Building Self-Healing Enterprise Workflows with AI-Driven Observability and Remediation

Abstract

The increasing complexity of enterprise IT ecosystems demands workflows that are not only resilient but capable of autonomously detecting and recovering from failures. AI-driven observability, combined with automated remediation engines, presents a promising pathway to realize self-healing enterprise systems. This review consolidates current research on anomaly detection, root cause analysis, and AI-based remediation strategies. While notable progress has been made, challenges such as explainability, training data scarcity, and robustness under dynamic environments persist. We conclude by outlining future research directions aimed at enhancing system adaptability, fairness, and trustworthiness, paving the way for more intelligent and resilient enterprise infrastructures.