Property Damage Research Articles

Contextualizing Representations of Violence in #Blacklivesmatter Coverage with Inductive Computational Content Analysis

ABSTRACT News coverage of social movements traditionally adheres to the protest paradigm, which focuses on violent and confrontational tactics instigated by activists. Recent work has called for a reevaluation of this theory through the lens of partisan media bias. We demonstrate how partisan media bias systematically shapes coverage of the summer 2020 Black Lives Matter protests that occurred in the aftermath of the police murder of George Floyd. Using an inductive computational content analysis, we surface words and phrases that best predict whether an article was published in a left-, center-, or right-leaning outlet. We then perform a qualitative close reading of those terms in context to determine how partisan media bias shapes depictions of the same events: namely, in representations of protestors and violence. Right-leaning media depict protestors as violent rioters, looters, and antifa members committing property damage, whereas center- and left-leaning media emphasize the peaceful nature of the protests motivated by the police killing of Floyd and systematic racism. We consider the implications of these findings on broader public opinion within media ecosystems.

Fault detection using machine learning based dynamic ICA-distributed CCA: Application to industrial chemical process

Unexpected accidents and events in industrial chemical processes have resulted in a considerable number of causalities and property damage. Safety process management in industrial chemical processes is critical to avoid and ensure casualties and property damage. However, due to the immense scope and high complexity of current industrial chemical processes, the traditional safety process management approaches cannot address these challenges to attain adequate fault detection accuracy. To address this issue, an innovative machine learning-based distributed canonical correlation analysis-dynamic independent component analysis (DICA-DCCA) approach is needed to improve the fault detection effectiveness of complicated systems. The (DICA-DCCA) model could potentially detect anomalies and faults in industrial chemical data by utilizing three essential statistics:Id2,Ie2and squared prediction error (SPE). The practical effectiveness of the proposed frameworks is evaluated and compared using a continuous stirred tank reactor (CSTR) framework as a standard benchmark study. The research findings present that the suggested (DICA-DCCA) approach is more resilient and effective in detecting abnormalities and faults than the ICA and DICA approaches with FDR 100 % and FAR 0 %. The implied research approach is robust, operational, and productive.

Assessment of the Susceptibility of Urban Flooding Using GIS with an Analytical Hierarchy Process in Hanoi, Vietnam

The incidence of floods is rapidly increasing globally, causing significant property damage and human losses. Moreover, Vietnam ranks as one of the top five countries most severely affected by climate change, with 1/3 of residents facing flood risks. This study presents a model to identify flood susceptibility using the analytic hierarchy process (AHP) in the GIS environment for Hanoi, Vietnam. Nine flood-conditioning factors were selected and used as initial data. The AHP analysis was utilized to determine the priority levels of these factors concerning flood susceptibility and to assess the consistency of the obtained results to develop a flood-susceptibility map. The performance of the model was found to be significant based on the AUC value for the obtained receiver operating characteristic (ROC) curve. The flood-susceptibility map has five levels of flood susceptibility: the area with a very high susceptibility to flooding accounts for less than 1% of the map, high- susceptibility areas for nearly 11%, moderate-susceptibility areas for more than 65%, low- susceptibility areas for about 22%, and very low-susceptibility areas for 2%. Most of Hanoi has a moderate level of flood susceptibility, which is expected to increase with urban expansion due to the impacts of urbanization. Our findings will be valuable for future research involving urban planners, and disaster management authorities and will enable them to make informed decisions aimed at reducing the impact of urban flooding and enhancing the resilience of urban communities.

Thermal and calorimetric investigations of some vegetative fuels

AbstractBushfires pose a significant threat to numerous countries, often causing vast property damages and loss of lives. Efforts to combat and manage these fires heavily rely on predicting the fires' rate of spread and intensity. A significant component of these predictions involves understanding the thermophysical characteristics of vegetative fuels. The accuracy of predictive models (especially physical models) also depends on obtaining precise thermophysical and combustion parameters. This research aims to provide a comprehensive set of thermal degradation and combustion parameters for surface and near‐surface fuel samples collected during prescribed fire experiment conducted in April 2022 in Little Desert National Park, Victoria, Australia. Firstly, fuel properties like fuel height, moisture content, bulk density, fuel load and heat of combustion were meticulously characterized for both surface and near‐surface samples. Then activation energies for degradation reactions were determined using the Flynn–Wall–Ozawa method and for the determination of pre‐exponential factors, in most cases these reactions closely aligned with a Second order model. This was followed by determination of other parameters such as heat of reaction, specific heat and conductivity. It was found that the density, activation energy and heat of combustion did not vary significantly across the six samples under question. The comprehensive set of obtained parameters will likely help to facilitate better predictions in fire propagation modelling.

Mid-century climate change impacts on tornado-producing tropical cyclones

Tornadoes are a co-occurring extreme that can be produced by landfalling tropical cyclones (TCs). These tornadoes can exacerbate the loss of life and property damage caused by the TC from which they were spawned. It is uncertain how the severe weather environments of landfalling TCs may change in a future climate and how this could impact tornado activity from TCs. In this study, we investigated four TCs that made landfall in the U.S. and produced large tornado outbreaks. We performed four-member ensembles of convective-allowing (4-km resolution) regional climate model simulations representing each TC in the historical climate and a mid-twenty-first century future climate. To identify potentially tornadic storms, or TC-tornado (TCT) surrogates, we used thresholds for three-hourly maximum updraft helicity and radar reflectivity, as tornadoes are not resolved in the model. We found that the ensemble-mean number of TCT-surrogates increased substantially (56–299%) in the future, supported by increases in most-unstable convective available potential energy, surface-to-700-hPa bulk wind shear, and 0–1-km storm-relative helicity in the tornado-producing region of the TCs. On the other hand, future changes in most-unstable convective inhibition had minimal influence on future TCT-surrogates. This provides robust evidence that tornado activity from TCs may increase in the future. Furthermore, TCT-surrogate frequency between 00Z and 09Z increased for three of the four cases, suggesting enhanced tornado activity at night, when people are asleep and more likely to miss warnings. All of these factors indicate that TC-tornadoes may become more frequent and a greater hazard in the future, compounding impacts from future increases in TC winds and precipitation.

Cyclone Intensity Estimation Using Deep Learning

Cyclone Intensity Estimation based on Deep Learning focuses on cyclone intensity estimation, enhancing early warning systems and empowering decision-makers to take proactive measures to safeguard vulnerable communities and infrastructure, which aims to minimize the devastating impacts of cyclones worldwide. The challenge in cyclone intensity prediction persists due to the complex and dynamic nature of these storms. Traditional methods fall short of capturing rapid changes. This research project leverages the power of deep learning to enhance the accuracy of cyclone intensity prediction by utilizing both satellite images and grayscale representations as input datasets. It involves preprocessing and feature extraction from satellite images captured during cyclonic events. Convolutional Neural Networks (CNN) are employed to automatically learn and extract relevant patterns from these complex datasets. In parallel, grayscale images derived from the original satellite images are utilized to capture essential structure information that contributes to cyclone intensity. The fusion of information from both modalities is achieved through a novel deep learning patterns that go beyond the capabilities of traditional intensity estimation methods, and architecture, fostering a comprehensive understanding of cyclonic patterns that go beyond the capabilities of traditional intensity estimation methods. A new approach seeks to automate cyclone estimation, streamlining timelines and increasing efficiency by merging deep learning with hurricane-focused satellite data. This approach might result in more precise forecasts, Reducing the amount of casualties and property damage in prone areas.

Harm from the drinking of people you know: A range of effects from different relationships.

To describe the range of effects experienced due to the drinking of people respondents know and analyze risk and protective factors for harm from the drinking of partners and household members, other relatives and friends and co-workers. Surveys of 2574 participants' experiences were obtained from two samples: 1000 people responded to random digitally dialled Australian mobile calls and 1574 participants responded from the Life in AustraliaTMpanel survey. Respondents were asked whether they had been negatively affected in the previous 12 months by the drinking of persons they knew who were 'a heavy drinker or drank a lot sometimes' and the nature of these harms. Weighted logistic regressions were used to analyze differences in rates of key negative outcomes from known others' drinking by gender, age and socio-economic status. Almost two thirds [60.2%; 95% confidence interval (CI) = 57.7%-62.7%] of participants reported having heavy drinkers in their lives and 21.8% (95% CI = 19.8%-23.9%) reported being negatively affected by the drinking of people they knew well in some way. Participants reported a gamut of effects, including, most commonly, adverse social effects: having to transport relatives and friends who had been drinking, role failure and faults, being emotionally hurt or neglected, serious arguments, family problems, having to care for drinkers and verbal abuse. Less commonly, respondents reported physical or sexual harm, property damage, financial stress and threats from others' drinking. Women (odds ratio = 1.49; 95% CI = 1.13-1.95), younger people, rural, Australian-born (vs. respondents born overseas in non-English speaking countries) and more frequent drinkers were more likely to report harm from a drinker they knew than their counterparts after adjusting for other variables in the model. Australians appear to be commonly adversely affected by the drinking of people they know. Harms from known drinkers are more likely to be experienced by women than men, particularly from the people they live with and other relatives.

Improving the system of protection of the rights of buyers of vehicles

The subject of the study is the existing system of protection of the rights of buyers of vehicles, which includes the prevention of offenses involving liability for misleading consumers about consumer properties or quality of goods (Article 14.7 of the Administrative Code of the Russian Federation) and for causing property damage by deception or abuse of trust (Article 7.27.1. of the Administrative Code of the Russian Federation), the commission of which involves, including for example, with the correction or reduction of the recorded mileage of the car. The connection of these actions with the technical condition of vehicles and road safety is considered, since any unauthorized actions with the technical condition of the vehicle are associated with a decrease in the level of safety of road users. The prevalence of this phenomenon requires an adequate response from the state's law enforcement system and the establishment of legislative barriers for this activity. The research methodology is based on general scientific (logical, systematic, analysis, generalization) and private scientific (comparative, formal legal) methods. The scientific novelty of the study lies in an integrated approach to studying the existing procedure for protecting the rights of buyers of vehicles at risk of being misled about the quality of the goods. The article substantiates the need to create a legal mechanism at the state level and on its basis an open mileage accounting system for used cars for citizens, which allows them to track their history from the moment they go on sale. It is proposed to give a legal assessment of the actions and establish liability for persons who, through the provision of services for correcting car mileage data, contribute to deceiving consumers, causing property damage to buyers, and are in direct connection with the decisions of vehicle owners on their maintenance and repair, which affects road safety.

Investigating the contributing factors of crashes on interstate bridges in Louisiana using latent class clustering and association rule mining

Drivers on long bridges on interstates often face unique challenges, such as restricted lane width, inadequate shoulders, and a lack of clear zones for safe recoveries. Studies on understanding the factors that contribute to crash severity on such high-risk sections of interstates are limited. This research study applies latent class clustering (LCC) to detect homogeneous clusters while accounting for unobserved heterogeneity in a dataset of 10,036 crashes that occurred over a 6-year period (2015–2020) on eight selected bridges. Utilizing the LCC method, the research identifies four optimal clusters in bridge crashes, characterized by attributes such as ‘4-lane,’ ‘6-lane,’ ‘single-vehicle crashes,’ and ‘unknown driver.’ The Association Rule Mining (ARM) approach is used to identify the important collective factors to visible injury (KAB – fatal, severe and moderate) and property damage only (PDO or no injury). In Cluster 1 (4-lane), KAB and PDO crashes differ in collision type and visibility conditions, with rear-end crashes linked to KAB and sideswipe crashes to PDO. Cluster 2 (6-lane) shows similar distinctions but lacks specific lighting associations for PDO. In Cluster 3 (single-vehicle crashes), KAB involves moderate traffic and low visibility, while PDO has lower speed limits and non-dry surfaces. Cluster 4 (unknown driver), despite overrepresenting hit-and-run cases, underscores challenges in injury crash data collection in high-volume mobility scenarios. The discussions of the findings on the severity factors in this study are expected to help traffic safety engineers, policymakers, and planners to identify effective safety countermeasures on major elevated sections.

Comprehensive investigation of crashes associated with citywide speed limit reduction in Seattle, Washington

Road traffic crashes are a leading cause of fatalities and injuries worldwide, prompting authorities to implement various strategies to enhance road safety. One common approach involves lowering speed limits in an effort to mitigate the severity of crashes and improve overall traffic safety. Seattle, Washington has been lowering its citywide default speed limits on municipal roads since 2016, among several cities in the US. This research paper presents a comprehensive investigation of the impact of these speed limit reductions on various safety indicators, such as the crash frequency, crash severity, crash characteristics, crash type, crash location, contributing factors, and overall traffic safety performance. The study employs a comprehensive dataset encompassing several years before and after the implementation of lower speed limits. Data analyses indicate varying changes in crash severity distribution between before and after periods in Seattle. Fatal and incapacitating crashes in Seattle decreased by 19% after the initial speed limit lowering in 2016 and further decreased by 3% after the second speed limit lowering in 2018. Pedestrian-involved crashes experienced a 26% significant decrease while speeding-related had a 16% decrease after the city implemented lowering speed limits. There was also a statistically significant 4% decrease in property damage only collisions and a 7% decrease in parked car crashes. Distracted driving / inattention-related crashes also decreased by 4%. The Mid-Block (but intersection related) crashes saw a significant decrease by 6%, and Driveway Junction related crashes decreased by 1%. The findings provide insights into the effectiveness of speed limit reductions in enhancing road safety in urban environments. The findings contribute to the existing body of knowledge on speed management strategies and their impact on traffic safety. The outcomes of this study can inform transportation policymakers and practitioners in Seattle and other urban areas as they make informed decisions regarding speed limit regulations and road safety measures.

An Approach of Multi Criteria Decision Making in Analyzing Rural Road Safety in Nashik District

Abstract: Rural road safety evaluations play a significant role in the transportation industry, and most fatal accidents happen on these types of roads. Road Safety Analysis (RSA), a process to evaluate a road's safety criteria and support overall road management decision-making, ensures the avoidance of loss of life and property damage. The "Pradhan Mantri Gram Sadak Yojana," which is primarily utilized in rural areas, is the main emphasis of this. In order to identify the most vulnerable roadways and implement mitigation measures, roads must be prioritized depending on safety ratings. This document presents the results of a study conducted to rank the safety criteria of Pradhan Mantri Gram Sadak Yojana (PMGSY) roads, a particular category of rural roads. A questionnaire survey was used for the study's analysis, and three MCDM methods were used to examine the relationships and interactions between the various factors. The study's findings indicated that the based on input from PMGSY engineers and contractors. From the perspective of the engineers in the Nashik area, we can determine the safety impact factor that influences the safety assessment of the PMGSY route.

영향한계강우량을 고려한 삼척시 호우특보 기준에 관한 연구

Heavy rain has a short duration of occurrence, causing many casualties and property damage. However, the heavy rain warning criteria are a single warning criterion that does not reflect local characteristics. Therefore, this study aims to study and propose heavy rain warning criteria for Samcheok City that consider the threshold rainfall to supplement the uniform decision-making system for heavy rain disasters. By analyzing the basic statistical values of the threshold rainfall grid for each advisory area in Samcheok City, the criteria for heavy rain warning for each advisory area were selected in four levels, and the applicability was examined by applying it to past rainfall events. As a result, this study visibly forecasted the expected risk level that changes depending on the rainfall event by advisory area, and categorized the advisories and warnings of heavy rain warning from the KMA were categorized into Warning and Danger levels. The results of this study are expected to be useful in securing more golden time for responding to heavy rain disasters and preparing efficient measures against heavy rain disasters.

Evaluation of Shear-Punched Surface Layer Damage in Three Types of High-Strength TRIP-Aided Steel

The damage properties in the shear-punched surface layer, such as the strain-hardening increment, strain-induced martensite fraction, and initiated micro-crack/void characteristics at the shear and break sections, were experimentally evaluated to relate to the stretch-flangeability in three types of low-carbon high-strength TRIP-aided steel with different matrix structures. In addition, the surface layer damage properties were related to the mean normal stress developed on shear-punching and microstructural properties. The shear-punched surface damage of these steels was experimentally confirmed to be produced under the mean normal stress of negative to 0 MPa. TRIP-aided bainitic ferrite (TBF) steel had the smallest surface layer damage, featuring a significantly suppressed micro-crack/void initiation. This was due to the fine bainitic ferrite lath matrix structure, a low strength ratio of the second phase to the matrix structure, and the high mechanical stability of the retained austenite. On the other hand, the surface layer damage of TRIP-aided annealed martensite (TAM) steel was suppressed next to TBF steel and was smaller than that of TRIP-aided polygonal ferrite (TPF) steel. The surface layer damage was also characterized by a large plastic strain, a large amount of strain-induced martensite transformation, and a relatively suppressed micro-crack/void formation, which resulted from an annealed martensite matrix and a large quantity of retained austenite. The excellent stretch-flangeability of TBF steel might be caused by the suppressed micro-crack/void formation and high crack propagation/void connection resistance. The next high stretch-flangeability of TAM steel was associated with a small-sized micro-crack/void initiation and high crack growth/void connection resistance.

A comparative analysis of risk factors in taxi-related crashes using XGBoost and SHAP

Taxis play a crucial role in urban public transportation, but the traffic safety situation of taxi drivers is far from optimistic, especially considering the introduction of ride-hailing services into the taxi industry. This study conducted a comparative analysis of risk factors in crashes between traditional taxi drivers and ride-hailing taxi drivers in China, including their demographic characteristics, working conditions, and risky driving behaviors. The data was collected from 2,039 traditional taxi drivers and 2,182 ride-hailing taxi drivers via self-reported questionnaires. Four XGBoost models were established, taking into account different types of taxi drivers and crash types. All models showed acceptable performance, and SHAP explainer was used to analyze the model results. The results showed that for both taxi drivers, risk factors related to risky driving behaviors are more important in predicting property damage (PD) crashes, while risk factors related to working conditions are more important in predicting person injury (PI) crashes. However, the relative importance of each risk factor varied depending on the type of crashes and the type of taxi drivers involved. Furthermore, the results also validated certain interactions among the risk factors, indicating that the combination of certain factors generated a greater impact on crashes compared to individual factors alone. These findings can provide valuable insights for formulating appropriate measures to enhance road safety for taxi driver.

Fusion of satellite and street view data for urban traffic accident hotspot identification

As the number of vehicles and the volume of traffic swell in urban centers, cities have experienced a concomitant increase in traffic accidents. Proactively identifying accident-prone hotspots in urban environments holds the promise of preventing traffic mishaps, thereby curtailing the incidence of accidents and reducing property damage. This research introduces the Two-Branch Contextual Feature-Guided Converged Network (TCFGC-Net) utilizing multimodal satellite and street view data. Designed to extract global structural features from satellite imagery and dynamic continuous features from street view imagery, the model aims to improve the accuracy of detecting urban accident hotspots. For the satellite imagery branch, we propose the Contextual Feature Coupled Convolutional Neural Network (Trans-CFCCNN) designed to extract global spatial features and discern feature correlations across adjacent regions. For the street view imagery branch, we develop the Sequential Feature Recurrent Attention Network (SFRAN) to assimilate and integrate dynamic scene features captured from successive street view images. We designed the Multi-Branch Feature Adaptive Fusion Structure (MBFAF) to aggregate different branch features for accurate identification of accident hotspots. Experimental results show that the model performs well, with an overall accuracy of 93.7 %. Ablation studies confirm that relative to standalone street view and satellite branch analyses, implementing multimodal fusion enhances the model's accuracy by 12.05 % and 17.86 %, respectively. The innovative fusion structure proposed herein garners a 4.22 % increase in model accuracy, outpacing conventional feature concatenation techniques. Furthermore, the model outperforms existing deep learning models in terms of overall efficacy. Additionally, to showcase the efficacy of the proposed model structure, we utilize Class Activation Maps (CAM) to provide visual interpretability for the model. These results suggest that the dual-branch fusion model effectively decreases false alarm occurrences and directs the model's focus toward regions more pertinent to accident hotspots. Finally, the code and model used for identifying hotspots of urban traffic accidents in this study are available for access: https://github.com/gwt-ZJU/TCFGC-Net.

Analisis Faktor Penyebab Banjir Di Kecamatan Jombang Dengan Model Sistem Dinamik

Cilegon City is an area that has a high level of vulnerability to various disasters, including natural disasters. Natural disasters are events that are natural and have the potential to cause damage and destruction to the environment, with impacts in the form of loss of life, material loss, and damage to infrastructure that has been developed over some time. Based on the Cilegon City Regional Disaster Management Agency (BPBD), flooding is a frequent natural disaster in Cilegon City. This research aims to identify the factors that cause flooding in the Jombang Subdistrict area. The research stages are observation, data collection, data processing with the causal loop diagram approach, data validation, data analysis, and conclusion. The results of the analysis show that the main factors causing flooding in Jombang Subdistrict are the intensity of rainfall, small drainage, and the small number of absorption wells. The consequences caused by floods are property damage, the spread of disease, economic losses, environmental damage, infrastructure disruption, and social disruption. Therefore, the solution to flooding in Jombang Subdistrict is to expand drainage, add the number of absorption wells, provide education to the community about environmentally friendly behavior, and green the environment. Keywords: Cilegon, Causal Loop Diagram, Dynamic System, Flooding

Impacts of Gorkha Earthquakes: A Comprehensive Study

Nepal is highly vulnerable to seismic activity. This review seeks to explore the impact of earthquakes on human populations in terms of casualties and injuries, as well as to pinpoint the risk factors associated with these outcomes. An earthquake measuring 7.8 on the Richter scale struck near Gorkha, Nepal, on April 25, 2015, resulting in the loss of nearly nine thousand lives, extensive property damage, and the destruction of numerous historical and cultural landmarks. Data sourced from the UNDRR, Ministry of Home Affairs, and the United States Geological Survey were analyzed using the ArcGIS software tool. The study offers a comprehensive examination of earthquake-prone regions at regional, national, and global levels. By analyzing recent significant earthquakes along with their associated losses and damages, it underscores the importance of precautionary measures and risk analysis in reducing vulnerability and enhancing preparedness among researchers and stakeholders.

Car Accident Detection and Alert System

Due to the high death and property damage rates from traffic accidents, road safety is a serious problem. This study suggests a novel traffic accident detection and warning system in response to this difficulty, with the goals of enhancing prompt response and minimizing damage. This system is made up of many different parts, such as an energy sensor, accelerometers, Arduino microcontrollers, GPS modules, and GSM modules. Combining these characteristics allows the system to differentiate between minor and severe occurrences, recognize and categorize collisions according to impact severity, and detect overturned vehicles. The system notifies designated projects in the precise accident zone of any serious accidents detected. The process includes developing algorithms, designing hardware, and implementing the system. Thorough testing is then done to assess the system's dependability and performance. The outcomes demonstrate how well the suggested approach works for precise accident detection and quick emergency response. This research uses technology to lessen the effects of traffic accidents and save lives, which advances road safety legislation.

Climate Change and Cholera: A Review Exploring the Association Between Storm Severity and Global Human Vibrio spp. Incidence.

Current global meteorological evidence suggests that climate change is a direct contributor to the increased severity of weather events, such as hurricanes and typhoons, putting billions of people at increased risk of physical harm, property damage, and destructive flooding. When flood water becomes stagnant, communities are more susceptible to a multitude of infectious diseases, ranging from mosquito-borne parasites to bacterial infections. In the case of Vibrio spp., a bacterial pathogen class with an infamous history of waterborne illness, recent data has demonstrated the ability for salt water, fresh water, and brackish water replication, creating ideal conditions for spread after a tropical cyclone. Vibrio spp. infections, specifically species V. vulnificus, V. parahaemolyticus, and V. cholerae, have shown significant upticks in the number of global cases, with some species increasing in incidence by more than 8-fold in the last 40 years. These outbreaks are contained mainly in areas susceptible to large-scale storm events and primarily occur during or immediately following the conclusion of local storm seasons, suggesting a potential correlation between the presence of tropical weather events and increases in the number of human Vibrio infections. In this review, we present a compilation of sources from tropical cyclone-susceptible areas, including the Americas, Asia, Africa, and Oceania, containing evidence demonstrating that recent outbreaks of Vibrio spp. are correlated with an increase in the severity of storms. Many of the included studies used modeling to forecast future trends of vibriosis and, therefore, directly link the effects of climate change with detrimental impacts to human global health.

Predicting types of human-related maritime accidents with explanations using selective ensemble learning and SHAP method

Maritime accidents frequently lead to severe property damage and casualties, and an accurate and reliable risk prediction model is necessary to help maritime stakeholders assess the current risk situation. Therefore, the present study proposes a hybrid methodology to develop an explainable prediction model for maritime accident types. Based on the advantages of selective ensemble learning method, this study pioneers to introduce a two-stage model selection method, aiming to enhance the predictive accuracy and stability of the model. Then, SHAP (Shapley Additive Explanations) method is integrated to identify effective mapping associations of seafarers’ unsafe acts and their risk factors with the prediction results. The results demonstrate that the model developed achieves good prediction performance with an accuracy of 87.50 % and an F1-score of 84.98 %, which benefits stakeholders in assessing the type of maritime accident in advance, so as to make proactive intervention measures.