Unexpected Incidents Research Articles

Unexpected Pathology During Pelvic Organ Prolapse Repair in an Urban Population.

This study quantifies the occult pathology risk among our urogynecologic patient population and highlights the importance of preoperative counseling, particularly in patients who have been underrepresented in prior studies. The aim of the study was to estimate unexpected gynecologic pathology incidence among a low-risk, racially, and ethnically representative patient population undergoing surgery for symptomatic pelvic organ (POP) prolapse. This was a retrospective study of patients undergoing hysterectomy for POP at an urban academic medical center. Patients with abnormal preoperative pathologic evaluation were excluded. Data were abstracted from the electronic health record, including demographic and clinical history, preoperative laboratory evaluation, and clinical risk factors for gynecologic malignancy. Two hundred ninety-nine nononcologic patients who underwent POP repair with hysterectomy were assessed. Thirty-six percent of patients identified as Hispanic, 17% as non-Hispanic Black, and 38% as non-Hispanic White. Twenty-three percent of patients reported abnormal uterine bleeding, 36% underwent a preoperative ultrasound examination, and 15% underwent endometrial biopsy. Two patients (0.9%) were diagnosed with endometrial carcinoma. Two patients (0.9%) were diagnosed with cervical dysplasia. No patients with concurrent oophorectomy or salpingectomy had ovarian or tubal pathology. The overall incidence of unexpected gynecologic pathology at the time of hysterectomy for prolapse was 1.7%. The rate of endometrial cancer in our cohort is on the higher end of previously published data, although absolute rates of unanticipated malignancy remain low. Future studies should examine the pathological findings of larger, racially, and ethnically diverse cohorts of patients undergoing POP surgery with hysterectomy, which can aid in providing relevant estimates for preoperative counseling.

Integrating spoken instructions into flight trajectory prediction to optimize automation in air traffic control

The booming air transportation industry inevitably burdens air traffic controllers’ workload, causing unexpected human factor-related incidents. Current air traffic control systems fail to consider spoken instructions for traffic prediction, bringing significant challenges in detecting human errors during real-time traffic operations. Here, we present an automation paradigm integrating controlling intent into the information processing loop through the spoken instruction-aware flight trajectory prediction framework. A 3-stage progressive multi-modal learning paradigm is proposed to address the modality gap between the trajectory and spoken instructions, as well as minimize the data requirements. Experiments on a real-world dataset show the proposed framework achieves flight trajectory prediction with high predictability and timeliness, obtaining over 20% relative reduction in mean deviation error. Moreover, the generalizability of the proposed framework is also confirmed by various model architectures. The proposed framework can formulate full-automated information processing in real-world air traffic applications, supporting human error detection and enhancing aviation safety.

The impact of regional emergency logistics response capacity on sustainable economic growth in China

To assess the impact of China’s regional emergency logistics response capabilities on economic growth sustainability, this study develops a set of comprehensive indicators. It applies the entropy weight method combined with the TOPSIS method to evaluate these capabilities across regions. Based on panel data from 30 provinces in China from 2012 to 2021, this study employs panel models, panel quantile regression, and panel threshold models to analyze the impact of emergency logistics response capability on sustainable economic growth. By measuring the regional emergency logistics response capability and examining its effects on the sustainability of economic growth, this research provides foreign investors with a clearer understanding of the ability of different regions in China to handle unexpected public incidents. This, in turn, helps reduce investment uncertainty, mitigate certain risks, and offers valuable insights for enhancing the efficiency of international investments in China. Key findings are: (1) Since 2018, the establishment of the Ministry of Emergency Management has significantly improved China’s regional emergency logistics capabilities, with the eastern region outperforming the central and western regions. (2) Enhancing China’s emergency logistics response capability helps improve the sustainability level of economic growth. This reflects how China’s emergency management departments can mitigate the adverse impacts of public emergencies on economic growth by transforming social logistics into emergency logistics. (3) As emergency response capabilities have strengthened, their impact on economic sustainability initially decreased but later increased.

Dynamic Response Characteristics and Pile Damage Identification of High-Piled Wharves under Dynamic Loading

In port dock engineering, high-piled wharves represent one of the primary structural forms. Damage to the foundation piles is a common issue, influenced by external loads such as impact forces during vessel berthing, slope deformations, and operational loads. This study focuses on the Jungong Road Wharf in Shanghai, utilizing FLAC 3D version 6.0 to conduct dynamic calculations under ship impact loading. The dynamic responses of the structure were analyzed, and various internal forces were extracted during the impact event. By combining concrete cracking failure criteria and fatigue damage theories, the effects of ship collisions on the cracking damage of high-piled wharf structures under different scenarios were assessed. Additionally, the applicability of modal flexibility in high-piled wharf scenarios was evaluated through finite element simulations. The results indicate that the dynamic amplification factor caused by dynamic loading is approximately 1.5, underscoring the necessity of considering this effect in the design and impact analysis of high-piled wharves. The impact loading significantly influences the bending moments of the piles, with inclined piles showing the greatest sensitivity. When a designed ship model collides with the high-piled wharf structure at a speed of 0.2 m/s, the tensile stress in the inclined piles reaches 87% of the ultimate tensile strength of the reinforcement. The impact loading has a relatively minor effect on the axial forces of the piles, a limited influence on the bending moments of the beams, but a considerable impact on the axial forces of the beams. Berthing by oversized vessels and unexpected incidents can lead to more severe damage to high-piled wharf structures. In the finite element simulations, modal flexibility effectively identified the locations of damage, with greater changes in modal flexibility correlating with increased damage severity.

A human visual attention analysis model for remote interaction interface of unmanned agricultural vehicles

When unmanned agricultural vehicles (UAVs) encounter unexpected failures or incidents during autonomous operations, a highly efficient remote human–machine interaction (HMI) interface is needed to assist in the manual joint control process. The highly complex cognitive mechanisms and visual characteristics of humans make it difficult to design efficient human–machine interfaces. Therefore, a human visual attention analysis model was proposed based on fully convolutional networks for the remote interaction interface of unmanned agricultural vehicles. First, visual attention experimental software combined with gaze-contingent real-time simulation methods was developed to build the dataset. Subsequently, the human visual attention analysis model (HVAAM) is constructed, and the corresponding training strategy is proposed, which is trained on the ILSVRC2012 standard and self-constructed datasets. The accuracy of the HVAAM analysis on the test dataset was validated. Ultimately, a human–machine interaction test of random failure warning for remote monitoring interface of unmanned plowing operation is performed with a self-developed unmanned electric tractor as the carrier. In the test dataset, the average analysis accuracy of HVAAM was 79.2 %, and the proposed training strategy led to a 20.7 % improvement in the model accuracy. In the unmanned electric tractor human–machine interaction test, HVAAM achieved 77.9 % accuracy in the visual attention analysis of the monitoring interface. This study can effectively facilitate the analysis, design, and optimization of remote monitoring interfaces for unmanned agricultural vehicles.

Design and application of an extracorporeal membrane oxygenation-specific line fixation device

Extracorporeal membrane oxygenation (ECMO) is an important treatment for extracorporeal cardiopulmonary life support for clinically critical patients. Currently, ECMO tubing is commonly fixed by tie-wraps or tourniquets, which have shortcomings such as easy loosening and potential damage to the tubing. Improper fixation of the catheter can lead to a series of adverse events, such as accidental disconnection of the tubing, rupture of the tubing, tubing folding, and air ingress into the tubing. In order to overcome the above problems, the research team of Zhongda Hospital, Southeast University invented a device for ECMO line fixation and obtained a national utility model patent of China (patent number: ZL 2019 2 2282849.3). The tool is mainly composed of several devices, including a line fixation clamp, a clip fixation device, and a base plate, which is uniquely designed and easy to operate. During ECMO therapy, this device ensures effective stabilization of the ECMO tubing, preventing unexpected incidents due to catheter loosening and facilitating the observation of ECMO catheter insertion markings. Pipeline can be effectively fixed to avoid the occurrence of accidents due to the loosening of the catheter, and at the same time, it is convenient to observe the placement scale of the ECMO catheter. The novelty and uniqueness of the fixation device materials also effectively prevent the occurrence of pressure injuries during its use.

Unexpected high incidence of parvovirus B19 nucleic acid detection in German blood donors in the winter/spring season 2023/2024.

In healthy adults, parvovirus B19 (PVB19) typically causes mild symptoms but can lead to severe complications in immunosuppressed individuals or those with high red blood cell turnover. Infection can occur through respiratory transmission or via transfusion, necessitating the testing of blood donations in Germany. Between 2015 and April 2024, we screened 2 105 755 blood donations for PVB19 using polymerase chain reaction. Incidence rates were calculated for three periods: pre-COVID-19 (2015-2020), during the pandemic (2020-2023), and post-COVID-19 (2023-2024). A total of 242 PVB19-positive donations were identified. In the first period, there were 101 positives out of 1 228 361 donations (incidence: 0.83/10 000). In the second period, four positives were found out of 621 222 donations (incidence: 0.06/10 000). In the third period, 137 positives were detected out of 235 088 donations (incidence: 5.35/10 000) with a striking increase of incidence between December 2023 and March 2024 (4.3-21.1/10 000 donations). Most people develop lifelong immunity after infection in childhood but the COVID-19 pandemic interventions, like masks and distancing, correlate with a decline in PVB19 infections in donors indicating an impact of hygiene measures on PVB19 infection rates.

Ontology-based inference decision support system for emergency response in tunnel vehicle accidents

Emergency response plans for tunnel vehicle accidents are crucial to ensure human safety, protect critical infrastructure, and maintain the smooth operation of transportation networks. However, many decision-support systems for emergency responses still rely significantly on predefined response strategies, which may not be sufficiently flexible to manage unexpected or complex incidents. Moreover, existing systems may lack the ability to effectively respond effectively to the impact different emergency scenarios and responses. In this study, semantic web technologies were used to construct a digital decision-support system for emergency responses to tunnel vehicle accidents. A basic digital framework was developed by analysing the knowledge system of the tunnel emergency response, examining its critical elements and intrinsic relationships, and mapping it to the ontology. In addition, the strategies of previous pre-plans are summarised and transformed into semantic rules. Finally, different accident scenarios were modelled to validate the effectiveness of the developed emergency response system.

Swift: A Data-Driven Flight Planning System at Scale

Flight planning, a pivotal challenge in the airline industry, strives to achieve economic and flexible scheduling of airplanes to serve designated flight itineraries. As the demand for air transportation soars, traditional planning methods can be inefficient in managing large-scale flights. Thus, we introduce Swift, a data-driven system tailored to enhance the scalability and effectiveness of flight planning. Swift primarily employs the bipartite graph model to derive optimal and economic flight plans for airlines. Our method not only minimizes the number of required planes but also ensures a balanced workload across these planes. Furthermore, Swift offers the capability of dynamic updates to flight plans in response to unexpected incidents at airports, such as bad weather conditions. Besides, Swift incorporates other functionalities like predicting future flight demand and monitoring real-time flight trajectories. Conference participants can interact with this system and explore our flight planning solution in real-world scenarios.

Application of TOPSIS Method in Evaluating the Performance of Insurance Companies: A Case Study

Insurance is a method for managing risk, as it protects against unforeseen financial losses. When you purchase insurance, you are safeguarding yourself against possible mishaps by having an insurance company compensate you or someone you designate. If you experience an accident and are uninsured, you may be responsible for all related expenses. Having appropriate insurance can help mitigate the various risks you might encounter throughout your life. Insurance not only covers unexpected incidents but also assists in paying for routine expenses such as annual medical checkups and dental visits. Furthermore, insurance companies negotiate discounts with healthcare providers, allowing their customers to pay the reduced rates. An insurance policy has two parties: the policyholder, who is the individual or company receiving the policy, and the insurer, who is the insurance company. The policyholder need not be the insured party. A person or business can become a policyholder by purchasing an insurance policy that protects another person or company. For instance, if a company buys life insurance for an employee, the employee is the insured party, and the company is the policyholder. Accounting, auditing, and actuarial standards must be carefully applied when examining insurance businesses' financial and other reports. Finding the business risks that insurance companies must deal with requires the use of financial information in particular. The management of insurance companies' technical reserves and guarantee reserve assets is crucial for reducing these risks. According to professional organizations and nations with free market economies, the main goal of financial statements in this situation is to give potential consumers the knowledge they need to make wise decisions. The economy depends on measuring the performance of insurance businesses. However, reliable performance evaluation can be difficult due to the unpredictability and complexity of the global market as well as the growing information flow. In certain situations, traditional performance measurement techniques might not produce appropriate findings. The fuzzy multi-criteria technique has been successfully used to address this issue.Critical proportional assessment is a multi-criteria decision-making analytical tool that aids in problem-solving. The COPRAS system has been developed as a tool for decision-making under specific contexts. This method calculates both the increase and decrease of criteria and improves the quality and quantity of criteria calculation. Compared to other methods, the COPRAS system can handle many criteria and offers a higher level of benefit and usage. Alternative taken as Equity and reserves, Business assets, Provision and liabilities, financial incomes, Cost of insurance, Cost of insurance. Evaluation preference taken as Life Insurance Corporation of India, HDFC Life Insurance Co. Ltd, Max Life Insurance Co. Ltd, ICICI Prudential Life Insurance Co. Ltd, Kotak Mahindra Life Insurance Co. Ltd, Aditya Birla SunLife Insurance Co. Ltd, TATA AIA Life Insurance Co. Ltd, SBI Life Insurance Co. Ltd, Bajaj Allianz Life Insurance Co. Ltd, PNB MetLife India Insurance Co. Ltd. From the result it is seen that SBI Life Insurance Co. Ltd is got the first rank where as is the Bajaj Allianz Life Insurance Co. Ltd is having the lowest rank.

Collaboration risk, vulnerability, and resource sharing in disaster management networks

AbstractLocal governments have the authority to implement emergency management and are the primary responders in emergencies. They are strategically positioned to lead the charge in disaster response. Similar to other wicked problems, disasters require comprehensive and complicated responses. Local emergency management networks facilitate the exchange of information and resources between local entities and their collaborating partners. Successful collaboration across local governments is imperative in unexpected and urgent incidents. This paper applies the institutional collective action (ICA) framework to investigate the effects of relational risk and vulnerability on the formation of networks. The relational risk perceived by each collaborative organisation and the vulnerability of the community are crucial factors in the establishment and maintenance of collaborative networks. In addition, the need for resource exchange also affects the formation of networks. Logistic regression estimates the effect of collaboration risk and vulnerability on network formation using data from the 2015 Seoul EM survey. The findings confirm that the perceived collaboration risk negatively influences the establishment of collaborative ties in networks, while the perceived vulnerability level positively affects the arrangement of collaborative networks. In addition, the respondent's need for resource exchange increases when building networks. These results imply that relieving the relational risk levels is important for facilitating inter‐organisational collaboration. Furthermore, the actors’ assessments of the vulnerability of the community influence their willingness to join networks to relieve external uncertainty and susceptibility. Lastly, organisations’ give‐and‐take relationships on information sharing and physical resource transmission have the potential to stimulate the establishment of collaborative networks.Points for practitioners Sustainable interconnectedness among functionally and horizontally fragmented organisations is crucial for dealing with disasters efficiently, but it does not come solely from emergency management planning itself. Understanding the risk mechanism embedded in an interdependent relationship should increase the potential benefits of a successful response. The degree of vulnerability in a community affects the level of risk perception when collaborating with other organisations. Beyond the internal capacity or assessment of their organisation, how they view their community in general influences their decision‐making differently when it comes to forging inter‐relational collaboration. Resource sharing in a timely manner is crucial during disasters. Each organisation has different negotiating powers and needs to consider building interdependent relationships.

START Triage Sebagai Istrumen Penanganan Korban dalam Setting Bencana

Unexpected incidents that result in significant loss or destruction are called disasters. In general, this event is closely associated with natural phenomena. Ultimately, though, this is frequently linked to incidents brought about by human intervention that entail drastic lifestyle modifications and significant harm. Since it is vital to allocate comparatively few resources to specific, urgent medical needs, triage is a crucial component of contemporary medical treatment. When a patient's medical demands and the resources that are available are not met in a timely or convenient manner, this task becomes critical. The foundation of Simple Triage and Rapid Treatment (START) is a quick evaluation of mental status, breathing, and perfusion (RPM). This work was written using a literature study methodology. Reference books (textbooks), scientific journals, and other pertinent sources provide the literature that is used. According to calibration and discrimination data, RPM (STARt triage) is a more reliable indicator of the blunt trauma survival rate than the Trauma Severity Score and the Revised Trauma Score. SALT triage is inferior to START (SMART) triage. On the other hand, employing START triage has been shown to increase time efficiency. It has been scientifically demonstrated that using START triage can result in suitable assessments for disaster management. Through the development of the Mass Casualty Triage Pack (SMART Triage Pack), STRAT triage has been used to enhance its efficacy.

Project Awakesure: Intelligent Drowsiness Detection Using Eye Tracking

Being sleepy or drowsy is referred to as being drowsy. A person who is sleepy may feel exhausted or lethargic and struggle to stay awake. People who are sleepy tend to be less attentive and may even nod off, though they can still be awakened. An increasing number of vocations nowadays call for sustained focus. In order for drivers to respond quickly to unexpected incidents, they must maintain a watchful eye on the road. Many road incidents are directly caused by tired drivers. In order to drastically lower the frequency of fatigue-related auto accidents, it is crucial to develop technologies that can identify and alert a driver to a poor psychophysical state. However, there are many challenges in developing systems that can quickly and accurately recognize a driver's signs of fatigue. Using vision-based technology is one technological option for implementing driver fatigue monitoring systems. The available driver drowsiness detection systems are described in this article. Here, we are assessing the driver's level of sleepiness utilizing his visual system. The automated system for preventing accidents and monitoring sleepy drivers developed for this study is based on detecting variations in the length of eye blinks. Our recommended technique makes use of the eyes' postulated horizontal symmetry property to identify visual changes in eye positions. Our novel approach precisely positions a standard webcam in front of the driver's seat to identify eye blinks. It will identify the eyeballs based on a specific EAR (Eye Aspect Ratio).

Macroscopic modeling of social crowds

Social behavior in crowds, such as herding or increased interpersonal spacing, is driven by the psychological states of pedestrians. Current macroscopic crowd models assume that these are static, limiting the ability of models to capture the complex interplay between evolving psychology and collective crowd dynamics that defines a “social crowd”. This paper introduces a novel approach by explicitly incorporating an “activity” variable into the modeling framework, which represents the evolving psychological states of pedestrians and is linked to crowd dynamics. To demonstrate the role of activity, we model pedestrian egress when this variable captures stress and awareness of contagion. In addition, to highlight the importance of dynamic changes in activity, we examine a scenario in which an unexpected incident necessitates alternative exits. These case studies demonstrate that activity plays a pivotal role in shaping crowd behavior. The proposed modeling approach thus opens avenues for more realistic macroscopic crowd descriptions with practical implications for crowd management.

Why Talking Is Not Cheap: Adverse Events and Informal Communication.

Healthcare management faces significant challenges related to upward communication. Sharing information in healthcare is crucial to the improvement of person-centered, safe, and effective patient care. An adverse event (AE) is an unintended or unexpected incident that causes harm to a patient and may lead to temporary or permanent disability. Learning from adverse events in healthcare is crucial to the improvement of patient safety and quality of care. Informal communication channels represent an untapped resource with regard to gathering data about the development of AEs. In this viewpoint paper, we start by identifying how informal communication played a key factor in some high-profile adverse events. Then, we present three Critical Challenge points that examine the role of informal communication in adverse events by (1) understanding how the prevailing trends in healthcare will make informal communication more important, (2) explaining how informal communication is part of the group-level sensemaking process, and (3) highlighting the potential role of informal communication in "breaking the silence" around critical and adverse events. Gossip, as one of the most important sources of informal communication, was examined in depth. Delineating the role of informal communication and adverse events within the healthcare context is pivotal to understanding and improving team and upward communication in healthcare organizations. For clinical leaders, the challenge is to cultivate a climate of communication safety, whereby informal communication channels can be used to collect soft intelligence that are paths to improving the quality of care and patient safety.

Investigating pedestrian-vehicle crashes on interstate highways: Applying random parameter binary logit model with heterogeneity in means

In the U.S., the interstate highway system is categorized as a controlled-access or limited-access route, and it is unlawful for pedestrians to enter or cross this type of highway. However, pedestrian-vehicle crashes on the interstate highway system pose a distinctive safety concern. Most of these crashes involve ‘unintended pedestrians’, drivers who come out of their disabled vehicles, or due to the involvement in previous crashes on the interstate. Because these are not ‘typical pedestrians’, a separate investigation is required to better understand the pedestrian crash problem on interstate highways and identify the high-risk scenarios. This study explored 531 KABC (K = Fatal, A = Severe, B = Moderate, C = Complaint) pedestrian injury crashes on Louisiana interstate highways during the 2014–2018 period. Pedestrian injury severity was categorized into two levels: FS (fatal/severe) and IN (moderate/complaint). The random parameter binary logit with heterogeneity in means (RPBL-HM) model was utilized to address the unobserved heterogeneity (i.e., variations in the effect of crash contributing factors across the sample population) in the crash data. Some of the factors were found to increase the likelihood of pedestrian’s FS injury in crashes on interstate highways, including pedestrian impairment, pedestrian action, weekend, driver aged 35–44 years, and spring season. The interaction of ‘pedestrian impairment’ and ‘weekend’ was found significant, suggesting that alcohol-involved pedestrians were more likely to be involved in FS crashes during weekends on the interstate. The obtained results can help the ‘unintended pedestrians’ about the crash scenarios on the interstate and reduce these unexpected incidents.

Revitalizing the Electric Grid: A Machine Learning Paradigm for Ensuring Stability in the U.S.A

The electric grid entails a diverse range of components with pervasive heterogeneity. Conventional electricity models in the U.S.A. encounter challenges in terms of affirming the stability and security of the power system, particularly, when dealing with unexpected incidents. This study explored various electric grid models adopted in various nations and their shortcomings. To resolve these challenges, the research concentrated on consolidating machine learning algorithms as an optimization strategy for the electricity power grid. As such, this study proposed Ensemble Learning with a Feature Engineering Model which exemplified promising outputs, with the voting classifier performing well as compared to the rainforest classifier model. Particularly, the accuracy of the voting classifier was ascertained to be 94.57%, illustrating that approximately 94.17% of its predictions were correct as contrasted to the Random Forest. Besides, the precision of the voting classifier was ascertained to be 93.78%, implying that it correctly pinpointed positive data points 93.78% of the time. Remarkably, the Voting Classifier for the Ensemble Learning with Feature Engineering Model technique surpassed the performance of most other techniques, demonstrating an accuracy rate of 94.57%. These techniques provide protective and preventive measures to resolve the vulnerabilities and challenges faced by geographically distributed power systems.

Developing a conceptual model for improving the software system reliability

In the Industry 4.0 environment, the software systems development methodology is rapidly evolving, flexible technologies and new programming languages are being applied. The development of the software industry has made the issue of the quality of software systems an urgent problem. A number of quality models have been proposed for determining the quality of software systems so far, and these models specify the parameters and criteria for evaluating quality. Software reliability is one of the key indicators among the quality parameters of software systems, as it quantifies software crashes which can bring down even the most powerful system, ensuring that software systems run correctly and unexpected incidents do not occur. The increasing difficulty of the software system, the expansion of the scope of issues assigned on them, and as a result, the significant increase in the volume and complexity of the software system have made the problem of the reliability of the software system even more urgent. The essence of the issue is to reveal the main factors affecting the reliability of software systems, demonstrate existing problems in this area and develop mathematical models for assessing reliability. Mathematical models estimate the number of errors remaining in the software system before commissioning, predict the time of occurrence of the next crash and when the testing process will end. It is necessary to comprehensively approach the issue of ensuring reliability at all stages of the life cycle of the software system. This paper proposes a conceptual model to solve this problem.

Revitalizing the Electric Grid: A Machine Learning Paradigm for Ensuring Stability in the U.S.A.

The electric grid entails a diverse range of components with pervasive heterogeneity. Conventional electricity models in the U.S.A. encounter challenges in terms of affirming the stability and security of the power system, particularly, when dealing with unexpected incidents. This study explored various electric grid models adopted in various nations and their shortcomings. To resolve these challenges, the research concentrated on consolidating machine learning algorithms as an optimization strategy for the electricity power grid. As such, this study proposed Ensemble Learning with a Feature Engineering Model which exemplified promising outputs, with the voting classifier performing well as compared to the rainforest classifier model. Particularly, the accuracy of the voting classifier was ascertained to be 94.57%, illustrating that approximately 94.17% of its predictions were correct as contrasted to the Random Forest. Besides, the precision of the voting classifier was ascertained to be 93.78%, implying that it correctly pinpointed positive data points 93.78% of the time. Remarkably, the Voting Classifier for the Ensemble Learning with Feature Engineering Model technique surpassed the performance of most other techniques, demonstrating an accuracy rate of 94.57%. These techniques provide protective and preventive measures to resolve the vulnerabilities and challenges faced by geographically distributed power systems.

STGAFormer: Spatial–temporal Gated Attention Transformer based Graph Neural Network for traffic flow forecasting

Traffic flow prediction is a critical component of Intelligent Transportation Systems (ITS). However, the dynamic temporal variations in traffic flow, especially in potential occurrence of unexpected incidents, pose challenges to the prediction of traffic flow. This paper proposes a Spatial–temporal Gated Attention Transformer (STGAFormer) model based Graph Neural Network(GNN), leveraging the encoder architecture of the transformer. The gated temporal self-attention in the model, a novel module, can improve the model’s ability to make long-term predictions and handle sudden traffic incidents by enhancing the extraction of both local and global temporal features. Additionally, this paper proposes a distance spatial self-attention module to extract spatial features, which employs thresholding to selectively identify crucial features from both nearby and distant regions. In this way, the model’s ability to assimilate critical spatial information is promoted. Moreover, our model incorporates a diverse range of inputs, including traffic flow attributes, periodicity, proximity adjacency matrix, and adaptive adjacency matrix. Experiments from four real datasets demonstrate that STGAFormer achieves state-of-the-art performance, especially the MAE value of the PeMS08 dataset experiment is improved by 3.82%. This method offers valuable insights and robust support for future transportation planning.