Failure Mode and Effect Analysis: Understanding Cause-Effect Relationships, Study notes of Engineering

Download Failure Mode and Effect Analysis: Understanding Cause-Effect Relationships and more Study notes Engineering in PDF only on Docsity! 1 Failure Mode and Effect Analysis Lecture 2-1 Cause-Effect Tools 2 FMEA Cause-Effect Analysis Cause-Effect Principle Cause-Effect Model Root Cause Analysis 2 3 FMEAProblem Solving • Typical problem-solving practices – Stopping too soon – The need to place blame – Root cause myth – Illusion of common sense and a single reality – Storytelling – Categorical thinking Reference: Apollo Root Cause Analysis, Dean L. Gano 4 FMEACauses if Ineffective Problem Solving • Incomplete problem definition • Unknown causal relationships • Focus on solutions Reference: Apollo Root Cause Analysis, Dean L. Gano 5 9 FMEA POTENTIAL PROBLEM LIKELY CAUSES LIKELY EFFECTS ACTIONS TRIGGER PREVENTIVE CONTINGENT - ADAPTIVE - CORRECTIVE Cause-Effect Model 10 FMEA POTENTIAL PROBLEM LIKELY CAUSES LIKELY EFFECTS ACTIONS TRIGGER PREVENTIVE CONTINGENT - ADAPTIVE - CORRECTIVE BUMP HEAD WALK INTO DOOR PAIN CAN’T SEE GET GLASSES WEAR HELMET REMOVE DOORS Initial Problem 6 11 FMEA POTENTIAL PROBLEM LIKELY CAUSES LIKELY EFFECTS ACTIONS TRIGGER PREVENTIVE CONTINGENT - ADAPTIVE - CORRECTIVE BUMP HEAD WALK INTO DOOR PAIN CAN’T SEE GET GLASSES WEAR HELMET REMOVE DOORS Problem becomes effect Cause becomes new problem 1st “Why” 12 FMEA POTENTIAL PROBLEM LIKELY CAUSES LIKELY EFFECTS ACTIONS TRIGGER PREVENTIVE CONTINGENT - ADAPTIVE - CORRECTIVE BUMP HEAD WALK INTO DOOR PAIN CAN’T SEE GET GLASSES REMOVE DOORSSURGERY NEAR-SIGHTED 7 13 FMEA POTENTIAL PROBLEM LIKELY CAUSES LIKELY EFFECTS ACTIONS TRIGGER PREVENTIVE CONTINGENT - ADAPTIVE - CORRECTIVE BUMP HEAD WALK INTO DOOR PAIN CAN’T SEE GET GLASSES REMOVE DOORSSURGERY NEAR-SIGHTED 2nd “Why” 14 FMEA HAVE WE FOUND ROOT CAUSE ? POTENTIAL PROBLEM LIKELY CAUSES LIKELY EFFECTS ACTIONS TRIGGER PREVENTIVE CONTINGENT - ADAPTIVE - CORRECTIVE BUMP HEAD WALK INTO DOOR FAIL EYE TEST CAN’T SEE CUT OUT STAR TREK GET GLASSES REMOVE DOORS NEAR-SIGHTED TOO MUCH T.V. 10 19 FMEARoot Cause Analysis Method • Once the problem or situation is defined, brainstorm the effect • From each effect, ask “Why” and continue until root cause is reached. Containers leak at mounting screw hole Suppliers leak test may not detect porosity leak Products are failing for contamination Suppliers have different leak test processes No standard process for supplier leak test WHY? WHY? WHY? WHY? 20 FMEARoot Cause Analysis Method - Steps 1. Identify Span of control and Sphere of Influence 2. Create a List of Undesirable Effects (UDE’s) Undesirable effects are negative by their own merit 3. Write the UDE’s on to Post-itTM Notes and place on a large sheet of paper UDE UDE UDE UDE UDE 11 21 FMEAUndesirable Effect (UDE) • Really Exists • Negative on its own merits Workers are Laid Off Net Profit is decreasing My boss is angry with me Absenteeism is rising Engineering and Production can’t agree I am late for work Undesirable on their own merits Neutral or marginal UDE or “fact of life?” 22 FMEACorrelation vs Cause-Effect • Cause-Effect diagram must not contain hidden correlations • What is the difference between Cause- Effect and Correlation? • How can “chasing correlations” lead you astray in a Cause-Effect analysis? 12 23 FMEARoot Cause Analysis Method - Steps 4. Connect UDE’s if one leads to, but not necessarily causes, the other. Lower UDE is the leader, upper is the follower. Test with IF - THEN 5. Determine if any steps are missing between the leader and the follower or if other contributing causes exist. Fill them in. Use ellipses to represent AND situations. UDE UDE UDE UDE UDE UDE UDE UDE UDE UDEIF THEN 24 FMEARoot Cause Analysis Method - Steps 6. Connect other UDE’s and build Cause-Effect chain downward from each. Continue adding missing steps and contributing causes UDE UDE UDE UDE UDE