Rene L. Bierbaum

Model-based reliability analysis

Testing has typically been a key means of detecting anomalous performance and of providing a foundation for estimating reliability for weapon systems. The objective of model-based reliability analysis (MBRA) is to identify ways to capitalize on the insights gained from physical-response modeling both to supplement the information obtained from testing and to better-understand test results. Five general MBRA processes are identified which can capitalize on physical response modeling results to make both quantitative and qualitative statements about product reliability. A case study that explores 1 of these 5 processes was completed and is described in detail. It had the benefits: MBRA can be used to determine a performance baseline against which current and future test results can be compared; during the design process, MBRA can provide tradeoff studies such that development time and required test assets can be reduced; MBRA can be used to evaluate the impact of production and part changes, as well as aging degradation, if they arise during the product life cycle; and MBRA lays the foundation to evaluate anomalies that are observed in a test program. Typically it has been challenging to determine how anomalous behavior can manifest itself under different-but still valid-conditions. One can use modeling to inject hypothesized behaviors under different conditions and observe the consequences.

Reliability assessment methodology for 1-shot systems

Much of the reliability literature focuses on analysis of continuously operating systems. However, there are many applications for which the system is a 1-shot device or a particular function of the system is 1-shot. For many of these applications the 1-shot device spends much of its life in dormant storage. A common example of this is a weapon system, although there are also numerous other examples such as fire suppression systems, nuclear power plant fail-safe systems, protection or deterrent systems, and various manned and unmanned spacecraft systems. This paper outlines some of the challenges of 1-shot device analysis. An existing 1-shot system reliability analysis approach, that for nuclear weapons, is described. Some of the other important characteristics that influence the approach are also highlighted. Although illustrated in the context of nuclear weapon systems, many of the aspects of the described approach are relevant to other applications.

Using statistical methods to assess a surveillance program

ABSTRACT Three metrics have been been developed to assess the National Nuclear Security Agency (NNSA) Surveillance Program against its objectives of detecting defects, determining margins and validating predictions. The surveillance metrics use statistical methods and are probabilities or confidences that produce quantitive assessments on a 0 to 1 scale—from no confidence that a given data stream achieves its surveillance program objectives to complete confidence that the data stream fulfills the objectives. These metrics may be compared and rolled up to support NNSA Surveillance Program management decisions.

A conceptual model for “inherent reliability” for nuclear weapons

Many people, when thinking about different stages of a particular devices life vis-à-vis defectiveness, use the notion of the “bathtub curve” as a model. However this model is not fully applicable for the class of systems referred to as one-shot or single-shot systems. Key attributes of these systems are outlined in [1]: they typically stay in dormant storage until called upon for one-time use. Common examples of one-shot devices are air-bags in vehicles, fire suppression systems, certain types of safety features in nuclear power plants, missiles, thermal batteries, and some stand-by systems. This paper will focus on a particular example of one-shot systems, nuclear weapons, but the concepts presented are relevant for one-shot devices in general. A new model will be proposed as an alternative to the bathtub curve for one-shot systems. The new model includes two regimes: birth defect dominated and time-dependent dominated. A short discussion of why a bathtub curve might mistakenly be inferred is included. Finally, the relationship between inherent and estimated reliability will be described in the context of this model.

A framework and taxonomy for the design and analysis of margins

There are many statistical challenges in the design and analysis of margin testing for product qualification. To further complicate issues, there are multiple types of margins that can be considered and there are often competing experimental designs to evaluate the various types of margin. There are two major variants of margin that must be addressed for engineered components: performance margin and design margin. They can be differentiated by the specific regions of the requirements space that they address. Performance margin are evaluated within the region where all inputs and environments are within requirements, and it expresses the difference between actual performance and the required performance of the system or component. Design margin expresses the difference between the maximum (or minimum) inputs and environments where the component continues to operate as intended (i.e. all performance requirements are still met), and the required inputs and conditions. The model Performance = f(Inputs, Environments? + ε (1) can be used to help frame the overall set of margin questions. The interdependence of inputs, environments, and outputs should be considered during the course of development in order to identify a complete test program that addresses both performance margin and design margin questions. Statistical methods can be utilized to produce a holistic and efficient program, both for qualitative activities that are designed to reveal margin limiters and for activities where margin quantification is desired. This paper discusses a holistic framework and taxonomy for margin testing and identifies key statistical challenges that may arise in developing such a program.

Defect types and surveillance strategies for one-shot items

This paper will describe some of the challenges and strategies for sampling and testing of complex one-shot systems. A taxonomy for defect types will be offered that informs the nature of the testing and analysis that should be done. In addition, some options for balancing and articulating risk will be summarized for the various surveillance programs described.