Engine and steering failure are minor, but it is likely to lead to a second accident if it occurs in a passenger ship or in a sea area where the ship moves frequently. The purpose of this study is to develop techniques to determine the situation of a passenger ship failure in order to support decision making by ship operators or real-time monitoring system operators. We investigated the current status of passengers and marine accidents in Korea. As a result, the number of passengers has been on the rise, while the number of passenger ship accidents relative to the number of registered ship has been consistently high. Therefore, it confirmed that policies are needed to reduce marine accidents on passenger ship. It then analyzed prior research on the investigation into the passenger ship's marine accident and how to determine the anomaly situation. Logistic regression and decision tree techniques were used to determine failure situations, and the judgment technique was developed by applying actual marine accident data. The accuracy of the judgment technique developed was analyzed to be high in the decision tree technique. The decision-making tree technique was applied to waters near Incheon, where the routes of South Korea's coastal and international passenger ships are widely distributed, and showed 91.8 percent accuracy. The developed judgment techniques could provide information to system operators who monitor passenger ships in real-time to make decisions and, if later, supplemented by the limits, could develop into a system to determine passenger ship failures in VTS areas.