FinOps–AIOps Fusion: Cost-Aware Anomaly Attribution for Microservice-Based Cloud Systems

Cloud-native microservice systems generate large volumes of operational telemetry and cloud billing data, making it increasingly difficult to balance system reliability and cost efficiency. Existing Artificial Intelligence for IT Operations (AIOps) systems focus on identifying technical anomalies, while Financial Operations (FinOps) systems provide cost visibility without linking costs to operational events. In this paper, a combined FinOps-AIOps hybrid model of cost-conscious anomaly attribution at the microservice level is proposed. The framework combines data from distributed tracing, system monitoring, and cloud billing in a fusion engine where the identified anomalies are correlated with their financial effect. The proposed system was tested within a simulated microservice environment based on Kubernetes with 20 services and a 7-day workload. Experimental results show that the framework achieves 92% anomaly detection accuracy with a precision of 0.90, a recall of 0.88, and an F1-score of 0.89. The model also attributes costs with approximately 85% accuracy, and cost-aware incident prioritization improves by 62% compared to AIOps-only models. The results show a statistically significant improvement in cost optimization efficiency without degrading detection performance (F1 deviation