An Intelligent Self-Healing AI Framework for Autonomous Error Detection, Root-Cause Diagnosis, and Adaptive Recovery in Distributed Software Systems
International Journal of Creative and Open Research in Engineering and Management · 27.05.2026 · 02 (05) · 1-14
DOI: 10.55041/ijcope.v2i5.800 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 27.05.2026. Verfasst von Rakesh Kumar Rakesh Kumar und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Modern software ecosystems are increasingly dependent on distributed services, cloud- native deployment, microservices, real-time data streams, and intelligent user-facing applications. While these architectures improve scalability and flexibility, they also introduce complex runtime failures such as service unavailability, API degradation, memory exhaustion, state inconsistency, configuration drift, and unpredictable workload spikes. Traditional error- handling mechanisms, including static exception handling, manual debugging, predefined retry policies, and rule-based monitoring, are often insufficient for highly dynamic environments where faults must be detected and resolved before they affect service continuity. This paper proposes an intelligent self-healing AI framework designed to autonomously detect software anomalies, identify probable root causes, select recovery strategies, and continuously improve recovery decisions through feedback learning. The proposed model integrates real-time telemetry collection, hybrid anomaly detection, causal diagnosis, reinforcement-learning-based recovery selection, and a knowledge-driven feedback loop inspired by autonomic computing principles. The framework is evaluated using a simulated distributed application environment containing API failures, memory leaks, latency spikes, database connection errors, and container restarts. Experimental analysis indicates that the proposed approach improves fault detection accuracy, reduces mean time to recovery, and increases service availability compared with traditional rule-based recovery and static monitoring methods. The system achieved 96.4% anomaly detection accuracy, reduced mean time to recovery from 8.7 seconds to 2.1 seconds, and improved recovery success rate to 94.8% under controlled test conditions. The study demonstrates that AI-driven self-healing systems can provide a practical pathway toward resilient, adaptive, and autonomous software operations for cloud, enterprise, and mission-critical applications. Keywords Self-Healing Systems; Artificial Intelligence; Autonomous Error Recovery; Anomaly Detection; Root-Cause Analysis; Reinforcement Learning; Cloud Resilience; Fault Tolerance; MAPE-K; Predictive Maintenance; Distributed Systems; Intelligent Automation.
Zitieren
APA
Kumar, R. K. R., Choudhary, S. C. S., & Shukla, S. S. S. (2026). An Intelligent Self-Healing AI Framework for Autonomous Error Detection, Root-Cause Diagnosis, and Adaptive Recovery in Distributed Software Systems. International Journal of Creative and Open Research in Engineering and Management, 02(05), 1-14. https://doi.org/10.55041/ijcope.v2i5.800
Vancouver
Kumar RKR, Choudhary SCS, Shukla SSS. An Intelligent Self-Healing AI Framework for Autonomous Error Detection, Root-Cause Diagnosis, and Adaptive Recovery in Distributed Software Systems. International Journal of Creative and Open Research in Engineering and Management. 2026;02(05):1-14. doi:10.55041/ijcope.v2i5.800
BibTeX
@article{Kumar2026,
author = {Kumar, Rakesh Kumar Rakesh and Choudhary, Sagar Choudhary Sagar and Shukla, Swatantra Shukla Swatantra},
title = {An Intelligent Self-Healing AI Framework for Autonomous Error Detection, Root-Cause Diagnosis, and Adaptive Recovery in Distributed Software Systems},
journal = {International Journal of Creative and Open Research in Engineering and Management},
year = {2026},
volume = {02},
number = {05},
pages = {1--14},
doi = {10.55041/ijcope.v2i5.800}
}
EUVIMED