Artificial Intelligence–Driven Predictive Analytics Framework for Data-Driven Decision Support in Complex Organizational Systems

Abstract

Artificial intelligence–driven predictive analytics has become an important analytical mechanism for supporting data-driven decision-making processes in complex organizational environments characterized by large volumes of operational data and dynamic decision variables. This study examined the effectiveness of an artificial intelligence–based predictive analytics framework in improving decision-support performance across data-intensive organizational systems. A quantitative research design was adopted to evaluate the relationship between predictive analytics capability and decision-support effectiveness using a dataset of 320 organizational observations collected from sectors including finance, healthcare, manufacturing, retail, and supply chain management. Descriptive statistical analysis indicated that organizations demonstrated moderate to high levels of predictive analytics capability with a mean value of 3.84, while decision-support effectiveness recorded a mean value of 3.91, suggesting a relatively strong presence of data-driven decision practices within the sampled organizations. Multiple regression analysis revealed that predictive analytics capability had a statistically significant positive relationship with decision-support effectiveness, with a standardized coefficient of β = 0.48 and a significance level of p < 0.001. The overall regression model demonstrated strong explanatory power with an R² value of 0.50, indicating that approximately 50% of the variance in decision-support effectiveness was explained by predictive analytics capability, data infrastructure maturity, and predictive model utilization. Subgroup analysis further revealed sectoral differences in predictive analytics performance, with finance, manufacturing, and supply chain organizations achieving higher predictive model performance scores exceeding 4.0 on the analytical scale. Predictive model evaluation results indicated strong analytical reliability, with the model achieving an overall predictive accuracy of 0.86 and an area under the ROC curve value of 0.88, confirming stable predictive performance across validation samples. The findings suggested that organizations possessing advanced predictive analytics infrastructures demonstrated significantly higher decision accuracy and forecasting performance compared with organizations with lower analytics maturity. Overall, the study demonstrated that artificial intelligence–driven predictive analytics frameworks play a significant role in enhancing data-driven decision-support effectiveness by enabling organizations to interpret complex datasets, generate reliable predictive insights, and improve strategic and operational decision-making outcomes within modern data-intensive organizational environments.