KIZAM

Artificial Intelligence in target and requirements management

In the vehicle development process, requirements form the basis for functions, principles of operations and solutions, right up to the manufacturing and testing concepts of the later product. The efficiency in the subsequent steps of the development process is decisively determined by the quality (unambiguousness, completeness, freedom from redundancy) of the requirements.

The increasing number of requirements due to new forms of drive or new regulations leads to a disproportionate increase in the complexity of requirements management, since each new requirement must be checked and coordinated for interactions with the large number of existing requirements. In addition, with each innovation, existing requirements have to be questioned, adapted or extended, and changes in standards can only be tracked with a time-consuming manual effort. Since classic requirements management methods are increasingly reaching their limits, future development processes in vehicle construction can only be designed efficiently if demanding tasks are automated and time-consuming and error-prone manual checks are reduced.

The aim of the KIZAM project is therefore to research the use of AI methods in requirements management in vehicle development. To this end, artificial intelligence should first of all make machine-readable, optimised and consistent requirements available. In addition, the use of AI also allows the fast and efficient comparison of all relevant requirements data and recognises interactions across requirements levels, process interfaces and system boundaries, which significantly increases the quality of the requirements catalogue.

The AI algorithms developed in the project base on existing systems and solutions and can be adapted quickly and easily if the project is successful. The use of AI in requirements management is expected to significantly improve the efficiency and speed of the vehicle development process.