The evaluation of maritime waterway risk is required to take further advance measure for reduction of ship collision accident. In general, a risk assessment of ship collision can be estimated by analyzing the level of risk for navigation to prevent collision. This paper proposes a framework of real-time multi-ship collision risk assessment, named Habour Traffic Hazard Map to identify encounter ships, using Marine Traffic Hazard Index from the large amounts of monitored vessels in a surveyed sea area. The methodology uses the influence of a ship to the neighborhood to detect high-risk area in busy waterway from a AIS data. The proposed index is validated by comparison with traffic congestion rate of surveyed area. To illustrate the effectiveness of the framework proposed, an experimental case study has been carried out within Busan Port, Korea. The result is shown as hazard map that is effective and efficient at showing the dynamic level of risk and predicting the potential collision locations. Therefore, this model can improve the safety and security of maritime waterway.