NASA PRA Database

1
NASA PRA Database

• Presented at NASA HQ PRA Workshop
• Rockville, MD
• October 26, 2000

2
Team Members

• Roger Boyer (SAIC-JSC)
• Joe Fragola (SAIC)
• Ed Zampino (GRC)
• Paul Bowerman (JPL)

3
Introduction

• Problem Statement
• NASA has no centralized PRA data set.
• Objective
• To create a data set to support the
  quantification of PRAs for NASA

4
Definitions

• Data Set
• Actual input used to quantify the basic and
  initiating events.
• Data Source
• Existing databases that potentially may be used
  to generate data sets.
• Database
• The combination of data and software to allow the
  collection, reduction, and management of data
  sets.

5
Issues

• Resources (Time/People/)
• Central Point (Website)
• Standard Failure Rates
• Schedule/Timeliness
• Standardize Nomenclature
• Data Set Definition

6
Data Set Definition

• Component/Device Type
• System/ORU/LRU/SRU/etc.
• Failure Mode
• Time/Demand/Degree of Failure
• Application/Environment
• Launch Vehicle
• Low Earth Orbit, Geosync Earth Orbit,
  Interplanetary

7
Similar Databases

• Nuclear Power Industry
• USA (IEEE 500, NPRDS, NUCLAAR)
• Spain (DACNE)
• Sweden (SKI)
• France (SRDF)
• Europe (EuReDa)
• Offshore Oil Industry
• OREDA

8
Relational Database
Device Types
Failure Modes
Applications/ Environment
9
Relational Example
Device Types

• Solenoid Valve
• Fails to Close on Demand
• Shuttle

Failure Modes
Applications/ Environment
10
Step 1

• Obtain Funding to perform Phase 1
• Organize Working Group
• From various NASA Centers with Shuttle, Station,
  LEO, areo, and interplanetary projects
• From consultants w/ PRA database experience
• Select Group Lead

11
Step 2

• Develop Project Plan (within 3 months)
• Phase 1 Develop IEEE 500 Equivalent for NASA
• Phase 2 Establish Website
• Phase 3 Initiate NASA specific data updating
• Note We must use the heritage of the past, but
  always remember that we are forecasting future
  performance.

12
Step 3

• Perform Phase 1 (within 9 months)
• Identify list of component types and failure
  modes for each application
• Determine what additional information is needed
• Identify experts to be used in failure rate
  estimating

13
Database Evolution
Device Type
Applications/ Environment
Heritage
Current
Future
Data Set must allow for Reliability Growth in
Development Systems
14

• Near Term Data System
• Elements
• Taximetrics structure covering all NASA program
  types
• Standard (prior consensus) data set for all
  failure modes for component types and
  applications in taxonomy showing range of
  credibility of failure rates and failure
  probabilities with best estimate point value
• Features
• Industry/NASA working group to generate Near Term
  System
• Use all existing data sources

• Visionary Data System
• Elements
• Data source library
• Standard data set
• Updating software
• Network interface to programmatic data sources
  (e.g., ISS, STS) HST
• Data source search and selection software (data
  mining software)
• Features
• Current data set (standard)
• Continual, periodic updating
• Program specific data set construction on-line

15
Recommendations

• Obtain Funding ASAP to perform Phase 1 (IEEE 500
  equivalent)
• Organize Working Group Select Group Lead
• From various NASA Centers with Shuttle, Station,
  LEO, areo, and interplanetary projects
• From consultants w/ PRA database experience
• Develop Plan and perform Phase 1 by December 2001