Cross-Checking Results of FMEA and FTA

Transcription

1 2016 ARS North America Blue Room, Begins at 9:10 AM, Thursday, June 23rd Cross-Checking Results of FMEA and FTA Nathaniel Ozarin, The Omnicon Group

2 The following presentation was delivered at the: PRESENTATION SLIDES International Applied Reliability Symposium, North America June 21-23, 2016: San Diego, California The International Applied Reliability Symposium (ARS) is intended to be a forum for reliability and maintainability practitioners within industry and government to discuss their success stories and lessons learned regarding the application of reliability techniques to meet real world challenges. Each year, the ARS issues an open "Call for Presentations" at and the presentations delivered at the Symposium are selected on the basis of the presentation proposals received. Although the ARS may edit the presentation materials as needed to make them ready to print, the content of the presentation is solely the responsibility of the author. Publication of these presentation materials in the ARS Proceedings does not imply that the information and methods described in the presentation have been verified or endorsed by the ARS and/or its organizers. The publication of these materials in the ARS presentation format is Copyright 2016 by the ARS, All Rights Reserved.

3 Introduction to Presenter Introduction to Presentation Nat Ozarin (MSEE, MBA) is a Senior Reliability Engineer, 15 years at the Omnicon Group Inc. FMEA and FTA results should be perfectly consistent with exceptions explained but they usually aren t. Omnicon Performs Every Kind of Reliability Analysis and Has Presented Papers on Several of Them Software tools can automate finding inconsistencies and improve the quality of both kinds of analysis. In Particular, Failure Modes and Effects Analysis (FMEA) and nalysis (FTA) on Hardware and Software for every kind of system. Develops Software Tools to Automate Reliability Analyses where None Exist e.g., Bent Pin Analysis (Connector FMEA), and to Expedite other Analyses Slide Number: 2 Humans can then determine legitimate inconsistencies and fix errors. Techniques can also be performed manually (but tediously and with additional human errors). Slide Number: 3 Agenda Top-Down and Bottom-Up Analyses Introduction 5 min Causes of Inconsistencies 10 min FTA Four Kinds of Inconsistences 10 min Using Visual Basic to Find Them 20 min Summary 5 min Questions 10 min Slide Number: 4 FMEA Slide Number: 5

4 Top-Down and Bottom-Up Analyses FMEA-FTA Approach Differences A Transfer Gate in FTA Is a Symbol that Represents the Head of an Internal Fault Tree that Contributes to Other Fault Trees. Transfer Gates Represent Fault Tree Diagrams Usually on Another Sheet, and Often Appear on Many FTA sheets. Slide Number: 6 FMEA Approach Considers Each Lowest-Level Component Failure Modes and Their System-Level Consequences at Each Phase of System Operation. Only One Component Fails at Any One Time (Sometimes One Failed Item Can Take Others with It). One Failed Member of a Redundant Set is a Latent Failure that Has No Immediate System Effects. Analysts Do Not Require Knowledge of System Hazard Analysis and Are (Generally) Not Experts on System Operations. Slide Number: 7 FMEA-FTA Approach Differences 1. Causes of FMEA-FTA Inconsistencies FTA Approach Considers Each Unacceptable System-Level Hazard at Each Phase of System Operation Performed by Different People Different Understanding of System and System-Level Effects Different Descriptions of Failure Modes and System Effects FTA Hazards Identified (Typically) as Catastrophic and Critical ) in a System Hazard Analysis. Multiple Component Failures (including All Members of Redundant Sets) May Contribute to Unacceptable Hazards. Analysts Require Expert Knowledge of System Operations and System Environment. Slide Number: 8 FMEA and FTA Have Different Ground Rules Single Failure vs. Multiple Failures Analyses May Have Different Lowest-Level Components Slide Number: 9

5 2. Four Kinds of FMEA-FTA Inconsistencies 2. Kinds of FMEA-FTA Inconsistencies (1) Missing Failure Modes in FMEA (3) Contradictory Severity Figures FTA Basic Events with Unacceptable Severities that Aren t Listed in the FMEA (2) Orphan Failure Modes in FMEA FMEA Failure Modes that are Basic Events in One or More Fault Trees where Fault Tree Severity is Greater than Basic Event Severity FMEA Failure Modes with Unacceptable Severity that Don t Appear as FTA Basic Events Interpretation of Apparent Discrepancies (1) and (2) Often Just an Omission in FMEA or FTA FMEA and FTA Scope May not be Identical Slide Number: 10 FMEA LRU Fail Mode Slide Number: Kinds of FMEA-FTA Inconsistencies 2. Kinds of FMEA-FTA Inconsistencies Interpretation of Apparent Discrepancies Type (3) Interpretation of Apparent Discrepancies Type (3) FMEA Failure Mode Insufficiently Severe Fault Tree Excessively Severe (not likely if FTA is based on Functional Hazard Analysis with Customer-Approved Severity) (more) FMEA LRU Fail Mode Slide Number: 12 FMEA Failure Mode May be Latent or Hidden by Redundancy, Resulting In Legitimately Lower Severity Incorrect Data Entry Error in Understanding System Severities FMEA LRU Fail Mode Slide Number: 13

6 2. Kinds of FMEA-FTA Inconsistencies 2. Kinds of FMEA-FTA Inconsistencies (4) Contradictory Severity Figures Interpretation of Apparent Discrepancies Type (4) FMEA Failure Modes that are Basic Events to One or More Fault Trees, with Each ssigned Less Severity Than the Basic Event Failure Severity Basic Event with Higher Severity Might be a Transfer Gate in Turn Affected by Another Basic Event with Higher Severity Basic Event Slide Number: 14 Fault Tree C Circuit Breaker Transfer Gate Slide Number: Kinds of FMEA-FTA Inconsistencies 3. Using Visual Basic To Find Inconsistencies Interpretation of Apparent Discrepancies Type (4) Obvious Advantages of Software For Cross-Checking Basic Event with Higher Severity Might be a Component Affecting Transfer Gate with Higher Severity Manually Set Up Worksheets in One Excel Workbook 1 One Worksheet For FMEA Data Fault Tree C Transfer Gate Circuit Breaker Slide Number: One Worksheet for asic Events One Worksheet for Transfer Gates One Worksheet for Software-Generated Results Slide Number: 17

7 FTA and Basic Events Example FTA and Basic Events Example FTA Fault Tree C FTA Fault Tree C 2 Transfer Gate 1 Transfer Gate 1 3 Transfer Gate 2 Transfer Gate n Basic Events Basic Events Basic Event n FMEA Slide Number: 18 Transfer Gate 2 Basic Events Basic Events 1 Transfer Gate n Basic Event n FMEA Slide Number: 19 FMEA Worksheet Example (1200 events) 1 Critical Fault Tree Worksheet Example 2 Use a Table! Slide Number: 20 Slide Number: 21

8 Transfer Gate Dependency Example TGate Worksheet Example 3 Slide Number: 22 Slide Number: 23 Setup and Run Software 4 Setup and Run Software 4 Find Basic Events in FTA but not in FMEA Find Critical Basic Events in FMEA but not in FTA Find Inconsistent Severity Assignments Copy/Paste Get from FTA Software Slide Number: 24 Slide Number: 25

9 Setup and Run Software 4 Setup and Run Software 4 Find Duplicate FMEA IDs Find Severity 1 Basic Events Affecting ONLY Severity 2 Fault trees Explained on Next Slide Paths of Basic Events with Severity 1 Basic Events with Apparently Incorrect Severity Slide Number: 26 Slide Number: 27 Setup and Run Software 4 Completing the Analyses Find Severity 1 Basic Events Affecting ONLY Severity 2 Fault trees Identify Legitimate Inconsistencies Flagged on Output Worksheet Apparent Error List Fault Tree C-21 Severity IS 1 So T23 severity should be 1 So this failure event should be 1 Slide Number: 28 Slide Number: 29

10 Completing the Analyses Summary Correct FMEA and FTA where Real Inconsistences are Flagged on Output Worksheet. All Inconsistencies between a System s FMEA and FTA Should be Identified and Errors Corrected. Update Cross Check Worksheets with Corrected FMTA and FTA Data, then Rerun Cross Check Software. Some Inconsistencies are Legitimate but Each Should be Explained in Worksheet Annotations. Will Take Several Iterations to Eliminate all Real Inconsistencies. Capturing Changes, and Results, Give Running History of Corrections. Upshot: Higher Quality Analyses in Final Report Greatly Improved Consistency, Accuracy, and Thoroughness. Slide Number: 30 Setting up Worksheets for Cross Checking by Software is Tedious but Very Effective Accurate, Fast, and Repeatable with Each Correction or Update. Can be Done Manually but Still Need to Set up Worksheets as You Would for Software. Greatly Improves Analysis Quality. Slide Number: 31 Where to Get More Information Contact Information Omnicon has written a technical paper on this topic and the associated cross check software. Please visit click Contact, and request the paper. (The paper is not currently among those that can be directly downloaded from the website.) Slide Number: 32 Nat Ozarin has been a senior engineering consultant for over fifteen years at The Omnicon Group Inc. His background includes hardware engineering, software engineering and development, systems engineering, and reliability engineering. He received a BSEE from Lehigh University, an MSEE from Polytechnic University of New York, and an MBA from Long Island University. He was named 2009 Reliability Engineer of the Year by the IEEE Reliability Society. nwozarin@omnicongroup.com The Omnicon Group Inc. 50 Engineers Rd Hauppauge NY Slide Number: 33

11 Questions Thank you for your attention. Do you have any questions? Slide Number: 34