The increasing use of Large Language Models (LLMs) offers significant opportunities across the engineering lifecycle, including requirements engineering, software development, process optimization, and decision support. Despite this potential, organizations face substantial challenges in assessing the risks associated with LLM use, resulting in inconsistent integration, unknown failure modes, and limited scalability. This paper introduces the LLM Risk Assessment Framework (LRF), a structured approach for evaluating the application of LLMs within Systems Engineering (SE) environments. The framework classifies LLM-based applications along two fundamental dimensions: autonomy, ranging from supportive assistance to fully automated decision making, and impact, reflecting the potential severity of incorrect or misleading model outputs on engineering processes and system elements. By combining these dimensions, the LRF enables consistent determination of corresponding risk levels across the development lifecycle. The resulting classification supports organizations in identifying appropriate validation strategies, levels of human oversight, and required countermeasures to ensure safe and transparent deployment. The framework thereby helps align the rapid evolution of AI technologies with established engineering principles of reliability, traceability, and controlled process integration. Overall, the LRF provides a basis for risk-aware adoption of LLMs in complex engineering environments and represents a first step toward standardized AI assurance practices in systems engineering.
Context: Autism spectrum disorder (ASD) leads to various issues in the everyday life of autistic individuals, often resulting in unemployment and mental health problems. To improve the inclusion of autistic adults, existing studies have highlighted the strengths these individuals possess in comparison to non-autistic individuals, e.g., high attention to detail or excellent logical reasoning skills. If fostered, these strengths could be valuable in software engineering activities, such for identifying specific kinds of bugs in code. However, existing work in SE has primarily studied the challenges of autistic individuals and possible accommodations, with little attention their strengths. Objective: Our goal is to analyse the experiences of autistic individuals in software engineering activities, such as code reviews, with a particular emphasis on strengths. Methods: This study combines Social-Technical Grounded Theory through semi-structured interviews with 16 autistic software engineers and a survey with 49 respondents, including 5 autistic participants. We compare the emerging themes with the theory by Gama et al. on the Effect of Neurodivergent Cognitive Dysfunctions in Software Engineering Performance. Results: Our results suggest that autistic software engineers are often skilled in logical thinking, attention to detail, and hyperfocus in programming; and they enjoy learning new programming languages and programming-related technologies. Confirming previous work, they tend to prefer written communication and remote work. Finally, we report a high comfort level in interacting with AI-based systems. Conclusions: Our findings extend existing work by providing further evidence on the strengths of autistic software engineers.
AI agents have recently shown significant promise in software engineering. Much public attention has been transfixed on the topic of code generation from Large Language Models (LLMs) via a prompt. However, software engineering is much more than programming, and AI agents go far beyond instructions given by a prompt. At the code level, common software tasks include code generation, testing, and program repair. Design level software tasks may include architecture exploration, requirements understanding, and requirements enforcement at the code level. Each of these software tasks involves micro-decisions which can be taken autonomously by an AI agent, aided by program analysis tools. This creates the vision of an AI software engineer, where the AI agent can be seen as a member of a development team. Conceptually, the key to successfully developing trustworthy agentic AI-based software workflows will be to resolve the core difficulty in software engineering - the deciphering and clarification of developer intent. Specification inference, or deciphering the intent, thus lies at the heart of many software tasks, including software maintenance and program repair. A successful deployment of agentic technology into software engineering would involve making conceptual progress in such intent inference via agents. Trusting the AI agent becomes a key aspect, as software engineering becomes more automated. Higher automation also leads to higher volume of code being automatically generated, and then integrated into code-bases. Thus to deal with this explosion, an emerging direction is AI-based verification and validation (V & V) of AI generated code. We posit that agentic software workflows in future will include such AIbased V&V.
Agrawal Naman, Ridwan Shariffdeen, Guanlin Wang
et al.
Large Language Models (LLMs) are becoming increasingly competent across various domains, educators are showing a growing interest in integrating these LLMs into the learning process. Especially in software engineering, LLMs have demonstrated qualitatively better capabilities in code summarization, code generation, and debugging. Despite various research on LLMs for software engineering tasks in practice, limited research captures the benefits of LLMs for pedagogical advancements and their impact on the student learning process. To this extent, we analyze 126 undergraduate students' interaction with an AI assistant during a 13-week semester to understand the benefits of AI for software engineering learning. We analyze the conversations, code generated, code utilized, and the human intervention levels to integrate the code into the code base. Our findings suggest that students prefer ChatGPT over CoPilot. Our analysis also finds that ChatGPT generates responses with lower computational complexity compared to CoPilot. Furthermore, conversational-based interaction helps improve the quality of the code generated compared to auto-generated code. Early adoption of LLMs in software engineering is crucial to remain competitive in the rapidly developing landscape. Hence, the next generation of software engineers must acquire the necessary skills to interact with AI to improve productivity.
Carl Philipp Hohl, Philipp Reis, Tobias Schürmann
et al.
Vehicle data is essential for advancing data-driven development throughout the automotive lifecycle, including requirements engineering, design, verification, and validation, and post-deployment optimization. Developers currently collect data in a decentralized and fragmented manner across simulations, test benches, and real-world driving, resulting in data silos, inconsistent formats, and limited interoperability. This leads to redundant efforts, inefficient integration, and suboptimal use of data. This fragmentation results in data silos, inconsistent storage structures, and limited interoperability, leading to redundant data collection, inefficient integration, and suboptimal application. To address these challenges, this article presents a structured literature review and develops an inductive taxonomy for automotive data. This taxonomy categorizes data according to its sources and applications, improving data accessibility and utilization. The analysis reveals a growing emphasis on real-world driving and machine learning applications while highlighting a critical gap in data availability for requirements engineering. By providing a systematic framework for structuring automotive data, this research contributes to more efficient data management and improved decision-making in the automotive industry.
Brownfield engineering work involving legacy systems, incomplete documentation, and fragmented architectural knowledge poses unique challenges for the effective use of large language models (LLMs). Prior research has largely focused on greenfield or synthetic tasks, leaving a gap in structured workflows for complex, context-heavy environments. This paper introduces the Discover-Define-Deliver (D3) Framework, a disciplined LLM-assisted workflow that combines role-separated prompting strategies with applied best practices for navigating ambiguity in brownfield systems. The framework incorporates a dual-agent prompting architecture in which a Builder model generates candidate outputs and a Reviewer model provides structured critique to improve reliability. I conducted an exploratory survey study with 52 software practitioners who applied the D3 workflow to real-world engineering tasks such as legacy system exploration, documentation reconstruction, and architectural refactoring. Respondents reported perceived improvements in task clarity, documentation quality, and cognitive load, along with self-estimated productivity gains. In this exploratory study, participants reported a weighted average productivity improvement of 26.9%, reduced cognitive load for approximately 77% of participants, and 83% of participants spent less time fixing or rewriting code due to better initial planning with AI. As these findings are self-reported and not derived from controlled experiments, they should be interpreted as preliminary evidence of practitioner sentiment rather than causal effects. The results highlight both the potential and limitations of structured LLM workflows for legacy engineering systems and motivate future controlled evaluations.
Lintang Suminar, Difa Ayu Balqist Ramadhani, Dhimas Endriyanto
The urge for urban public space evolved in increasingly dense cities. Tirtonadi Dam Park is a waterfront-designed urban park located on Bengawan Solo riverbank area. To support the activities, adequate infrastructure must be provided. The objective of this study was to assess Tirtonadi Dam Park's service quality as an urban public space and suggest methods to improve it to promote local communities' economic development. The variables—infrastructure, access & linkage, comfort & images, use & activity—were evaluated and compiled into 16 indicators. Based on user perceptions, Importance Performance Analysis (IPA) was employed to determine the performance and importance levels of predetermined variables. The findings demonstrated that vegetation, infrastructure for the disabled, and drainage and water systems all dropped into the "concentrate here" quadrant, necessitating further development to enhance quality and add more supporting facilities. Furthermore, the quality of security, environmental cleanliness, pedestrian paths, recreational facilities, informal sectors, lighting facilities, and transportation lies in the “keep up the good work” quadrant so that they should be maintained. Enhancing waterfront facilities with leisurely and recreational features can increase the number of visitors. Improving urban areas and incorporating the surrounding communities in all phases of development will be crucial, with the potential to enhance their economic circumstances.
Architecture, Architectural engineering. Structural engineering of buildings
An experimental investigation was conducted at Delft University of Technology to examine the behavior of eight statically loaded extended end plate moment connections up to collapse. The parameters investigated were the end plate thickness (10 mm, 15 mm, and 20 mm) and steel grade of the end plate (S355, S690). While the study was limited to a static test, this investigation intends to analyze the dynamic behavior of the research specimens (FS1 to FS4) using finite element methods. The multi-purpose software Abaqus was used to develop the 3D model. The mechanical properties of these connections, including strength, ductility, and energy dissipation capacity, are examined. The cyclic loading is applied according to the JGJ 101-96 standard specification. The finite element model was validated against experimental tests for both static and dynamic conditions, successfully reproducing moment-rotation curves and simulating ductile damage as well. The results indicate that increased plate thickness corresponds to improved stiffness and strength, while the use of higher steel grades introduces a delayed yield point and may reduce ductility, which must be balanced to optimize performance considering specific design requirements and loading conditions. Our findings align with previous findings and underscore the need for a better understanding of joint behavior under dynamic loading for seismic design since the strain rate at which load is applied significantly affects the material properties, which can significantly affect the performance of blast-resistant structures.
Architectural engineering. Structural engineering of buildings, Structural engineering (General)
Jocelyn Octavia Ongkowiyono, Geby Nathasha Tiffany Budianto, Elizabeth Ferren Armelia
et al.
Eco-friendly architecture (EFA) is a design approach to produce healthy and comfortable buildings. In the pandemic era and in years to come when working from home is a trend, a healthy and comfortable home is crucial because people spend their time at home. Thermal comfort is considered the most significant comfort factor for building occupants, especially at home. This paper reports a study on the relationship between thermal comfort, individual performance, and productivity while working at home. Data was collected qualitatively through observation, heat transfer calculation, and in-depth interviews. This study concludes that a thermally comfortable house that follows the EFA concept is an aspect that influences performance and productivity. However, thermal comfort is not the only aspect related to comfort. Habits and adaptation of occupants to certain conditions also affect comfort, which leads to good performance and productivity. Occupants feel comfortable doing office and household tasks at home due to their adaptation to the surrounding thermal comfort based on their preferences.
Architecture, Architectural engineering. Structural engineering of buildings
Joshua Owotogbe, Indika Kumara, Willem-Jan Van Den Heuvel
et al.
Organizations, particularly medium and large enterprises, typically rely heavily on complex, distributed systems to deliver critical services and products. However, the growing complexity of these systems poses challenges in ensuring service availability, performance, and reliability. Traditional resilience testing methods often fail to capture the intricate interactions and failure modes of modern systems. Chaos Engineering addresses these challenges by proactively testing how systems in production behave under turbulent conditions, allowing developers to uncover and resolve potential issues before they escalate into outages. Though chaos engineering has received growing attention from researchers and practitioners alike, we observed a lack of reviews that synthesize insights from both academic and grey literature. Hence, we conducted a Multivocal Literature Review (MLR) on chaos engineering to address this research gap by systematically analyzing 96 academic and grey literature sources published between January 2016 and April 2024. We first used the chosen sources to derive a unified definition of chaos engineering and to identify key functionalities, components, and adoption drivers. We also developed a taxonomy for chaos engineering platforms and compared the relevant tools using it. Finally, we analyzed the current state of chaos engineering research and identified several open research issues.
The landscape of software engineering is evolving rapidly amidst the digital transformation and the ascendancy of AI, leading to profound shifts in the role and responsibilities of software engineers. This evolution encompasses both immediate changes, such as the adoption of Language Model-based approaches in coding, and deeper shifts driven by the profound societal and environmental impacts of technology. Despite the urgency, there persists a lag in adapting to these evolving roles. By fostering ongoing discourse and reflection on Software Engineers role and responsibilities, this vision paper seeks to cultivate a new generation of software engineers equipped to navigate the complexities and ethical considerations inherent in their evolving profession.
The growing emphasis on studying equity, diversity, and inclusion within software engineering has amplified the need to explore hidden populations within this field. Exploring hidden populations becomes important to obtain invaluable insights into the experiences, challenges, and perspectives of underrepresented groups in software engineering and, therefore, devise strategies to make the software industry more diverse. However, studying these hidden populations presents multifaceted challenges, including the complexities associated with identifying and engaging participants due to their marginalized status. In this paper, we discuss our experiences and lessons learned while conducting multiple studies involving hidden populations in software engineering. We emphasize the importance of recognizing and addressing these challenges within the software engineering research community to foster a more inclusive and comprehensive understanding of diverse populations of software professionals.
Changyu Du, Sebastian Esser, Stavros Nousias
et al.
The conventional BIM authoring process typically requires designers to master complex and tedious modeling commands in order to materialize their design intentions within BIM authoring tools. This additional cognitive burden complicates the design process and hinders the adoption of BIM and model-based design in the AEC (Architecture, Engineering, and Construction) industry. To facilitate the expression of design intentions more intuitively, we propose Text2BIM, an LLM-based multi-agent framework that can generate 3D building models from natural language instructions. This framework orchestrates multiple LLM agents to collaborate and reason, transforming textual user input into imperative code that invokes the BIM authoring tool's APIs, thereby generating editable BIM models with internal layouts, external envelopes, and semantic information directly in the software. Furthermore, a rule-based model checker is introduced into the agentic workflow, utilizing predefined domain knowledge to guide the LLM agents in resolving issues within the generated models and iteratively improving model quality. Extensive experiments were conducted to compare and analyze the performance of three different LLMs under the proposed framework. The evaluation results demonstrate that our approach can effectively generate high-quality, structurally rational building models that are aligned with the abstract concepts specified by user input. Finally, an interactive software prototype was developed to integrate the framework into the BIM authoring software Vectorworks, showcasing the potential of modeling by chatting. The code is available at: https://github.com/dcy0577/Text2BIM
A systemic and configurable model for evaluating the global warming potential (GWP) of cut-stone building materials on the French market is developed and then used to benchmark performances against available low-carbon alternatives (cross-laminated timber (CLT) and slag concrete), for which ranges of GWP allocation models (regulatory and research-driven methods) are used to evaluate underlying uncertainties. Cut-stones stand out for their compliance to three key emission profile criteria in which industrial ecology roadmaps should anchor incentives for material selection: (1) a low margin of uncertainty on GWP values, (2) invariability of GWP magnitudes through time and (3) a high comparative performance with available alternatives. Assuming typically implemented load-bearing wall thicknesses (industry averages of 13, 20 and 24 cm for CLT, concretes and cut-stone, respectively) and high-probability scenarios for all materials, cut-stone assemblies are shown to be 1.43 and 2.73 times less impactful (GWP100) than CLT and slag concrete, respectively. Potential impacts of industrial applications at the parc scale are studied, showing that implementing cut-stone instead of concrete walls on 30% of new French collective housing projects over the 2025–50 period would result in a 2.77 Mt CO2e decrease in the embodied emissions of the parc, against 0.43 for slag concrete and 1.18 for CLT (high-probability). Policy relevance To ensure rapid implementation and tangible climate benefits, industrial roadmaps for transitioning away from carbon-intensive construction practices require swiftly deployable solutions (minimal research and development prerequisites, and the ability to leverage existing workforce in the construction sector) relying on secure supplies (domestically sourced materials, without intersectoral competition), and with high carbon and biodiversity performances readily monitored and ensured (localised low-impact sourcing with high extraction rates). With the potential resource and capabilities to specify and use cut-stone in buildings (i.e. widely available limestone deposits; a rich tradition of industrial know-how for the equipment and mechanisation of quarries for mass production; and industry average building typologies (four-storey height) suitable for masonry design), its use could be part of an overall strategy to reduce embodied emissions of buildings and meet France’s climate mitigation targets.
Architectural engineering. Structural engineering of buildings
Muhammad Awaluddin Hamdy, Baharuddin Hamzah, Ria Wikantari
et al.
Thermal comfort in buildings is determined by several aspects of the climate, such as external and internal wind speeds. Therefore, this research aims to analyze the effect of water elements and vegetation as microclimate modifiers in buildings, to obtain thermal comfort through air velocity and flow analysis. In this context, the field analysis emphasized microclimate parameters. Two cases were also encompassed, namely the interior space of a residential building and a shopping center. By using field measurements with quantitative methods, data were obtained through the analysis of the PMV (thermal comfort index Predicted Mean Vote), PPD (Predicted Percentage of Dissatisfied), and TSV (Thermal Sensation Vote). This experiment was conducted to determine the influential levels of the building water and vegetation on comfort and the thermal environment. Data analysis was also processed using a statistical approach, with airflow being simulated through CFD (Computational Fluid Dynamics) method. The results showed that the air movement occurring in the building to the comfort and thermal environment, through architectural elements, reduced the temperature and humidity in the room. This was due to the heat radiation outside the building, leading to an impact on the effective air temperature for the thermal sensation of visitors. In this case, the movement of air in the building with the placement of architectural elements, such as water, vegetation, and good ventilation, was important for various activities. These activities included the following, (1) providing positive value, (2) improving the quality of the indoor environment, (3) maintaining the stability of the thermal environment at the building scale, and (4) achieving a comfortable thermal sensation.
Architecture, Architectural engineering. Structural engineering of buildings
This paper introduces prompted software engineering (PSE), which integrates prompt engineering to build effective prompts for language-based AI models, to enhance the software development process. PSE enables the use of AI models in software development to produce high-quality software with fewer resources, automating tedious tasks and allowing developers to focus on more innovative aspects. However, effective prompts are necessary to guide software development in generating accurate, relevant, and useful responses, while mitigating risks of misleading outputs. This paper describes how productive prompts should be built throughout the software development cycle.
Heli Järvenpää, Patricia Lago, Justus Bogner
et al.
The rapid adoption of artificial intelligence (AI) and machine learning (ML) has generated growing interest in understanding their environmental impact and the challenges associated with designing environmentally friendly ML-enabled systems. While Green AI research, i.e., research that tries to minimize the energy footprint of AI, is receiving increasing attention, very few concrete guidelines are available on how ML-enabled systems can be designed to be more environmentally sustainable. In this paper, we provide a catalog of 30 green architectural tactics for ML-enabled systems to fill this gap. An architectural tactic is a high-level design technique to improve software quality, in our case environmental sustainability. We derived the tactics from the analysis of 51 peer-reviewed publications that primarily explore Green AI, and validated them using a focus group approach with three experts. The 30 tactics we identified are aimed to serve as an initial reference guide for further exploration into Green AI from a software engineering perspective, and assist in designing sustainable ML-enabled systems. To enhance transparency and facilitate their widespread use and extension, we make the tactics available online in easily consumable formats. Wide-spread adoption of these tactics has the potential to substantially reduce the societal impact of ML-enabled systems regarding their energy and carbon footprint.
Large Language Models (LLMs) have drawn widespread attention and research due to their astounding performance in text generation and reasoning tasks. Derivative products, like ChatGPT, have been extensively deployed and highly sought after. Meanwhile, the evaluation and optimization of LLMs in software engineering tasks, such as code generation, have become a research focus. However, there is still a lack of systematic research on applying and evaluating LLMs in software engineering. Therefore, this paper comprehensively investigate and collate the research and products combining LLMs with software engineering, aiming to answer two questions: (1) What are the current integrations of LLMs with software engineering? (2) Can LLMs effectively handle software engineering tasks? To find the answers, we have collected related literature as extensively as possible from seven mainstream databases and selected 123 timely papers published starting from 2022 for analysis. We have categorized these papers in detail and reviewed the current research status of LLMs from the perspective of seven major software engineering tasks, hoping this will help researchers better grasp the research trends and address the issues when applying LLMs. Meanwhile, we have also organized and presented papers with evaluation content to reveal the performance and effectiveness of LLMs in various software engineering tasks, guiding researchers and developers to optimize.
Sander Thuijsman, Gökhan Kahraman, Alireza Mohamadkhani
et al.
Supervisory control design of cyber-physical systems has many challenges. Model-based systems engineering can address these, with solutions originating from various disciplines. We discuss several tools, each state-of-the-art in its own discipline, offering functionality such as specification, synthesis, and verification. Integrating such mono-disciplinary tools in a multi-disciplinary workflow is a major challenge. We present Analytics as a Service, built on the Arrowhead framework, to connect these tools and make them interoperable. A seamless integration of the tools has been established through a service-oriented architecture: The engineer can easily access the functionality of the tools from a single interface, as translation steps between equivalent models for the respective tools are automated.