Bayesian Weapon System Reliability Modeling with Cox-Weibull Neural Network
This work addresses predictive maintenance for weapon systems, offering incremental improvements in reliability modeling through novel integration of neural networks and Bayesian methods.
The authors tackled the problem of predicting weapon system reliability by integrating system features into a Cox-Weibull model via neural networks and developing a Bayesian alternative with dropout methods, resulting in improved performance over traditional models like XGBoost and conditional Weibull estimation, as shown by metrics such as ROC AUC and F scores.
We propose to integrate weapon system features (such as weapon system manufacturer, deployment time and location, storage time and location, etc.) into a parameterized Cox-Weibull [1] reliability model via a neural network, like DeepSurv [2], to improve predictive maintenance. In parallel, we develop an alternative Bayesian model by parameterizing the Weibull parameters with a neural network and employing dropout methods such as Monte-Carlo (MC)-dropout for comparative purposes. Due to data collection procedures in weapon system testing we employ a novel interval-censored log-likelihood which incorporates Monte-Carlo Markov Chain (MCMC) [3] sampling of the Weibull parameters during gradient descent optimization. We compare classification metrics such as receiver operator curve (ROC) area under the curve (AUC), precision-recall (PR) AUC, and F scores to show our model generally outperforms traditional powerful models such as XGBoost and the current standard conditional Weibull probability density estimation model.