DESTination RAIL

Decision Support Tool for Rail Infrastructure

EU Project No. 636285

Analysis of Failures within Switches and

Crossings using Failure Modes and Effects
Analysis Methodology

Elias Kassa
Department of Civil and Environmental Engineering, NTNU
Trondheim, Norway
Intelliswitch Symposium, 28-30 August 2017, Copenhagen, Denmark
 Outline of Presentation
1. Background
2. FMEA methodology
3. Common Damage Mechanisms
4. Failure Classification
5. FMEA Failure Analysis
6. Conclusions

2
Background
 Turnout populations
Countries Track S&C S&C units • Belgium – 1.88 units/km
(km) population per track
kilometre • Sweden – 0.81 units/km
Belgium 6,500 12,200 1.88 < 5% of infrastructure
Italy 27,100 42,700 1.58
• France – 0.40 units/km
Netherlands 6,500 7,800 1.20
UK 31,100 25,800 0.83
• In Sweden over 12% of track
maintenance and 25% of track
Sweden 14,900 12,000 0.81
renewals are spent on S&Cs
France 65,100 25,600 0.40

• Network Rail is using about 17 % of the track maintenance budget and ca. 25
% of the track renewal budget in Switches and Crossings
• In addition, cost for disruption and delays in train operation are very high
4
 Background
Different researchers propose different remedies
to reduce maintenance costs:
1. reducing turnout population (# turnouts)
2. using more durable and advanced materials
3. optimizing turnout geometry (layout), support
stiffness (structure) and rail profiles
4. adopting preventative maintenance strategy
instead of corrective maintenance
Over 30% of the failure modes are related to rail
mechanical and track geometry failures
5
 Aim of this study
• The aim of this study is to identify and predict the potential failures
and failure risks based on historical data and failure occurrences

Identification of possible failure modes

Identify most The likely Determination of Better categorization of
critical failure corresponding different modes in terms of
components mechanism rectifications severity and criticality

Better understanding and used as input to enable

Optimised layout Improved/new More durable and Better approach to
and components S&C design advanced materials preventive maintenance
6
FMEA methodology
 FMEA methodology
Failure Modes and Effects Analysis (FMEA)
• FMEA is a procedure used to
– identify potential failure modes
– determine causes and effects of failure modes, and
– mitigate or remove its effects on system functional performance
based on the recorded data (past experience)

• FMEA/FMECA tools are well established method for safety and reliability
analysis of systems, or for product improvement of systems in aerospace,
nuclear, electronic, and automotive industries

8
 FMEA/FMECA standard tools
 FMEA consists of breaking a system down into specific data

• IEC-60812: procedure for failure modes and effect analysis (FMEA) from
electronic industry
• SAE-J1739: FMEA for automobile industry: Potential Failure Mode and
Effects Analysis in Design and in Manufacturing and Assembly Processes
• SAE ARP 5580: Recommended failure modes and effect analysis (FMEA)
practices for non-automobile application

9
 FMECA sheet
System: _____________________ Performed by:__________________
Ref. drawing no: ______________ Date:_______ Page __ of ___
Description of unit Description of failure Effect of failure Failur Severity Risk Comm
Ref Function Operatio Failure Failure Detectio On the On the e Rate Ranking Reducing ents
nal Mode Mode Cause/M n of subsyst system Measures
echanism failure em function
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

10
 Failure Modes and Effects Analysis
(FMEA)
• Questions that must be answered in a FMEA
1. Function: what is the function(s) of the components analyzed?
2. Failure mode: in what ways can the system fail in performing its
intended function?
3. Failure causes and failure mechanisms: what are the causes?
4. Consequence: what can happen when a failure occurs?
5. Failure frequency: how often the failure occurs?

11
 Definitions
Failure:
• The termination of the ability of an item to perform a required function
Example: a train unable to run over a switch with the intended speed
Function
• The normal or characteristic "operational tasks" to an item
Example: to guide train from one track “track 1” to a separate track “track 2”,
with a certain required speed
• For a unit to give this function, all the components must provide the intended
function
Example: switching machine drives the switch rails to the intended direction
by the help of stretcher bars

12
 Definitions, Cont’d
Failure mode
• The way in which an item fails to perform its required function
Example: Line blocked, derailment, switch functioning with reduced speed
Failure cause
• The circumstances or the causes that result to the failure to occur
Example: missing of rail fastening bolts, dry or contaminated switch rail
sliding chair, rail breakage or fracture
Failure mechanism
• Physical, chemical or other processes that causes failure
Example: wear, corrosion, plastic deformation, RCF

13
 Failure hierarchy for a turnout unit
Turnout Product

To guide the train in Primary: to guide the

the main route train to diverging route Function

Line Wheel Speed

blocked derailment restriction Failure mode
Rail mechanical
damage
Stretcher bar
damage
Failure cause
Switching rail
failure
Switch machine
failure

Sleeper and
14 fastenings failure Failure mechanism
Common Damage Mechanisms
 Common Damage Mechanisms

Fracture Plastic deformation

Wear RCF on stock rail Switch rail breakage

16
 Common Damage Mechanisms

Rail Head Cracks

Damage on fixed nose

crossing

Damage on swing
nose crossing Plastic deformation (lipping)

17
Failure Classification
 Failure classification by components
• Failure may be classified based on failing components

Failure cause/mechanisms in rail failure

• Rolling contact fatigue
• Wear
• Rail head deformation
• Rail head cracks
• Rail web cracks
• Transverse & Longitudinal rail foot cracks

19
 Failure classification by components
Failure cause in switching system
• Dry slide chair or baseplate
• Broken stretcher bar
• Switch Anchor Loosing
• Broken Bolts

20
 Failure classification by components
• Failure may be classified based on failing components
Components Failure causes/mechanisms
Rail Wear, rolling contact fatigue, plastic deformation, rail head
cracks, rail foot fractures, rail web cracks
Stretcher bar Stretcher bar bracket breakage
Switching machine Too much or too little power, unable to close the switch rail
against the stock rail
Sliding chair and Dry slide chair, rusty slide table or fully contaminated lubrication
rollers which blocks the movement of switch rail from sliding
Fastening system Missing bolts, damaged rail pad, broken base plate
Sleeper Rail seat deterioration, flexural cracking at the sleeper centre,
and transverse cracking at the fastening bolt
21
 Failure data analysis
• Example of failure data analysis based on the failed components

Failed Total Frequency

50 Frequency (%) Components Number (%)
40 Switch rail 1113 45.3
30 Slide chair 747 30.4
Ballast 194 7.9
20 Schiwag Roller 138 5.6
10 Stretcher bar 111 4.5
0 Stock rail 71 2.9
Crossing 33 1.3
Fishplate 24 1.0
Back Drive 18 0.7
Sleeper 5 0.2
Spacer Block 4 0.2
Sum 2458 100
22
 Failure data analysis, Cont’d
• Data assessment based on possible failure causes (mechanisms)
50 Frequency %
40
30
20
10
0

23
 Failure classification by severity
• Severity level is one way of failure classification method to categorise
the criticality of the effects on the function of item or component
Severity level Criticality nature
Category I - Catastrophic A failure which may cause death or total system loss
Category II - Critical A failure which may cause severe injury, major property
damage, or major system damage
Category III - Marginal A failure which may cause minor injury, minor property
damage, or minor system damage which will result in delay
or loss of availability or speed restriction
Category IV - Minor A failure not serious enough to cause injury, property
damage, or system damage, but which will result in
unscheduled maintenance or repair
24
 Failure data analysis, Cont’d
• Data assessment based on rectification
Total Frequency
Rectification Failed Components
Number %
De-iced 559 22.7 Switch rail, Slide chairs, Schiwag Roller, Back drive, Stretcher bar
Lubricated 445 18.1 Slide chairs, Schiwag Roller
Removed obstacle 427 17.4 Switch rail, Slide chairs, Stretcher bar, Back drive
Stretcher bar, Slide chairs (broken), Crossing (nose crack), Fish plate,
Replaced/Renewed 243 9.9
Switch rail, Stock rail, Sleeper, Space block, Ballast
Lift & Pack 190 7.7 Ballast
Grind 167 6.8 Switch rail, Stock rail, Rail weld
Adjusted 143 5.8 Schiwag Roller, Switch rail, Stretcher bar, Back drive, Slide chairs, Ballast
Cleaned 136 5.5 Slide chairs, Switch rail, Schiwag Roller
Weld repair 71 2.9 Switch rail, Stock rail, Crossing
Tightened 70 2.9 Slide chairs, Stretcher bar (nuts), Back drive, Fish plate
Gauged 7 0.3 Track gauge

25
Failure Analysis using FMEA
 FMEA analysis
• In the rail industry, the procedure used is based on Risk Priority Number
(RPN)
Risk Priority
Number • Occurrence (P): Failures Frequency
(RPN) = PxSxD
Distribution
Step 3: Detect a • Sensitivity Ratings (S): Very minor if no
Detection failure
number (D) mode immediate effect to Very high if results in
unsafe operation
Step 2: Step 1:
Sensitivity Probability • Detection (D): Easy to Hard to detect the
number number
(S) (P) failure
FMEA cycle
27
 FMEA analysis, Cont’d
Rating Meaning Range ( % )
Step 1. 1 No Effect OCCUR = 0
Occurrence 2 Low (few failure) 0 < OCCUR < 5
3 Moderate (occasional failure) 5 < OCCUR < 10
4 High (repeated failure) 10 < OCCUR < 20
5 Very high 20 < OCCUR
Rating Meaning
1 No Effect
Step 2. 2 Very Minor (no immediate effect or long term effect)
Sensitivity 3 Minor (affects little of the system)
4 Moderate (causes a less primary function failure)
5 High (causes a loss of primary function)
6 Very High (results unsafe operation or injuries)
28
 FMEA analysis, Cont’d
Rating Meaning
1 High
Step 3. Detection
2 Moderate
3 Low

• Risk Priority Number (RPN)

RPN = P x S x D

29
Failure Analysis Results
 Results
Failure causes / Occurrence Sensitivity Detection
RPN
Failure mechanisms Rate (P) Rate (S) Rate (D)
Obstructed (Iced, ….) 5 5 3 75
Dry chairs 4 5 3 60
Crack / broken rail 3 5 2 30
Voiding 3 3 3 27
Contaminated (Leaves,..) 3 3 3 27
Out of adjustment 3 4 2 24
Plastic deformation /Lipping 3 4 2 24
Wear 2 4 2 16
Loose/missing nuts 2 2 3 12
Squat, RCF 2 2 2 8
Track gauge variation 2 3 1 6
31
 Results, Cont’d
• Group 1. High Risk Priority 80
Number
60
– highest priority for preventive
maintenance 40
– components associated with
Group 1
these failure mechanisms need 20
new or improved design
0
0 20 40 60 80

32
 Results
Failure causes / Occurrence Sensitivity Detection
RPN
Failure mechanisms Rate (P) Rate (S) Rate (D)
Obstructed (Iced, ….) 5 5 3 75
Group
Dry chairs 4 5 3 60
1
Crack / broken rail 3 5 2 30
Voiding 3 3 3 27
Contaminated (Leaves,..) 3 3 3 27
Out of adjustment 3 4 2 24
Plastic deformation /Lipping 3 4 2 24
Wear 2 4 2 16
Loose/missing nuts 2 2 3 12
Squat, RCF 2 2 2 8
Track gauge variation 2 3 1 6
33
 Results, Cont’d
• Group 2. Moderate Risk Priority
25
Number
– the second priority for preventive 20
maintenance 15
– components may need some Group 2
10
improvements
5
Group 3
• Group 3. Low Risk Priority Number
0
– need to get rectified before 0 5 10 15 20 25
imposing a serious effect on the
system in long term

34
 Results
Failure causes / Occurrence Sensitivity Detection
RPN
Failure mechanisms Rate (P) Rate (S) Rate (D)
Obstructed (Iced, ….) 5 5 3 75 Group
Dry chairs 4 5 3 60 1
Crack / broken rail 3 5 2 30
Voiding 3 3 3 27
Contaminated (Leaves,..) 3 3 3 27 Group
Out of adjustment 3 4 2 24 2
Plastic deformation /Lipping 3 4 2 24
Wear 2 4 2 16
Loose/missing nuts 2 2 3 12 Group
Squat, RCF 2 2 2 8 3
Track gauge variation 2 3 1 6
35
 Failure occurrences vs. seasons

The largest number of failures occurred

• autumn period - contamination of rail
running surface by falling leaves
• winter period - switch obstruction by ice

36
 Conclusions
• Failure risk in turnouts has been assessed based on historical data
and occurrence of failures
• FMEA procedure has been applied to approach the classification of
critical failures in turnouts
• Two failure mechanisms are identified to critically affect the turnout
primary operation: switch obstruction and dry chair
• Several years of data, and wide range of data is required for an
accurate judgment
• Such kind of failure risk evaluation may support maintenance
planning and design improvement

37
Thank you for your attention!

Questions?