A STATISTICAL PERFORMANCE EVALUATION OF NETWORK TRAFFIC OPTIMIZATION IN CLOUD COMPUTING ENVIRONMENTS USING PREDICTIVE ARTIFICIAL INTELLIGENCE MODELS

This study evaluates the performance of network traffic optimization in cloud computing environments using predictive artificial intelligence (AI) models combined with rigorous statistical analysis. As cloud infrastructures continue to experience increasing traffic complexity due to applications such as big data analytics, Internet of Things (IoT), and real-time services, efficient traffic management has become critical for maintaining optimal performance and Quality of Service (QoS). Traditional traffic management approaches often fail to adapt to dynamic cloud environments, necessitating the adoption of intelligent, data-driven solutions.A quantitative experimental research design was employed, integrating machine learning techniques with statistical performance evaluation. Predictive models, particularly Random Forest Regression, were developed to forecast network traffic load based on historical performance metrics such as latency, bandwidth utilization, packet arrival rate, and throughput. Simulation experiments using CloudSim, alongside real-world network traces, were conducted to generate and validate datasets. Statistical tools including descriptive statistics, correlation analysis, regression analysis, and ANOVA were applied to assess the effectiveness of the proposed model. The results reveal a strong positive correlation (r = 0.987) between AI-predicted traffic load and network throughput, indicating high predictive accuracy. Regression analysis further shows that the model explains approximately 97.5% of the variance in throughput (R² = 0.975), with statistically significant results (p < 0.001). However, weaker relationships were observed between predicted traffic and other metrics such as latency and bandwidth utilization, suggesting the influence of additional network factors. The study concludes that AI-based predictive models significantly enhance network traffic optimization, particularly in improving throughput and enabling proactive resource allocation. It recommends the integration of AI-driven prediction with advanced network optimization techniques for holistic performance improvement in cloud computing environments.