Fraunhofer Institute for Experimental Software Engineering

Asset Administration Shells (AAS) and their associated submodels, which are essential components of Industry 4.0 are designed to ensure compliance and interoperability. It is crucial that AAS must be seamlessly integrated and compliant with industry standards for users in industrial manufacturing. As submodels serve as the core components of the AAS, ensuring their compliance with relevant standards is imperative. The manual validation process of submodels can be complex, time-consuming, and error-prone. This work builds on a previously developed framework that introduced recursive validation mechanisms, flexible input handling (including direct JSON input, API-based links, and file uploads for AAS and submodels), and adherence to Industrial Digital Twin Association (IDTA) submodel standards. In this paper, we extend the framework by incorporating real-time, MQTT-based event detection to enable automated validation of submodel updates. Furthermore, the proposed system supports the validation of multiple submodels within a single AAS as well as multiple AASs in parallel, enabling large-scale conformity testing. The test results are communicated to the user through an intuitive and user-friendly interface designed to support engineers and IT departments. Evaluation demonstrates that the system maintains an optimal validation time even as the number of tested submodels increases, ensuring the potential for large-scale use. This research improves the efficiency and accuracy of digital twin management, a key concern of Industry 4.0.

With the advent of generative LLMs and their advanced code generation capabilities, some people already envision the end of traditional software engineering, as LLMs may be able to produce high-quality code based solely on the requirements a domain expert feeds into the system. The feasibility of this vision can be assessed by understanding how developers currently incorporate requirements when using LLMs for code generation—a topic that remains largely unexplored. We interviewed 18 practitioners from 14 companies to understand how they (re)use information from requirements and other design artifacts to feed LLMs when generating code. Based on our findings, we propose a theory that explains the processes developers employ and the artifacts they rely on. Our theory suggests that requirements, as typically documented, are too abstract for direct input into LLMs. Instead, they must first be manually decomposed into programming tasks, which are then enriched with design decisions and architectural constraints before being used in prompts. Our study highlights that fundamental RE work is still necessary when LLMs are used to generate code. Our theory is important for contextualizing scientific approaches to automating requirements-centric SE tasks.

The use of machine learning models is increasing rapidly, leading to several safety challenges. One concern is the failure rate obtained from testing, as these are only valid at a certain confidence level. Therefore, many safety assessment models like fault trees or reliability block diagrams cannot be applied in a direct manner. Including confidence in extensions of these models increases the computational effort. Furthermore, their output is not a single-valued number and contains some confidence or uncertainty measurement.To overcome this problem, we present a new approach to retrieve a single-valued upper bound estimation for the failure rate in the case of dependent test data. Consequently, it allows the early usage of classical tool-supported techniques in the system development phase. The estimation is based on concentration inequalities and can also be seen as an optimal confidence level determination. Furthermore, the degree of dependency in the form of the minimal number of partitions in a proper cover is considered. As well as the amount of test data, which makes them more realistic than common independent data-based methods. Our simulation supports the high-confidence validity of these bounds and they therefore present a reasonable estimation for initial assessment in the system design phase.

The use of generative AI-based coding assistants like ChatGPT and Github Copilot is a reality in contemporary software development. Many of these tools are provided as remote APIs. Using third-party APIs raises data privacy and security concerns for client companies, which motivates the use of locallydeployed language models. In this study, we explore the tradeoff between model accuracy and energy consumption, aiming to provide valuable insights to help developers make informed decisions when selecting a language model. We investigate the performance of 18 families of LLMs in typical software development tasks on two real-world infrastructures, a commodity GPU and a powerful AI-specific GPU. Given that deploying LLMs locally requires powerful infrastructure which might not be affordable for everyone, we consider both full-precision and quantized models. Our findings reveal that employing a big LLM with a higher energy budget does not always translate to significantly improved accuracy. Additionally, quantized versions of large models generally offer better efficiency and accuracy compared to full-precision versions of medium-sized ones. Apart from that, not a single model is suitable for all types of software development tasks.

Fault trees are one of the most well-known techniques for safety analysis, which allow both quantitative and qualitative statements about systems. In the classical approach, deterministic failure probabilities for the basic events are necessary in order to obtain quantified results. Classical hardware-related failures and events can be obtained through testing, whereas software-dependent failures are harder to measure and identify, but are still possible to quantify when the implementation is given. In contrast, Machine Learning models lack this information, as their behaviour is not explicitly specified. Up to today, there are very few methods available to judge the worst-case performance of these models and predict their general performance.

Digital transformation is advancing across industries, enabling products, processes, and business models that change the way we communicate, interact, and live. It radically influences the evolution of existing systems of systems (SoSs), such as mobility systems, production systems, energy systems, or cities, that have grown over a long time. In this article, we discuss what this means for the future of software engineering based on the results of a research project called DynaSoS. We present the data collection methods we applied, including interviews, a literature review, and workshops. As one contribution, we propose a classification scheme for deriving and structuring research challenges and directions. The scheme comprises two dimensions: scope and characteristics. The scope motivates and structures the trend toward an increasingly connected world. The characteristics enhance and adapt established SoS characteristics in order to include novel aspects and to better align them with the structuring of research into different research areas or communities. As a second contribution, we present research challenges using the classification scheme. We have observed that a scheme puts research challenges into context, which is needed for interpreting them. Accordingly, we conclude that our proposals contribute to a common understanding and vision for engineering dynamic SoS.

Digital twins are virtual models of physical artefacts that may or may not be synchronously connected, and that can be used to simulate their behavior. They are widely used in several domains such as manufacturing and automotive to enable achieving specific quality goals. In the health domain, so-called digital patient twins have been understood as virtual models of patients generated from population data and/or patient data, including, for example, real-time feedback from wearables. Along with the growing impact of data science technologies like artificial intelligence, novel health data ecosystems centered around digital patient twins could be developed. This paves the way for improved health monitoring and facilitation of personalized therapeutics based on management, analysis, and interpretation of medical data via digital patient twins. The utility and feasibility of digital patient twins in routine medical processes are still limited, despite practical endeavors to create digital twins of physiological functions, single organs, or holistic models. Moreover, reliable simulations for the prediction of individual drug responses are still missing. However, these simulations would be one important milestone for truly personalized therapeutics. Another prerequisite for this would be individualized pharmaceutical manufacturing with subsequent obstacles, such as low automation, scalability, and therefore high costs. Additionally, regulatory challenges must be met thus calling for more digitalization in this area. Therefore, this narrative mini-review provides a discussion on the potentials and limitations of digital patient twins, focusing on their potential bridging function for personalized therapeutics and an individualized pharmaceutical manufacturing while also looking at the regulatory impacts.