[Objective] In order to improve the collision performance of the end elbow of the shoe-rail system and enhance the current collection quality, it is necessary to carry out the optimization research on the third-rail end elbow based on NURBS (Non-Uniform Rational B-Spline). [Method] Taking the end elbow of a certain metro line in Wuhan as example, an optimization design for the end elbow is carried out based on the concepts of parametric design and multi-objective optimization. According to the collision simulation of the end elbow of the shoe-rail system under the working condition of 120 km/h train running speed, the evaluation indicators of the current collection quality of the shoe-rail system under high speed working conditions are determined. The cross-section stretching curve of the end elbow is established through NURBS. A finite-element model of the curved end elbow is established by using the ANSYS Parametric Design Language. Taking the control point coordinates of the cross-section stretching curve as design variables, the optimization objectives are selected through main effect analysis, and the value range of the design variables is clarified. Based on NSGA-II (Non-dominated Sorting Genetic Algorithm II), a multi-objective optimization model of the curved end elbow is established, and the optimal solution is selected through the non-dominated sorting method and the min-max regret value method. [Result & Conclusion] According to the results of the main effect analysis, the bending-in collision force, bending-out collision force, total offline time, and the maximum impact acceleration are selected as the optimization objectives, and the maximum single offline time and the number of collisions are taken as the constraint conditions. After optimization, the current collection quality of the end elbow of the shoe-rail system is improved. When the train running speed is 120 km/h, the total offline time and the maximum single offline time are significantly optimized, with a reduction of 80.77% and 89.96% respectively. The corner entry impact force and the corner exit impact force are reduced by 32.26% and 44.03% respectively. When the train running speeds are 40 km/h and 80 km/h, all indicators are reduced by more than 20%.
Abstract The Basel Accord emphasizes the necessity of employing internal data models to manage key credit risk components, including Probability of Default (PD), Loss Given Default (LGD), and Exposure At Default (EAD). Among these, internal datasets are critical for estimating PD, a fundamental measure of borrower creditworthiness. Nevertheless, practical application often faces challenges due to incomplete datasets, which can skew analyses and undermine the accuracy of credit scoring models. Traditional approaches to addressing missing data, such as sample deletion or mean imputation, are widely used; however, they often prove insufficient for accurate prediction. Consequently, imputation methods are typically favored over deletion, as they allow for the full utilization of available data. Recent advancements have introduced more sophisticated techniques, such as Generative Adversarial Imputation Networks (GAIN), which utilize a generative adversarial network to model data distributions and impute missing values with greater precision than conventional methods. Building on these developments, this study proposes a novel imputation approach, SMART (Structured Missingness Analysis and Reconstruction Technique) for credit scoring datasets. SMART consists of two primary stages: first, it normalizes and denoises the dataset using randomized Singular Value Decomposition (rSVD), followed by the implementation of GAIN to impute missing values. Experimental results demonstrate that SMART significantly outperforms existing state-of-the-art methods, particularly in high missing data contexts (20%, 50%, and 80%), with improvements in imputation accuracy of 7.04%, 6.34%, and 13.38%, respectively. In conclusion, SMART represents a substantial advancement in handling incomplete credit scoring datasets, leading to more precise PD estimation and enhancing the robustness of credit risk management models.
In recent years, climate change and marine pollution have significantly degraded coral reefs, highlighting the urgent need for automated coral detection to monitor marine ecosystems. However, underwater coral detection presents unique challenges, including low image contrast, complex coral structures, and dense coral growth, which limit the effectiveness of general object detection algorithms. To address these challenges, we propose SCoralDet, a soft coral detection model based on the YOLO architecture. First, we introduce a Multi-Path Fusion Block (MPFB) to capture coral features across multiple scales, enhancing the model’s robustness to uneven lighting and image blurring. We further improve inference efficiency by applying reparameterization. Second, we integrate lightweight components such as GSConv and VoV-GSCSP to reduce computational overhead without sacrificing performance. Additionally, we develop an Adaptive Power Transformation label assignment strategy, which dynamically adjusts anchor alignment metrics. By incorporating soft labels and soft central region loss, our model is guided to prioritize high-quality, well-aligned predictions. We evaluate SCoralDet on the Soft-Coral dataset, achieving an inference latency of 9.52 ms and an mAP50 of 81.9. This surpasses the performance of YOLOv5 (79.9), YOLOv6 (79.4), YOLOv8 (79.5), YOLOv9 (78.3), and YOLOv10 (79.5). These results demonstrate the effectiveness and practicality of SCoralDet in underwater coral detection tasks.
Traditional obstacle avoidance algorithms usually use a single shallow application, such as sensor-based distance measurement or some logic judgment algorithm, which leads to problems such as the need to manually adjust the parameters first, the inability to recognize complex or unknown environments, and the recognition errors caused by significant noise errors. Therefore, to overcome these limitations, this paper combines convolutional neural network and obstacle avoidance algorithms. A model of obstacle avoidance method based on convolutional neural network established in this paper, and puts forward the theory of obstacle avoidance method based on convolutional neural network, which adopts MobileNet_v3 as the learning framework, roughly classifies all the obstacle maps into three categories, and then, through the research and application of six traditional obstacle avoidance algorithms, finally concludes that the model can be applied according to different kinds of obstacles. The model can learn and discriminate against different obstacle maps, thus improving the performance of obstacle avoidance and avoiding the limitations of traditional obstacle avoidance algorithms. Verified the effectiveness of each algorithm in various scenarios. A single shallow application of the problem is usually used to robotize the traditional obstacle avoidance algorithms, which provides an essential reference.
Highway spoil features a wide range of particle sizes and poor gradation, and their strength characteristics serve as key factors affecting the stability of spoil slopes. To explore the relationship between fractal characteristics of particle size and strength, this study examined 21 spoil sites along the Guizhou section of the Duyun‒Shangri-La Expressway. Based on the field investigation, particle size analysis tests, and laboratory shear tests, the particle composition and shear strength parameters of spoil were obtained. Based on fractal theory, the fractal characteristics of particle size under different gradation scaling methods were analyzed, and the relationship between shear strength parameters and the fractal dimensions of particle size was discussed. A formula for estimating the shear strength parameters of spoil was established. The results show that the particle composition of spoil varies significantly depending on their source and degree of weathering. Spoil from roadbeds and strongly weathered spoil generally exhibit unimodal distributions, while spoil from tunnels and with moderate weathering tend to exhibit bimodal distributions. The fractal dimension shows a strong correlation with both the internal friction angle and the fine particle content (mass percentage of particles smaller than 5 mm). When the fine particle content is ≤20%, the fractal dimension increases with the increasing of fine particle content, while the internal friction angle decreases with increasing fractal dimension. When the fine particle content exceeds 20%, the fractal dimension decreases with increasing fine particle content, and the internal friction angle increases with increasing fractal dimension. The proposed estimation formula for shear strength parameters was validated through case studies, and the strength parameters can be estimated based on particle composition, providing a reference for the stability assessment of spoil sites.
Yining Hong, Christopher S. Timperley, Christian Kästner
Machine learning (ML) components are increasingly integrated into software products, yet their complexity and inherent uncertainty often lead to unintended and hazardous consequences, both for individuals and society at large. Despite these risks, practitioners seldom adopt proactive approaches to anticipate and mitigate hazards before they occur. Traditional safety engineering approaches, such as Failure Mode and Effects Analysis (FMEA) and System Theoretic Process Analysis (STPA), offer systematic frameworks for early risk identification but are rarely adopted. This position paper advocates for integrating hazard analysis into the development of any ML-powered software product and calls for greater support to make this process accessible to developers. By using large language models (LLMs) to partially automate a modified STPA process with human oversight at critical steps, we expect to address two key challenges: the heavy dependency on highly experienced safety engineering experts, and the time-consuming, labor-intensive nature of traditional hazard analysis, which often impedes its integration into real-world development workflows. We illustrate our approach with a running example, demonstrating that many seemingly unanticipated issues can, in fact, be anticipated.
Hashini Gunatilake, John Grundy, Rashina Hoda
et al.
Empathy, defined as the ability to understand and share others' perspectives and emotions, is essential in software engineering (SE), where developers often collaborate with diverse stakeholders. It is also considered as a vital competency in many professional fields such as medicine, healthcare, nursing, animal science, education, marketing, and project management. Despite its importance, empathy remains under-researched in SE. To further explore this, we conducted a socio-technical grounded theory (STGT) study through in-depth semi-structured interviews with 22 software developers and stakeholders. Our study explored the role of empathy in SE and how SE activities and processes can be improved by considering empathy. Through applying the systematic steps of STGT data analysis and theory development, we developed a theory that explains the role of empathy in SE. Our theory details the contexts in which empathy arises, the conditions that shape it, the causes and consequences of its presence and absence. We also identified contingencies for enhancing empathy or overcoming barriers to its expression. Our findings provide practical implications for SE practitioners and researchers, offering a deeper understanding of how to effectively integrate empathy into SE processes.
From its first adoption in the late 80s, qualitative research has slowly but steadily made a name for itself in what was, and perhaps still is, the predominantly quantitative software engineering (SE) research landscape. As part of our regular column on empirical software engineering (ACM SIGSOFT SEN-ESE), we reflect on the state of qualitative SE research with a focus group of experts. Among other things, we discuss why qualitative SE research is important, how it evolved over time, common impediments faced while practicing it today, and what the future of qualitative SE research might look like. Joining the conversation are Rashina Hoda (Monash University, Australia), Carolyn Seaman (University of Maryland, United States), and Klaas Stol (University College Cork, Ireland). The content of this paper is a faithful account of our conversation from October 25, 2025, which we moderated and edited for our column.
Applications of Large Language Models (LLMs) are rapidly growing in industry and academia for various software engineering (SE) tasks. As these models become more integral to critical processes, ensuring their reliability and trustworthiness becomes essential. Consequently, the concept of trust in these systems is becoming increasingly critical. Well-calibrated trust is important, as excessive trust can lead to security vulnerabilities, and risks, while insufficient trust can hinder innovation. However, the landscape of trust-related concepts in LLMs in SE is relatively unclear, with concepts such as trust, distrust, and trustworthiness lacking clear conceptualizations in the SE community. To bring clarity to the current research status and identify opportunities for future work, we conducted a comprehensive review of $88$ papers: a systematic literature review of $18$ papers focused on LLMs in SE, complemented by an analysis of 70 papers from broader trust literature. Additionally, we conducted a survey study with 25 domain experts to gain insights into practitioners' understanding of trust and identify gaps between existing literature and developers' perceptions. The result of our analysis serves as a roadmap that covers trust-related concepts in LLMs in SE and highlights areas for future exploration.
Vladyslav PROTSENKO, Volodymyr MALASHCHENKO, Valentyn NASTASENKO
et al.
The article deals with mechanical engineering, and transport machines, namely the elevator brake mechanism structure. The article aims to study the number and location of redundant constraints in elevator brake mechanisms and to depict their impact on brake reliability and transportation safety. To study the structure of the mentioned mechanisms, we used classical methods of applied mechanics plus the circuit method of L. Reshetov. The structure of crane disc brakes with short-stroke DC electromagnet and long-stroke AC electromagnet mechanisms was analyzed and redundant constraints were identified. It was shown that the presence of redundant constraints causes friction torque oscillation and lead to load distribution unevenness between brake elements. Based on the provided analysis, construction improvement events should be implemented to remove the most dangerous redundant constraints.
Objective Aiming to propose effective methods and specific cases for capacity enhancement, the implementation experiences of capacity improvement renovation projects on existing Shanghai urban rail transit lines are systematically summarized. Method An in-depth analysis of the main reasons and influencing factors behind the insufficient capacity of existing Shanghai urban rail transit lines is conducted. Based on these analyses, the technical approach and five core strategies for capacity enhancement is outlined. Successful cases are introduced, including the achievement of 1-minute 50-second minimum departure interval on Line 9, the capacity enhancement renovation of Line 3 and Line 4, the expansion reconstruction of Line 5 from 4-car to 6-car trains, and the capacity enhancement renovation and double-track addition on Line 6. Result & Conclusion The key factors influencing line capacity include vehicle depot size, interval throughput, depot entry/exit capacity, and power supply capacity. Addressing these factors, a technical approach for capacity enhancement is established, which starts with passenger flow prediction and characteristics analysis, followed by facility and equipment capacity assessment, and train routing scheme design, ultimately leading to the development of scientifically sound renovation plans. Additionally, five major strategies are proposed: optimization of operational management, digging into and upgrading existing system capacity, signaling system upgrades or localized modifications of civil works, large-scale system overhauls, and line network overall optimization.
Currently, the research on the mechanical properties of rubber-modified asphalt mixtures primarily focuses on small-scale investigations, with insufficient exploration into the performance of rubber particles and their relationship with the mechanism and properties of modified asphalt mixtures. Limited studies have been conducted on large-scale rubber modification in asphalt mixtures. Due to frequent use and subsequent high damage to existing asphalt pavements, incorporating rubber-modified asphalt mixtures can partially alleviate premature deterioration. Dynamic modulus tests were conducted using MTS equipment under unconfined conditions to investigate the viscoelastic behavior of rubber-modified asphalt mixtures with high rubber content and elucidate the influence of rubber particle content on the elastic deformation and recovery capability. The dynamic mechanical properties of the mixtures were determined at different loading rates, temperatures, and types of rubber-modified asphalt mixtures. Based on the test data, variations in the dynamic modulus, phase angle, storage modulus, loss modulus, loss factor, and rut factor of the rubber-modified asphalt mixtures under different loading frequencies, temperatures, and types were analyzed. The results demonstrate the pronounced viscoelastic behavior of rubber-modified asphalt mixtures. The mixtures exhibit enhanced elasticity at low temperatures and high frequencies, while their viscosity becomes more prominent at high temperatures and low frequencies. Under constant test temperatures, an increase in load loading frequency leads to a higher dynamic modulus; conversely, a decrease in dynamic modulus is observed with increasing test temperatures. The dynamic modulus of ARHM-25 at a frequency of 10 Hz is found to be 12.99 times higher at 15 °C compared to that at 60 °C, while at 30 °C, the dynamic modulus at 25 Hz is observed to be 2.72 times greater than that at 0.1 Hz. Furthermore, the rutting resistance factors of the asphalt mixtures increase with loading frequency but decrease with temperature. The rutting factor for ARHM-13 at a frequency of 10 Hz is found to be 22.98 times higher at 15 °C compared to that at 60 °C, while at a temperature of 30 °C, the rutting factor for this material is observed to be 3.09 times greater at a frequency of 25 Hz than at 0.1 Hz. These findings suggest that rutting is most likely when vehicles drive at low speeds in hot weather conditions.
Smart transportation engineering is an important means to address urban transportation issues, and the continuous development and application of vehicle road collaboration technology provide new opportunities and possibilities for innovation in smart transportation. This article aims to explore the innovative application of vehicle road collaboration technology in smart transportation engineering, with a focus on analyzing its core value in real-time data communication and transportation system integration. At the same time, in-depth analysis will be conducted on the application challenges faced by vehicle road collaboration technology in smart transportation, and its innovative applications in autonomous vehicle communication, public transportation scheduling, emergency response systems, and traffic infrastructure monitoring will be demonstrated through specific cases. With the continuous acceleration of urbanization and the increasingly prominent traffic problems, smart transportation as an intelligent means of traffic management has received widespread attention. The rapid development of vehicle road collaboration technology provides technical support and innovative impetus for the implementation of smart transportation engineering. The research will start from the basic concepts of smart transportation and vehicle road collaboration technology, analyze the core value of vehicle road collaboration technology in real-time data communication and transportation system integration, and explore its application challenges and innovative application cases in smart transportation engineering, providing theoretical guidance and practical reference for the development of smart transportation systems.
Kansei Engineering (KE) is a user-centered design approach that emphasizes the emotional aspects of user experience. This paper explores the integration of KE in the case of a transportation company that focuses on connecting cargo owners with transportation providers. The methodology involves aligning the design process with the company's strategy, collecting and semantic scaling Kansei words, and evaluating website design through experimental and statistical analyses. Initially, we collaborated with the company to understand their strategic goals, using Use Case and Entity Relationship diagrams to learn about the website functionality. Subsequent steps involved collecting Kansei words that resonate with the company's vision. Website samples from comparable transportation companies were then evaluated by X subject in the survey. Participants were asked to arrange samples based on emotional feedback using a 5-point SD scale. We used Principal Component Analysis (PCA) to identify critical factors affecting users' perceptions of the design. Based on these results, we collaborated with designers to reformulate the website, ensuring the design features aligned with the Kansei principles. The outcome is a user-centric web design to enhance the site's user experience. This study shows that KE can be effective in creating more user-friendly web interfaces in the transportation industry.
Major cities worldwide experience problems with the performance of their road transportation systems, and the continuous increase in traffic demand presents a substantial challenge to the optimal operation of urban road networks and the efficiency of traffic control strategies. The operation of transportation systems is widely considered to display fragile property, i.e., the loss in performance increases exponentially with the linearly increasing magnitude of disruptions. Meanwhile, the risk engineering community is embracing the novel concept of antifragility, enabling systems to learn from historical disruptions and exhibit improved performance under black swan events. In this study, based on established traffic models, namely fundamental diagrams and macroscopic fundamental diagrams, we first conducted a rigorous mathematical analysis to prove the fragile nature of the systems theoretically. Subsequently, we propose a skewness-based indicator that can be readily applied to cross-compare the degree of fragility for different networks solely dependent on the MFD-related parameters. At last, by taking real-world stochasticity into account, we implemented a numerical simulation with realistic network data to bridge the gap between the theoretical proof and the real-world operations, to reflect the potential impact of uncertainty on the fragility of the systems. This work aims to demonstrate the fragile nature of road transportation systems and help researchers better comprehend the necessity to consider explicitly antifragile design for future traffic control strategies.
Ebube Alor, Ahmad Abdellatif, SayedHassan Khatoonabadi
et al.
Software engineering (SE) chatbots are increasingly gaining attention for their role in enhancing development processes. At the core of chatbots are Natural Language Understanding platforms (NLUs), which enable them to comprehend user queries but require labeled data for training. However, acquiring such labeled data for SE chatbots is challenging due to the scarcity of high-quality datasets, as training requires specialized vocabulary and phrases not found in typical language datasets. Consequently, developers often resort to manually annotating user queries -- a time-consuming and resource-intensive process. Previous approaches require human intervention to generate rules, called labeling functions (LFs), that categorize queries based on specific patterns. To address this issue, we propose an approach to automatically generate LFs by extracting patterns from labeled user queries. We evaluate our approach on four SE datasets and measure performance improvement from training NLUs on queries labeled by the generated LFs. The generated LFs effectively label data with AUC scores up to 85.3% and NLU performance improvements up to 27.2%. Furthermore, our results show that the number of LFs affects labeling performance. We believe that our approach can save time and resources in labeling users' queries, allowing practitioners to focus on core chatbot functionalities rather than manually labeling queries.