Hasil untuk "Industrial medicine. Industrial hygiene"

Sanjukta Ghosh

Industrial diagrams such as piping and instrumentation diagrams (P&IDs) are essential for the design, operation, and maintenance of industrial plants. Converting these diagrams into digital form is an important step toward building digital twins and enabling intelligent industrial automation. A central challenge in this digitalization process is accurate object detection. Although recent advances have significantly improved object detection algorithms, there remains a lack of methods to automatically evaluate the quality of their outputs. This paper addresses this gap by introducing a framework that employs Visual Language Models (VLMs) to assess object detection results and guide their refinement. The approach exploits the multimodal capabilities of VLMs to identify missing or inconsistent detections, thereby enabling automated quality assessment and improving overall detection performance on complex industrial diagrams.

Laura Hačková, Martina Gažarová, Mária Kijovská

Background Normal weight obesity (NWO) is defined as a phenotype in which individuals present with a body mass index within the normal range, yet exhibit an excessive proportion of body fat (> 28%). This condition is linked to elevated risks of metabolic and cardiovascular disorders. Although BMI remains a widely applied screening parameter, it does not capture the distribution of fat and lean tissue, which may result in misclassification and underestimation of health hazards. Objective This study sought to compare the body composition profiles of women classified as normal weight according to BMI but differing in adiposity levels, and to determine the diagnostic value of composite indices – fat mass index (FMI), fat-free mass index (FFMI), skeletal muscle mass index (SMMI), and the fat mass (FM)/fat-free mass (FFM) ratio – in identifying NWO phenotype and assessing nutritional status. Material and Methods A total of 402 female Caucasian volunteers aged 18.6-65 years were included in the study. Body composition was analyzed using the InBody 270 (MF-BIA). Results Among 402 participants, 235 fell within the normal-weight BMI range, and 62 of them fulfilled the criteria for the NWO phenotype. Relative to their normal weight (NW) counterparts, the NWO group displayed higher adiposity (%FM: 32.85 vs. 24.08%; FMI: 7.53 vs. 5.08 kg/m²; FM/FFM: 0.49 vs. 0.32, respectively), greater visceral fat accumulation (VFL: 8.68 vs. 5.43), and lower values of lean body mass (FFM: 41.93 vs. 45.22 kg; SMM: 22.76 vs. 24.79 kg). In NWO, BMI correlated only weakly with body fat percentage, whereas FMI and FM/FFM showed substantially stronger associations with an unfavorable body composition pattern. Conclusions BMI in isolation does not provide sufficient sensitivity to detect the NWO phenotype. Composite indices offer a more precise depiction of body composition and should be considered as complementary tools in both diagnostic procedures and metabolic risk prevention strategies. Their integration into clinical assessment protocols may facilitate earlier detection and targeted intervention.

Chen Li, Ruijie Ma, Xiang Qian et al.

Addressing the challenge of data scarcity in industrial domains, transfer learning emerges as a pivotal paradigm. This work introduces Style Filter, a tailored methodology for industrial contexts. By selectively filtering source domain data before knowledge transfer, Style Filter reduces the quantity of data while maintaining or even enhancing the performance of transfer learning strategy. Offering label-free operation, minimal reliance on prior knowledge, independence from specific models, and re-utilization, Style Filter is evaluated on authentic industrial datasets, highlighting its effectiveness when employed before conventional transfer strategies in the deep learning domain. The results underscore the effectiveness of Style Filter in real-world industrial applications.

F. Adetunji, A. Karukayil, P. Samant et al.

This paper addresses the challenges of vision-based manipulation for autonomous cutting and unpacking of transparent plastic bags in industrial setups, aligning with the Industry 4.0 paradigm. Industry 4.0, driven by data, connectivity, analytics, and robotics, promises enhanced accessibility and sustainability throughout the value chain. The integration of autonomous systems, including collaborative robots (cobots), into industrial processes is pivotal for efficiency and safety. The proposed solution employs advanced Machine Learning algorithms, particularly Convolutional Neural Networks (CNNs), to identify transparent plastic bags under varying lighting and background conditions. Tracking algorithms and depth sensing technologies are utilized for 3D spatial awareness during pick and placement. The system addresses challenges in grasping and manipulation, considering optimal points, compliance control with vacuum gripping technology, and real-time automation for safe interaction in dynamic environments. The system's successful testing and validation in the lab with the FRANKA robot arm, showcases its potential for widespread industrial applications, while demonstrating effectiveness in automating the unpacking and cutting of transparent plastic bags for an 8-stack bulk-loader based on specific requirements and rigorous testing.

M. Henderson, J. P. Edelen, J. Einstein-Curtis et al.

Industrial particle accelerators typically operate in dirtier environments than research accelerators, leading to increased noise in RF and electronic systems. Furthermore, given that industrial accelerators are mass produced, less attention is given to optimizing the performance of individual systems. As a result, industrial accelerators tend to underperform their own hardware capabilities. Improving signal processing for these machines will improve cost and time margins for deployment, helping to meet the growing demand for accelerators for medical sterilization, food irradiation, cancer treatment, and imaging. Our work focuses on using machine learning techniques to reduce noise in RF signals used for pulse-to-pulse feedback in industrial accelerators. Here we review our algorithms and observed results for simulated RF systems, and discuss next steps with the ultimate goal of deployment on industrial systems.

Sandeep Gupta

Secure and efficient communication to establish a seamless nexus between the five levels of a typical automation pyramid is paramount to Industry 4.0. Specifically, vertical and horizontal integration of these levels is an overarching requirement to accelerate productivity and improve operational activities. Vertical integration can improve visibility, flexibility, and productivity by connecting systems and applications. Horizontal integration can provide better collaboration and adaptability by connecting internal production facilities, multi-site operations, and third-party partners in a supply chain. In this paper, we propose an Edge-computing-based Industrial Gateway for interfacing information technology and operational technology that can enable Industry 4.0 vertical and horizontal integration. Subsequently, we design and develop a working prototype to demonstrate a remote production-line maintenance use case with a strong focus on security aspects and the edge paradigm to bring computational resources and data storage closer to data sources.

Nathan P. Lawrence, Seshu Kumar Damarla, Jong Woo Kim et al.

With the rise of deep learning, there has been renewed interest within the process industries to utilize data on large-scale nonlinear sensing and control problems. We identify key statistical and machine learning techniques that have seen practical success in the process industries. To do so, we start with hybrid modeling to provide a methodological framework underlying core application areas: soft sensing, process optimization, and control. Soft sensing contains a wealth of industrial applications of statistical and machine learning methods. We quantitatively identify research trends, allowing insight into the most successful techniques in practice. We consider two distinct flavors for data-driven optimization and control: hybrid modeling in conjunction with mathematical programming techniques and reinforcement learning. Throughout these application areas, we discuss their respective industrial requirements and challenges. A common challenge is the interpretability and efficiency of purely data-driven methods. This suggests a need to carefully balance deep learning techniques with domain knowledge. As a result, we highlight ways prior knowledge may be integrated into industrial machine learning applications. The treatment of methods, problems, and applications presented here is poised to inform and inspire practitioners and researchers to develop impactful data-driven sensing, optimization, and control solutions in the process industries.

Neave O'Clery, Juan Chaparro, Andres Gomez-Lievano et al.

What drives formal employment creation in developing cities? We find that larger cities, home to an abundant set of complex industries, employ a larger share of their working age population in formal jobs. We propose a hypothesis to explain this pattern, arguing that it is the organised nature of formal firms, whereby workers with complementary skills are coordinated in teams, that enables larger cities to create more formal employment. From this perspective, the growth of formal employment is dependent on the ability of a city to build on existing skills to enter new complex industries. To test our hypothesis, we construct a variable which captures the skill-proximity of cities' current industrial base to new complex industries, termed 'complexity potential'. Our main result is that complexity potential is robustly associated with subsequent growth of the formal employment rate in Colombian cities.

Elżbieta Szczepańska, Aleksandra Liedtka, Elżbieta Czech

Background. Functional food is a key element in the prevention and treatment of many diseases. The ingredients it contains, such as phytosterols that lower cholesterol, also have a preventive effect on type 2 diabetes, atherosclerosis and heart attack. Phenolic compounds have antioxidant, anti-inflammatory and antiviral properties. Xylo-oligosaccharides control insulin levels, and fibre lowers blood pressure, potentially reducing insulin resistance. These beneficial properties mean that there is an increasing interest in this kind of food. Objective. The aim of the study was to assess the state of knowledge and behaviour regarding functional food among adults and to answer the question whether there are differences between the state of knowledge and behaviour of women and men. Material and methods. The survey was conducted among 301 people, including 181 women and 120 men. The research tool was an original survey questionnaire. Results. The definition of functional food is known to 42.5% of people (47.5% of women and 35% of men), while the definition of prebiotic is known to 41.9% of people (43.1% of women and 40.0% of men). For 56.2% of respondents, the factor encouraging the consumption of functional food was a healthy lifestyle, and for 54.7% of them, the product composition was the main purchase criterion. Among functional products, cereals or muesli were most often consumed for breakfast by 35% of men and 55.8% of women, 42.5% of men and 33.7% of women were eaten oils for lunch. For dinner they most often consumed fruit teas, herbal teas, herbal mixtures, this answer was given by 25.8% of men and 29.8% of women. Conclusions. Knowledge of functional foods is unsatisfactory, and no differences in the knowledge of women and men have been observed. Consumption of functional food is generally low, and no differences in consumption have been observed between women and men.

Rui Zhang, Dandan Wei, Keliang Fan et al.

Abstract Background Endocrine function in patients with type 2 diabetes (T2DM) typically differs from those with normal glucose tolerance (NGT). However, few epidemiologic studies have explored how these differences impact the association between exposure to polychlorinated biphenyls (PCBs) and vitamin D levels. Methods This study included 1,705 subjects aged 18–79 years from the Henan Rural Cohort [887 NGT and 818 T2DM]. Linear regression was applied to evaluate the associations between PCB exposure and vitamin D levels. Quantile g-computation regression (QG) and Bayesian kernel machine regression (BKMR) were applied to evaluate the impact of PCB mixtures on vitamin D levels. Interaction effects of ΣPCBs with HOMA2-%β and HOMA2-IR on vitamin D levels were assessed. Results Plasma ΣPCBs was positively associated with 25(OH)D2 in the NGT group (β = 0.060, 95% CI: 0.028, 0.092). Conversely, in T2DM group, ΣPCBs was negatively associated with 25(OH)D3 and 25(OH)D (β = -0.049, 95% CI: -0.072, -0.026; β = -0.043, 95% CI: -0.063, -0.023). Similarly, both QG and BKMR analysis revealed a negative association between PCB mixture exposure and vitamin D levels in the T2DM group, contrary to the results observed in the NGT groups. Furthermore, the negative association of ΣPCBs with 25(OH)D2 and 25(OH)D disappeared or changed to a positive association with the increase of HOMA2-%β levels. Conclusions These findings suggest that decreased β cell function may exacerbate the negative effects of PCB exposure on vitamin D levels. Recognizing T2DM patients’ sensitivity to PCBs is vital for protecting chronic disease health.

Haoping Bai, Shancong Mou, Tatiana Likhomanenko et al.

Despite progress in vision-based inspection algorithms, real-world industrial challenges -- specifically in data availability, quality, and complex production requirements -- often remain under-addressed. We introduce the VISION Datasets, a diverse collection of 14 industrial inspection datasets, uniquely poised to meet these challenges. Unlike previous datasets, VISION brings versatility to defect detection, offering annotation masks across all splits and catering to various detection methodologies. Our datasets also feature instance-segmentation annotation, enabling precise defect identification. With a total of 18k images encompassing 44 defect types, VISION strives to mirror a wide range of real-world production scenarios. By supporting two ongoing challenge competitions on the VISION Datasets, we hope to foster further advancements in vision-based industrial inspection.

Gurjot Singh Bhatia, Yoann Corre, M. Di Renzo

Industrial environments are considered to be severe from the point of view of electromagnetic (EM) wave propagation. When dealing with a wide range of industrial environments and deployment setups, ray-tracing channel emulation can capture many distinctive characteristics of a propagation scenario. Ray-tracing tools often require a detailed and accurate description of the propagation scenario. Consequently, industrial environments composed of complex objects can limit the effectiveness of a ray-tracing tool and lead to computationally intensive simulations. This study analyzes the impact of using different propagation models by evaluating the number of allowed ray path interactions and digital scenario representation for an industrial environment. This study is realized using the Volcano ray-tracing tool at frequencies relevant to 5G industrial networks: 2 GHz (mid-band) and 28 GHz (high-band). This analysis can help in enhancing a ray-tracing tool that relies on a digital representation of the propagation environment to produce deterministic channel models for Indoor Factory (InF) scenarios, which can subsequently be used for industrial network design.

Hao-Wei Chung, Hui-Min Hsieh, Chung-Hsiang Lee et al.

Abstract Background Acute bronchiolitis and air pollution are both risk factor of pediatric asthma. This study aimed to assess subsequent exposure to air pollutants related to the inception of preschool asthma in infants with acute bronchiolitis. This study aimed to assess subsequent exposure to air pollutants related to the inception of preschool asthma in infants with acute bronchiolitis. Methods A nested case-control retrospective study was performed at the Kaohsiung Medical University Hospital systems between 2009 and 2019. The average concentration of PM10, PM2.5, SO2, NO, NO2, and NOX was collected for three, six, and twelve months after the first infected episode. Adjusted regression models were employed to evaluate the association between asthma and air pollution exposure after bronchiolitis. Results Two thousand six hundred thirty-seven children with acute bronchiolitis were included. Exposure to PM10, PM2.5, SO2, NO, NO2, and NOX in the three, six, and twelve months following an episode of bronchiolitis was found to significantly increase the risk of preschool asthma in infants with a history of bronchiolitis.(OR, 95%CI: PM10 = 1.517-1.559, 1.354–1.744; PM2.5 = 2.510-2.603, 2.148–3.061; SO2 = 1.970-2.040, 1.724–2.342; ; NO = 1.915-1.950, 1.647–2.272; NO2 = 1.915-1.950, 1.647–2.272; NOX = 1.752-1.970, 1.508–2.252) In a sensitive analysis of hospitalized infants, only PM10, PM2.5, SO2, and NO were found to have significant effects during all time periods. (OR, 95%CI: PM10 = 1.613-1.650, 1.240–2.140; PM2.5 = 2.208-2.286, 1.568–3.061; SO2 = 1.679-1.622, 1.197–2.292; NO = 1.525-1.557, 1.094–2.181) Conclusion The presence of ambient PM10, PM2.5, SO2 and NO in the three, six, and twelve months following an episode of acute bronchiolitis has been linked to the development of preschool asthma in infants with a history of acute bronchiolitis.

Laura N. Vandenberg, Swati D. G. Rayasam, Daniel A. Axelrad et al.

Abstract Background Understanding, characterizing, and quantifying human exposures to environmental chemicals is critical to protect public health. Exposure assessments are key to determining risks to the general population and for specific subpopulations given that exposures differ between groups. Exposure data are also important for understanding where interventions, including public policies, should be targeted and the extent to which interventions have been successful. In this review, we aim to show how inadequacies in exposure assessments conducted by polluting industries or regulatory agencies have led to downplaying or disregarding exposure concerns raised by communities; that underestimates of exposure can lead regulatory agencies to conclude that unacceptable risks are, instead, acceptable, allowing pollutants to go unregulated; and that researchers, risk assessors, and policy makers need to better understand the issues that have affected exposure assessments and how appropriate use of exposure data can contribute to health-protective decisions. Methods We describe current approaches used by regulatory agencies to estimate human exposures to environmental chemicals, including approaches to address limitations in exposure data. We then illustrate how some exposure assessments have been used to reach flawed conclusions about environmental chemicals and make recommendations for improvements. Results Exposure data are important for communities, public health advocates, scientists, policy makers, and other groups to understand the extent of environmental exposures in diverse populations. We identify four areas where exposure assessments need to be improved due to systemic sources of error or uncertainty in exposure assessments and illustrate these areas with examples. These include: (1) an inability of regulatory agencies to keep pace with the increasing number of chemicals registered for use or assess their exposures, as well as complications added by use of ‘confidential business information’ which reduce available exposure data; (2) the failure to keep assessments up-to-date; (3) how inadequate assumptions about human behaviors and co-exposures contribute to underestimates of exposure; and (4) that insufficient models of toxicokinetics similarly affect exposure estimates. Conclusion We identified key issues that impact capacity to conduct scientifically robust exposure assessments. These issues must be addressed with scientific or policy approaches to improve estimates of exposure and protect public health.

Benjamin Maschler, Tim Knodel, Michael Weyrich

Industrial transfer learning increases the adaptability of deep learning algorithms towards heterogenous and dynamic industrial use cases without high manual efforts. The appropriate selection of what to transfer can vastly improve a transfer's results. In this paper, a transfer case selection based upon clustering is presented. Founded on a survey of clustering algorithms, the BIRCH algorithm is selected for this purpose. It is evaluated on an industrial time series dataset from a discrete manufacturing scenario. Results underline the approaches' applicability caused by its results' reproducibility and practical indifference to sequence, size and dimensionality of (sub-)datasets to be clustered sequentially.

Yuanyuan Tian

Rapidly growing social networks and other graph data have created a high demand for graph technologies in the market. A plethora of graph databases, systems, and solutions have emerged, as a result. On the other hand, graph has long been a well studied area in the database research community. Despite the numerous surveys on various graph research topics, there is a lack of survey on graph technologies from an industry perspective. The purpose of this paper is to provide the research community with an industrial perspective on the graph database landscape, so that graph researcher can better understand the industry trend and the challenges that the industry is facing, and work on solutions to help address these problems.

Rani Marfuah, Endah Dwi Handayani

Introduction: Physical factors found in the workplace can have an impact on occupational health and safety; one example of these physical factors is high noise intensity. One of the workplaces that have high noise intensity is a steel manufacturing company. The purpose of this study is to determine the noise risk based on noise mapping and analyse efforts that have been made in the Continous Tandem Cold Mill area in a steel company in Cilegon. Methods: The method used in this study was descriptive method. The variables used were the results of noise intensity measurements. The data were collected by means of literature study, field observation and noise measurements. The data obtained were then analysed using a descriptive method and were used as a basis in developing noise mapping. Results: Based on noise mapping, the welder area has the highest noise intensity of 91.1 - 94 dBA. Efforts to control noise intensity that have been carried out in the company are administrative control and personal protective equipment. Conclusion:From the results of noise intensity measurements in the Continous Tandem Cold Mill area of a steel company in Cilegon, it can be concluded that the measurement point is 76% - 100% and that the noise measurement points exceed the threshold value stipulated in the Regulation of the Minister of Manpower of the Republic Indonesia Number 5 of 2018. However, the steel company in Cilegon has made several efforts to reduce the noise intensity. Keywords: noise, noise mapping, steel company

Zeenathnisa Mougammadou Aribou, Wee Tong Ng

Abstract Fumigants are gaseous pesticides or biocides which eradicate pests by suffocation or poisoning. Worker exposure to fumigants is mainly via inhalation, followed by dermal contact and ingestion, leading to various acute and chronic health effects. Implementation of appropriate workplace controls such as adequate ventilation, training and personal protective equipment ensure that exposure to fumigants are kept to the lowest level as practically possible. In addition, routine medical examinations also allow for doctors to identify and manage possible exposure to fumigants and ascertain workers’ fitness to work. While management guidelines after an acute exposure to such fumigants is clear and consistent, the guidelines on routine medical examination for fumigators is sparse. Components of the medical examinations vary according to the fumigant, workers are exposed to and its chronic health effects. Hence, this paper highlights the health hazards of commonly utilised fumigants; Methyl Bromide, Hydrogen Cyanide, Hydrogen Phosphide and Sulfuryl Fluoride; and outlines the guidance for routine medical examinations for exposed fumigators.

Thibaut Cuvelier

In the context of energy transition, industrial plants that heavily rely on electricity face more and more price volatility. To continue operating in these conditions, the directors become continually more willing to increase their flexibility, i.e. their ability to react to price fluctuations. This work proposes an intuitive methodology to mathematically model electro-intensive processes in order to assess their flexibility potential. To this end, we introduce the notion of reservoir, a storage of either material or energy, that allows models based on this paradigm to have interpretations close to the physics of the processes. The design of the reservoir methodology has three distinct goals: (i) to be easy and quick to build by an energy-sector consultant; (ii) to be effortlessly converted into mixed-integer linear or nonlinear programs; (iii) to be straightforward to understand by nontechnical people, thanks to their graphic nature. We apply this methodology to two industrial case studies, namely an induction furnace (linear model) and an industrial cooling installation (nonlinear model), where we can achieve significant cost savings. In both cases, the models can be quickly written using our method and solved by appropriate solver technologies.

Konrad Wolsing, Eric Wagner, Antoine Saillard et al.

The increasing interconnection of industrial networks exposes them to an ever-growing risk of cyber attacks. To reveal such attacks early and prevent any damage, industrial intrusion detection searches for anomalies in otherwise predictable communication or process behavior. However, current efforts mostly focus on specific domains and protocols, leading to a research landscape broken up into isolated silos. Thus, existing approaches cannot be applied to other industries that would equally benefit from powerful detection. To better understand this issue, we survey 53 detection systems and find no fundamental reason for their narrow focus. Although they are often coupled to specific industrial protocols in practice, many approaches could generalize to new industrial scenarios in theory. To unlock this potential, we propose IPAL, our industrial protocol abstraction layer, to decouple intrusion detection from domain-specific industrial protocols. After proving IPAL's correctness in a reproducibility study of related work, we showcase its unique benefits by studying the generalizability of existing approaches to new datasets and conclude that they are indeed not restricted to specific domains or protocols and can perform outside their restricted silos.