FMEA - PFMEA

DEFINITION

A Failure Mode & Effect Analysis is a primary

directive designed to identify the ways in which a product
or process can fail and provide input for the elimination or
reduction of the risk regarding those failures in order to
protect the customer.
 DESCRIPTION

A Failure Mode & Effect Analysis is:

 A structured approach to identify failures and the associated risk

of the failure
 A prioritizing tool for actions that should be taken to reduce risk
 An evaluating tool for validating design plan for a product or the
current control plan for a process
 A facilitating tool for documentation of a plan to prevent failures
from occurring
 OVERVIEW

A Failure Mode & Effect Analysis should be started:

 When new systems, products and processes are being designed.

 When existing designs or processes are being changed.

 When carry-over designs and processes will be used in new

applications or new environments.
 When preliminary drawings of a product are available.

 After completing a problem solving study.

 After system functions are defined, but before hardware is

selected.
 After product functions are defined, but before the design is
approved.
 FMEA Uses

 Key tool to improve a process in a pre-emptive manner, before failures

occur.
 Used to prioritize resources to insure process improvement efforts are
beneficial to the customer.
 Used to document completion of projects.
 Should be a “living” document; continually reviewed, amended, and
updated.
 Used to analyze new manufacturing processes.
 Used to identify deficiencies in the Control Plan.
 FMEA Uses (continued)

• Can establish and prioritize actions.

• Evaluates the risk of process changes.
• Guides the development of new manufacturing
processes.
• Helps set the stage for breakthrough.
 General Opinions about FMEA
 A FMEA is a disciplined analysis / method of identifying potential or
known failure modes.
It provides follow-up and corrective actions before the first production run
occurs.
 The threshold of the first production run is important because up to that
point modifying and changing the design is not a major event.
 The customer has the important role of defining the process, product and
procedure and so forth.
 The customer becomes involved through the letter of deviation, waiver of
change or some other formal notification.
 The result is to produce a defect-free product.

 It may be used as input for the product, assembly and or the Service
FMEA.
 Where Does Risk Come From?

Potential Safety Poor control

Unclear Customer
Hazards plans & SOP’s
Expectations

Poor
Assignable Process
Variation Capability
Cumulative Risk

Vague Raw Material

Workmanship Variation
Standards

Measurement Poorly
Machine Variation developed
Reliability (Online and Specification
QC) Limits
 WHEN FMEA

Timeliness : FMEA is meant to be a “Before – The Event” action, not an

“After – the – Event” Exercise.

 To achieve the best results, comprehensive FMEA must be done before a

failure mode has been unknowingly designed into a product/ process / service.
 TYPES OF FMEA(s)

 System FMEA

 Design FMEA

 Process FMEA

 Service FMEA
 COMPARISON OF FMEA(s)
S.No. Description System FMEA Design FMEA Process FMEA Service FMEA

To analyze systems and sub systems To analyze products before they are To analyze manufacturing and assy To analyze services before they
1 Uses
in the early concept design stage. released to manufacturing process reach the customer.

Failure modes ( tasks, errors,

Potential failure modes betw een the
Failure modes caused by design Failure modes caused by process or mistakes)
2 Focus functions of the system caused by
deficiencies assembly deficiencies caused by system or process
the system deficiencies.
deficiencies

A potential failure modes ranked by the RPN.

A poterntial list of system functions

A potertial list of critical and / or A potertial list of critical and / or A potertial list of critical and / or
that could detect potential failure
significant characteristics significant characteristics significant tasks or processes
modes.
A poterntial list of design actions to A poterntial list of design actions to A poterntial list of recommended
A poterntial list of bottleneck
eliminate failure modes safety issues eliminate failure modes, safety issues actions to address the critical and
processes or tasks
3 Output and reduce the occurrence and reduce the occurrence significant characteristics
A poterntial list to eliminate the causes
A poterntial list of parameters for
of failure modes, reduce their
apporpriate testing, inspection and or A potential list to eliminate the errors
occurrence and improve defect
detection methods
detection if CpK cannot be improved.

A poterntial list of recommended

A poterntial list of monitoring system /
actions for the critical and significant
process functions
characteristics
Helps select the optimum system estabilishes a priority for design Identifies process deficiencies and
Assists in the analysis of job flow
design alternative improvement actions offers a corrective action plan
Ientifies the critical and / or significant
Assists in the analysis of the system
Helps in determining redundancy. Documents the rationale for changes characteristics and helps in
and / or process
developing control plans
Provides information to help through
Helps defining the basis for system Estabilishes a priority of corrective
product design verification and identifies task deficiencies
level diagnostic procedures. actions
testing.
4 Benefit
Identifies critical or significant tasks
Increases the likelihood that potential Assists in the evaluation of design Assists in the analysis of the
and helps in the development of
problems w ill be considered. requirements and alternatives manufacturing or assembly process
control plans
Identifies potential system failures and
Helps identify and eliminate potential Estabilishes a priority for improvement
their interaction w ith other systems or Documents the raionale for changes
safety concerns actions
subsystems
Helps identify product failure early in
Documents the rationale for changes
the product development phase
Maximize the total process (system),
Objective / Maximize system quality, reliability, Maximize Design quality, reliability, Maximize the customer satisfaction
5 reliability, Cost maintainability and
goal Cost and maintainability Cost and maintainability through quality, reliability and service
productivity
 PROCESS FMEA BASIC PROCEDURE

•Define, Demonstrate and Maximize engineering solutions

•(i.e., Quality, reliability, maintainability, cost and productivity )

• The Engg. Solutions can not emphasize one of the elements at the expense of
the others.

• All of them satisfied at optimum so the process can be at maximum.

• To achieve this objective the process FMEA must base its requirements on
solid needs, wants and expectations of the customer.

• It may be a result of a QFD or an internal need for improvement or the results

of a design FMEA.

• First steps in conducting the FMEA should be include a feasibility study and or
a risk-benefit analysis.
 Basic types of process evaluation techniques

 Process capability studies.

 Mandatory process evaluation

Process capability studies

It is used to determine the inherent capability of specific elements of the
production processing.
Example : include machine acceptance capability.

Mandatory process evaluation

Mandatory evaluation points that are critical to the operation / or the customer.
 Identifications of Mandatory Process Evaluation

 Customer requirements

 Government regulations

 Internal engineering guidelines

 Design FMEA

 Industrial standards / guidelines

 Generally accepted practices

 Courts through product liability

 Evaluation TIPS

 Certification of operators.
Certification may be necessary for critical skills (ex: boiler operators / welders)

 Tool Proofing
Tools, jigs and Fixtures may be required to be validated.

 Critical process
It is defined by safety, customers or government regulations require evaluation
and prior approval.

 Test operation
Most complex testing operations require review and approval to ensure
accuracy.
 Process Features
 Standardization

 Test Provisions

 Interchangeability

 Accessibility

 Controls

 Technical data, procedures

 Transportability

 Producability

 Safety

 Reliability

 Software
 Step-by Step Process FMEA Analysis

There are two requirements to perform a process FMEA.

 Identification of the appropriate form.

 Identification of the rating guidelines.

The above requirements are not universal.

They are not standardized.

Each company’s requirements reflects the needs of the organization, the

product, and the concerns of the customer.
FORMAT OF PROCESS FMEA
 Potential Failure Mode (11)
Generally there are four categories of process failure modes.

 Testing and / or inspection

Accept or reject bad or good parts, respectively

 Assembly

Relational concerns, misoreiented parts and / or missing parts

 Receiving inspection

Why is the received part rejected?

 Manufacturing

 Visual characteristics

 Dimensional characteristics

 Design characteristics
To identify the potential failure mode one may think of the negative or loss of the function

Fails to open

Part leaking

Broken

No pressure

Cannot control speed

Hole is not round

Part undersized, oversized, porous, damaged

Another way to identify the failure mode anticipated is by asking the following questions

How could this process fail to complete its intended function?

Why could this part be rejected at this operation?

What does the customer find unacceptable?

How would the part not conform to specifications at this operation?

 Potential Effect(s) of Failure (12)
A potential effect of the failure is the consequence of its failure on the next process,
operation,

Product, Customer and or government regulations.

The questions usually asked are

 What does the customer experience as a result of the failure mode described?

 What happens? or the what is the ramifications of this problem of failure?

To identify the potential effects, some of the documents one review are

 Historical data

 Warranty documents

 Customer complaints

 Field service data

 Reliability data

 Feasibility studies

Similar current or past FMEAs (both process and design)

 Critical Characteristics (13)
The critical characteristics become of paramount importance because they define the
process requirements, sequences, tooling and anything that can affect the customer
or government regulations.

Examples of possible critical items may be

 Dimensions

 Specifications

 Tests

 Processes

 Tooling

 Usage

Critical Characteristics are identified when

Process requirements can affect safety

Process requirements can affect compliance with government regulations

Process requirements are necessary for special actions / controls

 Severity of Effect (14)
Severity is a rating indicating the seriousness of the effect of the potential
process failure mode.

There is direct correlation between effect and severity.

For example, if the effect is critical, the severity is high.

If the effect is not critical, the severity is very low.

 Potential Cause(s) of Failure (15)

 The cause of the process failure mode is the process deficiency that results in the failure mode.

 To do a good job of proper potential cause(s) of failure identification, one must understand both

the design and proces and ask the appropriate questions. Specificity is essential.

For Example,

 Would inadequate venting and gauging cause misruns, porosity and leakers?

 Would inefficient die cooling cause die hot spots?

 The basic question is ‘In what way can this system fail to perform its intended function?

 What circumstances could cause the failure?

 How or why can the part fail to meet its engineering specifications?
Failure mode can be caused by one or more individual components or by
 Hardware failure due to inadequate product design
 Improper selection of component parts
 Improper use of processes
 Inadequate control procedures
 Failure to enforce process and quality controls
 Improper installation, maintenance
 Lack of safety devices, environmental factors
 Misuse, abuse
 Alteration of the product
 Improper operating instructions
 Human error
 Improper choice of materials
 Stress concentrations
 Fatigue
 Corrosion, galvanic corrosion, uniform attack, crevice corrosion
 Hydrogen damage, pitting, blistering
 Decarbonization, abrasion and wear, shock and vibration
 Interaction with other components
 Interaction with components of other systems
 Interaction with the government
 Interaction with the customer
Examples if failure causes include

 Torque too high or low

 Air pressure too high or low

 Cure time too short or long

 Tool worn

 Speed not constant

 Human error
Occurrence (16)

 It is the rating value corresponding to the estimated number of frequencies and /

or cumulative

 Number of failures that could occur for a given quantity of parts produced with
existing controls.

 Generally, the process FMEA operates under the assumption of single-point

failure.

 A Single-point failure is defined as a component failure that would cause the

system failure and which is not compensated by either redundancy or an alternative
method.

 For example, single pieces of hardware and heavy loaded cable runs that place a
high degree of dependence on single components usually can be avoided through
the use of redundancy.

 In this case, installation of duplicate cables into a heavily loaded area can
minimize service disruption in the event of a cable cut.
Detection Method / Process Verification / validation, Existing Control
(17)
A method, test, or an engineering analysis. There are some of the first-level methods to
detect or prevent a failure in the process , part or in subsequent operations and or the
customer.

The objective is to detect a process deficiency as early as possible.

Two of the leading questions in the brain-storming process should be

 How can this failure be discovered?

 In what way can this failure be recognized?

Some of the most effective ways to detect a failure are

• Proven simulation techniques

• Mathematical modeling

• Proto type testing

• Design of experiments

• Process verification testing

• Specific product testing

 Detection (18)
Detection is a rating corresponding to the likelihood that the current process controls will
detect a specific root cause of a failure mode before the part leaves the manufacturing
area.

In addressing detection issues in the process FMEA, three items are very important.

 Do not make the assumption that the detection should be low, just because the
occurrence is low.

 If 100 percent automatic gauging is listed as a control,the effectiveness is based on

Condition of the gauge

Condition of the testing equipment

Calibration of the gauge and testing equipment

Variation of gauge and testing equipment ( based on R &R study )

Visual inspection - The FMEA team must consider
the effectiveness based on:

 100 % visual inspection is only 79 to 100 % effective depending on inspector

& conditions of inspection

 Who may perform the inspections?

 How is the consistency of evaluation going to be monitored ?

 The nature of failure – Is the failure obscure & calls for very high level of
inspection ability is required.

Is the failure so obvious that anyone can identify it?

 Risk Priority Number (19)

This number is the product of severity, occurance and detection.

The RPN defines the priority of the failure.

In the process FMEA one must always remember that the goal is to reduce the RPN in a specific

Way.

The specific way is through a reduction in

Severity

Occurrence

Detection
 Risk Priority Number (Continued)

• The output of an FMEA is the “Risk Priority Number”

• The RPN is a calculated number based on information
regarding:
– the potential failure modes,
– the effects, and
– the current ability of the process to detect the failures
before reaching the customer
• It is calculated as the product of three quantitative ratings,
each one related to the effects, causes, and controls:

• RPN = Severity x Occurrence x Detection

Effects
Effects Causes
Causes Controls
Controls
 Recommended Action (20)
No FMEA should be done without a recommended action.

The recommended action must be specific action(s) or it may be further studying.

The idea of recommended action in the design FMEA is to reduce the severity,
occurrence, detection, or all of these actions.

Design FMEA is performed to eliminate design deficiencies & therefore eliminate

Failure

FMEA team must prioritize those failure modes with highest RPN, the highest
severity, the highest occurrence.
Responsible Area or Person and Completion date (21)
Identify the responsible person / area and the target completion date for the
recommended action,

Action taken (22)

This is follow-up .

After the action has been taken, the effective date or completion date with a brief
description of the action should be entered.

Revised RPN (23)

After the actions are incorporated in the process, the FMEA team reevaluate the
consequences of severity, occurrence and detection.
 CONTROL PLAN

The intent of a process control plan is to control the

product characteristics and the associated process variables to
ensure capability (around the identified target or nominal) and
stability of the product over time.
Phases of the control plan
The control plan shall cover three distinct phases, as appropriate.

a) Prototype: A description of the dimensional measurements, material and

performance tests that will occur during building of the prototype. The
organization shall have a prototype control plan, if required by
the customer.

b) Pre-launch: A description of the dimensional measurements, material and

performance tests that occur after prototype and before full production. Pre-
launch is defined as a production phase in the process of product realization
which may be required after prototype build.

c) Production: Documentation of product/process characteristics, process

controls, tests and measurement systems that occur during mass production.
Each part shall have a control plan but, in many cases, family control plans
may cover a number of similar parts produced using a common process.
Control plans are an output of the quality plan.
Control plans are established at a part number level; but in many
cases, family control plans may cover a number of similar parts
produced using a common process. Control plans are an output of
the quality plan.

NOTE 1 It is recommended that the organization require its

suppliers to meet the requirements

NOTE 2 For some bulk materials, the control plans do not list most
of the production information. This information can be found in the
corresponding batch formulation/recipe details.
Elements of the control plan A control plan includes, as a minimum,
the following contents:

a)control plan number

b)b) issue date and revision date, if any
c)customer information
d)organization's name/site designation
e)Part number(s)
f)part name/description g) engineering change level
g)phase covered (prototype, pre-launch, production)
h)key contact
i)part/process step number
j)process name/operation description
k)functional group/area responsible
Product control
a)product-related special characteristics
b)other characteristics for control (number, product or process)
c)specification/tolerance
Process control
a)process parameters (including process settings and tolerances)
b)process-related special characteristics
c)machines, jigs, fixtures, tools for manufacturing (including
identifiers, as appropriate)
Methods
a)evaluation measurement technique
b)error-proofing
c)sample size and frequency
d)control method Reaction plan
e)reaction plan (include or reference)