Evaluation Indicators Research Articles

Traffic Flow Prediction: A Method Using Bagging-Based Ensemble Learning Model

For the development of the national economy, transportation is strategically significant. As the increasing ownership of automobiles, traffic jams are a common occurrence. Accurate prediction of traffic flow contributes to diverting traffic effectively and improving the quality of urban traffic, in turn improving the operation of the overall transportation system. The rapid development of artificial intelligence technologies, especially machine learning and deep learning, has provided effective methods for accurate prediction of traffic flow. Based on the above, in order to improve the accuracy of the prediction and to extend the application of machine learning and deep learning in the prediction of traffic flow, this study proposed a bagging-based ensemble learning model. Firstly, normalization method is used to preprocess the data. Subsequently, base prediction models including decision tree, random forest, logistic regression, convolution neural network, long short-term memory and multilayer perceptron are selected for training the prediction model, respectively. Finally, bagging-based ensemble learning method is used to integrate these base prediction models to further predict traffic flow. The results of comparison between the single base prediction models and the bagging-based ensemble learning model on the five evaluation indicators show that, for predicting the traffic flow, the bagging-based ensemble learning model outperforms the base prediction models. Meanwhile, this study explores the potential in the application of machine learning, deep learning, and especially bagging-based ensemble learning to predict traffic flow.

Performance Assessment of Health Visitor and Family Medicine Units in Babylon's Main Health Care Centers

Background: The Ministry of Health (MoH) has developed a standardized package of basic health services for all primary healthcare (PHC) facilities to enhance service quality. This study evaluates service availability and facility needs in the family health units in Babylon. Methods: This cross-sectional study was conducted in three family healthcare sectors (the first Al-Hillah sector, Al-Musayyib sector, and Al-Hashemiyyah sector), which includes 6 out of 23 primary healthcare centers in the Babylon Governorate. The data collection started on March 3, 2022, and ended on August 27, 2022, and was randomly chosen from all sectors. Results: According to this study, the family health unit had a poor mean score of 48.03±17.97. Similarly, the health visitor indicators had an inadequate level of assessment, with a mean score of 46.87±14.09. In conclusion, this revealed a study that the primary health care centers evaluated had poor scores in all the final evaluation indicators for the family health unit and the health visitor unit.

Safety-risk assessment system for prefabricated building construction in China

PurposePrefabricated building (PB) uses factory production and onsite assembly, which differs from traditional construction methods. This special construction approach may lead to dissimilar safety risks and challenges. Traditional safety assessment methods may not adequately and accurately assess the safety risks of PB construction. This paper aims to develop a new concept and methodology for targeted improvement in assessing PB safety risks.Design/methodology/approachRisk factors and indicators were established based on literature review and expert inputs. A structural equation model (SEM) was developed to investigate the relationships among three main risk categories: objects, workers and management. SEM analyzed the intricate associations between indicators and deepened understanding of safety risks. The model was tailored for China’s PB construction projects to enhance safety-risk management.FindingsThe cloud model evaluation validated the SEM model. A PB case study project tested and verified the model, evaluated its efficacy and quantified its safety performance and grade. We identified significant safety risk impacts across the three risk categories, safety-control level and specific areas that require improvement. The SEM model established a robust safety evaluation indicator system for comprehensive safety assessment of PB construction.Practical implicationsPractical recommendations provide valuable insights for decision-makers to enhance construction efficiency without compromising safety. This study contributed to the conceptual foundation and devised a novel method for evaluating safety performance in PB construction for safer and more efficient practices.Originality/valueThis study departed from the traditional method of calculating weights, opting instead for the SEM method to determine the weights of individual risk indicators. Additionally, we leveraged the cloud model to mitigate the influence of subjective factors in analyzing questionnaire survey responses. The feasibility and reliability of our proposed method were rigorously tested and verified by applying it to the PB case.

The Identification and Quantification of Hidden Hazards in Small Scale Reservoir Engineering Based on Deep Learning: Intelligent Perception for Safety of Small Reservoir Projects in Jiangxi Province

This study aims to enhance the detection and assessment of safety hazards in small-scale reservoir engineering using advanced image processing and deep learning techniques. Given the critical importance of small reservoirs in flood management, water supply, and ecological balance, the effective monitoring of their structural integrity is crucial. This paper developed a fully convolutional semantic segmentation method for hidden danger images of small reservoirs using an encoding–decoding structure, utilizing a deep learning framework of convolutional neural networks (CNNs) to process and analyze high-resolution images captured by unmanned aerial vehicles (UAVs). The method incorporated data augmentation and adaptive learning techniques to improve model accuracy under diverse environmental conditions. Finally, the quantification data of hidden dangers (length, width, area, etc.) were obtained by converting the image pixels to the actual size. Results demonstrate significant improvements in detecting structural deficiencies, such as cracks and seepage areas, with increased precision and recall rates compared to conventional methods, and the HHSN-25 network (Hidden Hazard Segmentation Network with 25 layers) proposed in this paper outperforms other methods. The main evaluation indicator, mIoU of HHSN-25, is higher than other methods, reaching 87.00%, and the Unet is 85.50%, and the Unet++ is 85.55%. The proposed model achieves reliable real-time performance, allowing for early warning and effective management of potential risks. This study contributes to the development of more efficient monitoring systems for small-scale reservoirs, enhancing their safety and operational sustainability.

The Optimal Drought Hardening Intensity and Salinity Level Combination for Tomato (Solanum lycopersicum L.) Cultivation under High-Yield, High-Quality and Water-Saving Multi-Objective Demands.

The extreme weather and the deteriorating water environment have exacerbated the crisis of freshwater resource insufficiency. Many studies have shown that salty water could replace freshwater to partly meet the water demand of plants. To study the effects of early-stage drought hardening and late-stage salt stress on tomatoes (Solanum lycopersicum L.), we conducted a 2-year pot experiment. Based on the multi-objective demands of high yield, high quality, and water saving, yield indicators, quality indicators, and a water-saving indicator were selected as evaluation indicators. Three irrigation levels (W1: 85% field capacity (FC), W2: 70% FC, W3: 55% FC) and three salinity levels (S2: 2 g/L, S4: 4 g/L, S6: 6 g/L) were set as nine treatments. In addition, a control treatment (CK: W1, 0 g/L) was added. Each treatment was evaluated and scored by principal component analysis. The results for 2022 and 2023 found the highest scores for CK, W2S2, W3S2 and CK, W2S4, W2S2, respectively. Based on response surface methodology, we constructed composite models of multi-objective demands, whose results indicated that 66-72% FC and 2 g/L salinity were considered the appropriate water-salt combinations for practical production. This paper will be beneficial for maintaining high yield and high quality in tomato production using salty water irrigation.

Evaluation of Urban Public Building Renovation Potential Based on Combination Weight Cloud Model—Case Study in China

Currently, urban renovation activities in China are booming. And promoting the renovation of public buildings is a key feature of urban renovation due to its large scale, high cost, and significant impact to the natural and social environment. To reduce the ambiguity and uncertainty in evaluating the potential for the renovation of existing public buildings, a renovation potential evaluation model integrating a game theory-based combination weighting method and cloud model theory is proposed. This paper constructs a comprehensive evaluation index system based on relevant standards and the literature. Game theory is used to optimize the weights obtained by AHP and entropy weight methods to obtain a combined weight. MATLAB programming is used to calculate the comprehensive cloud parameters of the evaluation index for the potential renovation of existing public buildings and therefore generate cloud Graphs. Through a case study in Nanjing, China, it was demonstrated that the combination weight cloud model can objectively reflect the relationship between the fuzziness and randomness of evaluation indicators for public building renovation potential. The visual expression of cloud Graphs can intuitively reflect the magnitude of renovation and renovation potential and the degree of uncertainty in evaluation results. The research result provides useful references for the sustainable utilization of building resources in the era of building.

Multi-objective human-robot collaborative disassembly line balancing considering components remanufacture demand and hazard characteristics

Disassembly enterprises face significant challenges in managing increasingly intensive workloads with a workforce alone. To address this, current study focuses on balancing a human-robot collaborative disassembly line. For the task operator allocation problem, a mapping constraint mechanism between the disassembly tasks and the task operators is established based on the remanufacture demand and hazard characteristics of components. A mixed-integer programming model is formulated, encompassing four optimization objectives: the number of workstations, the task operator idle balancing index, the demand index, and the hazard index. A novel modified teaching and learning optimization approach is proposed for the addressed human-robot collaborative disassembly line balancing problem. A two-layer encoding strategy is designed based on the solving characteristics of the problem and an embedded perturbation strategy based on randomly transforming the task operator is introduced to enhance the optimization performance of the proposed method. The validity of the mixed-integer programming model and the efficacy of the proposed algorithm are demonstrated through two small-scale human-robot collaborative disassembly case studies. The proposed algorithm is then applied to two real-life cases of human-robot collaborative disassembly lines with different scales. The results of the proposed optimization method are compared with other advanced algorithms, demonstrating its superior performance in both single-objective and multi-objective optimization based on multiple evaluation indicators.

Interpretability research of deep learning: A literature survey

Deep learning (DL) has been widely used in various fields. However, its black-box nature limits people's understanding and trust in its decision-making process. Therefore, it becomes crucial to research the DL interpretability, which can elucidate the model's decision-making processes and behaviors. This review provides an overview of the current status of interpretability research. First, the DL's typical models, principles, and applications are introduced. Then, the definition and significance of interpretability are clarified. Subsequently, some typical interpretability algorithms are introduced into four groups: active, passive, supplementary, and integrated explanations. After that, several evaluation indicators for interpretability are briefly described, and the relationship between interpretability and model performance is explored. Next, the specific applications of some interpretability methods/models in actual scenarios are introduced. Finally, the interpretability research challenges and future development directions are discussed.

A Framework for Monitoring the Effectiveness of Ecosystem-Based Adaptation Strategies Using Internet of Things and Machine Learning Techniques

Climate change poses a threat to the global ecosystem. Many countries adopt various approaches, including ecosystem-based adaptation (EbA), to address this problem. However, the assessment of the effectiveness of the EbA interventions is conducted manually, is resource-intensive, and is focused on short-term outputs. These limitations underscore a critical gap: the need for a comprehensive, automated system that enables long-term monitoring and predictive analysis. This study aimed to address this gap by developing an innovative framework that integrates Internet of Things (IoT) devices and machine learning (ML) algorithms to continuously monitor weather, hydrological, environmental, and other variables. We conducted a thorough analysis to design an appropriate framework. In addition, to obtain the relevant information and data, we conducted interviews with the local community and collected secondary data from various sources. The proposed framework consists of five layers: (i) EbA interventions; (ii) IoT-based key performance indicators (KPI) for monitoring and evaluation (M&E); (iii) primary data collection; (iv) data storage; and (v) application. As a proof of concept, we developed a system that supports early flood and drought alerts while simultaneously providing long-term evaluations of the effectiveness of EbA strategies. The developed system consists of IoT devices and a web application integrated with machine learning (ML). We set up and tested the IoT devices before deploying them in the study area. The devices capture data for two primary purposes: (1) short-term: flood detection and alerting, and (2) long-term: drought prediction and evaluation of EbA effectiveness through continuous data analysis. This research represents a significant advancement in the automation and long-term assessment of climate adaptation measures, offering a scalable and effective solution to disaster risk reduction.

Ontology-Based Landslide Knowledge Representation and Hazard Assessment Research

The occurrence of landslides is characterized by suddenness and unpredictability, with many complex relationships among various influencing factors. This study, by introducing the ontology of knowledge engineering, summarizes the characteristics of landslide knowledge and constructs a multidisciplinary, multi-level, and multi-perspective ontology of the landslide domain. Through in-depth analysis of existing landslide research systems and methods, the study defines the basic concepts, attributes, and their complex relationships and rules related to landslides. This research expresses landslide knowledge using the Web Ontology Language (OWL), enabling effective storage and retrieval of landslide knowledge, thus promoting the systematization and standardization of knowledge in the landslide domain. Furthermore, leveraging the high knowledge density feature of the constructed landslide domain ontology, this study extends traditional landslide evaluation and prediction methods by introducing more evaluation indicators based on the constructed landslide ontology. The comprehensive evaluation of landslide risk is conducted using the contribution-weighting method, providing new insights and tools for landslide risk management and decision-making.

Coupling Relationships and Driving Mechanisms of Water–Energy–Food in China from the Perspective of Supply and Demand Security

The rapid increase in population and economy, coupled with accelerated urbanization, is placing immense pressure on the water–energy–food (WEF) system. In this context, the water–energy–food nexus framework has emerged, recognizing the interdependencies and interactions among water, energy, and food systems, with the aim of optimizing resource management through cross-sectoral collaboration to promote sustainable development. Understanding the spatio-temporal differentiation patterns of the WEF nexus and elucidating the driving mechanisms behind changes in their coupling relationships is essential. This knowledge is crucial for ensuring the security of each subsystem and enhancing the overall sustainability of interconnected systems through coordinated efforts. To address these challenges, this study first established evaluation indicators for water, energy, and food security to quantify their levels and spatio-temporal dynamics. Subsequently, the degrees of coupling coordination within the WEF nexus were calculated. Finally, the WEF nexus’s spatial correlations were analyzed by using a spatial autocorrelation model. Spatial econometric models then identified key factors affecting its coordination. The results revealed significant spatial heterogeneity in water, energy, and food security across mainland China’s provinces. From 2002 to 2022, water security improved substantially in 87% of the provinces, while energy security began to improve in the eastern regions following a phase of high consumption. Food security saw significant enhancements, particularly in Inner Mongolia and the northeastern provinces. The overall coupling coordination of the WEF nexus improved across 30 provinces, progressing toward primary coordination. However, Henan and Anhui provinces experienced fluctuations in WEF nexus coordination. Spatial correlation analysis showed upward trends and increased clustering in WEF nexus coordination. Factors such as economic development and population positively influenced coordination, while economic agglomeration, education, and effective irrigation area had negative effects. This study elucidates the complex interconnections and key influencing factors within the WEF nexus, providing a reference framework and practical recommendations for equitable resource management.

UNet-based multi-organ segmentation in photon counting CT using virtual monoenergetic images.

Multi-organ segmentation aids in disease diagnosis, treatment, and radiotherapy. The recently emerged photon counting detector-based CT (PCCT) provides spectral information of the organs and the background tissue and may improve segmentationperformance. We propose UNet-based multi-organ segmentation in PCCT using virtual monoenergetic images (VMI) to exploit spectral informationeffectively. The proposed method consists of the following steps: Noise reduction in bin-wise images, image-based material decomposition, generating VMIs, and deep learning-based segmentation. VMIs are synthesized for various x-ray energies using basis images. The UNet-based networks (3D UNet, Swin UNETR) were used for segmentation, and dice similarity coefficients (DSC) and 3D visualization of the segmented result were evaluation indicators. We validated the proposed method for the liver, pancreas, and spleen segmentation using abdominal phantoms from 55 subjects for dual- and quad-energy bins. We compared it to the conventional PCCT-based segmentation, which uses only the (noise-reduced) bin-wise images. The experiments were conducted on two cases by adjusting the doselevels. The proposed method improved the training stability for most cases. With the proposed method, the average DSC for the three organs slightly increased from 0.933 to 0.95, and the standard deviation decreased from 0.066 to 0.047, for example, in the low dose case (using VMIs v.s. bin-wise images from dual-energy bins; 3D UNet). The proposed method using VMIs improves training stability for multi-organ segmentation in PCCT, particularly when the number of energy bins issmall.

Comprehensive evaluation of the implementation of close-type county medical alliance in Shandong Province using entropy weight TOPSIS method and non-integer rank sum ratio method

BackgroundTo improve the capacity of county medical and health services, China encourages all localities to pilot the close-type county medical alliance. In Shandong Province, medical alliances have been piloted in 47 counties, ranking first in the country. The objective of this study is to comprehensively evaluate the implementation of close-type county medical alliance in Shandong Province, identify the differences between different county regions, and analyze the reasons, so as to provide a reference for the construction of a new county medical and health service system with clear goals, powers and responsibilities, and division of labor.MethodsThe implementation of the close-type county medical alliance was comprehensively evaluated in 47 national pilot counties in Shandong Province using entropy weight TOPSIS method and non-integer rank sum ratio method. Variance analysis was used for comparison of the comprehensive evaluation results.ResultsThe weight coefficient of evaluation indicators was highest for information interconnection, at 18.06%, and lowest for orderly referral of patients, at 3.64%. There was no difference in results of the comprehensive evaluation of entropy weight TOPSIS method and non-integer rank sum ratio method. Comprehensively order the implementation status of each pilot county according to the relative paste progress, 13 counties Y5, Y11, Y14, Y16, Y20, Y25, Y26, Y27, Y28, Y32, Y33, Y40 and Y42 were ranked highest, whereas county Y37 was ranked lowest. Non-integer rank sum ratio method graded counties into three grades: excellent, good and average. Kruskal-Wallis nonparametric test showed that the difference between the grades was statistically significant (H\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\:\\:=\\:$$\\end{document}37.099, p\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\:<\\:$$\\end{document}0.001). Variance analysis based on comprehensive evaluation results showed that implementation status was not correlated with the county economic development level, the level of health resources input and the medical service ability of the lead hospital.ConclusionsOur findings indicated that the implementation of the close-type county medical alliance is significantly different between pilot counties, with a marked differentiation within the same urban area. Therefore, effective measures are recommended to reduce this gap, including promoting informatization empowerment of the county medical community, strengthening government responsibility and improving policy effectiveness.

Comparison of indicators to evaluate the performance of climate models

AbstractThe evaluation of climate models is a crucial step in climate studies. It consists of quantifying the resemblance of model outputs to reference data to identify models with superior capacity to replicate specific climate variables. Clearly, the choice of the evaluation indicator significantly impacts the results, underscoring the importance of selecting an indicator that properly captures the characteristics of a “good model”. This study examines the behaviour of six indicators, considering spatial correlation, distribution mean, variance and shape. Monthly data for precipitation, temperature and teleconnection patterns in Central America were utilized in the analysis. A new multicomponent measure was selected based on these criteria to assess the performance of 32 CMIP6 models in reproducing the annual seasonal cycle of these variables. The top six models were determined using multicriteria methods. It was found that even the best model reproduces one derived climatic variable poorly in this region. The proposed measure and selection method can contribute to enhancing the accuracy of climatological research based on climate models.

Ecosystem model of development of the unmanned aircraft systems industry

The article analyzes the structural transformations of Russian industry in the context of external constraints and the need to ensure technological sovereignty in high-tech industries. The scientific hypothesis of the work is the assumption of the importance of cross-sectoral coordination of subjects of industrial policy in order to develop the production of critical products. The digital industrial ecosystem of the unmanned aircraft systems (UAS) industry is considered as an object of research. The purpose of the research is to develop the foundations of the structure of the ecosystem model of the development of the UAS industry and substantiate the feasibility of its application in the interests of improving high-tech industries of the Russian Federation. The article offers the author’s interpretations of the concepts of “digital industrial ecosystem” and “cross-industry ecosystem”, substantiates the principles of functioning of the ecosystem model and its architecture. An ecosystem model of the development of the UAS industry was developed, including management bodies, technical infrastructure, organizational and economic tools, technological clusters and a network of participants. The main distinguishing features of the proposed model from the existing ones are: the distribution of ecosystem management bodies that ensure the autonomy of management at all horizontal and hierarchical levels of its organization; the ecosystem is considered as an instrument of the mechanism of strategic development of high-tech industries involved in the creation of civil-purpose UAS, and provides cross-industry interaction of participants in order to accelerate the development of industrial production and the formation of the UAS market. The ecosystem management model will ensure the possession of up-to-date data of actual and planned indicative evaluation indicators of the state and development of the industry on a time scale close to real, based on the principle of digital transparency.

A two-stage framework for site selection of deep-sea offshore wind-to-hydrogen projects based on GIS - MULTIMOORA method under Pythagorean triangular fuzzy environment

Constructing deep-sea offshore wind-to-hydrogen projects (DS-OWHP) is crucial to solve the problem of offshore electricity storage and consumption. Addressing the uncertainty of the decision-making environment, and the lack of a comprehensive indicator system, this paper proposes a two-stage decision framework based on Geographic Information System (GIS). In the first stage, GIS is used to analyze the suitability of study area and identify alternative sites. In the second stage, the fuzzy MCDM method is used to further optimize alternative sites. Firstly, a comprehensive evaluation indicator system is established, and the Pythagorean Triangular Fuzzy Number (PTFN) is used to describe evaluation information. Then, a combination of stepwise weight assessment ratio analysis (SWARA), entropy, and game theory is used to determine weights, and the Multi-Objective Optimization on the basis of a Ratio Analysis plus the Full Multipli-cative form (MULTIMOORA) method is used to rank alternative sites. Finally, the feasibility of framework is proved through a case study in Guangdong Province. The results show that average wind speed has the highest weight of 0.149, and alternative site A2 which is close to zhuahai is the optimal choice. Through sensitivity analysis and comparative analysis, the reliability and stability of proposed model is verified.

Research on mobile energy storage scheduling strategy for emergency power supply protection of post-disaster isolated loads

Aiming at the problem of insufficient power supply capacity of isolated loads in oceanic islands, a concept based on mobile energy storage and power conservation is proposed in this paper. Firstly, the energy flow and traffic flow characteristics of mobile energy storage are quantitatively analyzed, and an accurate mathematical model is established. Furthermore, the time-space distribution characteristics of load are carefully considered, and the Voronoi diagram and Tyson polygon method are introduced to determine cell differentiation. On this basis, combined with the power demand of load nodes and the energy storage characteristics of mobile energy storage vehicles, the evaluation indicators of cell energy and relative surplus energy of cells are designed, and the agent-cellular automata algorithm is used to schedule the mobile energy storage. The simulation results show that the power supply mode based on mobile energy storage can effectively improve the reliability of isolated loads. This paper provides a new perspective for solving the reliable power supply problem of isolated loads.

Evaluation of the Management Policy of BOS Funds in Elementary Schools

General background however, the effective management of these funds remains a significant challenge, particularly in certain regions. Specific background this study focuses on evaluating the management policy of BOS funds at an elementary school, highlighting the need for systematic assessment criteria that align with established standards. Knowledge gap despite the existence of frameworks for evaluating fund management, there is limited research on their practical application in specific educational contexts. The aim is to analyze the BOS fund management by employing evaluation indicators based on Wirawan's framework, which includes input, process, output, and impact assessments. The results reveal that while the school’s facilities and infrastructure are generally in good condition, delays in fund distribution have adversely affected maintenance efforts. Key factors influencing successful fund management include meticulous planning, effective coordination, and accurate reporting. Furthermore, the study finds that BOS funds significantly reduce operational costs, benefiting both educators and students. Novelty this research contributes novel insights into the evaluation process of educational funding, illustrating how structured assessments can improve management practices. The implications of this study suggest that better fund management can lead to enhanced educational outcomes and resource allocation, ultimately fostering an environment conducive to learning. Highlights: Evaluation of BOS fund management enhances accountability and resource allocation. Delays in fund distribution impact maintenance of school facilities. Structured assessments improve educational outcomes and support effective learning environments. Keywords: BOS , Educational Management, Evaluation, Operational Costs, Infrastructure

A new performance evaluation indicator for the LEE Jong-wook Fellowship Program of Korea Foundation for International Healthcare to better assess its long-term educational impacts: a Delphi study.

The Dr. LEE Jong-wook Fellowship Program, established by the Korea Foundation for International Healthcare (KOFIH), aims to strengthen healthcare capacity in partner countries. The aim of the study was to develop new performance evaluation indicators for the program to better assess long-term educational impact across various courses and professional roles. A 3-stage process was employed. First, a literature review of established evaluation models (Kirkpatrick's 4 levels, CIPP model, OECD DAC criteria) was conducted to devise evaluation criteria. Second, these criteria were validated via a 2-round Delphi survey with 18 experts in training projects from May 2021 to June 2021. Third, the relative importance of the evaluation criteria was determined using the analytic hierarchy process (AHP), calculating weights and ensuring consistency through the consistency index (CI) and consistency ratio (CR), with CR values below 0.1 indicating acceptable consistency. The literature review led to a combined evaluation model, resulting in 4 evaluation areas, 20 items, and 92 indicators. The Delphi surveys confirmed the validity of these indicators, with content validity ratio values exceeding 0.444. The AHP analysis assigned weights to each indicator, and CR values below 0.1 indicated consistency. The final set of evaluation indicators was confirmed through a workshop with KFIH and adopted as the new evaluation tool. The developed evaluation framework provides a comprehensive tool for assessing the long-term outcomes of the Dr. LEE Jong-wook Fellowship Program. It enhances evaluation capabilities and supports improvements in the training program's effectiveness and international healthcare collaboration.

A human reliability analysis method based on STPA-IDAC and BN-SLIM for driver take-over in Level 3 automated driving

Human factors play an important role in the take-over process of Level 3 (L3) automated driving. This paper combines Systems Theoretic Process Analysis (STPA) and Information, Decision and Action in Crew context (IDAC) for qualitative analysis and Bayesian Network (BN) and Success Likelihood Index Method (SLIM) for quantitative calculation to obtain the main performance shaping factors (PSFs) and evaluation indicators that cause human errors. Firstly, the STPA-IDAC method is used to analyze unsafe control actions (UCAs) for take-over process and form the mapping relationship of UCAs-IDA-PSFs. Secondly, the BN of human reliability analysis for take-over process is constructed based on the BN-SLIM method. Uncertainty in rates of PSFs and evaluation indicators is addressed in a probabilistic manner using expert opinions and empirical data. After diagnostic reasoning of BN, mean variation is used to identify the main PSFs and evaluation indicators. This method can effectively identify the main PSFs and evaluation indicators that cause human errors, facilitate risk assessment and management, and reduce the human error probability (HEP).