The rapid adoption of Artificial Intelligence (AI) and Generative AI (GenAI) has redefined the requirements of Continuous Integration and Continuous Delivery (CI/CD) pipelines in project management. Traditional CI/CD frameworks, though effective for conventional software development, often fall short in addressing the complexities of data dependencies, model retraining, dataset drift, and ethical considerations inherent to AI-driven systems. This study proposes and evaluates an analytics-powered, data-driven CI/CD framework tailored for AI project management (AI PM). Using mixed-method research design, the study compares traditional pipelines with analytics-enabled pipelines across key parameters including build frequency, deployment reliability, model performance, stakeholder satisfaction, fairness indices, and energy efficiency. Results reveal significant improvements in pipeline agility, project alignment, GenAI model accuracy, and sustainability, with statistical analyses confirming the robustness of outcomes. The findings emphasize the role of analytics not only as a monitoring tool but as a core driver of stability, transparency, and adaptability in AI delivery pipelines. This research contributes to bridging the gap between DevOps automation and the unique demands of AI PM, offering both theoretical insights and practical strategies for scalable and ethical GenAI deployment.
Maitra et al. (Wed,) studied this question.