LEGACY MAINFRAME MODERNIZATION THROUGH MACHINE LEARNING OPERATIONS AND INTELLIGENT DATA PIPELINES

Abstract

Legacy mainframe systems continue to support critical enterprise operations across banking, insurance, healthcare, government, and largescale organizations due to their reliability, security, and transaction processing capabilities. However, aging architectures, limited scalability, complex maintenance requirements, and integration challenges restrict their ability to support modern digital transformation initiatives. This paper proposes a legacy mainframe modernization framework through Machine Learning Operations (MLOps) and intelligent data pipelines by integrating cloud computing, artificial intelligence, automated data engineering, API-based integration, containerized deployment, and predictive analytics. The proposed methodology enables automated data extraction, transformation, migration, intelligent workload analysis, model deployment, and continuous optimization of modernized enterprise applications. Experimental evaluation demonstrates improvements in data processing efficiency, system scalability, operational automation, migration reliability, performance monitoring, and decision-making capability. The proposed framework provides a scalable and intelligent approach for transforming traditional mainframe environments into modern cloud-enabled enterprise platforms while preserving business continuity and operational reliability. Keywords— Mainframe Modernization, MLOps, Intelligent Data Pipelines, Cloud Migration, Machine Learning, Enterprise Transformation, Data Engineering, Legacy Systems.