Supporting students in the creation of requirements and functional specifications in interdisciplinary software development projects with the help of AI-based text generation tools

Abstract

In the field of software development, requirements specifications and functional specifications play an important role. These documents are used to specify functional and non-functional requirements and are frequently used in information technology. They facilitate effective communication and understanding between clients and software developers (client and contractor) and contribute significantly to the success of software development projects. In a concrete interdisciplinary student project (lane detection using AI image processing algorithms) from the field of business informatics at TH Wildau, the creation of a requirements specification from a specification using AI-supported (GPT based) text generation tools is examined. The aim is to improve the efficiency and accuracy of the transition of functional and non-functional requirements from the specifications to the functional specification. By using advanced text generation technologies, automated approaches can be developed to produce the requirements specification, reducing human errors and ensuring consistency of documentation. The results of this research focus on the potentials and challenges of text generation for the creation of requirements specifications and thus contribute to the further development of process management in software development.