Tanja E. J. Vos, Tijs van der Storm, Alexander Serebrenik
et al.
Software engineering is the invisible infrastructure of the digital age. Every breakthrough in artificial intelligence, quantum computing, photonics, and cybersecurity relies on advances in software engineering, yet the field is too often treated as a supportive digital component rather than as a strategic, enabling discipline. In policy frameworks, including major European programmes, software appears primarily as a building block within other technologies, while the scientific discipline of software engineering remains largely absent. This position paper argues that the long-term sustainability, dependability, and sovereignty of digital technologies depend on investment in software engineering research. It is a call to reclaim the identity of software engineering.
PURPOSE OR GOAL: This study investigates how GenAI can be integrated with a criterion-referenced grading framework to improve the efficiency and quality of grading for mathematical assessments in engineering. It specifically explores the challenges demonstrators face with manual, model solution-based grading and how a GenAI-supported system can be designed to reliably identify student errors, provide high-quality feedback, and support human graders. The research also examines human graders' perceptions of the effectiveness of this GenAI-assisted approach. ACTUAL OR ANTICIPATED OUTCOMES: The study found that GenAI achieved an overall grading accuracy of 92.5%, comparable to two experienced human graders. The two researchers, who also served as subject demonstrators, perceived the GenAI as a helpful second reviewer that improved accuracy by catching small errors and provided more complete feedback than they could manually. A central outcome was the significant enhancement of formative feedback. However, they noted the GenAI tool is not yet reliable enough for autonomous use, especially with unconventional solutions. CONCLUSIONS/RECOMMENDATIONS/SUMMARY: This study demonstrates that GenAI, when paired with a structured, criterion-referenced framework using binary questions, can grade engineering mathematical assessments with an accuracy comparable to human experts. Its primary contribution is a novel methodological approach that embeds the generation of high-quality, scalable formative feedback directly into the assessment workflow. Future work should investigate student perceptions of GenAI grading and feedback.
The quantum threat to cybersecurity has accelerated the standardization of Post-Quantum Cryptography (PQC). Migrating legacy software to these quantum-safe algorithms is not a simple library swap, but a new software engineering challenge: existing vulnerability detection, refactoring, and testing tools are not designed for PQC's probabilistic behavior, side-channel sensitivity, and complex performance trade-offs. To address these challenges, this paper outlines a vision for a new class of tools and introduces the Automated Quantum-safe Adaptation (AQuA) framework, with a three-pillar agenda for PQC-aware detection, semantic refactoring, and hybrid verification, thereby motivating Quantum-Safe Software Engineering (QSSE) as a distinct research direction.
Levitation control is one of the key technologies in maglev trains, and an accurate electromagnetic force distribution model is fundamental to the design of electromagnets and levitation controllers. The electromagnetic levitation force is generated bet-ween the pole faces of the electromagnets and the rails, and it is directly related to the magnetic field distribution within the levitation gap. Consequently, this force varies in real time, depending on the excitation current of the electromagnets and the levitation gap. However, due to the fringing effect, the magnetic field in the levitation gap is not uniformly distributed, and magnetic flux leakage occurs around the electromagnets. These factors cause significant errors in electromagnetic force calculations from existing models that assume an ideal magnetic field distribution. To address this issue, an optimized calculation model for electromagnetic force in single-point levitation systems was developed, based on a magnetic field distribution in the levitation gap derived via conformal mapping. Comparative simulation experiments were conducted to evaluate the proposed method against the model with an ideal electromagnetic force distribution and the finite element method. Results demonstrate that the proposed model significantly reduces calculation errors for both electromagnetic force and steady-state levitation current, with specific calculation errors not exceeding 4.2% and 3.2%, respectively, outperforming the finite element method. Platform tests further confirm that the proposed model results in steady-state levitation current calculation error not exceeding 5.6%.
This study aims to enhance the lateral stability performance of 25T passenger cars during operation. To this end, a secondary transverse damper model was constructed using Matlab/Simulink, based on an on-off skyhook damping semi-active control strategy. Dynamics simulation analysis was carried out in conjunction with Simpack, which specializes in multi-body dynamics. The analysis considered the influence of damper joint stiffness on the semi-active control effect, building on the traditional semi-active control model. A comparative analysis was performed to investigate the effects of passive control, traditional semi-active control, and semi-active control utilizing damper joint stiffness on the running smoothness, running stability and curve negotiation performance of the vehicles. The effect of time lag on the control effect was also examined. The simulation results indicate that the semi-active control significantly improves the lateral stability performance of the vehicles, but it reduces their nonlinear critical speed. Compared to passive control, traditional semi-active control and semi-active control utilizing joint stiffness resulted in a decrease in the lateral Sperling index by 14.5% and 15.8%, respectively, and a reduction in the nonlinear critical speed by 8.09% and 6.93%, respectively. Additionally, no significant effects were identified on the curve negotiation performance with these two control modes, indicating adequate safety margins. Study results also suggest that controlling the response time-lag of the system within 80 ms is essential to ensure the effect of semi-active control.
Foundation models (FMs), particularly large language models (LLMs), have shown significant promise in various software engineering (SE) tasks, including code generation, debugging, and requirement refinement. Despite these advances, existing evaluation frameworks are insufficient for assessing model performance in iterative, context-rich workflows characteristic of SE activities. To address this limitation, we introduce \emph{SWE-Arena}, an interactive platform designed to evaluate FMs in SE tasks. SWE-Arena provides a transparent, open-source leaderboard, supports multi-round conversational workflows, and enables end-to-end model comparisons. The platform introduces novel metrics, including \emph{model consistency score} that measures the consistency of model outputs through self-play matches, and \emph{conversation efficiency index} that evaluates model performance while accounting for the number of interaction rounds required to reach conclusions. Moreover, SWE-Arena incorporates a new feature called \emph{RepoChat}, which automatically injects repository-related context (e.g., issues, commits, pull requests) into the conversation, further aligning evaluations with real-world development processes. This paper outlines the design and capabilities of SWE-Arena, emphasizing its potential to advance the evaluation and practical application of FMs in software engineering.
Marc Bruni, Fabio Gabrielli, Mohammad Ghafari
et al.
Prompt engineering reduces reasoning mistakes in Large Language Models (LLMs). However, its effectiveness in mitigating vulnerabilities in LLM-generated code remains underexplored. To address this gap, we implemented a benchmark to automatically assess the impact of various prompt engineering strategies on code security. Our benchmark leverages two peer-reviewed prompt datasets and employs static scanners to evaluate code security at scale. We tested multiple prompt engineering techniques on GPT-3.5-turbo, GPT-4o, and GPT-4o-mini. Our results show that for GPT-4o and GPT-4o-mini, a security-focused prompt prefix can reduce the occurrence of security vulnerabilities by up to 56%. Additionally, all tested models demonstrated the ability to detect and repair between 41.9% and 68.7% of vulnerabilities in previously generated code when using iterative prompting techniques. Finally, we introduce a "prompt agent" that demonstrates how the most effective techniques can be applied in real-world development workflows.
Shavindra Wickramathilaka, John Grundy, Kashumi Madampe
et al.
The use of diverse mobile applications among senior users is becoming increasingly widespread. However, many of these apps contain accessibility problems that result in negative user experiences for seniors. A key reason is that software practitioners often lack the time or resources to address the broad spectrum of age-related accessibility and personalisation needs. As current developer tools and practices encourage one-size-fits-all interfaces with limited potential to address the diversity of senior needs, there is a growing demand for approaches that support the systematic creation of adaptive, accessible app experiences. To this end, we present AdaptForge, a novel model-driven engineering (MDE) approach that enables advanced design-time adaptations of mobile application interfaces and behaviours tailored to the accessibility needs of senior users. AdaptForge uses two domain-specific languages (DSLs) to address age-related accessibility needs. The first model defines users' context-of-use parameters, while the second defines conditional accessibility scenarios and corresponding UI adaptation rules. These rules are interpreted by an MDE workflow to transform an app's original source code into personalised instances. We also report evaluations with professional software developers and senior end-users, demonstrating the feasibility and practical utility of AdaptForge.
Qiaolin Qin, Ronnie de Souza Santos, Rodrigo Spinola
Context. The rise of generative AI (GenAI) tools like ChatGPT and GitHub Copilot has transformed how software is learned and written. In software engineering (SE) education, these tools offer new opportunities for support, but also raise concerns about over-reliance, ethical use, and impacts on learning. Objective. This study investigates how undergraduate SE students use GenAI tools, focusing on the benefits, challenges, ethical concerns, and instructional expectations that shape their experiences. Method. We conducted a survey with 130 undergraduate students from two universities. The survey combined structured Likert-scale items and open-ended questions to investigate five dimensions: usage context, perceived benefits, challenges, ethical and instructional perceptions. Results. Students most often use GenAI for incremental learning and advanced implementation, reporting benefits such as brainstorming support and confidence-building. At the same time, they face challenges including unclear rationales and difficulty adapting outputs. Students highlight ethical concerns around fairness and misconduct, and call for clearer instructional guidance. Conclusion. GenAI is reshaping SE education in nuanced ways. Our findings underscore the need for scaffolding, ethical policies, and adaptive instructional strategies to ensure that GenAI supports equitable and effective learning.
Mauro Marcelino, Marcos Alves, Bianca Trinkenreich
et al.
[Context] An evidence briefing is a concise and objective transfer medium that can present the main findings of a study to software engineers in the industry. Although practitioners and researchers have deemed Evidence Briefings useful, their production requires manual labor, which may be a significant challenge to their broad adoption. [Goal] The goal of this registered report is to describe an experimental protocol for evaluating LLM-generated evidence briefings for secondary studies in terms of content fidelity, ease of understanding, and usefulness, as perceived by researchers and practitioners, compared to human-made briefings. [Method] We developed an RAG-based LLM tool to generate evidence briefings. We used the tool to automatically generate two evidence briefings that had been manually generated in previous research efforts. We designed a controlled experiment to evaluate how the LLM-generated briefings compare to the human-made ones regarding perceived content fidelity, ease of understanding, and usefulness. [Results] To be reported after the experimental trials. [Conclusion] Depending on the experiment results.
Abstract Objectives Prehospital electrocardiogram (ECG) and administration of aspirin are evidence-based strategies for patients with acute coronary syndrome (ACS). However, emergency medical services (EMS) compliance in patients with suspected ACS varies widely. We sought to understand the barriers to prehospital ECG acquisition and aspirin administration for patients with suspected ACS. Methods In this qualitative study, we interviewed EMS clinicians at three geographically diverse United States (U.S.)-based EMS agencies. We interviewed practicing clinicians and quality and operations leaders at these agencies. Based on the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework, interviews were recorded, transcribed, and analyzed using a grounded qualitative approach with open coding. The Systems Engineering Initiative for Patient Safety (SEIPS) framework and a constant comparison technique were used to identify and refine themes. Results Twenty-five paramedics and 20 additional agency personnel participated. Median age was 41 (IQR: 34–51) years and 13 (29%) were female. Themes were organized using SEIPS and longitudinally through four phases of an EMS call. During the pre-arrival phase, staffing challenges, training quality, and dispatch may anchor EMS clinicians on a diagnosis. During the diagnosis and treatment phase, safety and communication barriers may take priority over care delivery. Additionally, EMS clinicians must allocate assets (e.g. whether to send an advanced life support unit) and financial resources; veteran EMS clinicians identified their experience whereas newer clinicians cited their recent education when making these decisions. Also, diagnostic uncertainty due to increasing patient complexity and atypical presentations contributed to diagnostic errors. During the response and transport phase, the scope of practice limits the use and interpretation of the ECG, with clinicians reporting that liberal use of ECG led to more rapid decision-making. Finally, in the after phase, personnel reported the “psychologically taxing” nature of the job contributing to biases, bad habits, and burnout. Performance feedback was desired for personal development, though currently perceived as infrequent and punitive. Conclusions Multiple, interrelated themes underscored the complexities of delivering evidence-based care to prehospital patients with ACS. Education in ECG interpretation, resource allocation, bias, and enhancing feedback may serve as strategies to address the identified barriers.
A. Lovska, Oleksandr Stanovskyi, Oksana Zharovа
et al.
The object of this study is the processes of occurrence, perception, and redistribution of loads in the body of a gondola car with reinforcing belts in the structure of side walls. In order to improve the strength of side walls of the gondola car body, it is proposed to strengthen them with additional belts. At the same time, it is reinforced with diagonal belts in three sections of the body on the side of the consoles, and in the middle section, 1/3 of the height from the lower strapping, with a horizontal belt. To determine the parameters for the execution of profiles in the reinforcing belts, the calculation of the gondola car body as a rod system was carried out. Based on the resulting values of bending moments, the moment of resistance of the cross-section of the profiles of the reinforcing belts was determined. The calculation of the strength of the body of the gondola car under the main modes of its loads in operation (I and III calculation modes) was carried out. It was found that the resulting stresses were 10.3 % lower than those occurring in a typical design of a gondola car body. The movement of the gondola car in the empty and loaded states was evaluated. A feature of the reported research results is that the improvement of the strength of the side walls of the gondola car body is achieved by increasing the rigidity of its frame. The field of practical use of the results is the engineering industry, in particular, railroad transport. The conditions for the practical application of results are the symmetrical distribution of reinforcing belts along the length of the gondola car body. This study results may contribute to improving the durability of gondola car bodies in operation, and accordingly to reducing costs for unscheduled repairs. Also, the findings could prove useful for designing modern structures of railroad cars
This paper focused on analyzing the influence of unsteady aerodynamic forces on the stability and safety of a specific intercity electric multiple unit (EMU) model subjected to strong crosswinds of 25 m/s on level open tracks. Based on their actual geometric and dynamic parameters, aerodynamic and multi-body dynamic models were respectively established for the 8-car configuration. The <italic>k</italic>-<italic>ω</italic> SST turbulence model was used to calculate the unsteady aerodynamic forces acting on each car. These calculation results were then incorporated into the multi-body dynamic model as external excitations, leading to the development of a fluid-solid coupling simulation model for subsequent calculations related to wind-induced carbody vibrations and safety. The results showed that the aerodynamic forces acting on each car under steady strong crosswinds exhibited obvious unsteady behaviors, with main frequencies varying from 0.5 to 7.0 Hz. The influence of these unsteady aerodynamic forces led to an increase in both the lateral and vertical Sperling indexes of each car, with the lateral Sperling index of the head car reaching 3.3, indicating a considerable deterioration in ride comfort. The collisions between the bodies of the head and tail cars and the lateral stops, under the action of large lateral forces and yaw torques, exacerbated lateral vibrations. Compared with scenarios without aerodynamic loads, the safety indexes of both the head car and tail car under strong crosswinds of 25 m/s were significantly increased, but still remained within safety limits. The wind-induced lateral vibrations of the head car could be mitigated to some extent by increasing the lateral stop clearance and increasing the damping coefficient of both the anti-yaw damper and the secondary lateral damper.
Eduard C. Groen, Kazi Rezoanur Rahman, Nikita Narsinghani
et al.
The farming domain has seen a tremendous shift towards digital solutions. However, capturing farmers' requirements regarding Digital Farming (DF) technology remains a difficult task due to domain-specific challenges. Farmers form a diverse and international crowd of practitioners who use a common pool of agricultural products and services, which means we can consider the possibility of applying Crowd-based Requirements Engineering (CrowdRE) for DF: CrowdRE4DF. We found that online user feedback in this domain is limited, necessitating a way of capturing user feedback from farmers in situ. Our solution, the Farmers' Voice application, uses speech-to-text, Machine Learning (ML), and Web 2.0 technology. A preliminary evaluation with five farmers showed good technology acceptance, and accurate transcription and ML analysis even in noisy farm settings. Our findings help to drive the development of DF technology through in-situ requirements elicitation.
Diana Robinson, Christian Cabrera, Andrew D. Gordon
et al.
What if end users could own the software development lifecycle from conception to deployment using only requirements expressed in language, images, video or audio? We explore this idea, building on the capabilities that generative Artificial Intelligence brings to software generation and maintenance techniques. How could designing software in this way better serve end users? What are the implications of this process for the future of end-user software engineering and the software development lifecycle? We discuss the research needed to bridge the gap between where we are today and these imagined systems of the future.
The object of research is the processes of emergence, perception, and redistribution of loads in the body of a universal railroad gondola with a cladding of corrugated sheets. To improve the strength of the sheets of cladding, it is proposed to strengthen the most loaded area in terms of height (1/3 of the bottom tie-up) with horizontal corrugations. Determination of the geometric parameters of the corrugation is carried out by the moment of resistance of the sheet. The dynamic load of the gondola body with improved cladding was determined by mathematical modeling. The fluctuations of the jump were taken into consideration, that is, the translational movements of the body relative to the vertical axis. The results of the solution of the mathematical model have made it possible to conclude that the studied dynamics indicators are within the permissible limits while the car movement is estimated as "excellent". The calculation was performed of the strength of the body of a gondola with improved cladding. It has been established that the strength of the gondola body under the main operating load modes is ensured. According to the results of calculations of static strength, the calculation was performed of the fatigue strength of the gondola body. It must be said that the fatigue strength of the body cladding increases by 3.7 % compared to the typical one. A feature of the results obtained is that the proposed improvement of the cladding can be carried out not only at the design stage but also during repairs of cars. The scope of practical use of the results includes the engineering industry, in particular railroad transportation. At the same time, the conditions for the practical application of the research results are compliance with the requirements for loading and unloading operations of gondola cars. The results of the current research will contribute to devising recommendations for the design of modern structures of gondolas and for improving the efficiency of their operation.
Introduction. Dynamic loads cause high contact stresses in the materials of wheels and rails, which leads to their wear and accumulation of contact fatigue damage. A number of approaches were proposed for calculations of contact fatigue and prediction of damage accumulation processes. One of them uses the Dang Van criterion. It involves time-consuming iterative procedures, so it is used to assess contact fatigue at individual points of wheel and rail. In order to simplify calculations and obtain more extensive information, the authors proposed an engineering method that allows determining the risk of contact fatigue damage for sub-contact layer nodes represented by a finite element model, with an unlimited number of wheel — rail contact implementations.Materials and methods. The approach is based on the use of the Dang Van criterion with the introduction of some assumptions. The amplitude of the largest shear stresses is taken as the amplitude of the shear stresses. The components of technological and operational residual stresses are set using the results of experimental studies or those obtained by calculation methods. Information about wheel — rail contact conditions was obtained by modeling of the vehicle movement. Calculations were performed considering changes in wheel and rail profiles during the modeling process caused by wear. Results. The test results were obtained for some cases of wheel — rail contact, as well as for determining the contact fatigue damage risk indicator for the wheel and rail with a sufficiently large number of contact implementations. The calculation results are presented in the form of isolines of the value of indicator for the points of the layer adjacent to the tread surfaces.Discussion and conclusion. The proposed approach can be used to determine the risks of contact fatigue damage in railway wheels and rails, in particular, to optimise wheel and rail profiles according to wear and contact fatigue criteria.