Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Managing requirements in the automotive industry entails unique challenges, demanding precision, timing, and adherence to strict safety standards such as ISO 26262. Generative AI can help with developing autonomous vehicles by generating and managing complex requirements that involve safety, functionality, performance, and user experience. These requirements must comply with the standards of ASPICE and ISO 26262, the current benchmarks for software quality and safety in the automotive industry. An AI Copilot for requirements management can assist with various tasks such as rating, summarizing, elaborating, converting, and translating requirements. It can improve the quality, speed, and cost of software development by reducing errors and rework.
A key challenge with using generative AI is the data that a model like GPT4 is trained on. The data is especially of concern when considering the vast amounts of requirements, documents, and internal standards that GPT4 isn't privy to. A key design pattern that can be used to augment the capabilities of an LLM (large language model) like ChatGPT is Retrieval Augmentation Generation (RAG). RAG architecture means that you can constrain natural language processing to your enterprise content sourced from vectorized documents, images, audio, and video.
Combining cognitive search with LLM allows organizations to provide inputs to the LLM prompt but doesn't train the model. In RAG architecture, there's no extra training. The LLM is pretrained using public data, but it generates responses based on information from the retriever.
For more information, see Retrieval Augmented Generation (RAG) in Azure AI Search and Azure/GPT-RAG GitHub Reference.
The Azure OpenAI Use your data feature makes the process for RAG even more efficient. For more information, see Quickstart: Chat with Azure OpenAI models using your own data.
Microsoft partners such Modern Requirements have a solution that can help customers called Copilot4DevOps. With the use of Azure DevOps, Azure OpenAI, and Cognitive Services, Modern Requirements Copilot4DevOps significantly enhances the productivity and efficiency of development teams.
Copilot4DevOps is an AI-powered tool that helps teams automate a large portion of the work. It helps teams to concentrate on strategic decision-making, which leads to faster authoring of requirements, improved analysis, and a reduction in errors.
In addition to individual productivity, Copilot4DevOps offers several organizational benefits. It enhances the quality of documentation, accelerates the time to market, and reduces the overall cost of projects. The AI assistance provided by Copilot4DevOps is a key feature that increases employee satisfaction and productivity. It also prioritizes security, offering the latest updates and protocols from Microsoft and OpenAI. If needed, the AI integration can be turned off at the admin level for privacy and security reasons.
With Copilot4DevOps, you can easily generate high-quality requirements from raw data. It allows you to analyze work item data for clarity, conciseness, coherence, correctness, courtesy, and conviction. It also helps in creating comprehensive use cases or user stories from the actor and user perspective, enhancing project clarity.
Finally, Copilot4DevOps offers the ability to convert your requirements data into Gherkin format, a language used for writing software behavior specifications. Overall, Copilot4DevOps provides a secure and controlled environment for managing your development requirements. It's a powerful tool that can greatly enhance your development process.