Kurt G. Vedros

Software Verification and Validation Plan Activities, 2011, Project Number: N6423, SAPHIRE Version 8

The SV&V Plan experienced changes over the past year to bring it into the operational software life cycle of SAPHIRE 8 and to maintain its sections on design features. Peer review of the SVVP with the former IV&V members identified the need for the operational use of metrics as a tool for quality maintenance and improvement. New tests were added to the SVVP to verify the operation of the new design features incorporated into SAPHIRE 8. Other additions to the SVVP were the addition of software metrics and the PDR and CDR processes. Audit support was provided for the NRC Technical Manager and Project Manager for the NRC OIG Audit performed throughout 2011. The SVVP is considered to be an up to date reference and useful roadmap of verification and validation activities going forward.

Preliminary Design Reviews Project: SAPHIRE 8

The Preliminary Design Review (PDR) is intended to be performed at the conceptual phase of a design request. The design request is initiated with a Design Specification document which includes a problem statement, design details, a design checklist and supporting documentation and/or projected sample output. In addition to this, the design specification has a chapter devoted to the completion of the Preliminary Design Review. This document describes the process of documentation of the PDR in the Design Specification.

Critical Design Reviews Project: SAPHIRE 8

The Critical Design Review (CDR) is intended to be performed at the phase of the design request immediately before proceeding to implementation of the design request. The design request is initiated with a Design Specification document which includes a problem statement, design details, a design checklist and supporting documentation and/or projected sample output. The document then records the process through the Preliminary Design Review (PDR) and on to the finalized design specification. In addition to this, the design specification has a chapter devoted to the completion of the CDR. This document describes the process of documentation of the CDR in the Design Specification.

Software Verification and Validation Plan Activities, 2011 -- SAPHIRE 8 Quality Assurance

SV&V Updates through FY11 An official update was released in February 2011. This update was to add a new automated test, to revise the SV&V wording to indicate the operational life cycle that SAPHIRE 8 has entered, and to provide design documentation for Inspection Planning Reports (which evolved into the Plant Risk Information e-Book (PRIB)) and Analysis Speed Improvements. Another update to the SV&V will be released in conjunction with this report in September 2011 to include the following: two new automated tests, the inclusion of metrics into the testing regime and references to the addition of the Preliminary Design Review and Critical Design Review processes.

Risk-Informed Decision Making for High-Consequence Outcomes

The issues faced during the operation of high-consequence facilities or systems require a blended quantitative and qualitative risk-informed approach to decision-making. For example, in the case of the operation of nuclear power plants or manned space missions, a risk-informed metric one might consider is the probability of loss of the plant or failure of the mission, where this probability is balanced by the consequences of the event. However, as we will discuss, one should focus on the observable event itself (e.g., the actual loss or failure) rather than the probability of the event. We will describe the general approach of making probabilistic decisions, including the focus on predictive distributions for outcomes such as a complex system failure.Copyright

A Modern Approach to Bayesian Inference for Risk and Reliability Analysis

Recent years have seen significant advances in the use of risk analysis in a variety of applications. Because risk and reliability models are intended to support these applications, it is critical that inference methods used in these models be robust and technically sound. The inference method described in this paper is that of Bayesian Inference. The inference we describe uses a modern computational approach known as Markov chain Monte Carlo (MCMC). MCMC methods work for simple cases, but more importantly, they work efficiently on very complex cases. Recently, with the advance of computing power and improved analysis algorithms, MCMC is increasingly being used for a variety of Bayesian inference problems. In the paper, an open-source program called OpenBUGS (commonly referred to as WinBUGS) is used to solve the inference problems that are described. The approach that is taken is to provide analysis “building blocks” that can be modified, combined, or used as-is to solve a variety of challenging problems. This paper provides an overview of guidelines for the evaluation of risk and reliability-related data. It is aimed at those familiar with risk and reliability methods and provides a hands-on approach to the investigation and application of a variety of risk and reliability data assessment methods, tools, and techniques.Copyright