Generative Artificial Intelligence for Software Development Using ISO/IEC 29110 Basic Profile: Gaps and Opportunities

Abstract

This study conducted a PRISMA-based systematic literature review of 52 primary studies published between 2022 and 2024 indexed in Scopus, Web of Science, and Springer to examine how generative artificial intelligence can be applied to software development in small and medium-sized software development enterprises adopting the ISO/IEC 29110 Basic Profile. The review identified 6 predominant generative artificial intelligence methods (e.g., large language models and prompt engineering), 32 key tools (e.g., ChatGPT and GitHub Copilot), 5 metric categories (efficiency, productivity, quality, precision, and user satisfaction), and 29 emerging roles (e.g., prompt engineering and machine learning engineers). The analysis reveals notable gaps in the application of generative artificial intelligence within specific process groups of the Basic Profile, particularly in Project Management (Project Plan Execution, Project Assessment and Control, and Project Closure) and Software Implementation (Implementation Initiation and Product Assembly). To address these gaps, this study proposes a structured alignment framework that maps the identified generative artificial intelligence methods, tools, metrics, and roles to the requirements of the ISO/IEC 29110 Basic Profile. Derived from a systematic synthesis of the literature, this framework highlights generative artificial intelligence driven opportunities that support standard-compliant adoption across the software development lifecycle and provides actionable guidance for small and medium-sized software development enterprises. Future research directions include empirical validation through industrial case studies and the integration of software quality assurance mechanisms within generative artificial intelligence enabled processes.

Ähnliche Arbeiten