{"id":13391,"date":"2025-02-03T12:35:01","date_gmt":"2025-02-03T11:35:01","guid":{"rendered":"https:\/\/www.iese.fraunhofer.de\/blog\/?p=13391"},"modified":"2026-03-09T13:42:00","modified_gmt":"2026-03-09T12:42:00","slug":"software-testing-test-case-generation-using-ai-llm","status":"publish","type":"post","link":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/","title":{"rendered":"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements"},"content":{"rendered":"

The verification and validation of software components are based on extensive testing. The required test cases to enable testing are derived from the specified requirements, which are then executed, and the results are compared with the acceptance criteria of the test cases. Even for relatively small systems, the derivation of test cases is a resource-intensive and therefore expensive endeavor. Assuming a conservative estimate of 5\u201310 minutes per test case, it may take more than twenty person-days of effort to write test cases for a system with around 500 requirements. By leveraging Large Language Models, (LLMs), we can increase the efficiency of test case generation.<\/p>\n

The development of complex systems starts with their requirement specifications. For dependable and safety-critical systems, this also includes safety requirements based on the safety analysis of the system. Deriving the test cases manually for these requirements is a time-consuming process. However, by leveraging LLMs, we can improve this process. As input for the LLM, a textual representation of the requirements is used, which is then autonomously transformed into test cases and scenarios in either plain-text format or any formal specification, such as ASAM Open Test Specification. The current best practice of test case reviews by test engineers can ensure the integrity and correctness of these test cases. By using LLMs, the work of the test engineer can be reduced by focusing on formulating test cases for edge cases and reviewing and refining the automatically derived test cases and scenarios.<\/p>\n

\n

Using Large Language Models can significantly reduce the time and costs needed to generate test cases.<\/p>\n<\/div>\n

LLM-based test case generator<\/h3>\n
We developed an LLM-based test case generator and applied it to a \u201cLane Keep Assist\u201d use case. Since LLMs may suffer from inherent uncertainties and quality deficits, our basic architecture includes quality and uncertainty evaluation. The table below shows an excerpt of the basic requirements for the chosen scenario, while Figure 1 illustrates the process of deriving test cases using our LLM-based approach.<\/div>\n
<\/div>\n\n\n\n\n\t\n\n\t\n\t\n\t\n\t\n\t
Requirement ID<\/th>Requirement ID Reqif<\/th>Category<\/th>Requirement Description<\/th>\n<\/tr>\n<\/thead>\n
1.1<\/td>R001<\/td>Lane Detection<\/td>The system shall detect lane markings on the road using cameras and\/or sensors. <\/td>\n<\/tr>\n
1.2<\/td>R002<\/td>Lane Detection<\/td>The system shall identify lane boundaries under various lighting and weather conditions.<\/td>\n<\/tr>\n
2.1<\/td>R003<\/td>Lane Departure Warning <\/td>The system shall provide a warning to the driver if the vehicle is unintentionally drifting out of the lane.<\/td>\n<\/tr>\n
2.2<\/td>R004<\/td>Lane Departure Warning <\/td>The warning shall be provided through visual, auditory, and\/or haptic feedback.<\/td>\n<\/tr>\n
3.1<\/td>R005<\/td>Steering Assistance<\/td>The system shall gently steer the vehicle back into the lane if it detects an unintentional departure.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n
\"Basic
Figure 1: Basic architecture: test case generator for software testing<\/figcaption><\/figure>\n
The requirements as input can be provided in either ReqIF, JSON, or CSV format. The LLM is used to generate test cases based on the given requirements. Since the data within the requirements may be confidential, we utilized our in-house deployed internal LLM tool, which does not expose information.<\/div>\n
\n

To maintain confidentiality of the requirements and the generated test cases, internally deployed LLM models are used.<\/p>\n<\/div>\n

Automated test case generation<\/h3>\n

Large language models generate their output based on prompts. For generating the test cases, one can start with a simple prompt, such as \u201cGenerate the test case for the following requirement.\u201d However, this may not yield the desired result. Studies have shown that LLMs are easier to work with when provided with prompts that are as concise as possible. Using the guidelines of the standards ISO 26262, we settled on a prompt that specifies in detail the expected output characteristics and attributes of a test case specification.<\/p>\n

Quality evaluation<\/h3>\n
Once we obtain the test cases using LLMs, it is essential to evaluate their quality automatically. Even though we plan to have the test cases evaluated by a test engineer, we can use this to judge the quality beforehand, thereby reducing the time needed by the test engineer for evaluation. Or triggering a new generation of the respective test case if quality defects are detected.<\/div>\n
<\/div>\n
For the quality evaluation, we settled on an evaluation based on content availability and correctness. From the standards (ISO 26262, ISO 29119, etc.), we extracted the attributes required for test cases that must be present. We then evaluated each generated test case to determine if the required attributes are present or missing. Based on that, we assessed content completeness using simple and compound metrics, as outlined below in Figure 2 and Figure 3.<\/div>\n

 <\/p>\n

\"Quality
Figure 2: Simple Quality of Conformance metric<\/figcaption><\/figure>\n
\"Quality
Figure 3: Compound Quality of Conformance metric<\/figcaption><\/figure>\n
The correctness of the generated test cases can be evaluated against the required criteria. This can either be done manually or automated. For manual evaluation, the criteria defined in the standards, such as ISO 26262 and ISO 29119, are used. The table below shows some of these criteria.<\/div>\n
<\/div>\n\n\n\n\n\t\n\n\t\n\t\n\t\n\t\n\t
Sl No<\/th>Criteria<\/th>Satisfied Yes\/No<\/th>Comment<\/th>\n<\/tr>\n<\/thead>\n
1<\/td>Language is simple and straightforward<\/td><\/td><\/td>\n<\/tr>\n
2<\/td>Steps are specific and detailed<\/td><\/td><\/td>\n<\/tr>\n
3<\/td>Steps are clear and unambiguous<\/td><\/td><\/td>\n<\/tr>\n
4<\/td>Consistent terminology and format used<\/td><\/td><\/td>\n<\/tr>\n
5<\/td>Inputs are clearly defined<\/td><\/td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n
Simple Quality of Content (QoC) and Compound QoC metrics, along with the correctness criteria, can be used to evaluate the quality of the generated test cases. This can even be automated in instances where human-written test cases (true test cases) are available. These test cases can be used to evaluate correctness using techniques such as fuzzy string matching. However, this can be replaced with more sophisticated techniques or even be based on LLMs.<\/div>\n

Uncertainty evaluation<\/h3>\n
Although LLMs have significantly advanced the domain of natural language processing, they still face challenges related to uncertainty. We evaluated the uncertainty of five LLM models, focusing on those that are deployable in-house. The five models evaluated are: Pixtral-12B, LLaMA2, LLaMA3.1 (8B & 70B), and Gemma:27B. The uncertainty evaluation was conducted using existing datasets, such as GSM 8k (for mathematical reasoning, evaluating the ability to solve arithmetic and algebraic problems), Business Ethics (a subset of the MML dataset that measures the model\u2019s understanding of ethical scenarios in business contexts), and Professional Law (a subset of the MML dataset that focuses on legal principles and professional reasoning). Figure 4 displays the results.<\/div>\n
\"Software
Figure 4: Performance comparison of LLMs<\/figcaption><\/figure>\n
\n

Out of all the evaluated models, LLaMA3.1 (70B) and Pixtral were found to be best performant.<\/p>\n<\/div>\n

Conclusion for software testing<\/h3>\n
In this work, we introduced a method to automatically generate test cases from requirements using LLMs. We further evaluated metrics to assess the quality of the generated test cases and evaluated the uncertainty of the LLMs. As the next step, we plan to automate the translation of test cases into ASAM Open Test Specification format and execute them.<\/div>\n
<\/div>\n
Every company specifies requirements in different ways: We are happy to generate insights on the improvement potential of our approach for your specific safety requirements shapes in case studies. Contact us today<\/a> to learn how a collaboration between Fraunhofer IESE and your company can be operationalized? Drop us a message and we\u2019ll arrange an introductory meeting, where we are happy to discuss your project and priorities.<\/div>\n

 <\/p>\n

References<\/p>\n

    \n
  1. ISO 26262 Road vehicles \u2013 Functional safety<\/li>\n
  2. ISO 29119 Software and systems engineering \u2014 Software testing<\/li>\n
  3. \n
    Agrawal, Pravesh, et al. „Pixtral 12B.“ arXiv preprint arXiv:2410.07073<\/i> (2024).<\/div>\n<\/li>\n
  4. \n
    Touvron, Hugo, et al. „Llama 2: Open foundation and fine-tuned chat models.“ arXiv preprint arXiv:2307.09288<\/i> (2023).<\/div>\n<\/li>\n
  5. \n
    Dubey, Abhimanyu, et al. „The llama 3 herd of models.“ arXiv preprint arXiv:2407.21783<\/i> (2024).<\/div>\n<\/li>\n
  6. \n
    Team, Gemma, et al. „Gemma 2: Improving open language models at a practical size.“ arXiv preprint arXiv:2408.00118<\/i> (2024).<\/div>\n<\/li>\n
  7. \n
    Cobbe, Karl, et al. „Training verifiers to solve math word problems.“ arXiv preprint arXiv:2110.14168<\/i> (2021).<\/div>\n<\/li>\n
  8. \n
    Hendrycks, Dan, et al. „Measuring massive multitask language understanding. (MMLU)“ arXiv preprint arXiv:2009.03300<\/i> (2020).<\/div>\n<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"

    The verification and validation of software components are based on extensive testing. The required test cases to enable testing are derived from the specified requirements, which are then executed, and the results are compared with the acceptance criteria of the…<\/p>\n","protected":false},"author":170,"featured_media":13490,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[177],"tags":[587,744,680],"coauthors":[722,741,691],"class_list":["post-13391","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-kuenstliche-intelligenz","tag-large-language-models-llm","tag-software-testing","tag-virtual-testing"],"yoast_head":"\nSoftware Testing: Using Large Language Models to save effort for test case derivation from safety requirements - Blog des Fraunhofer IESE<\/title>\n<meta name=\"description\" content=\"Software testing: Test case generation using generative ai (LLMs) can significantly reduce the time and costs needed to generate test cases.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements - Blog des Fraunhofer IESE\" \/>\n<meta property=\"og:description\" content=\"Software testing: Test case generation using generative ai (LLMs) can significantly reduce the time and costs needed to generate test cases.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/\" \/>\n<meta property=\"og:site_name\" content=\"Fraunhofer IESE\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/FraunhoferIESE\/\" \/>\n<meta property=\"article:published_time\" content=\"2025-02-03T11:35:01+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-03-09T12:42:00+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"800\" \/>\n\t<meta property=\"og:image:height\" content=\"375\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Jan Reich, Dr. Tobias Braun, Naveed Akram\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@FraunhoferIESE\" \/>\n<meta name=\"twitter:site\" content=\"@FraunhoferIESE\" \/>\n<meta name=\"twitter:label1\" content=\"Verfasst von\" \/>\n\t<meta name=\"twitter:data1\" content=\"Jan Reich\" \/>\n\t<meta name=\"twitter:label2\" content=\"Gesch\u00e4tzte Lesezeit\" \/>\n\t<meta name=\"twitter:data2\" content=\"6\u00a0Minuten\" \/>\n\t<meta name=\"twitter:label3\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data3\" content=\"Jan Reich, Dr. Tobias Braun, Naveed Akram\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/\"},\"author\":{\"name\":\"Jan Reich\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#\\\/schema\\\/person\\\/b44b2eb4f2663f2bf1ec125fa03e84f6\"},\"headline\":\"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements\",\"datePublished\":\"2025-02-03T11:35:01+00:00\",\"dateModified\":\"2026-03-09T12:42:00+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/\"},\"wordCount\":1161,\"publisher\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/software-testing-test-case-generation-using-ai.jpg\",\"keywords\":[\"Large Language Models (LLM)\",\"Software Testing\",\"Virtual Testing\"],\"articleSection\":[\"K\u00fcnstliche Intelligenz\"],\"inLanguage\":\"de\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/\",\"name\":\"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements - Blog des Fraunhofer IESE\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/software-testing-test-case-generation-using-ai.jpg\",\"datePublished\":\"2025-02-03T11:35:01+00:00\",\"dateModified\":\"2026-03-09T12:42:00+00:00\",\"description\":\"Software testing: Test case generation using generative ai (LLMs) can significantly reduce the time and costs needed to generate test cases.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#primaryimage\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/software-testing-test-case-generation-using-ai.jpg\",\"contentUrl\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/01\\\/software-testing-test-case-generation-using-ai.jpg\",\"width\":800,\"height\":375,\"caption\":\"Medicine doctor holding electronic medical record on tablet, Brain testing result, DNA, Digital healthcare and network connection on hologram interface, Science, Medical technology and networking.\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/software-testing-test-case-generation-using-ai-llm\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Startseite\",\"item\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/\",\"name\":\"Fraunhofer IESE\",\"description\":\"Blog des Fraunhofer-Institut f\u00fcr Experimentelles Software Engineering\",\"publisher\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"de\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#organization\",\"name\":\"Fraunhofer IESE\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2016\\\/08\\\/fhg_iese_logo.png\",\"contentUrl\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2016\\\/08\\\/fhg_iese_logo.png\",\"width\":183,\"height\":50,\"caption\":\"Fraunhofer IESE\"},\"image\":{\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/FraunhoferIESE\\\/\",\"https:\\\/\\\/x.com\\\/FraunhoferIESE\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/fraunhoferiese\\\/\",\"https:\\\/\\\/www.youtube.com\\\/c\\\/FraunhoferIESE\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/#\\\/schema\\\/person\\\/b44b2eb4f2663f2bf1ec125fa03e84f6\",\"name\":\"Jan Reich\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2024\\\/09\\\/reich_jan_web-96x96.jpg05a6de04fbbd4390f5055567746b4a23\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2024\\\/09\\\/reich_jan_web-96x96.jpg\",\"contentUrl\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/wp-content\\\/uploads\\\/2024\\\/09\\\/reich_jan_web-96x96.jpg\",\"caption\":\"Jan Reich\"},\"description\":\"Jan Reich has led the safety engineering department at the Fraunhofer Institute for Experimental Software Engineering IESE since 2024. This department focuses on systematic assurance, validation, and safety monitoring methods, which enable the safe market entry of innovative systems. The main focus is on autonomous systems that operate in complex environments. These systems often include innovations like machine learning. Previously, Jan Reich worked as an Expert Scientist at the institute. He handled the topic \\\"Dynamic Assurances for Connected Autonomous Systems.\\\" In the German lighthouse project \\\"PEGASUS Verification and Validation Methods (VVM)\\\" of the VDA flagship initiative \\\"Automated Connected Driving,\\\" he coordinated the safety argumentation framework. This framework aimed to approve highly automated driving systems. -- Jan Reich leitet seit 2024 die Abteilung f\u00fcr Sicherheitstechnik am Fraunhofer-Institut f\u00fcr Experimentelles Software Engineering IESE. Diese Abteilung konzentriert sich auf systematische Sicherungs-, Validierungs- und Sicherheits\u00fcberwachungsmethoden, die den sicheren Markteintritt innovativer Systeme erm\u00f6glichen. Der Schwerpunkt liegt auf autonomen Systemen, die in komplexen Umgebungen eingesetzt werden. Diese Systeme beinhalten oft Innovationen wie maschinelles Lernen. Zuvor war Jan Reich als Fachwissenschaftler am Institut t\u00e4tig. Er befasste sich mit dem Thema \u201eDynamische Versicherungen f\u00fcr vernetzte autonome Systeme\u201c. Im deutschen Leuchtturmprojekt \u201ePEGASUS Verification and Validation Methods (VVM)\u201c der VDA-Leitinitiative \u201eAutomatisiertes vernetztes Fahren\u201c koordinierte er den Sicherheitsargumentationsrahmen. Dieser Rahmen zielte darauf ab, hochautomatisierte Fahrsysteme zuzulassen.\",\"url\":\"https:\\\/\\\/www.iese.fraunhofer.de\\\/blog\\\/author\\\/jan-reich\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements - Blog des Fraunhofer IESE","description":"Software testing: Test case generation using generative ai (LLMs) can significantly reduce the time and costs needed to generate test cases.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/","og_locale":"de_DE","og_type":"article","og_title":"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements - Blog des Fraunhofer IESE","og_description":"Software testing: Test case generation using generative ai (LLMs) can significantly reduce the time and costs needed to generate test cases.","og_url":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/","og_site_name":"Fraunhofer IESE","article_publisher":"https:\/\/www.facebook.com\/FraunhoferIESE\/","article_published_time":"2025-02-03T11:35:01+00:00","article_modified_time":"2026-03-09T12:42:00+00:00","og_image":[{"width":800,"height":375,"url":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg","type":"image\/jpeg"}],"author":"Jan Reich, Dr. Tobias Braun, Naveed Akram","twitter_card":"summary_large_image","twitter_creator":"@FraunhoferIESE","twitter_site":"@FraunhoferIESE","twitter_misc":{"Verfasst von":"Jan Reich","Gesch\u00e4tzte Lesezeit":"6\u00a0Minuten","Written by":"Jan Reich, Dr. Tobias Braun, Naveed Akram"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#article","isPartOf":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/"},"author":{"name":"Jan Reich","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#\/schema\/person\/b44b2eb4f2663f2bf1ec125fa03e84f6"},"headline":"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements","datePublished":"2025-02-03T11:35:01+00:00","dateModified":"2026-03-09T12:42:00+00:00","mainEntityOfPage":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/"},"wordCount":1161,"publisher":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#organization"},"image":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#primaryimage"},"thumbnailUrl":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg","keywords":["Large Language Models (LLM)","Software Testing","Virtual Testing"],"articleSection":["K\u00fcnstliche Intelligenz"],"inLanguage":"de"},{"@type":"WebPage","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/","url":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/","name":"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements - Blog des Fraunhofer IESE","isPartOf":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#primaryimage"},"image":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#primaryimage"},"thumbnailUrl":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg","datePublished":"2025-02-03T11:35:01+00:00","dateModified":"2026-03-09T12:42:00+00:00","description":"Software testing: Test case generation using generative ai (LLMs) can significantly reduce the time and costs needed to generate test cases.","breadcrumb":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/"]}]},{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#primaryimage","url":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg","contentUrl":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg","width":800,"height":375,"caption":"Medicine doctor holding electronic medical record on tablet, Brain testing result, DNA, Digital healthcare and network connection on hologram interface, Science, Medical technology and networking."},{"@type":"BreadcrumbList","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/software-testing-test-case-generation-using-ai-llm\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Startseite","item":"https:\/\/www.iese.fraunhofer.de\/blog\/"},{"@type":"ListItem","position":2,"name":"Software Testing: Using Large Language Models to save effort for test case derivation from safety requirements"}]},{"@type":"WebSite","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#website","url":"https:\/\/www.iese.fraunhofer.de\/blog\/","name":"Fraunhofer IESE","description":"Blog des Fraunhofer-Institut f\u00fcr Experimentelles Software Engineering","publisher":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.iese.fraunhofer.de\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"de"},{"@type":"Organization","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#organization","name":"Fraunhofer IESE","url":"https:\/\/www.iese.fraunhofer.de\/blog\/","logo":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2016\/08\/fhg_iese_logo.png","contentUrl":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2016\/08\/fhg_iese_logo.png","width":183,"height":50,"caption":"Fraunhofer IESE"},"image":{"@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/FraunhoferIESE\/","https:\/\/x.com\/FraunhoferIESE","https:\/\/www.linkedin.com\/company\/fraunhoferiese\/","https:\/\/www.youtube.com\/c\/FraunhoferIESE"]},{"@type":"Person","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/#\/schema\/person\/b44b2eb4f2663f2bf1ec125fa03e84f6","name":"Jan Reich","image":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2024\/09\/reich_jan_web-96x96.jpg05a6de04fbbd4390f5055567746b4a23","url":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2024\/09\/reich_jan_web-96x96.jpg","contentUrl":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2024\/09\/reich_jan_web-96x96.jpg","caption":"Jan Reich"},"description":"Jan Reich has led the safety engineering department at the Fraunhofer Institute for Experimental Software Engineering IESE since 2024. This department focuses on systematic assurance, validation, and safety monitoring methods, which enable the safe market entry of innovative systems. The main focus is on autonomous systems that operate in complex environments. These systems often include innovations like machine learning. Previously, Jan Reich worked as an Expert Scientist at the institute. He handled the topic \"Dynamic Assurances for Connected Autonomous Systems.\" In the German lighthouse project \"PEGASUS Verification and Validation Methods (VVM)\" of the VDA flagship initiative \"Automated Connected Driving,\" he coordinated the safety argumentation framework. This framework aimed to approve highly automated driving systems. -- Jan Reich leitet seit 2024 die Abteilung f\u00fcr Sicherheitstechnik am Fraunhofer-Institut f\u00fcr Experimentelles Software Engineering IESE. Diese Abteilung konzentriert sich auf systematische Sicherungs-, Validierungs- und Sicherheits\u00fcberwachungsmethoden, die den sicheren Markteintritt innovativer Systeme erm\u00f6glichen. Der Schwerpunkt liegt auf autonomen Systemen, die in komplexen Umgebungen eingesetzt werden. Diese Systeme beinhalten oft Innovationen wie maschinelles Lernen. Zuvor war Jan Reich als Fachwissenschaftler am Institut t\u00e4tig. Er befasste sich mit dem Thema \u201eDynamische Versicherungen f\u00fcr vernetzte autonome Systeme\u201c. Im deutschen Leuchtturmprojekt \u201ePEGASUS Verification and Validation Methods (VVM)\u201c der VDA-Leitinitiative \u201eAutomatisiertes vernetztes Fahren\u201c koordinierte er den Sicherheitsargumentationsrahmen. Dieser Rahmen zielte darauf ab, hochautomatisierte Fahrsysteme zuzulassen.","url":"https:\/\/www.iese.fraunhofer.de\/blog\/author\/jan-reich\/"}]}},"jetpack_featured_media_url":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-content\/uploads\/2025\/01\/software-testing-test-case-generation-using-ai.jpg","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/posts\/13391","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/users\/170"}],"replies":[{"embeddable":true,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/comments?post=13391"}],"version-history":[{"count":14,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/posts\/13391\/revisions"}],"predecessor-version":[{"id":13495,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/posts\/13391\/revisions\/13495"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/media\/13490"}],"wp:attachment":[{"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/media?parent=13391"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/categories?post=13391"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/tags?post=13391"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.iese.fraunhofer.de\/blog\/wp-json\/wp\/v2\/coauthors?post=13391"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}