Failure Mode and Effect Analysis

1
Failure Mode and Effect Analysis
2
Learning Objectives

• Provide familiarization with FMEA principles and
  techniques.
• Summarize the concepts, definitions, application
  options and relationships with other tools.
• Learn how to integrate FMEA into your Company SOPs

3
Definition of FMEA

FMEA is a systematic design evaluation
procedure whose purpose is to

• 1. recognize and evaluate the potential failure
  modes and causes associated with the designing
  and manufacturing of a new product or a change to
  an existing product,
• 2. identify actions which could eliminate or
  reduce the chance of the potential failure
  occurring,
• 3. document the process.

4
FMEA

• FMEA Failure Modes and Effects Analysis
• FMEA is a systematic approach used to examine
  potential failures and prevent their occurrence.
  It enhances an engineers ability to predict
  problems and provides a system of ranking, or
  prioritization, so the most likely failure modes
  can be addressed.
• FMEA is generally applied during the initial
  stages of a process or product design.
  Brainstorming is used to determine potential
  failure modes, their causes, their severity, and
  their likelihood of occurring.
• FMEA is also a valuable tool for managing tasks
  during defect/failure reduction projects.

5
FMEA is Function-driven

• FMEA begins with a definition of the FUNCTIONS an
  item is supposed to perform. The inputs must
  come from several sources to be effective

6
Background

• Developed in early 60s by NASA to fail-proof
  Apollo missions.
• Adopted in early 70s by US Navy .
• By late 80s, automotive industry had implemented
  FMEA and began requiring suppliers do the same.
  Liability costs were the main driving force.
• Used sporadically throughout industry during
  1980s.
• Adopted by Seagate in 1996. Initial application
  in design centers. Now its time to apply FMEA
  to process applications in Seagate. Six Sigma is
  the catalyst.

7

• NASA used FMEA to identify Single Point Failures
  on Apollo project (SPF no redundancy loss of
  mission). How many did they find?
• 420
• and we thought we had No problems!

8
Types of FMEAs

System FMEA is used to analyze systems and
subsystems in the early concept and design stages.
SYSTEM
Design FMEA is used to analyze products before
they are released to production
DESIGN
Process FMEA is used to analyze manufacturing,
assembly and administrative processes
PROCESS
9
When Is the FMEA Started?
AS EARLY AS POSSIBLE
Guideline

• Do the best you can with what you have.

10
When to Start

• When new systems, products and processes are
  being designed
• When existing designs and processes are being
  changed
• When carry-over designs or processes will be used
  in new applications or environments
• After completing a Problem Solving Study, to
  prevent recurrence of a problem

11
Process FMEA Form

12
Process Failure Mode

• The potential failure mode is the manner in which
  the process could fail to perform its intended
  function.
• The failure mode for a particular operation could
  be a cause in a subsequent (downstream) operation
  or an effect in associated with a potential
  failure in a previous (upstream) operation.

PREVIOUS OPERATION
NEXT OPERATION
FAILURE MODE
EFFECT
CAUSE
13
Process Causes

• Process FMEA considers process variability due
  to

14
Current Controls

• Assessment of the ability of the control to
  detect the failure before the item leaves the
  manufacturing area and ships to the customer.
• Capability of all controls in the process to
  prevent escapes

Process Capability
Sampling
DoE
Testing
Gage RR
SPC
15
Types of Measures

Typically, three items are scored

• SEVERITY
• As it applies to the effects on the local system,
  next level, and end user
• OCCURRENCE
• Likelihood that a specific cause will occur and
  result in a specific failure mode
• DETECTION
• Ability of the current / proposed control
  mechanism to detect and identify the failure mode

16
Severity
17
Occurrence
18
Detection (Escape)
This is best thought of as Escape Potential - the
higher the score, the greater the problem
19
Risk Priority Number

• RPN O x S x D
• Occurrence x
• Severity x
• Detection

20
Basic Steps

• Develop a Strategy
• Form a FMEA team

21
Basic Steps

• 1. Develop a Strategy
• 2. Review the design/process
• Develop process map and identify all process steps

22
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• List all the value-added process
• For each process step, list process inputs
  (process characteristics

23
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes

24
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences of each
  failure mode

25
Basic Steps
1. Develop a Strategy 2. Review the design
/process 3. List functions 4. Brainstorm
potential failure modes 5. List the potential
consequences of each failure mode 6. Assign
severity (SEV) score
26
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences of each
  failure mode
• 6. Assign severity (SEV) score
• 7. Identify the cause(s) of each failure mode.

27
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences (effect) of
  each failure mode
• 6. Assign severity (SEV) score
• 7. Identify the cause(s) of each failure mode.
• 8. Assign occurrence (OCC) scores.

28
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences (effect) of
  each failure mode
• 6. Assign severity (SEV) score
• 7. Identify the potential cause(s) of each
  failure mode.
• 8. Assign occurrence (OCC) scores.
• 9. Identify current controls to detect the
  failure modes.

29
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences (effect) of
  each failure mode
• 6. Assign severity (SEV) score
• 7. Identify the potential cause(s) of each
  failure mode.
• 8. Assign occurrence (OCC) scores.
• 9. Identify current controls to detect the
  failure modes.
• 10. Assign an escaped detection (DET) score for
  each cause and control.

30
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences (effect) of
  each failure mode
• 6. Assign severity (SEV) score
• 7. Identify the potential cause(s) of each
  failure mode.
• 8. Assign occurrence (OCC) scores.
• 9. Identify current controls to detect the
  failure modes.
• 10. Assign an escaped detection (DET) score for
  each cause and control.
• 11. Calculate the Risk Priority Numer (RPN) for
  each line in the FMEA.

31
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences (effect) of
  each failure mode
• 6. Assign severity (SEV) score
• 7. Identify the potential cause(s) of each
  failure mode.
• 8. Assign occurrence (OCC) scores.
• 9. Identify current controls to detect the
  failure modes.
• 10. Assign an escaped detection (DET) score for
  each cause and control.
• 11. Calculate the Risk Priority Numer (RPN) for
  each line in the FMEA.
• 12. Determine the action to be taken.

32
Basic Steps

• 1. Develop a Strategy
• 2. Review the design /process
• 3. List functions
• 4. Brainstorm potential failure modes
• 5. List the potential consequences (effect) of
  each failure mode
• 6. Assign severity (SEV) score
• 7. Identify the potential cause(s) of each
  failure mode.
• 8. Assign occurrence (OCC) scores.
• 9. Identify current controls to detect the
  failure modes.
• 10. Assign an escaped detection (DET) score for
  each cause and control.
• 11. Calculate the Risk Priority Numer (RPN) for
  each line in the FMEA.
• 12. Determine the action to be taken.
• 13. Recalculate the RPNs based on the actions
  plans.

33
Shortcomings of RPN

Effectiveness
RPN
Occurrence
Failure Mode
Severity

• A 8 4 3 96
• B 4 8 3 96

SAME RESULT
34
Action Priority
35
INITIAL PROBLEM
POTENTIAL PROBLEM
WALK INTO DOOR
LIKELY CAUSE
LIKELY EFFECT
BUMP HEAD
CANT SEE
PAIN
TRIGGER
ACTIONS
PREVENTIVE
CONTINGENT - ADAPTIVE -CORRECTIVE
GET GLASSES
WEAR HELMET
REMOVE DOORS
36
1st WHY
PROBLEM BECOMES EFFECT
CAUSE BECOMES NEW PROBLEM
POTENTIAL PROBLEM
WALK INTO DOOR
LIKELY CAUSE
LIKELY EFFECT
BUMP HEAD
CANT SEE
PAIN
TRIGGER
ACTIONS
PREVENTIVE
CONTINGENT - ADAPTIVE -CORRECTIVE
GET GLASSES
WEAR HELMET
REMOVE DOORS
37
1st WHY
CANT SEE
POTENTIAL PROBLEM
WALK INTO DOOR AND
NEAR SIGHTED
LIKELY CAUSE
LIKELY EFFECT
BUMP HEAD
PAIN
TRIGGER
ACTIONS
PREVENTIVE
CONTINGENT - ADAPTIVE -CORRECTIVE
SURGERY
GET GLASSES
38
2ND WHY
CANT SEE
POTENTIAL PROBLEM
WALK INTO DOOR AND
NEAR SIGHTED
LIKELY CAUSE
LIKELY EFFECT
BUMP HEAD
PAIN
TRIGGER
ACTIONS
PREVENTIVE
CONTINGENT - ADAPTIVE -CORRECTIVE
SURGERY
GET GLASSES
39
2ND WHY
NEAR SIGHTED
POTENTIAL PROBLEM
CANT SEE
WALK INTO DOOR AND
LIKELY CAUSE
LIKELY EFFECT
TOO MUCH T.V.
BUMP HEAD
PAIN
TRIGGER
ACTIONS
PREVENTIVE
CONTINGENT - ADAPTIVE -CORRECTIVE
SURGERY
CUT OUT STAR TREK
HAVE WE FOUND ROOT CAUSE?
40
2ND WHY
NEAR SIGHTED
POTENTIAL PROBLEM
CANT SEE
WALK INTO DOOR AND
LIKELY CAUSE
LIKELY EFFECT
TOO MUCH T.V.
BUMP HEAD
PAIN
TRIGGER
ACTIONS
PREVENTIVE
CONTINGENT - ADAPTIVE -CORRECTIVE
SURGERY
CUT OUT STAR TREK
OR GONE TOO FAR !
41
Determining Level of Analysis
PRODUCT
SEAGATE DRIVE STXXXXX
SUBSYSTEMS
SPINDLE MOTOR
DRAWING OR SPEC REFERENCE
DESIGN FMEA
Oklahoma City
EFFECTIVENESS
OCCURRENCE
SEVERITY
S
O
E
RPN
PROCESS DESCRIPTION
POTENTIAL
POTENTIAL
POTENTIAL
CURRENT
FAILURE MODE
EFFECTS OF
CAUSE(S) OF
CONTROLS
FUNCTION
FAILURE
FAILURE
SPINDLE ROTATES
NO SPIN, OR DRIVE RUNS
DRIVE INOPERABLE
FAILURE OF FLEX
RESISTANCE
MEDIA AT FIXED RPM
IN REVERSE
SOLDER JOINT DUE
MEASUREMENT
TO INSUFFICIENT
AT INCOMING
STRAIN RELIEF
INSPECTION
Heres a Seagate Example
Handbook pg. 43
42
Determining Level of Analysis
PRODUCT
SEAGATE DRIVE STXXXXX
SUBSYSTEMS
SPINDLE MOTOR
DRAWING OR SPEC REFERENCE
DESIGN FMEA
Oklahoma City
EFFECTIVENESS
OCCURRENCE
SEVERITY
S
O
E
RPN
PROCESS DESCRIPTION
POTENTIAL
POTENTIAL
POTENTIAL
CURRENT
FAILURE MODE
EFFECTS OF
CAUSE(S) OF
CONTROLS
FUNCTION
FAILURE
FAILURE
SPINDLE ROTATES
NO SPIN, OR DRIVE RUNS
DRIVE INOPERABLE
FAILURE OF FLEX
RESISTANCE
MEDIA AT FIXED RPM
IN REVERSE
SOLDER JOINT DUE
MEASUREMENT
TO INSUFFICIENT
AT INCOMING
STRAIN RELIEF
INSPECTION
Cause becomes Failure Mode
Handbook pg. 43
43
Determining Level of Analysis
PRODUCT
SEAGATE DRIVE STXXXXX
SUBSYSTEMS
SPINDLE MOTOR
DRAWING OR SPEC REFERENCE
DESIGN FMEA
Oklahoma City
EFFECTIVENESS
OCCURRENCE
SEVERITY
S
O
E
RPN
PROCESS DESCRIPTION
POTENTIAL
POTENTIAL
POTENTIAL
CURRENT
FAILURE MODE
EFFECTS OF
CAUSE(S) OF
CONTROLS
FUNCTION
FAILURE
FAILURE
SPINDLE ROTATES
NO SPIN, OR DRIVE RUNS
DRIVE INOPERABLE
FAILURE OF FLEX
RESISTANCE
MEDIA AT FIXED RPM
IN REVERSE
SOLDER JOINT DUE
MEASUREMENT
TO INSUFFICIENT
AT INCOMING
STRAIN RELIEF
INSPECTION
Failure Mode becomes Effect
Handbook pg. 43
44
Determining Level of Analysis
PRODUCT
SEAGATE DRIVE STXXXXX
SUBSYSTEMS
SPINDLE MOTOR
DRAWING OR SPEC REFERENCE
DESIGN FMEA
Oklahoma City
EFFECTIVENESS
OCCURRENCE
SEVERITY
S
O
E
RPN
PROCESS DESCRIPTION
POTENTIAL
POTENTIAL
POTENTIAL
CURRENT
FAILURE MODE
EFFECTS OF
CAUSE(S) OF
CONTROLS
FUNCTION
FAILURE
FAILURE
WHY?
SPINDLE ROTATES
FAILURE OF FLEX SOLDER
NO SPIN, OR DRIVE
MEDIA AT FIXED RPM
JOINT DUE TO
RUNS IN REVERSE
INSUFFICIENT STRAIN
CAUSING DRIVE TO
RELIEF
BE INOPERABLE
PROVIDES A DEEPER LEVEL OF ANALYSIS BY ASKING YOU
FOR THE DESIGN CAUSES AND VERIFICATION OF
INSUFFICIENT STRAIN RELIEF
Handbook pg. 43
45
Whats Wrong With This Picture?

NUMBER OF PROCESS FAILURE CAUSES
46
Actions

A well-developed FMEA will be of limited value
without positive and effective corrective
actions.

• The design or process must be improved based on
  the results of the FMEA study.

47
Elements of FMEA

• Failure Mode Any way in which a process could
  could fail to meet some measurable expectation.
• Effect Assuming a failure does occur, describe
  the effects. List separately each main effect on
  both a downstream operation and the end user.
• Severity Using a scale provided, rate the
  seriousness of the effect. 10 represents worst
  case, 1 represents least severe.
• Causes This is the list of causes and/or
  potential causes of the failure mode.
• Occurrence This is a ranking, on a scale
  provided, of the likelihood of the failure
  occurring. 10 represents near certainty 1
  represents 6 sigma. In the case of a Six Sigma
  project, occurrence is generally derived from
  defect data.
• Current Controls All means of detecting the
  failure before product reaches the end user are
  listed under current controls.
• Effectiveness The effectiveness of each current
  control method is rated on a provided scale from
  1 to 10. A 10 implies the control will not
  detect the presence of a failure a 1 suggests
  detection is nearly certain.

48
FMEA is most effective when

• It is conducted on a timely basis
• and
• It is applied by a product team
• and
• Its results are documented

49
Integrating FMEA into SOPs
Example of how FMEA can be used in SCAR.
Section of SCAR procedure
FMEA can be used to identify the potential cause
of failure and determine whether the current
control is sufficient.
50
Link Tools Integration Tasks to Work Breakdown
Structure
The effort to integrate FMEA into SCAR procedures
should be translated into specific tasks in the
Work Breakdown Structure.
51
End of Topic

• Any Question ?