Why FMEA? · Failure Mode, Effects, and Criticality Analysis (FMECA). By the early 1960s, contractors for the U.S. National Aeronautics and Space Administration (NASA) were using

Why FMEA?

This technical paper will try to answer the question that many project managers and engineers

in Railways industry used to ask during a project. Why we need to implement the FMEA

analysis?

Despite of all effort necessary to carry out the FMEA analysis during the project in different

phases, more than a project document delivery, the FMEA is an opportunity to improve the

asset during the design phase and avoid many failures caused by error triggered during the

design and manufacturing phases. In addition, the FMEA is an important input to other

analysis, such as RCM and FRACAS as will be discussed in this paper.

Key words: FMEA, FMECA, DFMEA, PFMEA, Risk, V&V Test, Criticality.

1 - The FMEA concepts:

The FMEA was originally developed by the U.S. military during the 1940s and supported by

military specification beginning in 1949 (MIL-STD-1649, Procedures for Performing a

Failure Mode, Effects, and Criticality Analysis (FMECA). By the early 1960s, contractors for

the U.S. National Aeronautics and Space Administration (NASA) were using variations of

FMECA under a variety of names. In 1966 NASA released its FMECA procedure for use in

the Apollo program. Since then, the FMEA (Failure Mode and Effect Analysis) has become a

standard part of the design process and by the 1980´s, the aerospace industry has been in

continuous use through the present as a means of ensuring that hardware built for space

applications had the desired reliability characteristics.

Nowadays, the FMEA analysis is applied for different industries such as aerospace,

automotive, Oil and Gas, Energy, Electric and Electronic and Railways.

The FMEA is a qualitative reliability engineering method for systematically analyzing the

possible failure mode of each equipment and component, and identifying the possible failure

cause, the failure mode cause frequency, how the failure mode cause can be detected, as well

as the resulting consequence of the failure mode effect on quality, safety, health and

environment and asset performance.

Usually, during the FMEA, is carried out the risk assessment, which is a qualitative analysis

of each failure mode considering the likelihood (frequency) of the failure mode and the

severity of the failure consequence based on a Risk Matrix, which is defined the categories

for occurrences (likelihood or frequency) and severity. The FMEA application aims to avoid failures triggered in different asset phases such as

design, manufacturing (and transportation), installation (and commissioning) and operation.

However, most of the time, such failures will be detected and occur in the operational phase.

The best practice, whenever time and specialist are available, is to carry out a specific FMEA

for each asset life cycle phase such as SFMEA (System FMEA), DFMEA (Design FMEA),

PFMEA (Process FMEA), FMEA (operational Phase) as shows Figure 1.

Figure 1 - FMEA applied along Railways Asset life cycle

Source: Calixto, Eduardo at al 2017.

Therefore, since the SFMEA, DFMEA, PFMEA recommendation is successfully

implemented and avoid the systematic failures, the failure modes, which will occur during

operation time are caused by random failures or component degradation. In case of

degradation, the FMEA will be the basis for the RCM analysis (Reliability Centred

Maintenance), which will define the preventive maintenance task to detect and avoid such

failures. In addition, the FMEA will be the basis for the FRACAS (Failure Report Analysis

and Corrective Actions System) codification. In general terms, independently of the type of

FMEA, the basic FMEA steps are:

Figure 2 - FMEA general steps


2 - Risk Concept applied to FMEA

After understanding the failure concept, it´s important to understand how to prioritize

preventive actions during design and manufacture phases as well as preventive actions

concerning a group of equipment failures during the operation phase to detect and avoid

cause of equipment failures. Therefore, a criteria is necessary to enable to choose, which

failure mode has the highest priority to be avoided along the railway asset life cycle.

Therefore, an important concept concerning FMEA is the risk. By definition, the risk is the

combination of an event of hazard and its consequence. In order to analyse and evaluate the

risk, the qualitative and quantitative approach can be performed. In fact, when risk is assessed

and evaluated based on qualitative methods like FMEA, such assessment is performed

qualitatively based on specialist opinion regarding a risk matrix. In the risk matrix, the

frequency of occurrence and the severity of consequence criteria are defined. Depends on the

Railway company and asset, there must be different configurations of risk matrix, but in the

railway industry (Europe), mostly of companies follows the risk matrix defined in the

standard EN-50126. The figure 3 shows the risk matrix with four severity categories and six

frequency categories.

Figure 1 Risk Matrix

Source: EN-50126.

The additional risk concept is the criticality, which may be understood for some specialist as

RPN (Risk priority number) or as defined by MIL-STD-1629,1629, which the failure mode

criticality is defined as the multiplication of the component failure rate as well as the follow

indexes as demonstrated by the equation below.

Cr = Cp x Fr x Fm x T

Where:

Cp = Condition probability of failure

Fr =Failure Mode Ratio

Fm = part failure mode rate

T = Duration of application mission

Insignificant Marginal Critical Catastrophic

IV III II I

A Undesirable Intolerable Intolerable Intolerable

B Tolerable Undesirable Intolerable Intolerable

C Tolerable Undesirable Undesirable Intolerable

D Negligible Tolerable Undesirable Undesirable

E Negligible Negligible Tolerable Tolerable

F Negligible Negligible Negligible Negligible

Hazard Severity Level

Fre

quency o

f occure

nce

Frequent

Probable

Occasional

Remote

Improbable

Incredible

Depends on the type of rolling stock, infrastructure and signalling asset, different type of

definition applied to FMEA or FMECA can be implemented. In the cases of unsafe failure,

the detection is a very important aspect to be considered in order to enable the design

detectability improvement or even the inspection and preventive maintenance tasks.

3 - FMEA Workshop

In order to implement the FMEA analysis, it's recommended to organize the FMEA

workshop with a group of specialists to discuss about the asset failure modes, causes,

consequences, asses the and evaluate the risk and propose the risk recommendation. In many

projects, the FMEA analysis is carried out by an individual and revised by other specialist.

Despite of time saving, the main point to have an FMEA workshop is to have a chance to put

different specialist together to discuss about the probable failures of an asset and how to

avoid that. In fact, in many cases, there´s no structured and organized meeting during design,

which enable different engineers and specialist to discuss about the asset problems and

improvement together. In many cases, different engineers work isolated, send a document to

others, get some comment and sometime there are some meeting to justify the

recommendation implementation or not. The FMEA workshop gives the opportunity of all

asset design discipline interaction to deliver an integrated solution.

Before the FMEA workshop realization, the first step is to define the input information to

perform the FMEA workshop such as:

FMEA scope;

To define the system equipment list;

To define the functional relationship between these items;

To define how the main function is handled through these functional relationships;

To define the equipment identification number;

To define equipment and component list identification;

PFDs for all assets;

Equipment function, capacity and configuration;

Criticality (RPN number) or risk criteria;

Risk matrix definition;

Risk mitigation policy;

Complete FMEA spreadsheet definition.

Despite of the input information importance, the FMEA implementation process includes not

only the input information but also other critical FMEA success factor such as:

Specialist involvement during FMEA meeting and review;

Time and technical resources available during and after the FMEA meeting;

Managers support before, during and after the FMEA meeting;

Managers support FMEA recommendation implementation.

By considering the importance of FMEA for the asset design and performance during

operation phase, it´s necessary to be established an FMEA process, which is divided in four

stages with ten steps (Carson, Carl S, 2005). Such process will facilitate the FMEA

implementation with more chance to have a successful asset performance achievement as

shown in figure 3.

Figure 4 - FMEA process

Souce: Carson, Carl S, 2005. 4 - But Why FMEA? The question Remains!

After the concept description the main question remains. In order to apply the different

FMEA in different asset life cycle phases, which consume economic and human resources

it´s necessary to have the clear understanding what each type of FMEA enable to the asset

performance improvement and achievement. The first mindset that´s necessary to change

related to FMEA is that, the FMEA it´s not only a document that must be delivered to the

client or the supplier. The FMEA is an opportunity to improve the product since the

concept phase until the end of the design. 4.1 - FMEA applied to solve System Interface issues (SFMEA)

The system FMEA (SFMEA) is usually applied in high level configuration in order to have a

preliminary understanding about the system failures and effect in another system interface,

which is applied during concept or early design phase. In such level, it´s not possible to

propose specific recommendations about the equipment and component. However, the

SFMEA enable to have a first understanding about the system failure impact and interface

effects.

Despite the limitation of deep understanding in equipment and component level, the SFMEA

enables to look at system interfaces that is neglected in many projects.

Concerning the life cycle phase, the SFMEA is usually applied during concept phase or even

in the beginning of the design. During the design, because of the clear definition of all system

levels, it´s possible to carry out the FMEA in LRU level, similar to the operational phase,

which the FMEA is supposed to be applied whenever a system modification or an upgrade

take place. Therefore, in the operational phase, it´s advisable to update the FMEA as well as

the FRACAS system, which the codes are based on such FMEA. In general terms, some

examples of SFMEA failure modes are:

Total system loss;

Partial system loss;

Loss of input;

Loss of output;

Erroneous;

Inadvertent.

The figure 5 shows the TCMS (Train Control Management System) SFMEA simplified

example.

Figure 5 - TCMS SFMEA example (summarized)

Source: Calixto, Eduardo at al 2017. Once the SMEA is implemented and the interface issues are solved, why to go further and implement the DFMEA, PFMEA and FMEA? Let´s see in the next item.

4.2 - How to improve the Design and avoid the systematic failure occurrence during

early life phase (Design FMEA)!

The Design failure modes and effect analysis aim at identifying the asset (equipment and

component) failure modes triggered by design mistakes and their consequences during the

manufacturing or operation phase. In addition, for each failure mode the critical (RPN =

Occurrence x Severity x Detection) needs to be defined hierarchy in order to prioritize

preventive actions during the design. The DFMEA input information is the following:

DFMEA scope;

Identify items constituting the system (Boundary diagram);

Determine functional relationship between these items (Interface matrix);

Determine the equipment and component stressor factors (temperature, Humidity,

vibration, etc.);

Determine how the main equipment function (functional block diagram);

Equipment and component list identification;

Operational and environmental profile;

PFD of the assets (systems, equipment and component);


Criticality criterion (RPN number).

The DFMEA output information is the following:

Verification and Validation test Plan;

Verification and Validation report;

DFMEA report with recommendation;

DFMEA recommendation plan.

Some of the FMEA input information such as equipment list, system boundaries, risk criteria

and P&D are usually applied to many types of FMEA, safety and reliability engineering

analysis. However, the interface matrix and functional block diagram are important for

DFMEA as input information. The Interface matrix aims to establish the level of impact that

each equipment has on the system and other equipment, which enable to identify and classify

the equipment failure modes effect more precisely during the DFMEA. The figure 6 shows an

example of interface matrix.

Figure 6 - Interface Matrix


On figure 6, the level of impact is classified such as:

1 - Interaction does not affect the performance;

2 - interaction affects the performance;

Based on the figure 6, by looking the first column on the left and the second line at the top, in

case of failure of Diesel Engine, the locomotive will be unavailable. Such assessment can

also be applied for other different equipment by assessing the table from the left to the right

and classify the effect of each equipment in the locomotive and other equipment

performance.

After all input information is available, the DFMEA is implemented and the main design

cause of failures is detected, and the recommendation is defined. The figure 7 shows a

summarized DFMEA example of diesel engines. Note that the Diesel Engine is defined in the

equipment level in the DFMEA. That´s happened because the Integrator Company

responsible for the Locomotive design has no detailed information about the diesel engine

design in this design phase. Therefore, this company will request for the diesel engine

supplier to perform a detail Diesel DFMEA in LRU level and will follow up the

implementation of the DFMEA recommendation.

LocomotiveDiesel

Engine

Aux. Diesel

tank

Loco

Control

Transfer

fuel pump

Radiator

with fan

Locomotive

Diesel

Engine 2

Aux. Diesel

tank 2 2

Loco

Control 2

Transfer

fuel pump 2 1 2

Radiator

with fan 2 1

Figure 7 - DFMEA


During the DFMEA analysis, many of the recommendation are related to review the

calculation and establish and review procedures, which are the simplest to be implemented

during the design. The real challenge is when it´s necessary to carry out the verification and

validation (V&V) test. In order to have implemented the test V&V, it´s necessary two basic

steps such as:

To define precisely the V&V test;

To follow up and manage the test implementation and results. Some of the Verification (VE) and Validation (VA) analysis are:

Material Adequacy Assessment;

Deflection Analysis;

Low Cycle, High Load Analysis;

High Cycle Fatigue Analysis;

Fracture Mechanics Analysis;

Plastic Strain Analysis;

Thermal Modelling and Analysis;

Wear, Fretting Analysis;

Scuffing Analysis;

Creep Analysis;

Material or Lubrication Aging Analysis;

Contaminant Assessment;

Pressure Analysis;

Cold Weather Assessment;

Lubrication Distribution Modeling;

Reliability Assessment;

Dynamics Modelling & Analysis;

Item / Function Failure Mode

Potential

Effect(s)

of Failure

S

e

v

Cause

O

c

c

u

r

Current

Design

Controls

D

e

t

R

P

N

Recommend

ed

Action(s)

Responsibility

& Target

Completion

Date New

Sev

New

Occ

New

Det

New

RP

N

diesel engine early

mechanical

wear out in the

engine

(Components)

diesel

engine fails

and direct

impact on

lcomotive

availability

8

wrong

specification

and wrong

material

selection

during design

5

use of

proven

design,

specification,

supplier

confirmation

3 120

validation

based on

material

selection

procedure,

testing

designer,

supplier

8 4 3 96

diesel engine early electric

wear out in the

engine

(Components)

diesel

engine fails

and direct

impact on

lcomotive

availability

8

wrong

specification of

temperature

duty cylce

robustness

5

use of

proven

design,

specification,

supplier

confirmation

2 80

Cycle test designer,

supplier

8 4 2 64

diesel engine early

mechanical

wear out of the

engine

(components)

diesel

engine fails

and direct

impact on

lcomotive

availability

8

wrong

calculation of

power of the

engine5

use of

proven

design, load

calculation

(worst case),

supplier

confirmation

3 120

testing designer,

supplier

8 4 2 64

Bolted Joint Analysis;

Fluid or Air Flow Analysis;

Component Resonance Analysis;

Electrical Arching Analysis;

Lubricant Adequacy Assessment;

Interface Control Documents, Checks;

Transmission Error Analysis;

Torsional Vibration Analysis; The figure 8 shows an example of Verification and Validation test plan based on DFMEA result. Considering the diesel engine, the early wear out in internal component is triggered due to thermal cycle and overload. In addition to the V&V test definition it´s necessary to define the responsible and date to deliver the test result as well as the test procedures and test laboratories requirement and specification.

Figure 8 - V&V Test plan


Once the DMEA is implemented and the design issues are solved as well as the V& V test and preventive actions are planned and under control why to go further and implement the PFMEA and FMEA? Let´s see in the next item.

4.3 - How to avoid system failures during Manufacturing (PFMEA)?

The Process failure modes and effect analysis aim at identifying the failure modes, causes

and effect during manufacturing and assembly phase. The causes of process failures are bad

Item Failure

Mode

CauseStressor

FactorRPN

Recommended

Action(s)

VE

-

Hig

h a

nd

Lo

w T

erm

al

Lo

ad

An

aly

sis

VA

- T

est

Req

uir

em

en

t V

ali

dati

on

VA

- T

herm

al

Cycli

ng

(H

AL

T)

VA

- F

un

tio

nal

Test

An

aly

sis

(L

RU

)

VA

- C

om

po

nen

t R

eli

ab

ilit

y T

esti

ng

(A

LT

)

thermal

cycle loads

Engine

Component

failures

due

thermal

cycle

shocks

Low Cycle

Fatique

Termal

duty cycle

120

1:Verify the

design limtits

with supplyer

2: thermal

cycle test

(ALT & HALT)

X X X X

overload of

engine

early

mechanical

wear out of

the engine

Mechanical

Duty cycle

120

measure power

of compressor

during

functional

qualifikation

test

X X

DFMEA (Summurized) Verification (VE) and Validation (VA) test Actions

Diesel

engine

process quality control, wrong process parameters specified, wrong manufacturing

procedures, lack of references to the design and consequently lead to an early life failure

during the operational phase. In addition, the main causes of failures during assembly are lack

of procedures, wrong tools and human error.

The main product of PFMEA is a list of the most critical failures that must be avoided in the

during manufacturing and assembly phase. Therefore, before such phases, a list of PFMEA

recommendation need to be sent to the producer and the verification must be applied to avoid

early life failures during operation causes by manufacturing or assemble problems. Different

from other types of FMEA, the manufacturing process and assembly is out of the designer

control in many cases, or at least is led by different groups in the same company. The early

life failures triggered by manufacturing or assembly process are quite common in real life,

mainly the assembly errors. However, the assembly error on most of the cases can be fixed

when the equipment is reassembled, but in case of manufacturing error, the impact of the

asset life cycle cost can be very significative because it will be necessary to adjust the process

and produce the asset again.

The input of FMEA is also the same e design information listed such as:

PFMEA scope;

Manufacturing (Assembly) process equipment list identification;

Operational process and environmental profile;

PFDs for all manufacturing (Assembly) process assets (system, equipment and

component);

Manufacturing (Assembly) process equipment function, capacity and configuration;


The PMEA output information is the following:

THE PMEA report;

The PMEA recommendation (Human reliability analysis for human error, process

adjustment);

A good approach to check the influence of the manufactory and assembly is the so-called

HASS (High Accelerated Screes Test) test, which is performed after the assembly.

The figure 9 shows the Process FMEA applied for an asset assembly example. In this case,

the failure in trolley component will delay the assembly or even damage the product in case

of drops of the product from the trolley. In addition, the second line describes the trolley

frame assembly, which in case of failure can also lead to an accident and again drop of

product from the trolley.

Figure 9 - Process FMEA (Assembly)


Once the PMEA is implemented and the manufacturing and assembly process are under control why to go further and implement the FMEA concerning failures during the operation phase (random or degradation)? Let´s see in the next item.

4-4 - How to manage failures during Operation phase (FMEA)?

After implementing the DFMEA and PFMEA recommendation, it´s not expected to have

early life failures caused by error during the design and manufacture phases. In addition, the

designs supposed to be robust enough and consider all other interfaces, effects cause by other

systems as defined in the SFMEA. It´s also necessary other reliability engineering methods to

mitigate the early life failures. However, no matter how good is the design the asset will fail

someday. Therefore, the FMEA needs to consider the degradation and random failures and

how to detect and prevent it.

The FMEA must also be applied to investigate the assets, failures modes during the operation

phase caused by degradation or random failures. Therefore, the FMEA aims to mitigate risk

of such failures during the operation phase. In fact, the same concepts of the other types of

FMEA will also be applied here, but the focus will be the failure, which will occur due to

equipment and component degradation.

The input of FMEA is also the same e design information listed such as:

FMEA scope;

Equipment and component list identification;

Operational and environmental profile;

PFDs for all assets (system, equipment and component);



The FMEA output information is the following:

THE FMEA report;

The FMEA recommendation (Human reliability analysis for human error);

Input data to Reliability Centred Maintenance (RCM);

Input Information for the Failure Report and Corrective Actions System.

The RCM and FRACAS system will be described in another technical paper in the coming

months on the ECC website. The figure 10 shows an example of FMEA applied to a bogie

bearing (LRU) during the operational phase.

Figure 10 - FMEA (Bearing - Diesel Engine)


5 - Conclusion:

After all explanation above, including the examples, it´s not difficult to answer the question.

Why FMEA?

The FMEA will enable the following:

To arise and solve System Interface issues, failures (SFMEA);

To improve the asset design and avoid the systematic failure occurrence during early

life phase (DFMEA);

To avoid systematic failures during manufacturing and assembly processes (PFMEA);

The FMEA aims to risk mitigation of random and degradation failures during the

operation phase;

To create the basis for the RCM analysis and consequently the basis for the preventive

maintenance plan during the operation phase;

To create the basis for the FRACAS which will enable the warranty validation based

on LDA (Lifetime Data Analysis) as well as root cause analysis and corrective

actions.

But for who do not believe, thrust or like the FMEA, do not worry, you do not need the

FMEA analysis if you can say yes for the following conditions:

If your company does not need to be worried about the asset interface with other

assets or even have no interface,

If your company is able to deliver a perfect asset design without any error, which

dispense the verification and validation test;

If your company or your supplier manufactory and assembly process is the lack of

error;

If you know all preventive maintenance tasks to be taken place related to all failure

modes;

If you do not expect that your equipment fails during the whole life operation and will

have any other issue related to safety.

The aim of this technical paper was to explain the importance of the FMEA and its

application throughout the whole asset life cycle. Based on the best reliability engineering

practice FMEA is recommended for all projects, but need to be customized for each necessity

and demand of information which will depend on each asset project characteristic.

A more detailed information about FMEA and case studies applied to railway industry will be

provided in ECC books and in the website soon.