Hierarchical spline for time series forecasting: An application to Naval ship engine failure rate

TL;DR

This paper introduces a Bayesian hierarchical B-spline model for time series forecasting of naval ship engine failure rates, effectively handling imbalanced data and sharing structure, leading to accurate lifetime failure predictions.

Contribution

The paper presents a novel Bayesian hierarchical B-spline approach that models failure rates across different levels of ship engines, improving prediction accuracy over traditional methods.

Findings

Accurately predicted failure rates across multiple conditions

Effectively handled imbalanced categories and sharing structures

Demonstrated superior performance over existing models

Abstract

Predicting equipment failure is important because it could improve availability and cut down the operating budget. Previous literature has attempted to model failure rate with bathtub-formed function, Weibull distribution, Bayesian network, or AHP. But these models perform well with a sufficient amount of data and could not incorporate the two salient characteristics; imbalanced category and sharing structure. Hierarchical model has the advantage of partial pooling. The proposed model is based on Bayesian hierarchical B-spline. Time series of the failure rate of 99 Republic of Korea Naval ships are modeled hierarchically, where each layer corresponds to ship engine, engine type, and engine archetype. As a result of the analysis, the suggested model predicted the failure rate of an entire lifetime accurately in multiple situational conditions, such as prior knowledge of the engine.