In modern process industries, data-driven models are important tools for real-time monitoring when key performance indicators are difficult to measure directly. While accurate predictions are essential, reliable uncertainty quantification (UQ) is equally critical for safety, reliability, and decision-making, but remains a major challenge in current data-driven approaches. In this work, we introduce a diffusion-based posterior sampling framework that inherently produces well-calibrated predictive uncertainty via faithful posterior sampling, eliminating the need for post-hoc calibration. In extensive evaluations on synthetic distributions, the Raman-based phenylacetic acid soft sensor benchmark, and a real ammonia synthesis case study, our method achieves practical improvements over existing UQ techniques in both uncertainty calibration and predictive accuracy. These results highlight diffusion samplers as a principled and scalable paradigm for advancing uncertainty-aware modeling in industrial applications.
Kan Shimazaki, Yo Ishigaki, Kazunori Hayashi
et al.
Objectives: To evaluate the effects of carbon dioxide (CO2) concentrations in automobiles on driving performance. Methods: A driving simulator experiment was conducted with eight taxi drivers. The experiment was conducted under low CO2 concentration (<500 ppm) and high CO2 concentration (5,000 ppm) conditions. To evaluate driving performance and cognitive function, three measures were employed: a two-back task, an LED response task, and a driving assessment. The driving assessment used scoring criteria from the driving license proficiency test. Results: Poisson regression analysis showed that wobble (p=0.044), signal failure (p=0.045), contact (p=0.003), and wheel departure (p=0.005) were significantly increased under high CO2 concentration conditions. Generalized linear mixed model (GLMM) analysis showed that reaction time in the LED response task was significantly reduced under high CO2 concentration conditions (p<0.001). On the other hand, the GLMM analysis of the two-back task showed no significant effect of CO2 concentration (incorrect response rate: p=0.733, non-response rate: p=0.485). Conclusions: These results suggest that elevated CO2 concentrations may have a negative impact on driving behavior, especially skill-based driving behavior. On the other hand, the effects on cognitive tasks requiring working memory were limited. The results of this study suggest that managing CO2 concentration in vehicles is important for maintaining safe driving and raise the need for specific measures, such as the development of systems for measuring, predicting, and controlling CO2 concentration in vehicles, and the implementation of driver education programs.
Industrial safety. Industrial accident prevention, Medicine (General)
Industrial Control Systems (ICSs) are complex interconnected systems used to manage process control within industrial environments, such as chemical processing plants and water treatment facilities. As the modern industrial environment moves towards Internet-facing services, ICSs face an increased risk of attacks that necessitates ICS-specific Intrusion Detection Systems (IDS). The development of such IDS relies significantly on a simulated testbed as it is unrealistic and sometimes hazardous to utilize an operational control system. Whilst some testbeds have been proposed, they often use a limited selection of virtual ICS simulations to test and verify cyber security solutions. There is a lack of investigation done on developing systems that can efficiently simulate multiple ICS architectures. Currently, the trend within research involves developing security solutions on just one ICS simulation, which can result in bias to its specific architecture. We present ICS-SimLab, an end-to-end software suite that utilizes Docker containerization technology to create a highly configurable ICS simulation environment. This software framework enables researchers to rapidly build and customize different ICS environments, facilitating the development of security solutions across different systems that adhere to the Purdue Enterprise Reference Architecture. To demonstrate its capability, we present three virtual ICS simulations: a solar panel smart grid, a water bottle filling facility, and a system of intelligent electronic devices. Furthermore, we run cyber-attacks on these simulations and construct a dataset of recorded malicious and benign network traffic to be used for IDS development.
Autonomous control of multi-stage industrial processes requires both local specialization and global coordination. Reinforcement learning (RL) offers a promising approach, but its industrial adoption remains limited due to challenges such as reward design, modularity, and action space management. Many academic benchmarks differ markedly from industrial control problems, limiting their transferability to real-world applications. This study introduces an enhanced industry-inspired benchmark environment that combines tasks from two existing benchmarks, SortingEnv and ContainerGym, into a sequential recycling scenario with sorting and pressing operations. We evaluate two control strategies: a modular architecture with specialized agents and a monolithic agent governing the full system, while also analyzing the impact of action masking. Our experiments show that without action masking, agents struggle to learn effective policies, with the modular architecture performing better. When action masking is applied, both architectures improve substantially, and the performance gap narrows considerably. These results highlight the decisive role of action space constraints and suggest that the advantages of specialization diminish as action complexity is reduced. The proposed benchmark thus provides a valuable testbed for exploring practical and robust multi-agent RL solutions in industrial automation, while contributing to the ongoing debate on centralization versus specialization.
We present a novel reinforcement learning (RL) environment designed to both optimize industrial sorting systems and study agent behavior in evolving spaces. In simulating material flow within a sorting process our environment follows the idea of a digital twin, with operational parameters like belt speed and occupancy level. To reflect real-world challenges, we integrate common upgrades to industrial setups, like new sensors or advanced machinery. It thus includes two variants: a basic version focusing on discrete belt speed adjustments and an advanced version introducing multiple sorting modes and enhanced material composition observations. We detail the observation spaces, state update mechanisms, and reward functions for both environments. We further evaluate the efficiency of common RL algorithms like Proximal Policy Optimization (PPO), Deep-Q-Networks (DQN), and Advantage Actor Critic (A2C) in comparison to a classical rule-based agent (RBA). This framework not only aids in optimizing industrial processes but also provides a foundation for studying agent behavior and transferability in evolving environments, offering insights into model performance and practical implications for real-world RL applications.
This paper presents HERO (Hierarchical Testing with Rabbit Optimization), a novel black-box adversarial testing framework for evaluating the robustness of deep learning-based Prognostics and Health Management systems in Industrial Cyber-Physical Systems. Leveraging Artificial Rabbit Optimization, HERO generates physically constrained adversarial examples that align with real-world data distributions via global and local perspective. Its generalizability ensures applicability across diverse ICPS scenarios. This study specifically focuses on the Proton Exchange Membrane Fuel Cell system, chosen for its highly dynamic operational conditions, complex degradation mechanisms, and increasing integration into ICPS as a sustainable and efficient energy solution. Experimental results highlight HERO's ability to uncover vulnerabilities in even state-of-the-art PHM models, underscoring the critical need for enhanced robustness in real-world applications. By addressing these challenges, HERO demonstrates its potential to advance more resilient PHM systems across a wide range of ICPS domains.
Accurately predicting the lifespan of critical device components is essential for maintenance planning and production optimization, making it a topic of significant interest in both academia and industry. In this work, we investigate the use of survival analysis for predicting the lifespan of production printheads developed by Canon Production Printing. Specifically, we focus on the application of five techniques to estimate survival probabilities and failure rates: the Kaplan-Meier estimator, Cox proportional hazard model, Weibull accelerated failure time model, random survival forest, and gradient boosting. The resulting estimates are further refined using isotonic regression and subsequently aggregated to determine the expected number of failures. The predictions are then validated against real-world ground truth data across multiple time windows to assess model reliability. Our quantitative evaluation using three performance metrics demonstrates that survival analysis outperforms industry-standard baseline methods for printhead lifespan prediction.
Dr. A. R Arvind, Niteesh Balaji. K, Hemanth Sivaganesh. S. P
et al.
A well-planned plant layout is essential for ensuring safety, efficiency, and environmental sustainability in industrial operations. This study presents key safety guidelines focusing on risk reduction, accident prevention, and worker well-being. It highlights critical factors such as fire protection, electrical safety, machine guarding, ergonomics, and hazardous material handling, along with environmental controls like ventilation, noise reduction, and waste management. Provisions for emergency readiness, first aid, and safety signage are also included, all aligned with ISO standards , and the Indian Factories Act, 1948. Integrating these measures from the design stage helps minimize risks, enhance productivity, and ensure compliance, while supporting sustainable development and operational excellence in modern industrial facilities
David Chinalu Anaba, Azeez Jason Kess-Momoh, Sodrudeen Abolore Ayodeji
Health, Safety, and Environmental (HSE) standards are indispensable in industrial operations, safeguarding workers, protecting the environment, and ensuring operational efficiency. This paper comprehensively reviews HSE standards, focusing on their historical evolution, key components, implementation challenges, and future directions. The historical evolution section traces the development of HSE standards, from early regulatory responses to industrial hazards to establishing integrated frameworks encompassing health, safety, and environmental considerations. Key components of HSE standards include health standards addressing occupational health risks, safety protocols emphasizing accident prevention and emergency response, and environmental standards regulating pollution control and waste management. Integrated HSE management systems promote synergy across these domains, enhancing operational resilience. Challenges in implementing HSE standards are multifaceted, ranging from regulatory compliance issues and technological barriers to organizational cultural and economic constraints. Overcoming these challenges requires concerted efforts to enhance regulatory frameworks, adopt innovative technologies, foster a safety culture, and balance economic considerations with HSE commitments. Future directions in HSE practices emphasize AI, IoT, and digitalization innovations to enhance real-time monitoring and predictive capabilities. Policy recommendations focus on strengthening regulatory frameworks and promoting proactive approaches to HSE governance. Organizational strategies highlight the role of leadership, employee engagement, and continuous improvement in fostering a culture of safety and environmental stewardship. In conclusion, advancing HSE standards is crucial for promoting sustainable industrial practices and ensuring the well-being of workers and the environment. Embracing these standards mitigates risks and enhances operational efficiency and corporate responsibility in a rapidly evolving global landscape. Keywords: HSE Standards, Occupational Health, Safety Protocols, Environmental Protection, Regulatory Compliance.
The increase in energy demand due to industrial development and urbanization has resulted in the development of large-scale energy facilities. Republic of Korea’s petrochemical industrial complexes serve as prime examples of this phenomenon. However, because of complex processes and aging facilities, many of which have been in operation for over a decade, these industrial complexes are prone to process-deviation-related accidents. Chemical accidents in energy facilities involving high-pressure liquids or gases are especially dangerous; therefore, proactive accident prevention is critical. This study is also relevant to corporate environment, social, and governance (ESG) management. Preventing chemical accidents to protect workers from injury is critical for business and preventing damage to surrounding areas from chemical accidents is a key component of ESG safety. In this study, we collected accident data, specifically injury-related incidents, from Republic of Korea’s petrochemical industrial complexes, which are the foundation of the energy industry. We analyzed the causes of accidents in a step-by-step manner. Furthermore, we conducted a risk analysis by categorizing accident data based on the level of risk associated with each analysis result; we identified the main causes of accidents and “high-risk process stages” that posed significant risk. The analysis reveals that the majority of accidents occur during general operations (50%, 167 cases) and process operations (39%, 128 cases). In terms of incident types, fire/explosion incidents accounted for the highest proportion (43%, 144 cases), followed by leakage incidents (24%, 78 cases). Furthermore, we propose a disaster safety artificial intelligence (AI) model to prevent major and fatal accidents during these high-risk process stages. A detailed analysis reveals that human factors such as accumulated worker fatigue, insufficient safety training, and non-compliance with operational procedures can significantly increase the likelihood of accidents in petrochemical facilities. This finding emphasizes the importance of introducing measurement sensors and AI convergence technologies to help humans predict and detect any issues. Therefore, we selected representative accident cases for implementing our disaster safety model.
The prevention and effective management of industrial accidents is key to protecting public health, the environment, and economic stability. The comparison of online platforms and databases allows us to present a comprehensive picture of safety practices in different industrial sectors, as well as their efficiency and effectiveness. This will help to identify best practices and areas where further improvements are needed for the country concerned. As a result of our comparative analysis, targeted measures can be taken to prevent and manage accidents, reduce the number and severity of industrial accidents, and continuously improve industrial safety standards and practices. Furthermore, comparative analyses allow us to identify differences in industrial safety cultures and the lessons that can be learned from them. We provide suggestions for improving the efficiency and competitiveness of industrial processes while focusing on the safety of workers, the public, and the environment, as a key priority.
Dan Lehman, Tim J. Schoonbeek, Shao-Hsuan Hung
et al.
Recognizing errors in assembly and maintenance procedures is valuable for industrial applications, since it can increase worker efficiency and prevent unplanned down-time. Although assembly state recognition is gaining attention, none of the current works investigate assembly error localization. Therefore, we propose StateDiffNet, which localizes assembly errors based on detecting the differences between a (correct) intended assembly state and a test image from a similar viewpoint. StateDiffNet is trained on synthetically generated image pairs, providing full control over the type of meaningful change that should be detected. The proposed approach is the first to correctly localize assembly errors taken from real ego-centric video data for both states and error types that are never presented during training. Furthermore, the deployment of change detection to this industrial application provides valuable insights and considerations into the mechanisms of state-of-the-art change detection algorithms. The code and data generation pipeline are publicly available at: https://timschoonbeek.github.io/error_seg.
Juan M. Deniz, Andre S. Kelboucas, Ricardo Bedin Grando
This study explores human-robot interaction (HRI) based on a mobile robot and YOLO to increase real-time situation awareness and prevent accidents in the workplace. Using object segmentation, we propose an approach that is capable of analyzing these situations in real-time and providing useful information to avoid critical working situations. In the industry, ensuring the safety of workers is paramount, and solutions based on robots and AI can provide a safer environment. For that, we proposed a methodology evaluated with two different YOLO versions (YOLOv8 and YOLOv5) alongside a LoCoBot robot for supervision and to perform the interaction with a user. We show that our proposed approach is capable of navigating a test scenario and issuing alerts via Text-to-Speech when dangerous situations are faced, such as when hardhats and safety vests are not detected. Based on the results gathered, we can conclude that our system is capable of detecting and informing risk situations such as helmet/no helmet and safety vest/no safety vest situations.
Paula Fraga-Lamas, Tiago M Fernandez-Carames, Oscar Blanco-Novoa
et al.
Shipbuilding companies are upgrading their inner workings in order to create Shipyards 4.0, where the principles of Industry 4.0 are paving the way to further digitalized and optimized processes in an integrated network. Among the different Industry 4.0 technologies, this article focuses on Augmented Reality, whose application in the industrial field has led to the concept of Industrial Augmented Reality (IAR). This article first describes the basics of IAR and then carries out a thorough analysis of the latest IAR systems for industrial and shipbuilding applications. Then, in order to build a practical IAR system for shipyard workers, the main hardware and software solutions are compared. Finally, as a conclusion after reviewing all the aspects related to IAR for shipbuilding, it is proposed an IAR system architecture that combines Cloudlets and Fog Computing, which reduce latency response and accelerate rendering tasks while offloading compute intensive tasks from the Cloud.
The accident mortality rate of major accidents (MAs) show that China is still in the bottleneck period of accident prevention and control. To further promote the MAs prevention and control, this paper presents a novel major accidents evolution model from the theoretical perspective of information processing (IP). Firstly, based on the safety science paradigm of accident prevention and the emergency management paradigm of accident control, a safety information processing (SIP) process is proposed. Secondly, established the SIP model for different stages of accident prevention and control, which involves danger information processing (DIP), potential hazard information processing (PHIP), risk information processing (RIP), and emergency information processing (EIP). Thirdly, revealed the SIP of various management subject and the failure principle of accident prevention and control, that is, MAs occur under the premise of continuous failures of DIP, PHIP, RIP, and EIP under the social-technical system. Finally, the DPRE-IP model is proposed from the whole evolution path of “danger-potential hazard-risk-accident”. To demonstrate the viability of the model, this model is applied to the “6·13” Wenling major explosion accident. The results show that the proposed DPRE-IP model can provide new ideas for the formulation of accident prevention and control measures and accident analysis.
The article presents the results of the analysis of the main indicators of the supervisory and control activities of Rostechnadzor in the field of industrial safety at hazardous production facilities of the coal industry. It is revealed that with the implementation of constant state control (supervision), the share of bans and suspensions of the activities of hazardous production facilities increased by 1.5 times, while the injury and accident rates decreased significantly. This indicates an increase in the efficiency of supervision, and the level of industrial safety in the industry. The factors (reasons and circumstances) of prohibitions and suspensions of the activities of hazardous production facilities of the coal industry due to the violations of industrial safety requirements and mining operations are considered. Suspension and prohibition of the activities means that there is a potential risk of an emergency occurence. Considering the emphasis on stimulating risk prevention, characteristic assessments of the subjects of supervision indicated in the acts of scheduled field inspections carried out in 2021 as part of the state control and supervision of compliance by coal enterprises with industrial safety requirements are highlighted in the article. When analyzing the acts of scheduled inspections, attention was paid to the presence of identified violations that could provoke an accident or incident, leading to a deviation from the established mode of the technological process and possible causing harm. It should be noted that in the inspection acts, great emphasis is placed on monitoring the condition of technical devices: fences, signaling devices, blockages, crossings, stairs, platforms, railings, etc., what applies to the subjects of supervision by the Ministry of Labor of the Russian Federation. Therefore, in order to improve the level of industrial safety and the safety of mining operations at the coal industry enterprises, the emphasis should be placed on the adapted application of a risk-oriented approach to the subjects of supervision of a particular hazardous production facility. This approach will increase the level of industrial safety at the controlled enterprises and reduce the costs of the caused harm (damage).