Risk analysis-based reliability assessment approach under epistemic uncertainty using a dynamic evidential network

Probabilistic modeling is widely used in industrial practices, particularly for assessing complex systems’ safety, risk analysis, and reliability. Conventional risk analysis methodologies generally have a limited ability to deal with dependence, failure behavior, and epistemic uncertainty such as parameter uncertainty. This work proposes a risk-based reliability assessment approach using a dynamic evidential network (DEN). The proposed model integrates Dempster-Shafer theory (DST) for describing parameter uncertainty with a dynamic Bayesian network (DBN) for dependency representation and multi-state system reliability. This approach treats uncertainty propagation across conditional belief mass tables (CBMT). According to the results acquired in an interval, it is possible to analyze the risk like interval theory, and ignoring this uncertainty may lead to prejudiced results. The epistemic uncertainty should be adequately defined before performing the risk analysis. A case study of a level control system is used to highlight the methodology’s ability to capture dynamic changes in the process, uncertainty modeling, and sensitivity analysis that can serve decision making.

Evidential Networks for Evaluating Predictive Reliability of Mechatronics Systems under Epistemic Uncertainties

In reliability predicting field, the probabilistic approaches are based on data relating to the components which can be precisely known and validated by the return of experience REX, but in the case of complex systems with high-reliability precision such as mechatronic systems, uncertainties are inevitable and must be considered in order to predict with a degree of confidence the evaluated reliability. In this paper, firstly we present a brief review of the non-probabilistic approaches. Thereafter we present our methodology for assessing the reliability of the mechatronic system by taking into account the epistemic uncertainties (uncertainties in the reliability model and uncertainties in the reliability parameters) considered as a dynamic hybrid system and characterized by the existence of multi-domain interaction between its failed components. The key point in this study is to use an Evidential Network “EN” based on belief functions and the dynamic Bayesian network. Finally, an application is developed to illustrate the interest of the proposed methodology.