system is addressed as well as the automatic generation of adaptation options. Furthermore, the article provides concepts for analyzing the impact, effort and benefit of the generated adaptation options. Additionally, a multi agent architecture is presented for the implementation of the proposed adaptation approaches. The entire assistance concept was applied to a lab-size production machine to validate the applicability of the approach. [ABSTRACT FROM AUTHOR]