Infrastructure operation and maintenance (O&M) in data center environments is increasingly challenged by high alarm volumes, inconsistent incident descriptions and the need for timely root cause analysis (RCA). Traditional monitoring systems can collect and display alarms, but they often rely on engineers to manually interpret alarm context, filter non-actionable records and determine the underlying fault cause. This project designs and implements an infrastructure