A Process Model for AI‐Enabled Software Development: A Synthesis From Validation Studies in White Literature
Tugba Gurgen Erdogan,Haluk Altunel,Ayça Kolukısa Tarhan
DOI: https://doi.org/10.1002/smr.2743
2024-11-27
Journal of Software Evolution and Process
Abstract:An overview of the AI‐enabled software development process has been presented, represented as a concrete output of this systematic literature review (SLR) study. It offers a holistic view of development activities for software systems that include AI components. This process model is derived from 14 related validation studies that reported experiences in AI software development, particularly for ML components. It should be considered a first attempt to integrate a comprehensive overview process model. Context With the fast advancement of techniques in artificial intelligence (AI) and of the target infrastructures in the last decades, AI software is becoming an undeniable part of software system projects. As in most cases in history, however, development methods and guides follow the advancements in technology with phase differences. Purpose With an aim to elicit and integrate available evidence from AI software development practices into a process model, this study synthesizes the contributions of the validation studies reported in scientific literature. Method We applied a systematic literature review to retrieve, select, and analyze the primary studies. After a comprehensive and rigorous search and scoping review, we identified 82 studies that make various contributions in relation to AI software development practices. To increase the effectiveness of the synthesis and the usefulness of the outcome, for detailed analysis, we selected 14 primary studies (out of 82) that empirically validated their contributions. Results We carefully reviewed the selected studies that validate proposals on approaches/models, methods/techniques, tasks/phases, lessons learned/best practices, or workflows. We mapped the steps/activities in these proposals with the knowledge areas in SWEBOK, and using the evidence in this mapping and the primary studies, we synthesized a process model that integrates activities, artifacts, and roles for AI‐enabled software system development. Conclusion To the best of our knowledge, this is the first study that proposes such a process model by eliciting and gathering the contributions of the validation studies in a bottom‐up manner. We expect that the output of this synthesis will be input for further research to validate or improve the process model.
computer science, software engineering