Hasil untuk "Industrial safety. Industrial accident prevention"

Nurul Huda Mahmood, Onel L. A. Lopez, David Ruiz-Guirola et al.

Dependability - a system's ability to consistently provide reliable services by ensuring safety and maintainability in the face of internal or external disruptions - is a fundamental requirement for industrial wireless communication networks (IWCNs). While 5G ultra-reliable low-latency communication (URLLC) addresses some aspects of this challenge, its evolution toward holistic dependability in 6G must encompass reliability, availability, safety, and security. This paper provides a comprehensive framework for dependable IWCNs, bridging theory and practice. We first establish the theoretical foundations of dependability, including outlining its key attributes and presenting analytical tools to study it. Next, we explore practical enablers, such as adaptive multiple access schemes leveraging real-time monitoring and time-sensitive networking to ensure end-to-end determinism. A case study demonstrates how intelligent wake-up protocols improve event detection probability by orders of magnitude compared to conventional duty cycling. Finally, we outline open challenges and future directions for a 6G-driven dependable IWCN.

David Gringras

Safety benchmarks evaluate language models in isolation, typically using multiple-choice format; production deployments wrap these models in agentic scaffolds that restructure inputs through reasoning traces, critic agents, and delegation pipelines. We report one of the largest controlled studies of scaffold effects on safety (N = 62,808; six frontier models, four deployment configurations), combining pre-registration, assessor blinding, equivalence testing, and specification curve analysis. Map-reduce scaffolding degrades measured safety (NNH = 14), yet two of three scaffold architectures preserve safety within practically meaningful margins. Investigating the map-reduce degradation revealed a deeper measurement problem: switching from multiple-choice to open-ended format on identical items shifts safety scores by 5-20 percentage points, larger than any scaffold effect. Within-format scaffold comparisons are consistent with practical equivalence under our pre-registered +/-2 pp TOST margin, isolating evaluation format rather than scaffold architecture as the operative variable. Model x scaffold interactions span 35 pp in opposing directions (one model degrades by -16.8 pp on sycophancy under map-reduce while another improves by +18.8 pp on the same benchmark), ruling out universal claims about scaffold safety. A generalisability analysis yields G = 0.000: model safety rankings reverse so completely across benchmarks that no composite safety index achieves non-zero reliability, making per-model, per-configuration testing a necessary minimum standard. We release all code, data, and prompts as ScaffoldSafety.

Lucy M. Collins, Éilis J. O’Reilly, Joan Omosefe Osayande et al.

Parkinson’s disease (PD) is an age-related neurological disorder with increasing incidence and modifiable risk factors. People exposed to pesticides have up to a 2-fold higher risk of developing PD. Use of personal protective equipment (PPE) when using pesticides can lower an individual’s exposure. We examined awareness of the relationship between pesticides and PD risk in individuals working/living on farms in Ireland. We also investigated the practice of behaviours aimed at mitigating exposure, such as using PPE. An online survey was completed by a sample of the farming community (<i>n</i> = 707) attending agricultural fairs, and included demographics, lifetime/current residence/work on farms, pesticide contact, PPE use, PD diagnosis, and awareness of pesticide–PD association. Among participants, 51% worked/lived on farms and 62% reported contact with pesticides. Only 69% of those with pesticide contact reported using PPE, with gloves (57%) and masks (50%) most commonly used. Only 22% were aware of an association between PD and pesticides, and awareness did not increase PPE use. Among people with PD, only 40% had knowledge of the risk. We found that in a highly agricultural economy, occupational exposure to pesticides is common, but mitigation behaviours are not optimal. Educational campaigns to improve awareness of health risks from pesticides and to encourage PPE use could lower the personal and healthcare burden of PD and other health outcomes.

Aminu Darda’u Rafindadi, Bishir Kado, Abdurra’uf M. Gora et al.

This systematic review examines caught-in/between accidents in construction, revealing complex safety challenges involving machinery errors, vehicle incidents, loading mistakes, and structural collapses. The analysis highlights significant risks, including heavy equipment rollovers, trench cave-ins, and material shifts, with injuries ranging from minor to fatal. Despite the critical nature of these accidents, existing research demonstrates notable gaps, particularly in understanding long-term worker health impacts, economic consequences, and nuanced risk factors. Most studies insufficiently explore correlations between worker experience, age, and accident susceptibility, while gender-specific risks remain poorly documented. Training inadequacies and safety protocol non-adherence emerge as primary contributors to these incidents. This review identifies a pressing need for standardized, comprehensive safety interventions that address technological, human, and organizational factors. Recommendations include stricter safety regulations, enhanced training programs, advanced safety technologies, and robust support systems for workers. By fostering a holistic safety culture and addressing research gaps, the construction industry can potentially mitigate caught-in/between accidents, ultimately protecting worker well-being and improving overall productivity.

Tim Schreiter, Andrey Rudenko, Jens V. Rüppel et al.

Successful adoption of industrial robots will strongly depend on their ability to safely and efficiently operate in human environments, engage in natural communication, understand their users, and express intentions intuitively while avoiding unnecessary distractions. To achieve this advanced level of Human-Robot Interaction (HRI), robots need to acquire and incorporate knowledge of their users' tasks and environment and adopt multimodal communication approaches with expressive cues that combine speech, movement, gazes, and other modalities. This paper presents several methods to design, enhance, and evaluate expressive HRI systems for non-humanoid industrial robots. We present the concept of a small anthropomorphic robot communicating as a proxy for its non-humanoid host, such as a forklift. We developed a multimodal and LLM-enhanced communication framework for this robot and evaluated it in several lab experiments, using gaze tracking and motion capture to quantify how users perceive the robot and measure the task progress.

Anastasia Sergeeva, Claudia Negri-Ribalta, Gabriele Lenzini

Although extended reality(XR)-using technologies have started to be discussed in the industrial setting, it is becoming important to understand how to implement them ethically and privacy-preservingly. In our paper, we summarise our experience of developing XR implementations for the off-highway machinery domain by pointing to the main challenges we identified during the work. We believe that our findings can be a starting point for further discussion and future research regarding privacy and ethical challenges in industrial applications of XR.

Younggun Kim, Ahmed S. Abdelrahman, Mohamed Abdel-Aty

Ensuring the safety of vulnerable road users (VRUs), such as pedestrians and cyclists, is a critical challenge for autonomous driving systems, as crashes involving VRUs often result in severe or fatal consequences. While multimodal large language models (MLLMs) have shown promise in enhancing scene understanding and decision making in autonomous vehicles, there is currently no standardized benchmark to quantitatively evaluate their reasoning abilities in complex, safety-critical scenarios involving VRUs. To address this gap, we present VRU-Accident, a large-scale vision-language benchmark designed to evaluate MLLMs in high-risk traffic scenarios involving VRUs. VRU-Accident comprises 1K real-world dashcam accident videos, annotated with 6K multiple-choice question-answer pairs across six safety-critical categories (with 24K candidate options and 3.4K unique answer choices), as well as 1K dense scene descriptions. Unlike prior works, our benchmark focuses explicitly on VRU-vehicle accidents, providing rich, fine-grained annotations that capture both spatial-temporal dynamics and causal semantics of accidents. To assess the current landscape of MLLMs, we conduct a comprehensive evaluation of 17 state-of-the-art models on the multiple-choice VQA task and on the dense captioning task. Our findings reveal that while MLLMs perform reasonably well on visually grounded attributes, they face significant challenges in reasoning and describing accident causes, types, and preventability.

Theofanis P. Raptis, Andrea Passarella, Marco Conti

Energy efficiency and reliability are vital design requirements of recent industrial networking solutions. Increased energy consumption, poor data access rates and unpredictable end-to-end data access latencies are catastrophic when transferring high volumes of critical industrial data in strict temporal deadlines. These requirements might become impossible to meet later on, due to node failures, or excessive degradation of the performance of wireless links. In this paper, we focus on maintaining the network functionality required by the industrial, best effort, low-latency applications after such events, by sacrificing latency guarantees to improve energy consumption and reliability. We avoid continuously recomputing the network configuration centrally, by designing an energy efficient, local and distributed path reconfiguration method. Specifically, given the operational parameters required by the applications, our method locally reconfigures the data distribution paths, when a network node fails. Additionally, our method also regulates the return to an operational state of nodes that have been offline in the past. We compare the performance of our method through simulations to the performance of other state of the art protocols and we demonstrate performance gains in terms of energy consumption, data delivery success rate, and in some cases, end-to-end data access latency. We conclude by providing some emerging key insights which can lead to further performance improvements.

Huajie Liang, Di Wang, Yuchao Lu et al.

With the advancement of Industry 4.0, intelligent manufacturing extensively employs sensors for real-time multidimensional data collection, playing a crucial role in equipment monitoring, process optimisation, and efficiency enhancement. Industrial data exhibit characteristics such as multi-source heterogeneity, nonlinearity, strong coupling, and temporal interactions, while also being affected by noise interference. These complexities make it challenging for traditional anomaly detection methods to extract key features, impacting detection accuracy and stability. Traditional machine learning approaches often struggle with such complex data due to limitations in processing capacity and generalisation ability, making them inadequate for practical applications. While deep learning feature extraction modules have demonstrated remarkable performance in image and text processing, they remain ineffective when applied to multi-source heterogeneous industrial data lacking explicit correlations. Moreover, existing multi-source heterogeneous data processing techniques still rely on dimensionality reduction and feature selection, which can lead to information loss and difficulty in capturing high-order interactions. To address these challenges, this study applies the EAPCR and Time-EAPCR models proposed in previous research and introduces a new model, Time-EAPCR-T, where Transformer replaces the LSTM module in the time-series processing component of Time-EAPCR. This modification effectively addresses multi-source data heterogeneity, facilitates efficient multi-source feature fusion, and enhances the temporal feature extraction capabilities of multi-source industrial data.Experimental results demonstrate that the proposed method outperforms existing approaches across four industrial datasets, highlighting its broad application potential.

Chao Chen, G. Reniers, N. Khakzad

Abstract Recent catastrophic accidents in China and the USA urge and justify a thorough study on current & future research trends in the development of modeling methods and protection strategies for prevention and mitigation of large-scale escalating events or so-called domino effects in the process and chemical industries. This paper firstly provides an overview of what constitutes domino effects based on the definition and features, characterizing domino effect studies according to different research issues and approaches. The modeling approaches are grouped into three types while the protection strategies are divided into five categories, followed by detailed descriptions of representative modeling approaches and management strategies in chemical plants and clusters. The current research trends in this field are obtained based on the analysis of research work on domino effects caused by accidental events, natural events, and intentional attacks over a period of the past 30 years. A comparison analysis is conducted for the current modeling approaches and management strategies to pose their applications. Finally, this paper offers future research directions and identifies critical challenges in the field, aiming at improving the safety and security of chemical industrial areas so as to prevent and mitigate domino effects.

Ashvini Shrikrushna Hirve, Siddhant Sanjay Jaiswal

This paper explores the integration and capabilities of the ESP32-CAM, a versatile development board combining the ESP32 microcontroller with the OV2640 camera module. Emphasizing its cost-effective solution for Wi-Fi-enabled cameras, the ESP32-CAM boasts a potent 32-bit microcontroller and microSD card support, making it ideal for diverse IoT projects. Its applications in security surveillance, particularly in DIY security projects and home automation, further underscore its versatility in remote monitoring and surveillance. Security is a top priority for the ESP32-CAM, featuring a secure boot to authenticate firmware and robust encryption protocols ensuring secure Wi-Fi communication. Over-The-Air (OTA) updates enhance security by allowing remote firmware updates while maintaining data integrity. Access control measures, strong credentials, regular updates, and network segmentation fortify security at both device and network levels. The paper delves into the critical application of eye glass detection in industrial settings, emphasizing its role in ensuring occupational safety, regulatory compliance, and accident prevention. The ESP32-CAM's proactive approach in identifying non-compliance in hazardous areas contributes significantly to workplace safety and productivity. Integration with access control systems adds an extra layer of security, ensuring that only individuals with proper eye protection gain access to specified areas. Key findings highlight the ESP32-CAM's contributions, including its impact on occupational safety enhancement, integration with access control systems, data insights for safety analytics, emergency prioritization, and the development of customized training programs. In conclusion, the ESP32-CAM emerges as a crucial technological solution for enhancing safety, security, and productivity in industrial settings, showcasing its multifaceted benefits and contributions to creating a safer and more efficient working environment.

Andrea Fernández García, Rick Kye Gan, José Antonio Cernuda Martínez et al.

BACKGROUND Technological disasters in Asia have significant public health and environmental implications, but there is limited epidemiological analysis of these events. This study aims to characterize the epidemiological profile of technological disasters in Asia from 2000 to 2021, focusing on morbidity and mortality trends. METHODS A retrospective descriptive observational analysis was conducted using data from emergency events database (EM-DAT), DesInventar, NatCAt, and Sigma. The study categorized disasters into transport, industrial, and miscellaneous accidents. Statistical analyses were used to examine frequencies, trends, and correlations among the different disaster types. RESULTS From 2000 to 2021, Asia experienced 2333 technological disasters, with transport accidents being the most frequent (55.77%), followed by industrial (26.10%) and miscellaneous accidents (18.13%). The overall trend showed a statistically significant decrease in the frequency of these disasters and in average mortality and injury rates. The study highlighted the varying impact of different disaster types, with industrial accidents causing the highest fatality and affected rates despite being less frequent than transport accidents. CONCLUSIONS The study indicates a declining trend in the frequency and severity of technological disasters in Asia, reflecting improved safety measures and disaster management. However, the high impact of industrial accidents underscores the need for targeted prevention strategies.

Víctor Manuel Ramos-García, Josué Aarón López-Leyva, Ana Paola Balderrama-Carmona et al.

The occupational health and safety of agricultural workers is a topic that has a direct impact on the agricultural sector worldwide. For this reason, investigations into ergonomic factors are relevant to the health and safety of agricultural workers. In this study, nine variables of the physical–ergonomic dimension were analyzed to determine which factors represent occupational risks for agricultural workers in Mayo Valley, Mexico. A sample of 200 people was considered. The sample was separated by gender and divided into groups according to age. A closed-ended survey was developed and validated to assess physical ergonomics variables using a five-level Likert scale. Using Principal Component Analysis, it was found that there are physical ergonomic variables that affect male agricultural workers more than female workers (the risk of carrying heavy objects, PE3, and the risk of performing repetitive movements, PE4). It was also found that certain physical ergonomic variables are not perceived as hazardous by agricultural workers (the risk of using inappropriate materials, PE9). In addition, various research findings are discussed to determine the implications and recommendations for improving the occupational health and safety of agricultural workers in Mayo Valley, Mexico.

Alida LANGLOIS

This article explores workplace safety dynamics amid demographic shifts, highlighting the increase in aged workers. It addresses their unique safety challenges, particularly with machinery. The article examines the physical and cognitive effects of ageing on safety and technological adaptation. Findings show aged workers' experience reduces risks but highlights challenges with new technologies and complacency. It suggests tailored training, ergonomic adjustments, and health strategies to support the ageing workforce.

Razin Farhan Hussain, Mohsen Amini Salehi

The Industry 4.0 revolution has been made possible via AI-based applications (e.g., for automation and maintenance) deployed on the serverless edge (aka fog) computing platforms at the industrial sites -- where the data is generated. Nevertheless, fulfilling the fault-intolerant and real-time constraints of Industry 4.0 applications on resource-limited fog systems in remote industrial sites (e.g., offshore oil fields) that are uncertain, disaster-prone, and have no cloud access is challenging. It is this challenge that our research aims at addressing. We consider the inelastic nature of the fog systems, software architecture of the industrial applications (micro-service-based versus monolithic), and scarcity of human experts in remote sites. To enable cloud-like elasticity, our approach is to dynamically and seamlessly (i.e., without human intervention) federate nearby fog systems. Then, we develop serverless resource allocation solutions that are cognizant of the applications' software architecture, their latency requirements, and distributed nature of the underlying infrastructure. We propose methods to seamlessly and optimally partition micro-service-based application across the federated fog. Our experimental evaluation express that not only the elasticity is overcome in a serverless manner, but also our developed application partitioning method can serve around 20% more tasks on-time than the existing methods in the literature.

Arthur Amorim, Trevor Kann, Max Taylor et al.

Industrial control systems (ICSs) increasingly rely on digital technologies vulnerable to cyber attacks. Cyber attackers can infiltrate ICSs and execute malicious actions. Individually, each action seems innocuous. But taken together, they cause the system to enter an unsafe state. These attacks have resulted in dramatic consequences such as physical damage, economic loss, and environmental catastrophes. This paper introduces a methodology that restricts actions using protocols. These protocols only allow safe actions to execute. Protocols are written in a domain specific language we have embedded in an interactive theorem prover (ITP). The ITP enables formal, machine-checked proofs to ensure protocols maintain safety properties. We use dynamic attestation to ensure ICSs conform to their protocol even if an adversary compromises a component. Since protocol conformance prevents unsafe actions, the previously mentioned cyber attacks become impossible. We demonstrate the effectiveness of our methodology using an example from the Fischertechnik Industry 4.0 platform. We measure dynamic attestation's impact on latency and throughput. Our approach is a starting point for studying how to combine formal methods and protocol design to thwart attacks intended to cripple ICSs.

Hyungmin Kim, Donghun Kim, Pyunghwan Ahn et al.

While deep neural networks have achieved remarkable performance, data augmentation has emerged as a crucial strategy to mitigate overfitting and enhance network performance. These techniques hold particular significance in industrial manufacturing contexts. Recently, image mixing-based methods have been introduced, exhibiting improved performance on public benchmark datasets. However, their application to industrial tasks remains challenging. The manufacturing environment generates massive amounts of unlabeled data on a daily basis, with only a few instances of abnormal data occurrences. This leads to severe data imbalance. Thus, creating well-balanced datasets is not straightforward due to the high costs associated with labeling. Nonetheless, this is a crucial step for enhancing productivity. For this reason, we introduce ContextMix, a method tailored for industrial applications and benchmark datasets. ContextMix generates novel data by resizing entire images and integrating them into other images within the batch. This approach enables our method to learn discriminative features based on varying sizes from resized images and train informative secondary features for object recognition using occluded images. With the minimal additional computation cost of image resizing, ContextMix enhances performance compared to existing augmentation techniques. We evaluate its effectiveness across classification, detection, and segmentation tasks using various network architectures on public benchmark datasets. Our proposed method demonstrates improved results across a range of robustness tasks. Its efficacy in real industrial environments is particularly noteworthy, as demonstrated using the passive component dataset.

Heyuan Huang, Xingyu Lou, Chaochao Chen et al.

Cross-Domain Recommendation (CDR) have received widespread attention due to their ability to utilize rich information across domains. However, most existing CDR methods assume an ideal static condition that is not practical in industrial recommendation systems (RS). Therefore, simply applying existing CDR methods in the industrial RS environment may lead to low effectiveness and efficiency. To fill this gap, we propose DIIT, an end-to-end Domain-Invariant Information Transfer method for industrial cross-domain recommendation. Specifically, We first simulate the industrial RS environment that maintains respective models in multiple domains, each of them is trained in the incremental mode. Then, for improving the effectiveness, we design two extractors to fully extract domain-invariant information from the latest source domain models at the domain level and the representation level respectively. Finally, for improving the efficiency, we design a migrator to transfer the extracted information to the latest target domain model, which only need the target domain model for inference. Experiments conducted on one production dataset and two public datasets verify the effectiveness and efficiency of DIIT.

Taejun Lee, Keanseb Woo, Panyoung Kim et al.

Korea’s industrial accident rate ranks high among Organization for Economic Co-operation and Development countries. Moreover, large-scale accidents have recently occurred. Accordingly, the requirements for management and supervision in industrial sites are increasing; in this context, the “Act on Punishment of Serious Accidents, etc.” has been enacted. Aiming to prevent such industrial accidents, various data collected by devices such as sensors and closed-caption televisions (CCTVs) are utilized to track workers and detect hazardous substances, gases, and fires at industrial sites. In this study, an industrial area requiring such technology is selected. A hazardous situation event is derived, and a dataset is built using CCTV data. A violation corresponding to a hazardous situation event is detected and a warning is provided. The events incorporate requirements extracted from industrial sites, such as those concerning collision risks and the wearing of safety equipment. The precision of the event violation detection exceeds 95% and the response and delay times are under 20 ms. Thus, this system is believed to be used at industrial sites and for other intelligent industrial safety prevention solutions.

Michelle Duren, Bryce Corrigan, Ryan David Kennedy et al.

Perceived safety is recognized throughout the mode choice literature as a key barrier to cycling, yet its constructs are poorly understood. Although commonly understood to relate to crash and injury risk and sometimes vulnerability to crime, health impact assessments identify numerous other pathways through which cycling can negatively impact health. This study leverages a nationally representative survey of U.S. adults in 2022 to assess a set of eleven factors as potential components of perceived cycling safety. We use principal component analysis to identify components of perceived cycling safety and then employ principal component regression to assess these components in relation to predicting unsafe cycling perception. We identify five key dimensions of perceived safety. Specifically, we found that perceived bicycling safety can be encompassed in the following components: (1) contaminant exposure, (2) injurious collision risk, (3) street conditions, (4) weather conditions, and (5) crime risk. In evaluating each identified component, we found that injurious collision risk and street conditions were the most predictive of considering cycling as unsafe. We further develop an understanding of how differences in cycling behavior, such as using cycling for commuting purposes, may contribute to differences in how cycling safety components coalesce into perceived safety.