FMEARef Guide
FMEARef Guide
Reference Guide
The subject matter contained herein is covered by a copyright owned by Ford Motor Company, Dearborn, MI.
© 2003, Ford Motor Company
Define FMEA FMEA (Failure Mode and Effects Analysis) is a structured group of activities, which:
FMEAs (Design, Process, etc.) are primarily concerned with problem prevention, aimed at
reducing customer dissatisfaction for all customers - consumer, service, etc.
For additional information regarding FMEA processes, tools and software, visit
http://www.quality.ford.com/cpar/fmea/.
Purpose of By implementing FMEAs properly, failures, design/process changes, and workloads are more
FMEA evenly distributed and occur prior to Job #1.
When to There are three basic cases for which FMEAs are generated or updated, each with a different
Generate an scope or focus:
FMEA
• Case 1: New designs (component or system), new technology, or new process. The scope of
the FMEA is the complete design, technology, or process.
Introduction The FMEA Working Model helps team members perform FMEAs. The Working Model
is consistent with the SAE Standard J1739. It is a "flow diagram" of the FMEA thought
process overlaid on an FMEA form.
Although there are minor differences in the forms for the different types of FMEAs, the
thinking process is the same for all types of FMEAs.
The key steps of the Working Model are grouped into three Steps. These Steps are briefly
addressed on the following page and will be discussed in depth as each type of FMEA is
discussed. Using the Working Model is mandatory. However, before using the Working
Model you must:
Step 1
Step 2
Step 3
Working If the Failure Mode cannot be eliminated, then continue with Step 2, which consists of the
Model Step 2 following steps:
Working If the Causes cannot be eliminated, then continue with Step 3, which consists of the
Model Step 3 following steps:
DFMEA The DFMEA supports the design process by reducing the risk of failure by:
Objective
1. Identifying all potential Failure Modes and their Severity
2. Identifying and resolving all Causes by design actions
3. Testing the design to make sure all Failure Modes/Causes are eliminated or their
Occurrence (O) is significantly reduced, prior to release
Team The DFMEA Core Team consists of the design engineer (lead) and the related
manufacturing/process engineer. The Core Team then identifies others to be members of
the Support Team (e.g., customer service, suppliers, global test operations, and corporate
quality).
Scope and The Scope of the DFMEA is the boundary or extent of the analysis and defines what is
Boundary included and excluded.
Diagram
1. Consider a system, sub-system, or component that can be represented as a "block".
2. Represent other components in the high-level assembly, including all system attachments
and mechanisms.
3. Identify relationships with:
a. Other systems
b. Manufacturing/assembly tools
c. Service/customer adjustment
4. Construct a boundary diagram using the "block".
5. Identify the scope by drawing a dotted line around those "blocks" that are to be included.
"Blocks" outside the dotted line are excluded from the analysis of the FMEA.
FPDS Timing DFMEA starts at Strategic Intent <SI> and complete "First Pass" by Program Approval <PA>.
DFMEA Step 1
Define The first step in DFMEA is to define Function, which must be written in a measurable/verifiable
Function format. A description of Function should answer the question: "What is this item supposed to
do?" Function is the design intent or engineering requirement of the item being analyzed.
Identify Failure For each identified Function, you must identify what can go wrong. Failure Modes fall
Modes of the within these four categories:
Function
• No Function—not operational
• Partial/Over/Degraded Function—not all of the function is operating
• Intermittent Function—occasionally does not function
• Unintended Function—unexpected function occurs due to system interaction
Identify the Having documented the Failure Modes, we must now consider the Effects of each Failure Mode.
Effects of the The use of the DFMEA Effects List is highly recommended to evaluate the effects on the part,
Failure Modes the high-level assembly, the system, the vehicle, the customer, and governmental regulations.
Assess the Once the effects have been documented in the Effects Column of the FMEA form, they need
Severity of the to be evaluated for Severity. Severity is linked to the effect of failure, and each effect
Effect on a identified in the list must be assigned a Severity rating. These should be noted in parentheses
scale of 1 to 10 following each Effect. Severity is a relative ranking, within the scope of the individual
FMEA. Each effect must be assigned a Severity rating using the Severity Rating Table.
Selection of The highest Severity identified in the Effects column is entered in the Severity Column.
Severity and
Classification For Severity values of 9 or 10, where safety and/or government regulations are affected:
Note: For a particular Failure Mode, there is only one Severity rating in the Severity column of
the FMEA form.
Recommend The FMEA team tries to eliminate the Failure Mode. If it is not possible, continue to Step 2, Cause
Actions of Failure Mode.
Address Cause This Step addresses the Cause (or design weakness), which results in the Failure Mode.
Identify all of Potential Cause of failure is defined as an indication of a design weakness, the consequence of
the Causes of which is the Failure Mode. Hence, Causes should be written as a design weakness, which
the Failure should be corrected.
Mode
When brainstorming potential Causes of each failure, begin with the following assumptions:
Use Fishbone Diagram or Why Ladder to identify all the Causes, and remember that Root
Causes:
Note: The Root Cause is the reason for the primary non-conformance and is the item that
requires change to achieve permanent preventive/corrective action.
Assess how After all the Causes of the Failure Modes have been identified, determine how often each
often the Cause occurs on a scale of 1 to 10. Occurrence can be defined as the likelihood that a
Causes occur specific Cause will occur during the design life (e.g., 150,000 miles or 10 years). Refer to
the following Occurrence Rating Table that is used to estimate the failure rate and/or
criteria to develop a rating for each Cause.
Classification When a Failure Mode/Cause combination has a Severity rating of 5 to 8 and an Occurrence rating
of 4 or higher, a potential Significant Characteristic exists. In the classification column, enter the
letters "YS" when a potential Significant Characteristic is identified. At this time, the PFMEA
team should start considering special plant inspections to catch "YS" related errors.
Actions to Address all Causes by recommending design actions, which will either eliminate the Cause or
eliminate the lower its occurrence. If no action is taken, explain why. Remember that Causes address Failure
Cause Modes, not Effects.
DFMEA Step 3
Current Design There are two types of Design Controls/features to consider:
Controls
• Prevention: Prevent the Cause/Mechanism or Failure Mode/Effect from occurring, or reduce
the rate of Occurrence. These are design actions.
• Detection: Detect the Cause/Mechanism or the Failure Mode, either by analytical or physical
methods, before the item is released to production. These are tests.
Detection Detection rating is the team's assessment of the test being conducted to identify Cause and/or Failure
Mode. Detection is a relative ranking within the scope of the individual FMEA. In order to achieve a
lower ranking, generally the planned design control (e.g., validation, and/or verification activities)
has to be improved. Detection ratings are given in the following table.
Note: When several detection controls (or tests) are listed for a Cause, use the best (or lowest) rating.
Risk Priority Compute the Risk Priority Number (RPN) which equals Severity times Occurrence times
Number (RPN) Detection (S x O x D), for each Cause and enter in the RPN column.
Recommended The purpose of recommending actions in a DFMEA is to eliminate potential Failure Modes or
Actions reduce their occurrence. The team should prioritize actions based on those Failure Modes with:
Robustness Linkages
Introduction Robustness Linkages have been added to the FMEA process to significantly reduce vehicle
campaigns, enhance the corporate image, reduce warranty claims, and increase customer
satisfaction. These Robustness Linkages primarily emanate from the P-diagram, which identifies
the five Noise Factors. These factors need to be addressed early to make the design insensitive to
the noise factors. This is the essence of Robustness. It is the engineer's responsibility to ensure
that the Robustness Linkages are captured in the engineering documentation.
P-Diagram
Noise Factors
1. Piece-to-Piece Variation
2. Changes Over Time/Mileage
3. Customer Usage & Duty Cycles
4. Environmental
5. System Interactions
Response
Signal Factor System (Ideal
Function)
Interface The interface matrix is a recommended Robustness tool that acts as an input to a DFMEA.
Matrix The interface matrix identifies and quantifies the strength of system interaction by:
The interface matrix is an input to the Potential Causes/Mechanisms Failure column of the
DFMEA and also to the P-diagram.
Headlamp Housing
Grill Opening Panel
Hood
Screws/Fasteners
Headlamp
Wiring/Electrical System
Connector
Body
-1 -2 -1 -1 2 -2
Grill Opening Panel
Hood -1 -2 -1 -1 12
Headlamp Housing -1 -1 -1 -1 2
Headlamp 2 -2 -1 -1 2
Screws/Fasteners
Wiring/Electrical System
Connector
Body
P E P: Physically touching
I M I: Information Exchange
E: Energy Transfer
M: Material Exchange
Numbers in each corner represent the above interface types, with values denoting the
following:
Robustness The diagram below shows how the FMEA process has a robustness focus. The Robustness
Linkage tools are inputs into the DFMEA process:
Diagram
FMEA with
Robustness
Linkages
Note: Boundary Diagram, Interface Matrix, and P-Diagram should be completed before starting
the FMEA process. The Robustness and Reliability checklist is the output of the FMEA.
PFMEA PFMEA:
Objective
• Is a technique employed to analyze the manufacturing and assembly processes
• Focuses on eliminating potential Failure Modes of the manufacturing and assembly processes
PFMEA consists of three Steps similar to DFMEA except that now the focus is on process
function/purpose as opposed to component function.
Team The PFMEA Core Team includes representatives from manufacturing/process engineering and the
appropriate design engineer. The Core Team identifies others to be members of the Support Team
(e.g., technical specialists, suppliers, maintenance, production, and next process engineer).
Without properly determining the scope, the FMEA may either go too far or not far enough. The
process flow must be created for the PFMEA.
Characteristic A part characteristic is a feature about the part, which when compromised, generates a Failure
Matrix Mode. It may be a dimension, such as a hole’s inside diameter, or it may be a specification such as
material composition. Once all the Failure Mode’s associated part characteristics are identified,
they must be summarized and communicated to the corresponding PFMEA team. These
characteristics are then placed on a Characteristic Matrix, aligning them individually with each of
the process steps where they can potentially be compromised. It is important to understand that
anytime one of these part characteristics is compromised, the associated DFMEA Failure Mode is
reintroduced along with its effects. Therefore, the part characteristic is the link between the
DFMEA and PFMEA.
FPDS Timing PFMEA should start at Strategic Conformation <SC>, and complete First Pass by Product
Readiness <PR>.
PFMEA Step 1
Process The function/purpose of a process operation is what is achieved at that operation. It should:
Function/
Requirement • Describe what the operation is there to do
• Be precisely expressed in a measurable format
Identify For each identified process purpose, the PFMEA team must identify what can go
Potential wrong. Potential Failure Mode is defined as the manner in which the process could
Failure Mode potentially fail to meet the process requirements and/or design intent as described in the
Process Function/Requirements column. Like DFMEA, Failure Modes fall within four
groupings that are useful as thought starters:
• No Function—not operational
• Partial/Over/Degraded Function—not all of the function is operating
• Intermittent Function—occasionally does not function
• Unintended Function—unexpected function occurs due to system interaction
Effects of Once the team has identified and listed the potential Failure Modes, it should next consider the
Failure Mode effects of each Failure Mode.
Potential Effects of Failure are defined as the effects of the Failure Mode on the customer(s). The
customer(s) in this context could be the next operation, subsequent operations or locations, the
dealer, and/or the vehicle owner. Use the PFMEA Effects List to identify all effects.
Severity Once the Effects have been documented, they need to be evaluated for Severity (refer to the
Severity Table on the next page). Severity is linked to the Effect of failure, and each effect
identified in the list must be assigned a Severity rating. These should be noted in parentheses
following each Effect. Failure in a process can affect the product (customer) and/or the process
(manufacturing/assembly). The team must analyze the nature of the Effect using the PFMEA
Severity Table.
Selection of The highest Severity identified in the Effects column is entered in the Severity column.
Severity and
Classification • For product related Severity of 9 or 10, enter ∇ (Critcal Characteritics or Inverted Delta) in
the classification column.
• For process related Severity of 9 or 10, enter OS (Operator Safety) in classification column.
Note: For a particular Failure Mode, there is only one Severity rating in the Classification column
of the FMEA form.
Recommended Recommending actions to eliminate the Failure Mode completes Step 1 of the Working Model.
Actions If the Failure Mode cannot be eliminated, continue with the analysis in Step 2.
Introduction PFMEA Step 2 begins with the same Failure Mode. The Causes are a process and/or
design weakness, which results in the Failure Mode.
Identifying Potential Cause of failure is described in terms of something that can be corrected or
Causes controlled. When brainstorming potential Causes, there are two assumptions to begin with.
These two assumptions are not mutually exclusive, but each must be considered sequentially.
Rating the Occurrence is the likelihood that a Specific Cause/Mechanism will occur. Occurrence Ratings
Occurrence for in PFMEA are the same as they are in DFMEA. Refer to the DFMEA/PFMEA Occurrence
each Cause Table on page 8.
Classification Once Occurrence has been determined for each Cause, the need for a Special
Characteristic has to be established. For customer/product related Severity of 5 to 8 and
Occurrence of 4 to 10, enter SC (Significant Characteristic) in the Classification column.
For manufacturing/process related Severity of 5 to 8 and Occurrence of 4 to 10, enter HI
(High Impact) in the Classification column.
Note: Criticality is defined as the product of Severity x Occurrence (S x O).
Recommended All Causes identified in the PFMEA should be addressed. All ∇ and SC related Causes
Actions must have Special Controls (inspections) in the Plant to assure no defective parts leave the
plant. If the Failure Mode and their Causes cannot be eliminated, proceed to Step 3.
PFMEA Step 3
Current There are two types of process controls/features to consider:
Process
• Type 1: Prevention: Prevent the Cause/Mechanism or Failure Mode from occurring,
Controls
or reduce their rate of Occurrence – could be a design and/or process actions.
• Type 2: Detection: Detect the Cause/Mechanism and lead to corrective actions –
generally this is an inspection conducted in the plant.
How to Identify The preferred approach is to first use Type 1 controls if possible. The initial Occurrence
Process rankings will be affected by the Type 1 controls provided they are integrated as part of the
Controls process intent. The initial Detection rankings will be based on the Type 2 current process
controls, planned for production operation. The Occurrence ranking may need to be
revised after the identification of Type 1 prevention controls.
Detection Detection rating is the team's assessment of the inspection being conducted in the plant to
identify Cause and/or Failure Mode. If several detections (or inspections) controls are listed
for a particular Cause, use the best (or lowest) rating. Manual/visual inspection has poor
ratings (6-10). Detections for ∇ and SC are called Special Controls and should be excellent
(non manual) inspections with low ratings. The objective of Special Controls is to ensure
that no defective parts leave the plant. Refer to the following table for detection ratings.
Risk Priority Compute the Risk Priority Number (RPN) that equals Severity times Occurrence times
Number (RPN) Detection (S x O x D), for each Cause and enter in the RPN column.
Recommended The purpose of recommend actions in a PFMEA is to eliminate potential Failure Modes or
Actions reduce their occurrence. The team should prioritize actions based on those Failure Modes with:
Special All products and processes have features described by characteristics that are important
Characteristics and need to be controlled. However, some characteristics, called Special Characteristics,
required extra effort to minimize failure. Refer to the table on the following page.
4. High-Impact Characteristics (HI) are related to a process parameter that does not
affect the product but may have an impact on the operation of the process itself or
subsequent operations. These are Failure Modes with a Severity rating of 5 to 8 and
Occurrence of 4 to 10. HI does not require special control.
PFMEA) Focus
Design YS A potential Significant Severity = 5 - 8 and Highlight for
Characteristic (Initiate Occurrence = 4 - 10 PFMEA Team
PFMEA) Focus
Design Blank Not a potential Critical Severity < 5 Not Required
Characteristic or
Significant Characteristic
Process ∇ A Critical Characteristic Severity = 9, 10 Special Control
Required*
Process SC A Significant Severity = 5 - 8 and Special Control
Characteristic Occurrence = 4 - 10 Required*
Process HI High Impact Severity = 5 - 8 and Emphasis
Assembly Effect
Manufacturing/
Occurrence = 4 - 10
Process OS Operator Safety Severity = 9, 10 Safety Sign-Off
Special Special Controls are those manufacturing and assembly process methods, administrative
Controls actions, techniques, and tests beyond the normal and customary controls used to prevent
and/or contain product defects prior to release for manufacturing or assembly.
Control Plans A Control Plan is a written description of the system for controlling production processes.
It describes a producer's quality planning actions for a specific product or process. Control
Plans contain all applicable Critical and Significant Characteristics.
CFMEA CFMEA is similar to the Design and Process FMEAs, including Robustness Linkages.
Overview CFMEA is used to analyze concepts at early (or pre-hardware) stages and is typically
performed for systems and sub-systems, but it can also be performed on components.
CFMEAs are critical to the advanced technology process and are an integral part of the
Implementation Readiness process that must be completed by the Strategic Intent <SI>
gateway.
CFMEA Step 1
CFMEA Once the team is established and the CFMEA scope is set, it is time to move on to Step 1.
Objectives Step 1 for CFMEA is fundamentally the same as both the DFMEA and PFMEA.
In CFMEA, Failure Modes, Effects and Severity are determined the same as in DFMEA
and PFMEA. For recommended actions, CFMEA provides more opportunity to take
actions in Step 1, because it is so early in the development cycle. In Step 1, changes:
CFMEA Step 2
Determine After Step 1, if Failure Modes have not been eliminated, then complete Step 2, which is
Causes and similar to Step 2 of DFMEA and PFMEA.
Rank
Occurrences • CFMEA Causes are determined in a manner similar to DFMEA and PFMEA, except
that the Cause refers to a deficiency in the concept, rather than in the hardware.
• Ranking occurrences is done the same as in DFMEA and PFMEA, except that
occurrence in a CFMEA requires creative team analysis, as historical data are limited
or non-existent.
• CFMEA teams may take actions at Step 2. There is the possibility of no Occurrence
number, which leads to a rating of 10, so the action is to get some data.
Analyzing the interfaces and interactions is especially important. A major benefit of the
Concept FMEA is the identifying of potential Failure Modes caused by interactions that
must be addressed before the concept can be approved and implemented.
Controls As in DFMEA and PFMEA, Step 3 is completed for all Failure Modes that still exist after
completing Steps 1 and 2. Step 3 of the CFMEA will look fairly familiar, but the controls
are quite different in a CFMEA.
Since CFMEA is done prior to hardware solutions, typical controls, may include:
• Computer simulation
• Mathematical modeling
• Load calculations
• Environmental testing
• Laboratory testing
FPDS Timing CFMEA's are started at Kick-Off <KO> and should be completed prior to Strategic Intent <SI>.
Introduction The inputs and outputs for each type of FMEA and how they are interrelated are shown in
the following diagram. FMEA is an important part of Ford product development and
manufacturing, and serves as an important link between many elements of the system.
Remember there are many other information sources available that are valuable resources
for FMEAs. Consult the FMEA Handbook for further information regarding FMEA's.
For assistance in completing an FMEA, use the FMEA Checklist.