Factory Production Line Research Articles

A Seq2Seq transformation strategy for generalizing a pre-trained model in anomaly detection of rolling element bearings

The monitoring of the health of Rolling Element Bearings (REBs) in the rolling mill process was recently automated through signal processing and machine learning technologies. However, large data sets are required, and existing methods require precise labels. Moreover, the scarcity of equipment anomalies challenges their practical identification. In addition, machine learning approaches require human resources to manage machine learning models because all REBs in the production line are not identical. Thus, we propose herein a Seq2Seq Transformation Strategy inspired by the accurate performance of such the methods in Natural Language Processing (NLP) and Computer Visioning (CV). Our strategy aims to use an anomaly detection model trained on one signal to detect anomalies in other signals, reducing the need for model management in practical scenarios. Specifically, we considered time series vibration signals collected from the REBs to be analogous to linguistic data, which are sequential and time variant. In essence, our strategy generates synthetic data from a vibration signal represented as the target domain by using a transformer. Subsequently, a model trained by another signal called the source domain is employed to detect the conditions of the target domain. The goal of transformation is to minimize the differences between the distributions of the synthetic and source data. We assumed that the target conditions could be identified from the synthetic data by the source model if the detection model trained on the source domain was precise. Our strategy was experimentally evaluated on six large, unlabeled data sets collected over 1 year from a rolling mill production line in a steel factory. The strategy significantly improved the accuracy of anomaly detection, which had initially been unacceptable when the source model was directly applied to the detection of the target conditions.

Suborganellar resolution imaging for the localisation of human glycosylation enzymes in tobacco Golgi bodies.

Plant cells are a capable system for producing economically and therapeutically important proteins for a variety of applications, and are considered a safer production system than some existing hosts such as bacteria or yeasts. However, plants do not perform protein modifications in the same manner as mammalian cells do. This can impact on protein functionality for plant-produced human therapeutics. This obstacle can be overcome by creating a plant-based system capable of 'humanising' proteins of interest resulting in a glycosylation profile of synthetic plant-produced proteins as it would occur in mammalian systems. For this, the human glycosylation enzymes (HuGEs) involved in N-linked glycosylation N-acetylglucosaminyltransferase IV and V (GNTIV and GNTV), β-1,4-galactosyltransferase (B4GALT1), and α-2,6-sialyltransferase (ST6GAL) were expressed in plant cells. For these enzymes to carry out the stepwise glycosylation functions, they need to localise to late Golgi body cisternae. This was achieved by a protein targeting strategy of replacing the mammalian Golgi targeting domains (Cytoplasmic-Transmembrane-Stem (CTS) regions) with plant-specific ones. Using high-resolution and dynamic confocal microscopy, we show that GNTIV and GNTV were successfully targeted to the medial-Golgi cisternae while ST6GAL and B4GALT1 were targeted to trans-Golgi cisternae. Plant cells are a promising system to produce human therapeutics for example proteins used in enzyme replacement therapies. Plants can provide safer and cheaper alternatives to existing expression systems such as mammalian cell culture, bacteria or yeast. An important factor for the functionality of therapeutic proteins though are protein modifications specific to human cells. However, plants do not perform protein modifications in the same manner as human cells do. Therefore, plant cells need to be genetically modified to mimic human protein modifications patterns. The modification of importance here, is called N-linked glycosylation and adds specific sugar molecules onto the proteins. Here we show the expression of four human glycosylation enzymes, which are required for N-linked glycosylation, in plant cells. In addition, as these protein modifications are carried out in cells resembling a factory production line, it is important that the human glycosylation enzymes be placed in the correct cellular compartments and in the correct order. This is carried out in Golgi bodies. Golgi bodies are composed of several defined stacks termed cis-, medial and trans-Golgi body stacks. For correct protein function, two of these human glycosylation enzymes need to be placed in the medial-Golgi attacks and the other two in the trans-Golgi stacks. Using high-resolution laser microscopy in live plant cells, we show here that the human glycosylation enzymes are sent within the cells to the correct Golgi body stacks. These are first steps to modify plant cells in order to produce human therapeutics.

Implementations of Digital Transformation and Digital Twins: Exploring the Factory of the Future

In the era of rapid technological advancement and evolving industrial landscapes, embracing the concept of the factory of the future (FoF) is crucial for companies seeking to optimize efficiency, enhance productivity, and stay sustainable. This case study explores the concept of the FoF and its role in driving the energy transition and digital transformation within the automotive sector. By embracing advancements in technology and innovation, these factories aim to establish a smart, sustainable, inclusive, and resilient growth framework. The shift towards hybrid and electric vehicles necessitates significant adjustments in vehicle components and production processes. To achieve this, the adoption of lighter materials becomes imperative, and new technologies such as additive manufacturing (AM) and artificial intelligence (AI) are being adopted, facilitating enhanced efficiency and innovation within the factory environment. An important aspect of this paradigm involves the development and utilization of a modular, affordable, safe human–robot interaction and highly performant intelligent robot. The introduction of this intelligent robot aims to foster a higher degree of automation and efficiency through collaborative human–robot environments on the factory floor and production lines, specifically tailored to the automotive sector. By combining the strengths of human and robotic capabilities, the future factory aims to revolutionize manufacturing processes, ultimately driving the automotive industry towards a more sustainable and technologically advanced future. This study explores the implementation of automation and the initial strides toward transitioning from Industry 4.0 to 5.0, focusing on three recognized, large, and automotive companies operating in the north of Portugal.

Garment production line optimization using production information based on real‐time power monitoring data

AbstractThe implementation of Fourth Industrial Revolution technologies at a manufacturing site requires analysis of the site status based on real‐time information, optimization of the processes, and onsite execution. However, in labor‐intensive industries, such as the garment‐manufacturing industry, it is extremely difficult to develop smart factories based on real‐time onsite information because such industries are accustomed to managing labor using empirical expert judgment; moreover, they require rapid production circulation and are dependent on human‐related factors. In this study, we developed an optimization simulator using onsite real‐time production information that provides decision‐making support to maximize the productivity of a garment production plant. As an optimization method, a genetic algorithm was used to incorporate operator relocation into various garment production line variables. Through the developed simulator, field managers can predict and optimize productivity with simple operation in connection with production information. The application of the simulation optimization process to an actual production line in an Indonesian garment factory indicated that the simulator can improve productivity by 34.8%. The results of this study will provide guidance regarding the application of industrial information integration in labor‐intensive industries using methods that can systematically support decision‐making to achieve optimal productivity.

Actual NOX and SOX removal rates in the atmospheric environment of concrete permeable blocks containing TiO2 powders, coconut shell powders, and zeolite beads

This study examined actual nitrogen oxide (NOX) and sulfur oxide (SOX) removal rates of concrete permeable blocks in the atmospheric environment. To give the concrete permeable block a NOX and SOX removal function, TiO2 submicron particle or coconut shell powder was embedded in the surface layer, and zeolite bead was embedded in the base layer. Using an actual factory production line, concrete permeable block specimens were prepared by adding TiO2 submicron particles and coconut shell powder to the surface layer of the concrete permeable block. The prepared concrete permeable block specimens were placed in a parking lot, and then NOX and SOX removal rates were directly measured for 6 months on the concrete permeable block specimens placed in the parking lot. Test results showed that the clear day removal rates of NOX and SOX for the concrete permeable block specimens embedded with TiO2 submicron particles during the day was 84% and 70%, respectively. Compared to the concrete permeable blocks embedded with coconut shell powder, the NOx removal rate of the specimens embedded with TiO2 submicron particles was 1.45 times higher, while the SOx removal rate was similar. However, during cloudy days and nights, the NOX removal rate of the concrete permeable block specimens embedded with TiO2 submicron particles was zero. On the other hand, NOX and SOX removal rates of the concrete permeable block specimens embedded with coconut shell powder and zeolite bead were maintained at approximately 51% and 77% for 6 months, respectively, in the atmospheric environment regardless of weather and day light conditions. Consequently, the coconut shell powder added to the surface layer of the concrete permeable block is a practically promising material in reducing the NOX in the atmospheric environment regardless of weather and day light conditions.

Bottleneck Detection in Modular Construction Factories Using Computer Vision.

The construction industry is increasingly adopting off-site and modular construction methods due to the advantages offered in terms of safety, quality, and productivity for construction projects. Despite the advantages promised by this method of construction, modular construction factories still rely on manually-intensive work, which can lead to highly variable cycle times. As a result, these factories experience bottlenecks in production that can reduce productivity and cause delays to modular integrated construction projects. To remedy this effect, computer vision-based methods have been proposed to monitor the progress of work in modular construction factories. However, these methods fail to account for changes in the appearance of the modular units during production, they are difficult to adapt to other stations and factories, and they require a significant amount of annotation effort. Due to these drawbacks, this paper proposes a computer vision-based progress monitoring method that is easy to adapt to different stations and factories and relies only on two image annotations per station. In doing so, the Scale-invariant feature transform (SIFT) method is used to identify the presence of modular units at workstations, and the Mask R-CNN deep learning-based method is used to identify active workstations. This information was synthesized using a near real-time data-driven bottleneck identification method suited for assembly lines in modular construction factories. This framework was successfully validated using 420 h of surveillance videos of a production line in a modular construction factory in the U.S., providing 96% accuracy in identifying the occupancy of the workstations and an F-1 Score of 89% in identifying the state of each station on the production line. The extracted active and inactive durations were successfully used via a data-driven bottleneck detection method to detect bottleneck stations inside a modular construction factory. The implementation of this method in factories can lead to continuous and comprehensive monitoring of the production line and prevent delays by timely identification of bottlenecks.

Industrial Internet Identity Resolution+5G Full Connection Digital Factory Research

In this paper, in order to help improve the information capacity of modern factories, a full connection digital factory for industrial Internet identity resolution is built, and digital technology is used to perform the digital interconnection of the production equipment. The full connection digital factory is an advanced factory with extensive connection of production units, deep integration of IT–OT, full release of data value, and improvement of the total factor productivity. This is enabled by the full integration and use of the new generation technologies represented by 5G, edge computing, and identity resolution to reach new factory level, workshop level, and production line level production scenarios. Based on the integration of infrastructure, information technology, and platform service resources, a full connection digital factory architecture is constructed. The fusion application of 5G, edge computing, and mark resolution technology in transformer discrete manufacturing digital plant is then discussed. The challenges and development trends are summarized. Finally, it is demonstrated that the full connection digital factory of identity resolution is a possible future development direction.

Changes in microbial community during the factory production of sweet dumplings from glutinous rice determined by high-throughput sequencing analysis

To study the source of microorganisms in sweet dumplings factory produced from glutinous rice, samples including raw materials, semifinished and products, and surface of facilities, were collected from a local factory production line. Viable cell count showed that the presence of more than 4 CFU/cm2 of Staphylococcus aureus and Bacillus cereus, lactic acid bacteria and total plate count in the soaking and milling tank, milling pipeline, and temporary storage tank. The most significant increase in microbial counts were observed in the soaking and milling procedure of raw materials. Thereafter, five samples were subjected to high-throughput sequencing (HTS) analysis, including two raw materials japonica (RJ) and indica (RI) flours, two semifinished products milled rice milk (MR) and mixed dough (MD), and one final product sweet dumpling (SD). The predominant genera of bacteria in SD were Leuconostoc, Chryseobacterium and Lactococcus, and the predominant fungi were Apiotrichum, Aspergillus and Geotrichum. The rice flour added in MR significantly affected the bacterial community of sweet dumpling product, but this influence was not shown at the fungal level. The results showed that the combination of HTS and viable cell count can effectively analyse the changes in microbial load and diversity during the production of sweet dumplings.

사업주, 경영책임자등의 관리⋅감독 강화를 위한 화재예방, 소방시설 설치⋅유지 및 안전관리에 관한 법률 개정에 관한 연구

In January 2022, the Serious Accidents Punishment Act went into force in Korea. According to The Korea Times (2022.2.11.) article “Sampyo Industry booked as 1st firm accused of violating new workplace disaster law,” workplace disaster law is “aimed at better protecting workers from industrial accidents.” According to the same article, “under the law, owners and CEOs of companies with five or more employees can face a minimum one-year prison sentence or a fine of up to 1 billion won (US$ 832,708) in the event of serious work place disaster.” Things are not very different in large-scale fires at factories. The fire accident which occurred on August 21 2018, within the Namdong Industrial Complex in Incheon, is an example of administrative negligence. According to The Korea Times (2018.8.21.) article “9 killed in electronics factory fire west of Seoul,” “[t]he fire occurred at around 3:43 p.m. at an electronics goods factory production line that reportedly handles hazardous materials. Some workers jumped out of the building before firefighters arrived, and lost their lives. Dozens of firetrucks were sent to the scene located inside the Namdong industrial complex.” It is clear that the CEO and managers of the factory bungled the management of its fire-facilities. According to The New York Times (2020.4.30.) article “‘We Are Not Learning Lessons’: South Korea Mourns an Avoidable Disaster,” “[m]ajor fires, many of them linked to lax safety standards, have plagued South Korea for decades.” Therefore, the Act on Fire Prevention and Installation, Maintenance, and Safety Control of Fire-Fighting Systems should be strengthened to prevent negligent behavior by the management, and prevent disasters, in line with the “Serious Accidents Punishment Act.”

Towards achieving a high degree of situational awareness and multimodal interaction with AR and semantic AI in industrial applications

With its various available frameworks and possible devices, augmented reality is a proven useful tool in various industrial processes such as maintenance, repairing, training, reconfiguration, and even monitoring tasks of production lines in large factories. Despite its advantages, augmented reality still does not usually give meaning to the elements it complements, staying in a physical or geometric layer of its environment and without providing information that may be of great interest to industrial operators in carrying out their work. An expert’s remote human assistance is becoming an exciting complement in these environments, but this is expensive or even impossible in many cases. This paper shows how a machine learning semantic layer can complement augmented reality solutions in the industry by providing an intelligent layer, sometimes even beyond some expert’s skills. This layer, using state-of-the-art models, can provide visual validation and new inputs, natural language interaction, and automatic anomaly detection. All this new level of semantic context can be integrated into almost any current augmented reality system, improving the operator’s job with additional contextual information, new multimodal interaction and validation, increasing their work comfort, operational times, and security.

Airborne LTA Nanozeolites Characterization during the Manufacturing Process and External Sources Interaction with the Workplace Background.

Engineered nanoscale amorphous silica nanomaterials are widespread and used in many industrial sectors. Currently, some types of silicon-based nanozeolites (NZs) have been synthesized, showing potential advantages compared to the analogous micro-forms; otherwise, few studies are yet available regarding their potential toxicity. In this respect, the aim of the present work is to investigate the potential exposure to airborne Linde Type A (LTA) NZs on which toxicological effects have been already assessed. Moreover, the contributions to the background related to the main emission sources coming from the outdoor environment (i.e., vehicular traffic and anthropogenic activities) were investigated as possible confounding factors. For this purpose, an LTA NZ production line in an industrial factory has been studied, according to the Organisation for Economic Cooperation and Development (OECD) guidelines on multi-metric approach to investigate airborne nanoparticles at the workplace. The main emission sources of nanoparticulate matter within the working environment have been identified by real-time measurements (particle number concentration, size distribution, average diameter, and lung-deposited surface area). Events due to LTA NZ spillage in the air during the cleaning phases have been chemically and morphologically characterized by ICP-MS and SEM analysis, respectively.

Determination of Grasping Position and Posture for Handling Robots Based on CNN Considering the Direction of End-Effector Insertion.

Bin picking is a parts handling process in factory production lines. There are various challenges in automating bin picking. In particular, the more complex the object shape is, the more difficult it becomes to recognize the position and posture of the object using a camera and to determine the appropriate position and posture of the end-effector. Using a Convolutional Neural Network (CNN), a deep learning technique for image recognition, the characteristics of the grasping objects, and grasping with a high success rate would expected. However, due to the limited number of simulation trial, the estimated position and posture is not always the optimal grasping posture. In this study, we modify the position and posture of the end-effector for handling robots using deep learning. Several candidates for position and posture of the end-effector are estimated using three types of CNNs. For each candidate for position and posture of the end-effector, the actual grasping position is determined using the balance between the estimated grasping possibility and the posture of the end-effector. Experimental results showed that integration of the CNN result and posture of the end-effector improved grasp success rate.

Contemporary Campus Life: transformation, manic managerialism and academentia by Keyan G Tomaselli

Reviewed by: Contemporary Campus Life: transformation, manic managerialism and academentia by Keyan G Tomaselli Robert Morrell (bio) Keyan G Tomaselli (2021) Contemporary Campus Life: transformation, manic managerialism and academentia. Cape Town: HSRC Press Keyan Tomaselli has written an unusual book that focuses on the state of universities in South Africa and in this respect, builds on a recent literature that includes Jonathan Jansen’s As by Fire and Adam Habib’s Rage and Rebels. Tomaselli draws on his work experiences of campus life at Rhodes, Wits, UKZN and the University of Johannesburg. His book is partly satirical, largely anecdotal but there is no doubting its intention – to identify, highlight and critique the many processes in universities that are the bane of academics’ lives. Michael Chapman in his Foreword identifies the paradox that exercises the author: ‘Tomaselli asks, implicitly, why the aspirations of new beginnings – the end of apartheid, the end of the Cold War – so easily permitted our entrapment in new bureaucratic “structures with no soul”’ (no page number). The book is concerned with changes in and to the South African university system. Tomaselli’s location as a cultural studies and media scholar is evident in his reflections both in the style of writing and the selection of evidence. The book is based on pieces he wrote primarily for the UKZN Griot over the period of a decade. His column took the voice of a commentator writing as ‘a West African poet, praise singer and wandering musician’ who is the repository of oral tradition (10). This enabled irreverence and seems to have provided a cloak of protection from official censure. Tomaselli begins with a prescription: ‘The academy anywhere should be a dialogically managed, open-ended organism, open to reset, rather than managed as a homeostatic machine-led factory production line’ (11). This identifies the academic labour process as a key focus of the work. What [End Page 87] follows are keen insights but often delivered in a language that labour sociologists may not find immediately accessible. The title of the book both conceals and reveals. The book does not contain ethnographic description and first-hand accounts from students and staff, descriptions of campus activity – from lecture halls to sports fields, from pubs to internet cafes. It does, on the other hand, have some of this description but not in a concentrated way that enables a big picture of South African universities to emerge. The sub-titles of the book are eye-catching but not necessarily revealing. ‘Transformation’? The book does indeed reflect on how ‘transformation’ has become a frequently used and abused term that dominates debate and is at the centre of a decolonisation politics. ‘Manic managerialism’? Management is generally treated as a destructive force, the binary opposite of academics and Tomaselli does discuss the capers of management in terms that invite speculation of irrationality. The final sub-title, ‘Academentia’, a word unknown to me before reading this book, draws on a similar trope, of madness, of being out of control. Tomaselli uses this term to refer to the current state of universities while signalling his suspicion of the direction that universities are being taken by Management. The titles of chapters are like newspaper headlines – designed to grab attention. They don’t always alert readers to the exact domain of the chapter. Here are the titles of the book’s ten chapters. ‘Hacking through Academentia’; ‘Cash Cows, E-Cow-nomics and branding’; ‘The backlog syndrome’; ‘Of science and souls’; ‘Of Bulls and Bears’; ‘Publications, Rankings and Abacus management’; ’Writing Africa and Identity: shifting (our)selves’; ‘Of Colonialism and Capture: cartoons’, ‘Blackface and social critique’; ‘Culture can kill’, and ‘The academentia sunrise’. There are no holy cows for Tomaselli. This book is therefore a commendably fearless engagement with many prickly contemporary issues. Tomaselli’s engagement also displays his lifetime attachment to radical views and his rejection of policies that entrench inequality (and stupidity). Managerialism is a constant focus of Tomaselli’s critique. He laments the decision-making powers now taken by Management and efforts to control all aspects of university life. He quotes a correspondent from UKZN who defines performance management as a ‘nasty corporate American import which in reality only...

Semantic product personalization based on the CogniPy environment

Purpose: The aim of the publication was to visualize the process of creating knowledge on the example of the CP Factory production line. For this purpose, the data contained in the relational databases and data of the operating production line were used. This data was converted to the form required by the CogniPy environment to create a semantic ontology for product personalization. Design/methodology/approach: With the available software based on ontologies and knowledge charts, the possibility of common human-computer reasoning has been opened, especially in production management. Findings: During the activities carried out, the on-conceptualization of the knowledge contained in the production management system was brought to an ontological form. At the moment, the information contained in the database is not very different from the data existing in the relational database. However, further modeling of ontology can be directed towards the creation of rules, logic and axioms of production processes. Practical implications: The operations and transformations performed presented the operation of CogniPy in the process of creating ontology and materializing the graph and queries. The created ontology takes the form of a universal set of knowledge that makes it open to wide integration with other systems. Originality/value: The publication shows the possibilities of using the CogniPy environment in the construction of ontology and semantic product personalization. Keywords: ontology, semantics, CogniPy, CP Factory.

A hybrid method based on P and P′ control chart for identifying hotspots

AbstractThe terms “hotspots,” “blackspot,” or “high‐risk” are normally used for an area on a road where the number of car accidents are far beyond the normal level and, more importantly, the causes of the incident always link to one or many specific reasons. Identifying these spots is helpful and may assist people in authority to implement corrective actions. To do the aforementioned, some decision‐support tools could be useful since they enable us to monitor the indicators that may lead to a car accident, same as the procedure engineers use to control production lines in factories. In this regard, control charts are utilized to identify blackspots. Although the method is low cost, and provides quick detection and statistical reliability level and, more importantly, uses online applications; however, it advised that the rate quality control method could be used on roads with the same geometry and traffic conditions. This research develops a hybrid method, based on accident type, for improving rate quality control method to overcome the shortcomings. Furthermore, the suggested framework considers crash severity and traffic volume that have not been taken into account in the rate quality control method yet. The case study evaluates how this framework is successful to identify blackspots, and based on the result in the real environment, it reveals that the proposed method could detect and identify the hotspots.

Efficiency of fishmeal and fish oil processing of different pelagic fish species: Identification of processing steps for potential optimization toward protein production for human consumption

Most fish meal and fish oil production facilities run with outdated processes, producing low-value products, typically not intended for human consumption. The aim of the study was to perform a detailed analysis of the current fishmeal and oil production processes of capelin (Mallotus villosus), and compare the key locations of both fattier and leaner pelagic raw material to identify which steps need to be improved for the production of higher-value products. Results indicated inefficient draining and concentration during the production process and ineffective break-down of the raw material, as lipid separation was ineffective in all studied species. Although the raw material initially differed in lipid content (2%–20%, depending on the species), all fishmeal produced resulted in high-lipid fishmeal (9%–14%). Chemical composition variations in the solid steams entering the dryers suggested that drying these streams individually could lead to more process and product flexibility, which can result in higher-value products, such as fish meal and protein powders for human consumption. Practical applications Traditional production lines of fishmeal and fish oil factories have remained the same for decades, resulting in low-quality products with a low market price. Therefore, optimization and redesign of the production processes are needed to increase the product quality. This study analyzed the lipid quality and water content in all processing steps. The current evaluations will help to shift the traditional fishmeal and fish oil production process from low-quality to high-quality products intended for human consumption by presenting a detailed analysis of the production process not available in the literature to date. Our results can act as reference values for other fishmeal and fish oil producers to assess their production quality. That includes identification of bottlenecks affecting the efficiency and effectiveness of their production processes.

Investigation of compressed air losses on production cost for Mosul dairy factor

A realistic field study was conducted on compressed air production stations in the Mosul Dairy Factory / Iraq, to determine the quantities of field losses resulting from the production of compressed air in the factory, and to determine the effect inlet temperature, pressure ratio, electric motor and leakage in the production line on the production cost, as all data were taken from the factory, and preparing a mathematical program to find and calculate the leaks in the compressed air in all parts of the factory production line. In addition to that the total cost of the compressed air losses during the production line was calculated. The results were listed in the form of quadrants that determined the relationship of the variables to losses in the production line. The typical lifetime compressed air costs in the year had been calculated and the percentage of each of the previous effects in the calculation of the saved cost, the results shows that the amount of the leakage, electric motor efficiency, intake temperature, and the reducing of setting pressure is about 16%, 54%, 25% and 5% respectively. This work leads us to another future work in which how to address these losses in compressed air in the factory production line and work to reduce them by technical means in order to reduce the cost wasted due to these losses.

Causal Features Extraction for Workpiece

In order to reduce the cost, computer vision technology is introduced into the measurement of workpiece size and shape on the factory production line. At present, the most widely used solution is the neural network model based on big data. However, the lack of data and the high cost of data processing also greatly limit the practical application of this aspect. The method of feature extraction brings challenges to the real-time, rotation invariance, and anti-noise of online detection. In this paper, firstly, Harris operator is used to extract feature points quickly. Then a two-layer scale space based on causality is constructed to filter the noise and project downward to obtain the robust feature position, which provides a basis for subsequent processing.

CONTEXT: An Industry 4.0 Dataset of Contextual Faults in a Smart Factory

Abstract Cyber-physical systems in smart factories get more and more integrated and interconnected. Industry 4.0 accelerates this trend even further. Through the broad interconnectivity a new class of faults arise, the contextual faults, where contextual knowledge is needed to find the underlying reason. Fully-automated systems and the production line in a smart factory form a complex environment making the fault diagnosis non-trivial. Along with a dataset, we give a first definition of contextual faults in the smart factory and name initial use cases. Additionally, the dataset encompasses all the data recorded in a current state-of-the-art smart factory. We also add additional information measured by our developed sensing units to enrich the smart factory data even further. In the end, we show a first approach to detect the contextual faults in a manual preliminary analysis of the recorded log data.

Object Sorting using Faster R-CNN

In a factory production line, different industry parts need to be quickly differentiated and sorted for further process. Parts can be of different colors and shapes. It is tedious for humans to differentiate and sort these objects in appropriate categories. Automating this process would save more time and cost. In the automation process, choosing an appropriate model to detect and classify different objects based on specific features is more challenging. In this paper, three different neural network models are compared to the object sorting system. They are namely CNN, Fast R-CNN, and Faster R-CNN. These models are tested, and their performance is analyzed. Moreover, for the object sorting system, an Arduino-controlled 5 DoF (degree of freedom) robot arm is programmed to grab and drop symmetrical objects to the targeted zone. Objects are categorized into classes based on color, defective and non-defective objects.