Adaptive AI-Driven Platform Modernization: Automating Enterprise Migration Workflows Using Large-Scale Language Models and Policy-Aware Cloud Orchestration

Abstract

The increasing complexity of enterprise IT environments has created significant challenges in modernizing legacy systems while ensuring policy compliance, efficient resource utilization, and minimal operational disruption. This study proposes an adaptive AI-driven platform modernization framework that integrates large language models (LLMs) with policy-aware cloud orchestration to automate enterprise migration workflows. The framework is evaluated through simulations and real-world case studies across small, medium, and large-scale application workflows. Results indicate substantial improvements in workflow efficiency, with task completion times reduced by 33–40%, and policy compliance, with violations reduced by over 90% compared to traditional automation methods. Additionally, resource utilization efficiency increased by 20–36%, and task failures decreased by 80–85%, highlighting enhanced reliability and fault tolerance. The framework demonstrated high scalability, dynamically adapting to changing workloads, resource constraints, and policy requirements. These findings confirm that combining predictive LLM reasoning with intelligent orchestration enables enterprises to conduct complex migrations with predictable performance, operational resilience, and policy adherence. This research contributes a practical methodology for automating enterprise modernization while addressing key challenges in compliance, efficiency, and scalability.