Sarcouncil Journal of Engineering and Computer Sciences

Abstract: Agile product development has become a dominant approach for managing uncertainty and accelerating value delivery in competitive digital markets, yet its effectiveness is often constrained by limited use of data-driven decision support. This study examines how machine learning–driven business intelligence (ML-BI) systems enhance agile product development by transforming operational, customer, and performance data into actionable insights. Using a quantitative, system-oriented research design, data were collected from agile product teams across multiple development cycles and analyzed through supervised machine learning models integrated within business intelligence platforms. The results show that advanced machine learning techniques, particularly ensemble and non-linear models, significantly improve predictive accuracy compared to traditional analytical approaches. Empirical findings further indicate that ML-BI adoption reduces time-to-market, improves product quality, increases customer satisfaction, and enhances sprint reliability and delivery consistency. Distributional and multivariate analyses confirm that ML-BI systems act as integrative mechanisms aligning process efficiency with outcome-oriented objectives. Overall, the study demonstrates that embedding intelligent business intelligence into agile workflows strengthens data-driven agility, supports proactive decision-making, and enables continuous product improvement in dynamic environments.