Risk Assessment Of Accidents Research Articles

Research on fire accident prediction and risk assessment algorithm based on data mining and machine learning

Forest fire is a kind of natural disaster that is destructive, easy to spread, and difficult to extinguish. It greatly harms the balance of ecosystem and human life and property. The prediction and risk assessment of forest fire accidents can find forest fires as early as possible and then take corresponding remedial measures; the loss of forest fires will be minimized. This paper first collected the relevant data information in the Lesser Hinggan Mountain region, established the fire driving factor database, and then analyzed the impact and distribution characteristics of forest fire driving factors such as temperature and rainfall. Finally, a fire accident prediction model was built based on a deep neural network, and the model’s performance was compared with the SVM and RF models. The analysis results show that this paper’s fire accident prediction model is more accurate.

Анализ состояния охраны труда и промышленной безопасности на АО «Разрез Колыванский»

A risk-oriented approach at hazardous production facilities is a part of critical measures to maintain the safety of the production environment affecting both human health and performance during the working process and to minimize accidents and production hazards. The objective assessment of all stages of the production process is an important aspect of the enterprise's activities. Enterprises should immediately introduce and implement comprehensive measures of production safety that must simultaneously minimize threats to human health (occupational risk assessment — the development and implementation of unified criteria of accident risk assessment) and prevention of potential hazards at each workplace in particular and at the enterprise in general (establishing the industrial safety culture and awareness of personal responsibility for the workflow safety, compliance with industrial safety requirements). The latter is the key principle as it concerns cross-industry integration, i.e., the basic level of knowledge and skills obtained during the learning process and the practical experience of working with industrial equipment, and specifics of the enterprise’s operation. The article provides the analysis of outcomes of work in the sphere of occupational and industrial safety at Razrez Kolavansky JSC with the application of a risk-oriented approach. At the same time, the impact of education (qualification) on injuries of the employees of Razrez has been considered.

Investigation of scaling-down experiments for accidental CO2 leakage dispersion risks in constant-pressure transportation pipelines within the CCUS process

Within the framework of strategies to prevent global warming, the safety of CO2 transportation pipelines requires thorough hazard analysis. Accidental leakage during long-distance CO2 transport, influenced by the size of the leakage opening, typically occurs under constant pressure conditions in the CCUS process. This paper introduces an innovative scaled experimental apparatus with an elliptical orifice, designed to simulate leakage due to free rupture in pipelines. The goal is to understand the risk characteristics of constant-pressure leakage in transportation pipelines. The results indicate that during leakage, the pressure drop in the source CO2 pipeline is less than 0.3013 MPa/s, which can be considered constant pressure leakage. The Mach disk size dm has a linear relationship with the elliptical leakage orifice ab. The scaled experiments also show that temperatures can drop to as low as −78°C, indicating significant low-temperature risks. This scaling experiment provides a new approach for predicting and mitigating diffusion risks in long-distance pipeline transportation by examining the relationship between pipeline length and orifice diameter. The research data and results contribute to a deeper understanding of the safety aspects of long-distance transportation of CO2 under constant pressure leakage conditions, providing a basis for risk assessment of leakage accidents and environmental protection measures.

Risk assessment of hazmat road transportation accidents before, during, and after the accident using Bayesian network

Risk exists in the whole process of before, during and after the accident of hazardous materials (hazmat) transportation, which usually lead to serious consequences. Risk assessment is a significant way to identify hidden dangers and their interactions, which can curb accidents happen. The purpose of this study is to analyze the risk factors of hazmat road transportation accidents through entire process risk assessment and propose management strategies. Considering nodes dependency, i.e. accident type and rescue time, this study applies the concept of Tree Augmented Naive Bayes (TAN) and built a hazmat road transportation Bayesian network (HRT-BN). Mutual information and model validation are carried on to better facilitate convenient inference and diagnosis of risk nodes and their interactions. Results show that in causal inference with single factors, human risk factors are more significant, which are likely to lead to leakage accident with slight consequences and short rescue time (s2). Meanwhile, with coupling effect of multiple factors, accident types become more diverse with serious consequence. In diagnostic inference, the rescue time is linked to seasonal and regional factors, e.g., the rescue time in summer is longer. Based on these analysis, targeted recommendations have been proposed at the end. This study provides a whole angle of view to assess risks in hazmat transportation, which has significant practical feasibility for risk management and planning of hazmat road transportation.

Digital twin-based modeling of natural gas leakage and dispersion in urban utility tunnels

Background Unexpected leakage accidents of the natural gas pipeline inside urban utility tunnels can pose great threats to public safety, property, and the environment. It highlights the modeling of natural gas leakage and dispersion dynamics, especially from a digital twin implementation perspective facilitating effective emergency response in a data-driven way. Methods In this study, a digital twin-based emergency response framework for gas leakage accidents in urban utility tunnels is proposed. Within this framework, the data-calibrated gas concentration prediction (DC-GCP) model is developed by integrating the Lattice Boltzmann Method (LBM) with data assimilation (DA) techniques. This combination enables accurate spatiotemporal predictions of gas concentrations, even with a prior or inaccurate gas leakage source term. Specifically, we develop a high-performance LBM-based gas concentration prediction model using the parallel programming language Taichi Lang. Based on this model, real-time integration of gas sensor data from utility tunnels is achieved through the DA algorithm. Therefore, the predicted results can be calibrated by the continuous data in the absence of complete source term information. Furthermore, a widely used twin experiment and statistical performance measures (SPMs) are used to evaluate and validate the effectiveness of the proposed approach. Results The results show that all SPMs progressively converge towards their ideal values as calibration progresses. And both the gas concentration predictions and the source term estimations can be calibrated effectively by the proposed approach, achieving a relative error of less than 5%. Conclusions This study helps for dynamic risk assessment and emergency response of natural gas leakage accidents, as well as facilitating the implementation of predictive digital twin in utility tunnels.

Retraction Note: Preliminary risk assessment of in-vessel leakage accident in ITER

Retraction Note: Preliminary risk assessment of in-vessel leakage accident in ITER

Comprehensive analysis of the mining accident forecasting and risk assessment methodologies: Case study – Stanterg Mine

Purpose. This research aims to outline a methodology for accident forecasting and risk assessment in mining operations using the Stanterg Mine as a case study. It emphasises the crucial role of reliable mining accident reporting and accurate data processing in forecasting accidents and effectively managing risks during mining operations. Methods. The research paper analyzes various methodologies for forecasting mining accidents using Excel and Simple Linear Regression Method (SLRM) to analyse data selected from the Stanterg Mine accidents. Findings. The forecast indicates that on average there are about 3 accidents per month at the Stanterg Mine. The analysis, based on a one-month study of 42 reported accidents, and assuming a steady production rate, suggests an increased risk of accidents. This is supported by a thorough assessment using a 3×3 risk assessment matrix tailored to the Stanterg Mine. Stope mining is highlighted as the most hazardous area, associated with risks ranging from moderate to extreme levels. Originality. Mining accident analysis at the Stanterg Mine involves an examination of the incidents, including factors leading to accidents, encountered hazards and their consequences. Accident forecasting entails studying historical data, identifying patterns, and using predictive modelling to anticipate future incidents. This proactive approach enables mining companies to proactively address risks and take preventative measures, reducing the probability of accidents. Practical implications. The systematic processing and analysis of mining accidents has revealed valuable insights into the practical application of risk assessment in mining operations. The examination of accidents at the Stanterg Mine provides researchers with crucial knowledge for effective risk assessment and management in the mining sector.

Hydrogen Jet Flame Simulation and Thermal Radiation Damage Estimation for Leakage Accidents in a Hydrogen Refueling Station

With the rapid development of hydrogen energy worldwide, the number of hydrogen energy facilities, such as hydrogen refueling stations, has grown rapidly in recent years. However, hydrogen is prone to leakage accidents during use, which could lead to hazards such as fires and explosions. Therefore, research on the safety of hydrogen energy facilities is crucial. In this paper, a study of high-pressure hydrogen jet flame accidents is conducted for a proposed integrated hydrogen production and refueling station in China. The effects of leakage direction and leakage port diameter on the jet flame characteristics are analyzed, and a risk assessment of the flame accident is conducted. The results showed that the death range perpendicular to the flame direction increased from 2.23 m to 5.5 m when the diameter of the leakage port increased from 4 mm to 10 mm. When the diameter of the leakage port is larger than 8 mm, the equipment on the scene will be within the boundaries of the damage. The consequences of fire can be effectively mitigated by a reasonable firewall setup to ensure the overall safety of the integrated station.

A deep neural network algorithm-based approach for predicting recovery period of accidents according to construction scale

Despite ongoing safety efforts, construction sites experience a concerningly high accident rate. Notwithstanding that policies and research to reduce the risk of accidents in the construction industry have been active for a long time, the accident rate in the construction industry is considerably higher than in other industries. This trend may likely be further exacerbated by the rapid growth of large-scale construction projects driven by urban population expansion. Consequently, accurately predicting recovery periods of accidents at construction sites in advance and proactively investing in measures to mitigate them is critical for efficiently managing construction projects. Therefore, the purpose of this study is to propose a framework for developing accident prediction models based on the Deep Neural Network (DNN) algorithm according to the scale of the construction site. This study suggests DNN models and applies the DNN for each construction site scale to predict accident recovery periods. The model performance and accuracy were evaluated using mean absolute error (MAE) and root-mean-square error (RMSE) and compared with the widely used multiple regression analysis model. As a result of model comparison, the DNN models showed a lower prediction error rate than the regression analysis models for both small-to-medium and large construction sites. The findings and framework of this study can be applied as the opening stage of accident risk assessment using deep learning techniques, and the introduction of deep learning technology to safety management according to the scale of the construction site is provided as a guideline.

Quantitative risk assessment of leakage accident of crude oil storage tank based on fuzzy Bayesian network and improved AHP

Leakage accidents of crude oil storage tanks (LACOST) occasionally occur during the production and storage processes of the petroleum and chemical industry, significantly impacting lives, the environment, and private property. To enhance the risk assessment of LACOST, our study sought to construct a fuzzy Bayesian network (FBN) through expert evaluation based on an improved analytic hierarchy process (AHP). Subsequently, the societal risk of LACOST was analyzed in conjunction with the surrounding population density. Applying the proposed method to a crude oil storage depot in China revealed that incorporating the improved AHP significantly enhances the FBN's risk assessment capability, leading to more accurate predictions of LACOST likelihood. Furthermore, the importance of basic events was assessed, thereby effectively and reliably identifying critical events of LACOST. The rationality of the layout of buildings and population density in the oil depot was assessed through societal risk analysis. Collectively, our findings demonstrated that the proposed method can effectively identify changes in both LACOST probabilities and consequences, enabling decision-makers to optimize risk management strategies and achieve efficient resource allocation.

Risk assessment of shield tunnel collapse accident based on fuzzy analytic hierarchy process

Risk assessment of shield tunnel collapse accident based on fuzzy analytic hierarchy process

Risk assessment of unmanned aerial vehicle accidents based on data-driven Bayesian networks

With the popularity of unmanned aerial vehicles (UAVs) in different application scenarios, a series of potential risks and challenges have emerged. Few quantitative studies in the existing literature analyze the risk factors that affect and cause UAV accidents. Therefore, this study aims to reveal the dependency relationships among the risk factors associated with UAV accidents by using a data-driven tree-augmented naïve Bayesian network (TAN-BN) method. Firstly, UAV accident reports from transportation departments of various countries are collected and analyzed to establish a dataset on the risk factors inducing different UAV accidents. Then, a data-driven UAV risk analysis model is constructed through data learning and training approaches to reveal the interdependency of risk factors and their comprehensive impacts on the severity of UAV accidents. Sensitivity analysis is comprehensively conducted to show the rationality of the proposed TAN-BN. The results reveal that the risk factors “other aircraft”, “observer errors”, “non-powerplant”, “preconditions for unsafe acts”, “powerplant”, and “adverse weather” are the key factors resulting in UAV accidents. In addition, “observer errors”, “other aircraft”, “non-powerplant” and “adverse weather” are more likely to incur severe “accidents”. Some potential proactive plans and measures are proposed to help prevent the occurrence of UAV accidents.

A novel method for analyzing the leakage and diffusion of hydrogen: First arrival time distribution

Hydrogen energy plays an important role in achieving the carbon peak and carbon neutrality goals. However, owing to its flammability and explosiveness, hydrogen leakage during transportation and use can easily cause severe consequences. Therefore, research on the safety of hydrogen leakage is particularly important. This study proposes a novel method for analyzing the diffusion behavior of hydrogen leakage called first arrival time distribution (FATD). The spatial distribution of the first arrival times of specific concentration values (such as explosion limit and alarm concentration) at each coordinate point in the calculation domain was obtained by processing the concentration field of hydrogen leakage and diffusion via computational fluid dynamics simulations. The hydrogen leakage and diffusion processes were simulated for multiple scenarios using Fluent software, and the method of obtaining the FATD based on the simulations was detailed. The study also analyzed the hydrogen leakage and diffusion processes as well as the effects of mesh quantity and sampling time interval on the FATD; the FATD was additionally applied in actual scenarios to obtain an alarm time distribution for guiding sensor deployment. The results show that the FATD intuitively displays the hydrogen diffusion and evolution processes via specific concentration values in the calculation domain while describing the detailed influences of obstacles on the hydrogen diffusion process. Moreover, the FATD can serve as an indicator of grid independence analysis. This provides a new method for understanding diffusion behavior, reference for accident risk and consequence assessments, and safety guidance for the utilization of hydrogen energy.

A Review on the 3D Cartographic and Spatiotemporal GIS Models for Safety of Accidents in Deep Underground Coal Mines

The aim of this study was the safety risk assessment of accidents that occur because of production dynamics in deep coal mining, one of the highly dangerous business class, with spatiotemporal GIS. In this study, accidents that occur at certain times during production in underground mining workplaces are discussed based on evaluations of spatial and temporal dimensions. Accordingly, various analyses have been handled using criminology science tactics and strategies such as 80–20, kernel density, space–time cube, and hot spot, which have been adapted to GIS tools. Implementations have been realized on 3D underground cartographic models covering accident data from deep coal mine workplaces in the Zonguldak-Kozlu basin in Turkey between 2019 and 2021, at an average elevation of – 400 m. Space–time cube, kernel density, and time series analyses suggest that the Annual Bonus Incentives, which are repeated every year and given before the new year, increase the psychosocial risks for employees. It has been determined that this situation can lead to problems such as poor concentration among miners and loss of focus in some shifts. Additionally, significant relationships were found between the number of accidents and monthly production progress rates as a temporal parameter. The results demonstrate that applying a reactive approach with spatiotemporal GIS to accidents in deep coal mines is becoming increasingly important.

Predicting maritime accident risk using Automated Machine Learning

Machine learning (ML), particularly, Automated machine learning (AutoML) offers a range of possibilities for analysing large volumes of historical maritime accidents data with advanced algorithms for integrating predictive analytics in operational and policy decision-making for improving maritime safety. This study explores historical data of maritime accidents in Norwegian waters over 40 years. The data has been utilised for analysing five major maritime accident categories: grounding, contact damage, fire or explosion, collision, and heavy weather damage. A total of 29 classification ML algorithms were trained, and the Light Gradient Boosted Trees Classifier was found to be the best-performing with the highest predictive accuracy. The three most impactful factors for accident risk are the category of navigation waters, phase of operation, and gross tonnage of the vessel. Based on the feature effect results, vessels sailing in narrow coastal waters, in the along-the-way operational phase, and fishing vessels are highly vulnerable to grounding relative to other types of accidents. The results can be used as input for the entire procedure of risk analysis, from hazard identification to quantification of accident consequences, and the best-performing ML algorithm can be utilized in developing a decision support system for real-time maritime accident risk assessment.

К ВОПРОСУ О НЕОБХОДИМОСТИ РАЗРАБОТКИ МЕТОДИКИ ОЦЕНКИ РИСКА АВАРИЙ НА АВТОЗАПРАВОЧНЫХ КОМПЛЕКСАХ

The article justifies the danger of gas stations combining a gas station and buildings for servicing vehicles and their owners in the event of fires and explosions, including those on high-speed highways. The development of the most destructive accident in recent years at a gas station located in a complex with a car maintenance station is described. An analysis of accident risk assessment methods at fire hazardous facilities was carried out, on the basis of which it was concluded that it was necessary to develop a methodology for assessing accident risk at gas stations, taking into account the peculiarities of their development.

Fuzzy-Based Road Accident Risk Assessment

It is necessary to extensively investigate the causes of road accidents with the utmost precision to harness future technological advancements, such as autonomous driving and intelligent accident prevention systems. Nevertheless, since most accidents are attributed to simple human errors, unraveling the complex root-cause factors poses a considerable challenge. This is where fuzzy logic can offer a potential solution: it is essential to understand even seemingly straightforward errors, such as speeding, to identify external factors that could play a pivotal role in future accident prevention. A more in-depth examination and comprehension of elements like road curvature, slope, and their correlation with accidents are necessary. Additionally, it is crucial to explore how the frequency of accidents on specific road segments varies under diverse weather conditions. This article analyzes which curves can be considered more dangerous and the factors that render them risky. The fuzzy model presented in this article is primarily capable of estimating the risk of a given road segment based on its curvature characteristics. The model results presented in the article indicate that sections of the road can become more risky due to multiple curves and curves with a radius of less than 80 m. The model assesses risk based on the physical characteristics of road segments, primarily the curvature radius, while, typically, other road risk assessment models rely on traffic volume and accident counts.

Risk assessment of construction safety accidents based on association rule mining and Bayesian network

Risk assessment of construction safety accidents based on association rule mining and Bayesian network

Risk assessment of water inrush accident during tunnel construction based on FAHP-I-TOPSIS

This study introduces an innovative risk assessment method, namely FAHP-I-TOPSIS, for evaluating water inrush risks during tunnel construction. Its primary objective is to address the challenges posed by various fuzzy and uncertain factors in risk assessment, aiming to enhance the reliability of on-site risk evaluation. The method combines the Fuzzy Analytic Hierarchy Process for weight calculation and the Ideal Solution Similarity Ranking Preference Interval Technique for risk interval computation. Firstly, a comprehensive risk assessment framework is established, categorizing factors influencing tunnel water inrush into geological factors (Ni), construction technology and management (Mi), and the inherent vulnerability of the tunnel itself (Vi). These three indicator layers are further subdivided based on corresponding risk factors. The method is applied to the risk analysis of water inrush in the Shijingshan Tunnel in Zhuhai, analysing the risk weights of 16 influencing factors. It is observed that cracks and fault zones have the most significant impact on the occurrence of the incident. Through a comparison with traditional TOPSIS and VIKOR methods, which calculate a single risk value, the study demonstrates that FAHP-I-TOPSIS effectively integrates various fuzzy and uncertain factors, providing a reasonable level of risk assessment. It enhances adaptability to on-site errors, significantly improving the robustness of water inrush risk assessment during tunnel construction and sustainable management.

Risk assessment based on fault tree analysis and analytic hierarchy process in rural areas of Thailand

Introduction The importance of understanding the factors contributing to road accidents at schools cannot be overstated. This study aims to determine the risk of accidents in situations that could lead to accidents near schools. Methods A total of 10 kindergarten to junior high schools were selected for the study. The research used the Haddon Matrix to classify factors at risk of accidents, risk assessment through fault tree analysis, and the analytic hierarchy process (FTA-AHP) techniques. Fourteen sub-criteria were defined for risk identification, risk probability analysis, and risk assessment of the 10 roads. The likelihood of each event was analyzed using the AHP technique for all schools with an expert choice program. RI (random index) was calculated, and CR (consistency ratio) < 0.10 was considered satisfactory. Results The possibility of human accidents ranked highest in three areas: 1) Risk perception in SC 01, 03, and 02, with probabilities of 69.30%, 61.90%, and 57.4%, respectively. 2) The likelihood of accidents from vehicles/equipment, with the highest probabilities in a) Handling (SC01) at 64.70%, b) Braking (SC07) at 61.90%, and c) Lighting (SC03) at 57.80%. 3) The likelihood of accidents from the environment, with the highest probabilities in 1) driving at excessive speeds in areas SC01, 06, 03, and 09, which were 43.60%, 40.90%, and 40.00%, respectively. Conclusions The impacts of all three main factors were as follows: a) humans had the highest impact in the SC01 area (77.90%), b) vehicles/equipment had the highest impact in the SC01 area (75.90%), and c) the highest environmental impact in the SC01 area was 69.90%. The accident risk assessment revealed the highest risk score in three areas: 1) human risk perception, 2) environment with driving at excessive speeds, and 3) vehicle/equipment, including lighting, braking, and handling.