Answers are grounded in this paper and cite their sources.
Shows improved reliability assessment with bivariate degradation models, suggesting better analysis in complex systems.
Traditional degradation models often rely on a single performance characteristic (PC) to assess reliability, which may overlook complex interactions in systems where two PCs influence reliability. In bivariate degradation modeling, marginal processes are typically modeled using the same stochastic processes, commonly Wiener, inverse Gaussian (IG), or gamma processes. However, these processes may not fully capture scenarios with distinct marginal behaviors, increasing the risk of model misspecification. To address this limitation, this paper introduces a bivariate Tweedie exponential dispersion process (BTEDP) model with a shared frailty factor to capture inter-PC correlation. The BTEDP model encompasses commonly used processes as special cases, offering broad applicability. Parameters are estimated using a combination of saddlepoint approximation and Monte Carlo expectation-maximization (MCEM) algorithms. By approximating the first-passage distribution of the BTEDP with the Birnbaum-Saunders (BS) distribution, we derive the reliability function for products with bivariate PCs. A simulation example and several case studies are performed to demonstrate the flexibility and applicability of the proposed model. Results demonstrate that the BTEDP model has a significant advantage in capturing the complex degradation behavior of two PCs, providing a versatile tool for reliability analysis.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: