Fig. 3. Working process5. Results and Discussion5.1. Accuracy comparison Fig. 4 is the result of HS model predicting the total lifetime failure rate of 98 ship engines. Spots are points of failure with 98 data. Considering the missing data in the early period due to ­warranty repair (as in section 3.1), they are bathtub-formed, as some previous literature has suggested (as in section 2.2). Among the models introduced in section 2.2, Prophet and Arima are selected as baseline models each for its high accuracy and popularity, respectively. Accuracy comparison was performed in two ways considering the model’s application situation. In general, when it is necessary to introduce a new ship engine or predict the ship engine in use, refer to the data of the same engine type. The prediction accuracy of the ship engine is used as a reference to predict the spare parts of the ship engine in use, and the prediction accuracy between the engine type and the ship engine is used as a reference when introducing a new ship engine. Therefore, first (Table 1), the accuracy of ship engine predicted and ship engine actual values were compared, and second (Table 2), the accuracy of engine type predicted and ship engine actual values were compared. RMSE (Root Mean Square Error) was used for the error measure.
Table 1. RMSE (Ship engine predicted VS Ship engine actual)