Bottlenecks are a major cause of productivity decline in manufacturing systems. In the PDAM metal box industry at PT. MNO, the production process frequently encounters constraints that reduce output. This study employed discrete event simulation using ProModel software. The research stages included field observation, process time data collection, data distribution analysis with Stat::Fit, and model validation through Chi-Square and Mann-Whitney tests. The simulation revealed a significant bottleneck in the assembly workstation, with utilization reaching 93.64%, while production output was limited to 22 units during a 7-hour working shift. In contrast, the finishing station experienced idle time. An improvement scenario involving the addition of one workstation and one operator in the assembly and finishing processes is recommended to reduce production flow imbalances. This study confirms the effectiveness of discrete event simulation as an analytical tool for production system optimization and highlights opportunities for further development through digital twin integration and metaheuristic optimization