Professional Documents
Culture Documents
Maintainability Prediction: MI L-HDBK-472
Maintainability Prediction: MI L-HDBK-472
24
MILITARY
STANDARDIZATION
MAINTAINABILITY
May
1966
HANDBOOK
PREDICTION
DEPARTMENT
WASHINGTON
A?IL- HDBK-472
MA INT.\INABILITY
:4 Jltiy ]966
OF DEFENSE
D,C. 20360
PREDICTION
1. Thls maintainability
handbook. V.as developed
\\itb es~abiisheu
proce[lur~o
?-~fcnqe ir, :!(,eord~ce
)
-.
InpIIlc!nn
This publication
ivas approted
in the mi]itar~ standardization
of
and
profidcs
information
on current maintainahilit!
It i~i]] prn~idr valuable information
and guidanw
and production
of
with the drsikm, de~)clopment,
requiring
a high order of maintamability.
Ever~ effort
b} the I)epartn)cnt
waintqi~~hilit~,
.....
. .. .
nrdirtion
MIL-HDBK-472
24 May 1966
FOR EWORD
The purpose of this Maintainability
Prediction
Handkmok is w familiarize
project managers
and design engineers
with current maintainability
prediction procedures.
To achieve this objective,
particular
care has been
exercised
in selecting
and including only those procedures
which are currently
used in predicting
the maintainability
of equipment and systems.
The hi@lights of each maintainability
prediction
procedure
are presented
m a clear,
lucid and irttell igible manner and include useful supplementary
information
applicable
to specific procedures,
using the following format.
1.0
GENERAL
Philosophy,
Assumptions
and Summary
Applicability
Point of Application
Basic
Parameters
Information
of Measure
Required
Data Basis
Correlation
Between Predicted
2. ()
ANALYTIC
KXJNDATION
3.0
APPLICATION
and observed
Values
ii
MIL-nDBK-472
24 May 1966
Prediction
facilitates
anearly
Assessment
of thematurity
of the maintainability
design and enables eariy decisions concerning
the compatibility
of a proposed
&sign with specified
requirements
or the choice of better alternatives.
The maintainability
prediction
procedures
I and III are applicable
solely to
Proccdurcs
II and J\ can be used for
systems
and equipments.
electronic
all systems
and expnprnemts.
in applying procedure
II to non-clectromc
equipments
the appropriate
task times must be estimated.
In conclusion,
and production
bility.
.,.
Ill
MIIm HDBK-472
24 May 1966
TABLE
OF CONTENTS
Page
INTRODUCTION
The Need for blaintainability
Prediction
Definition of Maintainability
Basic Assumptions
and interpretations
Elements of Maintxiinability
Prediction
summary
MMNTAINABILITY
PROCEDURE
SECTION
1,1
1.2
1.3
1.4
1.5
1.6
1.7
2.3
1-1
1-1
Phi losoph!,
Assumptions
and Summar.v
Appl]cabjlity
Point of Application
Basic Parameters
of h!easure
Information
Required
Data Basis
Correlation
Between Predicted
and Observed
2,0
1-1
i-l!
1-2
l-~
Values
Structure
of Time Elements in Fix of Malfunction
Elemental
Activities
2. 2.1
Assumptions
Relating to Elemental Activities
2. 2.2
Major Characteristics
of Elemental Activities
Synthesis of Time Distributions
2.3, 1 Modes for Synthesizing
Distributions
of Time
2. 3.2
Cumulative
Time Distributions
and the Monte Carlo
Method
SECTION
3. I
3.2
PROCEDURES
1.@ GENERAL
SECTION
2.1
2.2
PREIXCTION
Techniques
300
}-2
1-3
1-4
1-4
1-4
]-12
1-14
1-14
1-16
1-16
1-17
APPHCATION
Preliminary
Procedure
Steps of the Prediction
1-19
Procedure
1-19
MIb~BK-472
24 hhy 1866
TABLE
OF CONTENTS
(Continued)
Page
ADDENDUM
A CALCULATION
OCCURRENCE
ADDENDUM
B DEFINITIONS
PROCEDURE
SECTION
1.1
1.2
1.3
1.4
1.5
1.6
1.7
2. z
-.
TERMS
2.0
3.0
AI-44
2-1
GENERAL
ANALYTIC
Al-37
2-1
II
2-2
2-3
2-3
2-3
2-3
2-5
2-5
2-5
2-6
2-6
2-7
2- 7
2-8
FOUNDATION
Basic Considerations
aad Definitions
2.1.1
Applicable
Maintenance
Taska
2.1, 2 Applicable
Flmctional
Levele
2.1.3
Replaceable
Item
principles
of the hiaintahability
Prediction
SECTION
3.1
OF MAINTAINABILITY
Philosophy,
Assumptions
and Summary
Applicability
Point of AppU cation
Basic Parameters
of Measure
1.4.1
Corrective
hlaintenance
(Part A)
1.4.2
Corrective
Maintenance
(Part B)
1.4.3
Preventive
Maintenance
(Part B)
1.4.4
Active Maintenance
(Part B)
1.4.5
Maintainability
Index (M 1)
Information
Required to Apply This Procedure
Data Basis
Correlation
Between Predicted
and Observed Values
SECTION
2.1
1.0
OF THE PRO134BIXJTIES OF
OF ELEMENTAL
ACTMTIES
Rocedure
2-8
2-8
2-9
2-1o
2-1o
2-11
APPLICATION
PART A Corrective
Maintenance
Rediction
Procedure
3.1.1
Maintainability
Prediction
Durtng Final Design Stage
3.1.1.1
Prediction
Procedure
3.1.1.2
Determining
Functional
Levels of the
Equipment or &stem
3.1.1.3
IMermin!ng
Functional
Levels at wMch
Maintenance
Features
are Effective
v
2-11
2-12
2-12
2-12
2-14
MI& EDBK-472
M May l@36
TABLE
OF CONTENTS
(Continti)
Page
3.1.1.4
3.1.1.5
3.1.1.6
3.2
SECTION
1.5
1.6
1.7
1.0
2.4
2.5
2.6
2.0
2-27
2-27
2-29
3-1
III
3-1
GENERAL
ANALYTIC
Gene-l
3.1.1
3.1.2
3-1
3-2
3-2
3-2
%3
3-3
3-3
3-3
3-4
Values
3-5
FOUNDATION
SECTION 3.0
3.1
2-26
Philosophy,
Assumptions
and Summary
Applicability
Point of Application
Basic Parameters
of Measure
1.4.1
preventive
Maintenance
1.4.2
Corrective
Maintenance
No=&non
RuwIred
Data Baei~
Correlation
Between %edicted
and observed
SECTION
2.1
2.2
2.3
2-15
2-26
htntnia.inabili~
PROCEDURE
1.1
1.2
1.3
1.4
Prediction
Obtain@
Predicted
Pammeters
Calculations
3-5
3-6
3-9
3-9
3-11
3-11
3-12
3-12
Times
APPLICATION
3-14
Approach
Determination
*p-by-Siep
314
3-14
3-14
of Sample Size
Procedure
for Calculating
Vt
Sample
Size
TWL-HDBK-472
24 May 1966
TABLE
OF CONTENTS
(Continued)
Page
3.2
3, 3
3.4
Sb?p-by-Stap
Procedure
for Calculating
Step-by- Step Procedure
for Performing
Calculstio~
of Maintenance
Indices
ADDENDUM
Task Sample
Ta 8k Prediction
2.0
GENERA L
ANALYTIC
3.0
A3- 33
A3-47
A3-!i2
4-1
Values
,OUNDATION
General
Theoretical
Considerations
2. 2.1
End Kern Identification
2.2.2
qeratiunal
Ihtnctton
2.2.3
Coirective Maintenance
Action Definition
2.2.4
Ma Ifurction Detection Analysts
2.2.5
Prevcntite
Maintenan&
Task Time Analysis
2. 2.6
Corrective
Mai,,tenance
Task Time Analysis
2.2.7
Tots? Maintenance
Task Time Analysis
.
SECTION
A3-33
4-1
Philosophy,
Assumptions
and summary
Applicability
Point of Application
Basic Parameters
of Measurement
Information
Required
Da* Basis
Correlation
Between Predicted
and Observed
SECTION
2.1
2.2
1.0
IV
3-16
?-IF
3-30
APPLICATION
4-1
4-2
4-2
4-2
4-2
4-3
4-3
4-3
4-3
4-7
4-7
4-7
4-7
4-8
4-8
4-9
4-12
4-14
BIBLIOGRAPHY
vii
MIb HDBK-472
24 May 1!466
LIST OF ILLUSTRATIONS
Table
1-2
1-3
1-4
1-5
1-6
1-7
l-b
1-9
1-1o
1-11
2-1
3-1
3-2
3-3
Page
i-ieadmg
List of Categories
and Elemental
Acttvtties
of .4ctive
Repair Ttme
Fitted Dlstrlbutions
for Complet~on Times for Elemental
Acti\ities
S}-nthesis of Time Distributions
Parameters
of Distributions
of Elemental Activities
Proposed
Format for Determining
the Cumulative
Time
III stribution
Frohability
of Completing
Eiemental Activity of Final
Malfunction Test by Designated
Time
Combining Matrix for Active Repair Time S.ynl.hesis
Observed
Distributions
of System Final Test Time and
System bgistic
Time
Combining Matrix for System Down Time Synthesis
Occurrence
Probability
Functions for the Elemental
Activities
System Readouts
Average Parts Failures
per Part per 106 Hours by Part
Category
Maintenance
Task Times
Interchange
Time
Element Times Based on the Work Factor System
An Example of the Use of Element Times for Determining
Interchange
Time
Part Class Failure Distribution
and Sample
Task Prediction,
R-7801
Maintenance
Analysis,
R-7801 /FPS-20
Size
1-13
!-!5
1-20
i-24
1-27
1-34
AI-36
AI-43
2-18
2-20
2-23
2-24
2-25
3-17
3-23
3-24
.,.
Vlll
MIL-HI)BK-472
24 May 1966
.-
LIST OF ILLUSTRATIONS
Figure
A
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
1-9
1-10
2-1
2-2
~- :
2-4
2- 5
2-6
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-B
3- fl
Figure
Page
Heading
Comparison
Matrix of hfaintainabilit~
Prediction
Procedures
Distribution
of Malfunction Active Repair Time for the
AN; ASB-4
Distribution
of System Down Time for the AN/ASB 4
D~stribution of Malfunction
Active Repair Time for the
ANr ARc-34
Distribution
of System Dow~ Time fcr the AN,; ARC- 24
Distribution
of Malfunction
Active Repair Time for
the MD- 1
Distribution
of System Down Time for the hlD-1
Structure
of Time Elements
in Fix of Malfimction
Distributions
of Times Required w Perform
Repair Actions
Resultants
Synthesized
From t a, t~andtc
Probability
of Occurrence
of System Logtstic Time
Functional
Worksheet
~ui-ktdleei
Worksheet
Worksheet
Worksheet
System
4
1-5
1-6
1-7
1-8
1-9
1-1o
l-lx
1-18
1-18
1-33
2-13
2-16
2-32
?-33
2-34
2-35
3-8
3-1o
3-19
3-20
3-21
3-26
3-1o
Comparison
of Population
ar. rl Sampling Mean Distributions
.Sample Size Nomograph
Maintainability
Prediction
Form
Maintenance
Analysis
Continuation
%eet
Maintenance
Diagram AN/GRT-3
AN/ FPS-20,
Transmitting
System
Amplifier-Mixer,
AM-1347 /FP&20
Expanded Iiew, Plate Power Supply
Plate supply, Block Diagram
Nomograph - Downtime
4-1
4-2
4-3
4-4
4-5
Procedural
Flow Block Diagrnm
Related Constraint
Matrix
End Item/Corrective
Maintenance
End Item/Preventive
Maintenance
End !tem/Operational
Functional
4-6
4-14
4-15
4-16
4-17
ix
Action
Action
Matrtx
Matrix
Matrix
3-27
3-28
3-29
3-32
MIL-HDBK-472
24 My 1966
INTRODUCTION
MAINTAINABILITY
PREDICTION
DEFINITION
foIlows :
OF MAINTAINABILITY:
$lM~n~n~i~ity
Ml bSTD-
77 ?i wimes
is a characteristic
maintainability
as
of
MIL-HDBK-472
24 May 1966
BASIC ASSUMPTIONS .4ND INTERPRETATIONS:
Each maintainability
prediction
procedure
included in this handbook depends upon the use of recorded reliability
and maintainability
data and experience
which have been obtained from comparable
It is also
systems and components under similar conditions of use and operation.
customary
to assume the applicability
of the principle
of transferability.
This
assumes that data which accumulate
from one system can be used to predict the
maintainability
of a comparable
system which is undergoing,
design, development,
or study.
This procedure
is justifiable
when the required degree of commonality
between systems can be established.
Usually during the early design phase of the
life cycle, commonality
can only be inferred on a broad basis.
However, as the
design becomes refined, during later phases of the life cycle, commonality
is
extendable
if a high positive correlation
is established
relating to equipment
functions,
to maintenance
task times, anti to leveis of mainte.aance.
Alt.ho@
the four techniques
contmned in this handbook have been proposed and appear to
fit certain applications,
it should be borne in mind that they have not truly ben
tested for generality,
for cxmsi stency one to another,
or for most other criteria
dealing with broad applicability.
It should also be borne in mind, though, that
experience
has shown that the advantages
greatly outweigh the burden of making
a prediction.
For that reason,
it is not the purpose of this document to deter
further research
or inquiry.
ELEMENTS OF MMNTAIN.ABILITY
PR ED! CTJON TECHNIQUES:
Each maintain.~i II+.. __~.-I{~+i~ +~ohiq,.o ,~+ili?~t mrnn-dvre~ whj~h are ~cifically
designed tO
EUL
A&.,
~.bw.-...,
,uti.
ti.
--
--------
~..--
Failure
(b) Repair
rates
of components
time required
at the specific
at the maintenance
assembly
prediction
methods
le~el of interest.
level involved.
There are many sources which record the failure rate of parts as a function of
use and environment.
This failure rate, is expressed
as the number of failures
per unit of time.
A typical measure
is failures
per 106 hours. The major
advantage of using the failure rate in maintainability
prediction
calculations
is
that it provides an estimate of the relative frequency of failure of those components
which are utilized in the design.
Similarly,
the relative frequency of failure
of components
at other m aintafnable
levels can be determined
by employing
standard re[iabili~
prediction
techniques
using parts failure rates.
Failure
rates can also be utilized in applicable
regression
equations for calculating
the
maintenance
action time.
Another use of the failure rate is to weight the repair
times for various categories
of repair activity,
in order to provide an estimate of
its contribution,
to the total maintenance
time,
MIIrHDBK-472
24 May 1966
-
..
I=
I -,
1-
MI~HDBK-472
24 M8y 1966
PROCEDURE
1.0
GENERAL
This procedure
is used to predict system downtime of airborne electronic
and
elecrro-mechanical
systems involving modular replacement
at the flight-line.
Just as a masonry
building depends upon the brick as its basic building block,
the procedure
relies on the Elemental
Activity as the fundamental
element of
downtime from which other mea sures of downtime are developed through a process
of synthesis
of time distributions.
Activity is a simple maintenance
action of short duration and
The Elemental
relatively
small variance which does not vary appreciably
from one system to
another.
An example of a basic elemental
activity would be the opening snd
shutting of a door, or opening and closing of a radome on an aircraft.
It should
be obvious that the performance
time does not depend upon the construction
of
the house or aircraft
provided that the door or radome are similar.
Therefore
if one should record the times required
over many trials,
to open and close a
door, he should be able to calculate the mean time, ~, and the standard
deviation,
C, of this Elemental Activity.
This is precisely
what has been done
in this prediction
procedure
for various Elemental
Activities
which in total
ccm.pr!se Lie basic Categories
nf Active Repair Time.
These activities
are
listed in Table 1-1 and the corresponding
recommended
values of ~and
@
for the Elemental
Activities
of each category
are shown in Table 1-2.
The technique
of using the basic building block, namely the Elemental
Activt~,
and building the structure
step by step to include other measures
of downtime is
explained
in 2.0 Analytic Foundation
and detailed in 3.0 Application.
1.1
Philosophy,
Assumptions
snd Summary
1-1
.
MIbHDBK-472
24 May 1966
Downtime but rather concludes with the determination
of System Downtime.
The
reason is that Total System Downtime is the result of combining the distributions
Initial Delay
of .System Downtime with Initial DelaJ. The methods of determining
require
some further refinement
before they will be considered
for Inclusion.
In summary,
tbe philosophy of the entire prediction
procedure
is based OD the
As already discussed
the synthesis
principles
of synthesis
and transferability.
principle
involves a buildup of dowmtlmes,
step by at.ep, progressing
from the
distribution
of downtimes
of Elemental
Activities
through various stages culminating finally with the Distribution
of System Downtime.
Tbe transferability
principle
embodies the concept
of system can be applied to similar
systems
under
environment
to predict system maintainability.
1.2
Applicability
This maintainabili@
prediction
procedure
is applicable
to predict flight-ltne
maintenance
of airborne
electronic
and electro-mecha.nical
systems involving
modular replacement
at tbe flight-line.
The procedure
may also be used for
maintainability
prediction
in echelons of maintenance
other than flight-line
such
as field or depot by extension of formulae
through further development
work,
+= \~~~.~& ~+-her ~!e~emt~l
~fi++~,itte~.
= rcqutrcd,
..-...
1.3
The technique can be applied at any time after the design concept has been established,
provided the essential
data enumerated
in 1.5 entitled Information
Required
is
available.
1.4
Basic
Parameters
of Measure
Information
Required
In order 10 perform
k available,
a maintainability
prediction
1-2
the following
information
must
MIbimBK-#a
24 Hay )9W
(b) Number
(c)
of flight-line
replaceable
comwnents
mmpments
Number of pmmme-rddnlng
(g)
Number
(h) Nmne
oontabhg
adjuu&nenta
urried.
ommeckm.
of teat potnts.
of apeetal
(1) Number
of each type.
teet equipment
(designed
specifictiy
of zna.gnetrons.
of durations
of verage
o)
Estimates
*)
Mimriixig scha&les
!0: spesatiorw
IU2dmaintenance
all ahtfta and all breaka for hmch, cof!ee, etc.
(1) Estimates
for intervals
debrteftng.
oecuphd
mleaion.
by tmachethled
peraonncd
activities
tncludtng
such aa
In the ortgtnd
development
of the prediction
procedure,
data wexw employed from
ma Munction repairs
on the AN/ASB-4 Bombtng and IUavi&ption Syutem( wed in the
B-52 bomber).
In testing and refining the prediction
system,
data were used from
seven other systems:
AN/APN
AN/ARC
AN/ARC
A N/AIC
89
34
65
10
AN/APX
AWARN
MD-1
1-3
25
21
MIIr HDBK-472
24 May 1966
Thus it is expected that this prediction
procedure
electronic
or electro-mechanical
systems
similar
1.7
Values
Correlation
Between
Predicted
and Observed
ANALYTIC
2.1
Structure
observed
and predicted
values
Time and System Downtime
FOUNDATION
of Time
Element6
in Fix of Mdfunctton
Dountime
(c) System
number
consiBts
of: ~~
system
System
System
comprtses:
@tatic
Time.
Repair Time.
Ftnal Test Tfme.
Malfunction
1.
2.
from
Initial Delay.
Sywtem Downtime.
i.
2.
3.
(d)
Downtime
is viewed
Repatr
Malfunction
Malfunction
Time consists
of Malfunction
Time
of:
Repair
Administrative
Active
Repair
Time.
Ttrne.
~/The calculation
of Total System Downtime is not shown in this prooedure
because it depends on the distribution
of tfmes of Initial May and the
procedure
requires
more development.
1-4
and the
MIL-XDBK-472
24 hlay 1966
Io.oo
6.00
6.00
S.oo
4.00
3.00
2.00
1.00
0.6
0.6
WI
0.4
0.3
02
alo
De
~6
.05
.04
.03
.02
.01
0.01
0.1
0-s I
IO
30
50
70
PROBABILITY OF COMPLETION
Figure
1-1.
Distribution
of fila)function
1-5
..
Active
90 95
9899
99.9 99.99
WITHIN TIME t
Rfpair Time
ML-H DBK-472
24 hby 1966
4J
0.01
0.1 0.5 I
10
30
50
70
PROBABILITY OF COMPLETION
F@ire
1-2. Distribu~iofi
of System
1-6
9095
WITHIN
wm
m,
99.99
YIME t
Down Time
MWHD13K-472
24 May 1966
PROBABILITY OF COMPLETION
Figure
1-3.
Distribution
of Malfunction
1-7
WITHIN TIME t
Active Repair
Time
MILHDBK-472
24 May 1966
6.0
5.0
40
3.0
2.0
z
K
=
o
z
I.0
4 0.8
:
+ 0.6
0.5
0.4
0.3
0.2
0.1
oh
0.1
0:5 I
10
30
50
70
PROBABILITY OF COMPLETION
Figure
1-4.
D~strlbut!on
of System
1-8
90 9s
*a 99
99.9 99 99
WITHIN lIME t
Down Time
MIGMDBK-472
24 May 1966
100
8.0 t
0i-ii
1 I 1 1 1 i
me
m-
I
b
PREDICTED
t-
2.0
1.0
Oa
0.6
05
0.4
0.3
0.2
0.I
0.01
0.1
0.3 I
10
30
50
PROBABILITY OF COMPLETION
Figure
1-5,
Distribution
of Malfunction
1-9
90 95
70
WITHIN
Active
9699
99.9 99.99
TIME t
Repair
MXL- HDBK-4 ?2
24 May 1966
100.0
80.0
60.0
30D
40.0
30D
20J2
lox)
B.o
6L)
So
40
3.0
2.0
I.0
0.8
0.6
0.4
0.3
0.2
0.1
0.01
0.1 0.s I
10
30
50
PROBABILITY Of COMPLETION
Figure
1-6.
Distribution
of System
1-10
70
90 95 98 w
WITHIN TIME t
Down Time
99.9 99.99
!$
k
------
1-11
MI L- HDBK-42
24 May 1966
Active Repair
(e) hlalfunction
1.
-.
2.
4.
L.
G.
(f)
Time is comprised
of:
Preparation
Tim~.
Alalfunction Verification
Time,
Fault Location Time.
Part Prc)curement
Time.
Repair Time,
Final Malfunction Test Time.
Elemental
Activities:
Each of the maintenance
categories
of (e)
above consists
of a series of Elemental Activities.
as shoun in
Iahle 1-1 and discussed
in 2.2.
_ Elemental
-.
Activities
To facilitate
analtsls and minmlize variations
in the performance
time required,
each malfunction
rnwntenance
category is subdivided into smaller maintenance
As (Iisrussed
It] 1. u these represent
actions,
labellcd Elemental
Activities.
rclati{$clj simple md brief maintenance
act]ons,
which require a short time to
pc rfo rm.
Therefore,
in the prediction
procedure
x origin~ly
developed,
the
assumption
was that Elemental
Acti\ritics were normally
distributed.
However,
after further refinement
of the proccdurc
three optional distributions
are available
ior use in a prediction
as follows:
@ee Table 1-2. )
(~) The fitted
normal
lng-norms!
(c) A corrected
distribution.
distribution.
time log-normal
distribution.
The determination
of which Elemental Activities
are normall}
distributed,
and
\vhich Iog normally,
is based on the h}~thesis
that Elemental Activitieshaving
standard deviations
less than an arithmetic
mean of one hour, are representative
of activities
of a routine nature and are assumed to be normall~ distributed.
This is due to the fact that execution time is not significantly
influenced
by
changes of personnel,
characteristics
or surrounding
events.
On the other hand, Elemental
Activities
having standard
deviations
greater than
than one hour, are conthe arithmetic
mean, or an arithmetic
mean greater
sidered both as more complex and as containing manv possible subactivities,
in such a case
all of ~vhich may not need to be performed
to define the achvlt}.
1-12
MI L- HDBK-472
24 hfay 1966
I
TABLE 1-1
AND FLFMFNTAL
ACTIV!TTES
I
Elemental
Categrrry
Sjstcrr
turr{
ActIvIty
on,
narrmup,
cl PIUB lime
Gaming
I
!
access
~Acl,L!tv~o.
Activlt~
and couders
as ncccssar).
component
stabilizatlotr.
raoorne.
test equipment
Checking
Prnerrrmg
matntcnance
components
dIals
particular
and remstalllrtg
Obtalnmg
covers
and or Tech
Iother
than radome).
Or&rs.
records.
IrI antlclpatlon
of need.
equipment.
cd,.
IxlI:4tw2ns
Obscr}mg
scttlng
awatt]ng
10 $erifv maliunchona
inherenll)
test problems
or checks
Ieaks
tlusl$t!
Fault
self -ewdent
from
meter
uniI(s)/suburtil
Swilchmg
and/or
wil(hln~
anrl/or
anti
bl~klrrg
on ground.
2
3
5
SymPIOm(IIJ.
specifically
Ior
tiIs
equ!pmerrl.
6
7
observation.
analysis
srttlngs
only (from
of controie.
knowledge
~exrlence).
rcadmga.
Removmg
iiemovmg
s.wnptom
symptoms
by mental
dIsPlsys M djfltrent
Inler?sretmg
n(,l reproducible
4
or nOn-eSt91ent
Interpreting
Lnterpretmg
REPAIR lTMr
OF Aml!rE
substikttmg
suhslitutirrg
ciwchng
in shop.
tmtt(s)ltrubttnlt
{s).
part(s).
parsa.
check.
3 WISUS) mle~lty
Cbcckmg $Ill!ages.
cnntmulty.
Wawfnrms
andinr
aipal
tracing
ConsulilnR
Ttw+ Orders.
CorrlerrtnE
wl!h Tech Reps or other mamtenunce
personrrel.
Performtrrg
standard
tom problem (n].
laolattnK
l:sin~
pressure
dcslgncri
sm.c~flcally
Part
CSbtammg
Obtaln]ng
replacement
component
from a[rcraft
spares or tool &x
replacement(s)
from &nch,
ahop. or pre-iseue
stock.
Prcmutwmcnt
CMtatrrlrrg
replacement
Atlemptlng
10 Obtain
Rcpalr
Replacing
unlt(rs)/subun!
parts.
Correct
Making
M.!klng
wtg improper
atdjustmcnls
adjustments
Baktng
mogrretmt.
procurement.
Rcpstrm~
Final
Mal-
functmn
Test
crrmporrcnt(sj
replacement
Replacing
Precautionary
...
.;
1
leak.
hctton
chcckou!
or delectl.e
2
3
tfs).
mstallatton
activity
and repsir
wirtng
hy cmmlballzation.
Unsvallable.
Crsmpcmrnt(sl
plug- kn mxrnactlon(s).
m sircraft.
In shop.
repstr
5
(Include?
t trees
S- called
fau!t
Ioca!lon,
parl
6
7
not vprlft~d).
8
or eonnerllons.
folloulrtg
I
2
completion
of repair.
l-I::
14flL- HDBK-472
24 May 1966
tbe tendency is for the applicable distribution
b be skewed to the right and is
assumed to have log normally distributed
completion times.
Table 1-2 shows fitted distributions
of completion
times for Elemental
Activities.
It is recommended
that the distribution
parameters
shown in this fable, which
are denoted by an asterisk,
be used in any prediction.
Predictions
based on
maintenance time.
these distributions
yield the best empirical
fit to reported
in those cases when a prediction
of txme time, rather than the time reported by
repair personnel,
is desired,
the distribution
of corrected
time is used.
(See columns 7 & 8 of Table 1-2. )
2. 2.1
A.ssumpticmc
The selection
assumptions:
RelatLng to E!ement~
and phrasing
of Elemental
.4ctivitics
Activities
is based
on the following
Activities
of an Elemental
design or support
in any maintenance
Activity correlates
facilities.
category
are independent
2.2.2
Major
Characteristics
of Elemectai
Activities
i-14
r;;i,
$&!:
... .
+
MIL-11D13K-j7Z,
24 May 19GG.J!.
,
,.. ..
; .
T&OLK 1.2
,. .,.
.,,.
.,,,,4.
.,.
G,
. :;.
-..,.
,:
FIT7EIIDISTRJBUTK)NSFOR COMPL@ON
TtAfESFOR ELEMENTAL A~ti
,!
.,
,!, !#,
+
,...
, ,.
Dlotrlbufhm
3
Cormcled
Tlmc
Number
(Img
.4
0.068
0.085
0.606
0. 3R4*
0. S:!6
0.196
0.448
0,104*
0.0s0
0.127
0.Slrl
0,204
0.21%
o.195*
0.112
0.Octl
0.0G9*
0.ltts
0.693
0.3!)7
0.329
i.394
0.309
0.059
0.263
0.416
(j.
4X
0.605
1.236
0.171
0.i3?
1
2
3
4
5
6
7
@
~
Jl)
11
12
0.010
0.019
0.333
0.141
0.821
0.324
0.436
0.181
0,)40
0.R07
G.ool
o.04U
o.25!*
fl.172
o.7BFI
0.346
0.480
0.191
0.213
1.owl
11
1.4
!1. se?
O. 6R3
ij.?~l
~.420
O..A.
0.960
!5
P. :12(I*
0.)62*
3
4
s
6
7
n
0.102
0.665*
o.23s
O.27S
0.330
0.140
0.070*
0 107
1
2
3
4
5
G
7
Prep2ra2
llom
.,
Md-
Iundon
Vtrf(i
c8110n
Fault
LocalIon
- (\
0. 4G6
o.0:1(1
o.z3a*
0. 31s
0.171
0.199 0.144
I I-1.iliz
PW4
Proou*-
ment
3
.,
K@
0.344
0.135
1
1
2
3
4
,.,
r,
1
8
I 0.313*
;.394
0. .&qfi
o.W41
0.415
0.997
1.416
0.7S2
0.810
0.s18
0.s0:
().0s5
0.541
0.862
0.701
0.747
Normsl)
o.Onz
o.5s6
0.198
0.200
0.190
n.109
11.058
0.008
o.ri43
0.64s
0.643
0.643
1.011
0.643
0.643
1.011
0.0.391,409
0.485 0.63:1
0.2rIl?0.W18
G.137* 1.215
[.047* I.o.wl
).269 0.520
).024 1.345*
0.O.?*
0.469
0.?Co
0.1!33
1.019
0.261
0.023
1.011
0,641
1.011
1.011
1.011
0.643
1.011
).010 n.100
1,201
).2G3 0.686
!.
OIW* 0.955
).532 0.808
).221* G.ei:~
}.293* o.66t*
).125 0.865
).077 1.095
).w? 0.!t65
).
240 0.444*
).
342 0.785
J,321, 0.78!l
1.WIG* 1.044
1.u15 0.477
0.010
0.008
0.256
0,0S6
0,672
~>:::
0.1:2
0,074
0.491
0.233
0.333
0.413
0,364
O.277
0
1,011
0.643
1.011
1.011
*
. . nll
. .
1.011
1o11
1.011
1.011
1.011
0.643
0.643
1.?11
0.643
).012
).249
1.277
1.161
1.086
0.675
0.S08
0.649
0,012
0.242
0.269
0.1s7
1.011
rJ.643
0.647
().643
t.239
).203
).051
1.
253*
1.
7s0
1.
%9
1 534
1.002
0.231
0.138
1.011
0.534
o.6s7
1.023
0.6GI
0.54>
o.93n
).008
1.122
O.726
n.997*
0.749
l,4GFG
0.U29
0.284
1.011
0,049
0.246
0324
1.238
0.515
[ 0,461
0.64?
1.011
0.643
0.643
1.011
1.011
J_
9A
leg normal
diet
rtbkbns,lhe
dlstrlbutlon
whl+
M Mterld(
1-15
marked
wtth
,..
1
MIL-HDBK-472
24 May 1966
For a majority of the Elemental
Activities,
the probability
of occurrence,
P,
is predicted
for a particular
system by solving an equation generated by a
Table 1-10 list~ the occurrence
probamultipk-linear
refgramion
antdysis.
bility function, p, for a number of Elemental
Activities.
In these equations,
tbe dependent vsxiatdes
are based on observations
of comparable
systems,
and
the independent
variables
~e certain qusmtitatfve
characteristics.
The occurrence
of multiple Elemental
Activities
is a function of the probabilities
of the Constituent
Activities;
thus, the probabili~
of conjunctive
occurrence
of
multiplied
two independent
activttiea
is the product of the indtvldual probabilities
by the probability
of non-occurrence
of all the remaining
activities.
For three
independent
activities
(A, B and C), the probability
of joint occurrence
of any
*O such as A and B and a non-occurrence
of the third C would be:
In this type of notation the bar over a letter indicates the probability
of nonoccurrence
and a plain capital letter is the probability
of occurrence.
2.3
Byntbesis
of Time
Distributions
Distributions
modes,
There are four synthesis
fail.
These are described
as:
depending
of Time
on the manner
in which a systam
CM
1-16
MILr HD13K-472
24 May 1966
Llnequal Sampling, Adding Iariates:
This applies when only onc
failure occurs and the others will probably occur at a rate corresponding
to their occurrence
probabilities
(the relative frequency,
determined
from failure rate data).
In this case, various combinations
of the
repair action will take place.
(310de 2, Figure 1-9. )
This applies when only one
Equal Sampllng,
Not Adding Variates:
failure can occur at a time and each component ha6 an equal probability
of failure.
Only one of the repair actions is possible to correct
system failure.
(Mode 3, Figure 1-S. )
Unequal %rnplin~,
Not Adding Variates:
This is applicable when oniy
one of the components
can fail at a time. in the order of their relative
probability
of failure,
The repair consists of perfomn ing only one
reptir
action at a time, since failures
wi!l probably occur one at a time,
in accordance
with the relative frequency.
(Mode 4, Figure 1-9. )
the basic principles
of synthesis.
Note that three
Figures 1-H and 1-9 illustrate
norms! distributions
of time are shorn in Figure 1-8.
Two of these distributions,
while the distributions
to time tc for the
te and tb, overlap to a certain degree,
third does not, to any extent.
Figure 1-9 shouts the resulting
synthesis
for each of the four synthesizing
modes.
Note that resultant
mode 2 is skewed to the right.
This is due to the effect of tc,
wjth its larger mean and its own distribution,
intermingling
with the sampling of the
other two. This is the mode which can be expected to occur in a complex system.
~. ~. ~ ~~ulative
Time
D*@rlb~ti~n~
~d
Method
cumulative
distributions
such as Figure 1-1. to
This involves the selection
of times, t, by
value of the abscissa
and reading Lhe corresponding
random selection of probability
is performed
by
1-17
----
Mlb HDBK-472
24 May 1966
F we
1.8.
Distributions
of Times
Required
to Perform
Relulr
&uQ!0
!/<(02
!,%
Figure
1-9.
Res~tan~s
%nthesized
1-18
From to,
to and tc
Act ons
MJb HDBK-472
24 May 1966
using a table of random numbers,
or by utilizing a random number generator
when
a computer
is employed,
to select as many values asneedcd
for the prediction
procedure
for the desired parameters.
Typical
of the mrthod
is the s}mthesizing
Times.
In this case
of Elemental
the normal or log norrrlaI distribuof random times is nbtalned as explained
tions as required,
and the selection
in ~Section 3.0 .application,
APPLICATION
?. 1 Prelimina~
(a)
Procedure
in terms
of its constituent
input and
fb) Compute the failure rates of the system and of the flight replaceable
components
listed in the }~gend ~f Tab!e 1-1o vti!izing accephbie
standard
reliability
prediction
procedures.
(c!
Finf~ :he number of system readouts and calculate the readout factor
as outlined in Table 1-11 of Addendum A. This readout factor is used
of system
final
to multiply each value of time given in the distribution
test time in Table 1-6.
(d) Estin]ate
to be
3.2
an a~erage
flight length
(in hours)
during
which
the
system
operatccl,
Procedure
probability
of each Elemental Activity as
Step (1). Compute tie occurrence
in Table 1-10
outlined in .4ddenaum A, The act]vit]es
marked with an asterisk
cannot occur with any other activity wilhin the category;
therefore
they should
be omitted from the calculations
in Steps (2), (3), and (4) below.
Step (2). Determine
a!one by multiplying
the probability
of occurrence
i~s computed tolai occurrence
is
.. .
>--
~.
HDBli-4/i!
24 May 1966
1
i
i
i
P
,
E
I
b
;
(
(
8
;
.:. ;
..
.
---1
.
1
--L.
.
---
1-20
MIhllIX3K-472
24 Hay J 966
of the complements
Elemental
Activities
of the probabilities
of occurrence
&termine
occurrence
of arty No activltiee
Step (3). Within each category,
which can occur together.
Use the following fomnula for act!vltiesAl , A2, . . . An,
where, Al and AZ are the activities
whose joint probability
is to be computed:
%obabtlitieg
~ 1%
will
be
considered
b be zero.
(s). Within each category add the probabilities calculated in Steps (2), (3),
ti Table
1-10.
an
in a
MIL- HDBK-472
24 May 1966
P(AI)
.30
P(A2)
.40
P(A3)
.10
P(A4)
.05
(,0081)
( = 1%., omit)
(.0021)
( = 1%., omit)
(.0009)
( = !%. , omit)
.95
Obtained
by adding the
normalized
P(A21
(.10) (1.176)
= . 12, normalized
P(A3)
(.05) (1,176)
= . 06, normalized
P(A4)
Total
rim
Step (7).
Complete
(a) Insert,
Table
Probabi 1ity
1-4 as follo~s:
Steps
(3) and
(4)
above.
Ml L- HDBK-472
A May 1966
(b) Determine~ia
distribution
and Clap
Orpia
of the multiple
acti~tities
for the
and ala
Bzl
/38
from the following equations:
>
log normal,
(d) Complete
occurrence
(e)
/L I or if ~1
~1
<
I hour,
and N means
(see 2. 2).
The ~ 2 and
at
The equations
independence
values
normsJized
in Columns
7 and 8)
H=
in Column 7 and 8 will be identical
tn those
in Coiumns
1-23
htIL-~DBK-472
24 bflly 1966
TABL& 1-4{~tlnued)
1
Bade
Uemeold
Calegmry
Dltiri
Aetivtty
Number
butloo
butlon
Pmrunetera
PI
u,
0w8. )
(hrs.
LN
LN
LN
N
N
LN
N
0.213
0.60? 1.000
c.351
t).
344
0.466
0,430
0.541
0.582
0.960
0.603
0.162
0.320
0.140
Fault
(CO*
10
11
tirmed)
12
katloo
mtiti-
13
14
15
1
ckcLlrrewe
Pmbsbuny
motrt-
button
PsramelerO
Crz
fbra. )
0.077
0.507
0.240
0.466
0.582
0.396
0.320
0.430
0.54)
1.044
0.162
G.0L2
1.0M6
1.095
1.Mt5
9.644
-.
- -.
.
.
1
2
3
4*
--
.-.
0.022 c. $x!
0.313 O. 238
0.315 0.171
0.199 0.144
0.313
0.238
0.31s
0.171
0.144
0.199
.
-..
Rrpatr
0.394 0.5 8
2
3
9.360
0.066
0.055
0.41s
0.993
1.416
0.752
0.810
0.541
0.862
0.701
5
6
7
B
LN
LN
N
LN
N
LN
N
0.603
0.747
1.114
LN
1.002
1.122
0.055
0.997
0.862
0.468
0.747
o 239
0.203
0.066
0.2s3
0.993
1.269
0.752
0.476
1.030
--
---
Canoot
occur
wltb
sny other
activity.wltiln
the
J-25
cahgvry
MIL-HDBK-472
24 hhy 1966
Step (6).
This
step tranafoms
a normal
variable
with mean~zand
staadard
!09 ti
z,
C2
log p*
(lor parameters
of Table 1-4).
designated
LN in Coiumm 5
The values of/L2 and a2 me obtained from Columns 7 and 8 of Table 1-4.
The tl values are gtven in Column 1 of Table 1-5.
The cumulative
probiabtlitles
of completing
an Elemental
Activity by a designated
time tl are obtained by
normal
distribution
and obtatntng each desired
entering a table of the cumulative
probabili~
opposite the entry Z,.
These probabilltiea
are then entered in Cohnnn 2
of Table 1-5.
ColumxI 3 of Table 1-518 determined
by multiplying
the values in
Column 2 by the probability
of occurrence
of the Elemental
Activity which is listed
in Coh.urm 6 of Table 1-4.
Ibat is, mu!tipl y each cumulative
probabi 11ty of completing an Elemental
Activity
in a given time ( tl) by the probability
of occurrmce
of
categoxy.
the activity.
Thh3 1s to be done for each maintenance
Step (9). For each of the twenty discrete
values of time ( tl) listed Ln Column 1
of Table 1-5 sum the corresponding
Elemental
Activity probabilities
shown in
amurntng
there are 8 Elemental
Actlvit{es comprising
Column 3. For example,
the Category of Prepamition
each Elemental
Activity will have an individual
sheet such as Is illustrated
by Table 1-5.
Therefore,
there will be eight
probability
values,
one on each ahet?t, for each of the twenty discrete
values of
time (tI} listed [n Column 1. Summing each of these eight probabilltie6
(Column
3),
1-26
-
MIL-HDBK-472
24 May 1966
~
DA
II
u-
MXL-HDBK-472
24 May 1966
TABLE
1-6
-
PROBABILITY
OF COMPLETTNC ELEMENTAL
ACTTW~
OF FINAL MALFUNCTION TEST BY DESIGNATED TIME
f
~Time
(H~ur~)
Probability
0.01
0.009
0.02
0.022
I
I
0.03
0.043
I
\
t
0. 0s
0,07
0.083
0, 180
I
4
0.10
1
0.250
0.15
0.270
0.20
0.380
0.30
I
1
0.580
0,40
0.700
0.50
0.790
0.60
0.840
0.80
0.890
0.90
0.905
1.00
0.934
1.50
0.982
2.00
0.993
3.00
0.999
5.00
1.000
10.00
1.0;0
log Normal
Distribution
-==i
1-28
MIL-?KDBK-472
24 May 1966
-the d.istrhutlon
of malfunction
active
repair
time by
using
the combining matrix (Table 1-?), the inetruction8 thereon, and the Bix
category distritmtiona
plotted in Step (9), above.
(]2).
The following
selected probability.
of random
numbers
a=
time,
t:
/S
F (L )is a randomly
P = 0.027
administrative
10
To obtain
where:
ie u~ed to compute
-Q in [i - F(t)]
t=
where:
a table
equation
X3
Q and
- 0.233
X2
An acceptable
method
is to use
+ 0.521 X + 0.230
onci
.
0,310
- 0,064
X 1s a value of malfunction
repair
time
eelected
in Step (11) i
Choose a minimum of 200 Xs and for each value compute the corresponding
U
AISO select the mme number of F ( t ) ~Iues by H~pling
by me~e
and ~ vaiucs.
U6ing the three setg of values,
Q , B , and F(t),
of a table of mndnm numbers.
we obtain 200 (or more) values of t from the above equation.
Wp
time
where:
1-29
repair
time
plus
administrative
Mlb HDBK-472
24 May 1966
TABLE 1-7
COMBINT14G MATRIX
FOR ACTIVE
REPAIR
TIME
SYNTHESIS
(X = Occurrence)
Instructlorl:
For each row, uu.m randomly
eelectcd samplee of time* frwm tho~e
maintenance
categories
denoted by an x,
ThiB summation
will be
executed (M) (P) ttrnes for each row, where M is the desired number of
samples
of malfunction
active repair tlrne, and P is the indicated probability
of occurrence
of each mw.
Mai-
Row
Nm.~ber
I
t
1
2
3
4
5
6
8
9
10
11
12
function
Verification
Prep aralton
x
I
x
x
x
x
x
x
x
I
x
x
x
x
x
x
x
1
x
x
x
x
x
x
v..
x
x
x
x
~
xl
x
x
x
x
x
x
x
x
x
x
x
x
x
x
0.022
0.064
0.173
I o. 0%
x
xl
x
x
x
x
0.018
0:032
0.011
0.008
0.173
0.011
0.008
0.011
0.050
x
19
20
nlo
..
1 0.012
J
x
1 0.077
I
x
I
x
x
x
0.170
0.022
0.029
Y.
x
1-30
time
MILHDBK-472
24 May 1966
~or
is the median
~1 is the ~ystem
of the predicted
f~il~re
ra@; ~d
distribution
of mal.f unction
active
avera&
repair
tlrne;
flight len@.
When malfunction
repair time is mlItiplied by 0.95N, system repair time 16
during which
obtained.
The factor of 0.95 accounts for an overlap time observed
two or more malfunctions
are king repaired
concurrently.
Step 04).
times obtained
jn Step (]3).
Its probabll!ty
of occurrence
Sep (15). ThiB step involves system final test time.
was observed
to be about 0.5.
l%e probability
di~tribution
of final test times as
shown m Table 1-8 was obgerved mainly from AN/ASB-4 data.
A system readout
of the time completion
factir,
as computed !n Table 1-11, is used as a multiplier
probabilities
of Table 1-8 when a different
system 16 being considered. The kger
the number of readouts,
the higher the probability
of completion
of system ftnal
tegt h a given time.
.-.
numbers
hefiveen 00 and 49, and betwcn
50 and 99, respectively).
below
@nd including)
in f , gpaces &ve
line so ad x in r ~ spaces
1-31
Place an x
line 30.
MILr HDBK-472
24 May 1966
TABLE
OBSERVED
DISTRIBU110?W
1-8
OF SYSTEM
LOGISTIC
FI_NAL TEST
TIME
System
Time, t,
liours
in
0.15
0.20
0.30
0.40
0.50
0.60
0.80
0.90
1.00
1.50
2.00
3.00
0.50
0.020
0.055
0.60
0.80
0.90
1.00
1.50
2.00
3.00
4.00
5.00
6,00
8.00
10.00
15.00
20.00
40.0
50.0
Go. o
150.0
0.172
0.31
n,4s
56
0.728
0.790
0.830
0.945
0.980
1.000
O.
1-32
Logistic
Time
Probability
of
Time, t, or Less
0.010
0.020
0.071
0.08
G. i~
0.17
0.22
0.28
0.32
0.37
0.40
0.42
0.43
0.57
0.78
0.82
0.90
0.94
1.00
24
0.18 -
May
0.15 -
0.1
m!
0.05
01
NUMBER
Figure
~v
OF COMPONENT
1-10.
Probability
z
TYPES
REPLACEABLE
of Occurrence
1-33
AT
FLIGHT-LINE
of System
Logistic
LEVEL
Time
1966
MILIIDBK-472
24 May 1966
TABLE 1-9
MATRIX FOR SYSTEM ~W
COMBIWNC
(x
r
Syntheota
Number
1
~
system
S@em
System
Repmr
Final
Logictic
Time
Text
Ttme
I
1
I
I
6
3
10
11
12
13
14
15
16
17
lb
19
20
21
en
*23
24
25
26
27
2B
2:
30
31
32
33
34
35
36
37
38
39
40
x
x
x
a
x
x
x
a
x
x
x
x
x
x
Y.
x
x
x
x
x
x
x
41
42
43
44
45
46
47
48
49
50
1
53
Ii
54
5s
56
57
5*
59
60
61
62
63
64
65
66
67
6B
69
70
71
72
?3
74
7s
76
77
7b
79
130
81
82
83
84
8s
x
x.
x
Syxtem
/ SYUm
Lopsllc
TIrxxe
Time
?::
51
52
x
x
fv8~,
Time
x
x
Synthesis I
Number /
I
TIME SYNTHESIS
Occurrelme)
Ttmr
T
2
3
4
5
6
7
I
,
1
x
x
98
99
100
1-34
x
x
...
x
x
x
x.
x
x
x
i
1
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
!
I
x
x
lx
I
86
87
RR
89
90
91
92
93
94
9s
96
97
MIb~BK-472
24 May 1966
step (18) . Select a quantity,
Q, of an integral number of hundreds
(no fewer
than 200) random times from the distribution
of system repair times, Step (14).
Select O/2 random times from the distribution
of system final test times,
(step 15). %Iect Q Nzo r~~m
times from tie digtrib~ion
of lo~stic
time
plotted in Step (16). Put these times in place of Xs in Table 1-9 and add
them across each row to obtin
system dowtimes.
Plot the distribution
of
system downtimes and draw the best fitting line through the points.
1-35
MIL-HDX-472
24 May 1966
1-36
ieft blank.
MIL- HDBK-472
24 hIay 1966
ADDENDUM
CALCULATION
OF THE PROBABILITIES
OF OCCURRENCE
A CTMTIES
The probabilities
of occurrence
oc~~rrence
probability
~ctions,
OF ELEMENTAL
of Elemental
Activities,
calculated
are given in Table 1-10.
These
from the
functious
determined,
for the moat part, by subjective
selection of those system
chararteriatms
wh lch are deemed logically responsible
for the occurrence
of
the activity.
The selected characteristics
were then used as the independent
variables
in a multiple-linear
regression
analysis,
with tbe dependent variable
being the observed occurrence
probability.
The foIlowlng statements
concerning
the occurrmce
probability
functions must be complied with:
were
(a) If P
t9
a calculated
occurrence
of any term
probabfl!ty,
O, delete
For explanation
of the symlmls ueed in the occurrence
refer to the legend of Table 1-10.
Al- 37
and if P <0,
that term
probability
set P O.
and
functions,
MIL-HDBK-42
24 May 1966
LLEMESTAL
Svtim
dtds
AC T2VIll
Lurn-oo,
-win-@.
BWW
d ceurneru aa necesaq
lICUIU eOmpm-74
3
~nimg
GeImIU
md
ClOaIU
MCem
eowrt
Par-
wallt~
mabti lu4t0n.
4*/A,
rebmr
L,t/A,
rctnetd;tq
(l, -A,,)
)8s5-0.022 n, -Q3t9
A,,ol, -11
I
Mdfunciton
Verlf
1-
Chnew:ng
Indlrat,
m? ofdy
USIIW Irst
rqutpmeni
le..
tlon
2
JO
to verlf)
mdlu4ctt004
Ialwewly
produ:tbl~ on WOWS
1:
or
12
AW+
Perf~mmg
MM
~-13
Altcn-@Ing
? Uf,
pmblem~
USIM
CPH42
drDv&
0065 V441
letit.
tumve
wpmmbl.
10 dmrvve
or mo-emateal
14
D07
R-
chcckc.
qwpmen!
tern
@eelficdtv
0997
O*Y5 Ip,,) -
1.627
(P,l]
*,O
for thlc
aymrrn
It
FmlI
I Ocu,e.
+
Mti
IW
YWJS! !tietrlt)
check.
1149
1P,,) f~, )
0.036
{?,.1
0670
(P,,)
IL
0.2!0
1?
2.
lF
1s
0.0?
20
0.012
lP, I
0030
0623
I
I
(%)
I
N,,
2 134
(p, ) -
0473
N,,
I
I
CMnm
cccur
with
my
IX*?
=ttvltv
wt!b!n
Ibr
CKCOIV
.41-3R
MIL-HDBK-472
24 May 1966
STEP
U(,
CYCCURKEY4CE
PROBAB!LtlW YV14CWN
AC7W2TY
X LEMEYJTAL
a??r rP,J
21
22
24
25
10
Qots
26
11
0.5 lf~
12
0.002 (+
26
13
{?~
14
moo4
so
15
[0.,,3 ,PJ
O.M 1P*)
-o o v?)
amz +
nz?l
W4,1 - O.w
(*#l - 0.0s7
1~~) - 0 0s$
- 0.oo1 (MJI
&)
0.096 (~,)
I-*,,
33
0.02
1.4 444*O)
s
37
onoo3t41 0.0?7 y
n
40
QO06 I%J
41
{P,,) {P,J
any ohr
=tMyy
-ltUa
O*FY
Al-39
0.024
a352 ~+ * 0022] M*
8
xl
wtth
T!
32
CM601 occur
0017
(?,)
42
002!
IP,ol
11
38
-0.03
* 0025
MIL-HDBK-472
24 May 1966
TABLE
( AS
to
1-10
be expressed
LEGEND
as failures
N,
- Number
of flight-line
N2
- NMber
system.
of different
t-ypes of flight-llnc
N3
- Number
of readouts
in system,
N,
- Number of system
test problem.
N5
- Number
of different
N~
- Number
dynamic
(electrical
of connectors
pressure
integrity.
N7
- Number
Na
- Number
of circuit
N9
- Number
of test points
N,.
N,,
N12
replaceable
readouts
types
components
in system.
replaceable
as determined
whose function
of spares
parameters
per hour)
aboard
mechanical)
(excluding
monitored
In
by use of Table
is to evaluate
cmried
or
components
1-11,
a standard
the aircraft.
which maintain
built-in
by built-in
test scopes)
meters.
in the system.
specifically
Lf yes,
N,.
Lf no,
Nlo = o.
contain
any information
N,l = 1.
If no,
Nll = O.
any auditov
If yes,
N12 = 1.
lf no,
N,z = O.
devices,
devices,
as determined
determined
In T*1c
(continued)
AI-40
used
= 1.
If yes,
contain
1-11.
MIL- HDBK-472
24 May 1966
TABLE
- Is special
otherwse
N13
1-10-
LEGEND
(Continued)
not
If yes, N,, = I
If no,
N13 = 0,
?4,4
- Does
Anticipated
average
if yes,
Nl~
Lf no,
N,4
o.
flight length
10
(in hours).
~15
td,6
N}7
N,.
lfN8
tester
go-no-go
boxes ?
tester
1.
If yes,
N16
T.fno,
N,6
=0,
go-no-go
If yes,
N17 s 1,
If no,
N17 O.
- Does system
N20
- Probability
of occurrence
Figure 1-10.
N21
- Readout
factor,
contain
one or more
-magnetrons?
of lo~stlc
as calculated
in Table
A1-41
t irne,
6-1.
as determined
by he usc of
MIL-HDBK-472
24 May 1966
TABLE
LEGEND
(Conttnud)
Pn
- Probability
Al
- Failure
AZ
replaceable
components
located
AS
components
for
A.
- Summation of failure
operation is reflected
components
whose
15
A~
- Summation of failure
contain adjustment(s).
A,
components
which
parts,
or both.
A8
- Summation of failure
lamps and hardware).
of flight-line
replaceable
parts
(excluding
19
replaceable
parts
for which
A12
A13
calculated
1-10-
in the nt~
step of Table
1-10
r*ee of fltght-hne
by a readout.
rales
rates
of flight-llne
~hceabie
replaceable
components
which
replaceable
component
A1-42
TABLE
MI L- KDBK-472
24 May 1966
1-11
S}STEM READOUTS
(A)
Quantity
I.
(B)
Normalized
Weight ing
Factor
(c)
Product
(A) (B)
Visual
A.
Information
----
Devices
----
0.05
Circuit
m *
Monitors
rim
n m_
n E_
(i. e. , voltage-current,
power-pressure,
and
frequency indicators)
C. CWhd~
o. 03+
0.003 (n-1)*
Ray IX S@avs
D. Optics
-------------
----
0.14
-----
----
0.05
----
----
0.03
----
----
0.03
-----
Il. Auditory
A.
Audio
B.
Transmitter
Sidetone
Total =
(Number of System
TcAaJ =
(Readout Factor)
Readouts)
x$ This ~cightin~
monitors
listed
f=tor
must be m~tip]ied
in Column
(A).
AI-43
by q
of t~e
ClrCUit
MIb HDBK-472
24 May 1966
ADDENDUM
DEFINXTIONS
OF MAINTAINABILITY
TERMS
1.
Preparation
maintenance
2.
Malfunction
previously
verfficadon
ttme ts the time spent testing
reported
synptoms
of malfunction.
3.
4.
Part procurement
time is the time spent by the maintenance
or trying to procure,
necessary
replacement
items.
5.
G.
?wffilf~l~tiuil finai kd
Lime is ihe time spent com-irm mg that the malfunction
in question
has been corrected,
after which time no further maintenance
is performed
on that malfunction.
i.
Malfunction
8.
Q
. .
k~stic
time is al! replacement
the ma.intermncc man is engaged
active
repajr
the system
of Items
to observe
man in procuring,
1 through
6, above.
procurement
time, except that time when
in the procurement
acliv~tv.
10.
I.n]tlal
delay time is the time between the moment the equipment
available
for maintenance
and the moment work is commenced.
11.
becomes
Al-44
MI LHDEIK-472
24 May 1966
12.
System achnirdstrative
time Is all system
matnteaance
ttme and log fatic time.
13.
14.
15.
16.
17.
test problem
is an evaluation
of the performance
of a system,
Standard
or any part of it, conducted by setting psmmeter~
into the system; the
parameters
are operated on and the result obtatned from system readouts.
AI-45
downtime
other
than active
prl.mafily
MIL- HD13K-472
24 May 1966
PROCEDURE
1.0
II
GENERAL
This maintainability
prediction
procedure
describes
the methods and techniques
Preventive
and Active Maintenance
which are used to predjct Corrective,
parameters.
As appljed in this procedure
corrective
maintenance
time includes only actwd
repair time which js the period when repair work is in progress.
Therefore,
time or log!stlc
of measure
as admtrdstrattve
it exclude~ such parameters
Mtidmance.
in deflnltiorm
of Corrective
time, ek. , which me usually considered
Similarly,
preventive
maintenance
time includes ordy the actual active time which
is required
for rcp~r
during a preventive
maintenance
shut down. It does not
include preventive
ma.tntenance
time wldch is expended while equipment
is
in operation.
Active rnaintcnance
combines
both corrective
and preventive
maintenance
because it includes the time when both of these activities
are actually being
performed,
The details imd parameters
of measure
are discussed
in 1.4.
Corrective
Maintmance.
There are two methods which are presented
for predicting
Th~ first method described
in Part A of this procedure
results in a maintainability
prwiict~on expressed
in hours becauae it utilizes
tabulated maintenance
task repair
times,
recorded
in hours,
which have been established
from past experience.
Those data are discussed
in 1.6 and tabulated in Table 2-2.
The second method,
does not use tabulated task timeshstiti
explained in Part B of this procedure,
it utilizes estimates
of m?n-hours
required
to perfom
a maintenance
task which
are Insed OI? past experience
)1 ~ analysis of the design with respect to malnten~ce.
The WO different
measures,
one in terms of hours which is representative
of
of manpower
aetunl elapsea
Ezne, and the d.her in man-hours
which is a measure
required
to complete a maintenance
activity h a given ttrne, have of necee!sity
rcsulteci in the development
of a different
symbology
for each method.
These are
mplained
in Parts A and B respectively
of this procedure
and summarized
in 1.4.
However,
once the repair times have been established
either in hours or man-hours
the actual ;~redi~tion uroce(tures
for hth Parts A and B are very similar
since
each uses LvOrk sneets whicn closely resemble
each other.
2-1
~HDBK-472
24 May 1966
Part A 1s solely concerned
with corrective
maintenance,
during the final design
singe of the product development
cycle, and describes
the method of obtxintng
a prediction
of the mean time to repair expressed
in hours when the information
&scribed
in 1.5 is availnble.
~/
Part B cover6 active maintenance
which includes both preventive
and corrective
maintenance
and details the methods for obtaining an estimate
of the mean time
expressed
in man-hours
for performing
botk types of maintenance.
1.1
Philosophy,
Assumptions
and Summary
of measure
in the field of MaMatnablHty
are
The two most important
parameters
required.
the durzttfon of downtime due to maintenance
and the number of personnel
Each are important
measures
of matntainabllity
and ideally,
both should be kept
at a minimum
if possible.
However,
for a critical mission,
the number of maMenance man-hours
required
may not be as important
as minimizing
the time required
to repair regnrdiess
of the number of men involved or the inefficiency
of their
utilization.
Conversely,
when downtime is not of paramount
significance,
the number
of ma-hours
becomes an important
pammeter
of measure
and control.
This preof measure,
diction procedure
outlines the methods of predicting
both parameters
the results
of which can be utilized for design improvement
or other evaluations.
7%1s procedure
aasumes that repair
times,
applicable
to certain equipment
of Hke equipments
and devices.
types, can be used to predict the maintainability
ittabtdatee
certain repair times which can be used as described
tn
Therefore,
2.0 and 3.0 to develop a prediction
of corrective
maintenance
in hours.
bother
assumption
is that reasonably
guod estimates
can be made of the
maintenance
tasks required
for corrective
and preventive
maintenance
~ analyzing
that the number of hours
basic features
of the design.
Moreover,
it is asaumed
or man-hours
required
to perform
these tasks for both corrective
and preventive maintenance
can be estimated
from the information
developed.
A fundamental
philosophy in any case, is that the magnitude
of the repair time,
for a discrete
repair,
is the sum of the individual maintenance
task times which
are required
for its completion.
Seven such maintenance
asks
are assumed to
effect the magnitude of maintenance
time.
These are:
Localization,
Isolation,
Disassembly,
interchange,
Reassembly,
Alignment and Check Out. The procedure
~/ Adaptations
of the ksic
prediction
method for each of the earlier
stages of the
product development
cycle arc detailed in MIL-M-23313A
and NAtSH7PS 94324.
These include:
Early Conceptual,
Equipment Planning,
Earl: Development
and
bite Development
Stages.
2-2
w
MIL-HDBK-472
24 hfay 1966
also assumes
that each of the maintenance
task times are a function of the
For e=mple,
it is postulated
that it should take
method and level of repair.
longer to replace a pan whicJ is wired than is required
to repiace a part which
is plugged in. Also that at the part level of repair it takes longer to perform
a repair task than at the subassembly
or equipment
level because Aesa time
iB required
for the discrete
steps involved at the latter levels,
Accordingly,
nine functional
lewe}a of repair are described.
These are: System, Subsystem,
Equipment,
Croup, Unit, Assembly,
Subassembly,
Stage and Part.
In summary
the assumptions
and philosophy of this procedure
are applied in
maintmwme
measured
in hours, or active maintenance
predicting
either corrective
time consisting
of corrective
and preventive
maintenance
measured
in man-hours.
The techniques
of application
are described
in 2.0 and 3.0.
1.2
Applicability
This procedure
can be applied to predict the corrective
maintenance
time, as
outlined in Part A, and the active maintenance
time, aa described
in Part B,
of shipboard
and shore electronic
equipment and systems.
It can also be used
equipments
and systems of other
to predict the maintainability
of comparable
branches
of the armed services
provided there is similarity
of design, we and
application.
This procedure
may also be applied to predict the maintainability
of mechanical
equipments
or systems if the maintenance
task times for the applicable
functional
levels can be established.
~)
1.3
Point
of Application
The maintainability
prediction
methods
only during the final design stage.
1.4
Basjc
The basic
follows :
1.4.1
Parameters
parameters
Corrective
described
in this procedure
are applicable
of MeRsure
of measure
Maintenance
are as
(Part A)
to this procedure
tasks
and functlond
2-3
is the Equipment
levels,
Repair
MIL-HDBK-472
24
M8y 1966
repair times
distributions
and is expressed
as follows:
= MTTFi
where:
A =
average
part failure
rate
times
in failures
to perform
times
where:
times.
maintenance
action
distribution:
(2-2)
MTTR
follow a l~g-nerma!
ERT
106 hours.
a corrective
follow an exponential
ERT = 0.69
per
(2-1)
d!stribut!cn
MtTR
0! repair
times:
(2-3)
MTTR
(2-4)
(M TTRG) is:
(2-5)
2-4
MXL-HDBK-472
24 hf8y 1966
1,4.2
Corrective
Maintenance
~e applicable
formulation
(Mc) in man-boure
fE:
(Part
B)
the mean
to obtatn
corrective
maintenance
time
(2-6)
where:
A=
Me
1.4.3
average
part
failure
rates
tn failures
per
10 hours.
required
to perform
a correctl
= the man-hours
task,
(Corrective
Maintenance
Time)
Preventive
Maintenance
(Part
maintenance
as:
:
ve maintenance
B)
time
(Rp) is expressed
(fMp)
@-7)
Zf
where:
action.
Mp = fie man-hours
required
(Preventive
Maintenance
Time)
of occurrence
f = the frequency
per 106 hours.
1. 4.4
Active
Maintenance
(Part
to perfomn
of preventive
a preventive
maintenance
maintenance
actions
B)
This parameter
is expressed
as the mean active maintenance
sum of the preventive
and corrective
maintenance
man-hours
maintain a product for a apec!fied period,
divided by the total
and corrective
maintenance
tasks required
during that time.
ii is expressed
as follows:
(2-8)
2-5
MI~HDBK-472
24 May 1966
where:
= mean
actl ve mainLmance
Wc
= mean
corrective
maintenance
time
(resulting
Wp
= mean preventive
maintenance
time
(during
~~
2 f
Li
= operating
tj
= calendar
Maintainability
1. 4.5
failures
time during
time
time
time,
ti)
tj)
mtes
of occurrences
the period
in operating
fmnn time,
of preventive
maintenance
tj
inventory
Index (MI)
(xx)
iict~ +
(Zf)
in operational
Iiptj
(2-9)
MI =
i
where:
t = operating
~C, ~p,
1.5
Information
time
tit tji
Required
X A, Zf
to @ply
defined.
This Prcmdure
packaging arrangement
muet lM Imown
breaiwbwn into tbe various equipment
subaeaembllea,
and parts can be
2-6
MI~HDBK-472
24 May 1966
(b) Diagnostics:
The diagnostic procedure
that would be followed
in the event of failure of each part m the equipment,
including
application
of all designed fault location features,
must be available.
(c) Repair
failure
Methods:
The repair methods
of each part must be known.
of all parts
and environmental
categorized
stresses
under
by
which each
Data Basis
operating
BeMecn
and maintenance
and planned
Predicted
maintenance
and Observed
equipment
personnel
Jalues
NDkXDBK-472
24 May 1966
maintenance
resuIts.
These equipments
from destroyers
to submarines.
2.0
ANA LYTIC
were
helng ut[lized
in many
ship types
FOUNDATION
Active Maintenance
time consists
of two basic components,
namely,
Corrective
and Preventive
Maintenance
time.
Corrective
Maintenance
ie the Maintenance
performed
to restore
an item to a satisfactory
condition by providing
correction
of a malfunction
which has caused degradation
of the item below the specified
performance.
Preventive
Maintenance
is the maintenance
performed
to retain
an item in satisfactory
operational
condition by providing systematic
inspection,
detection,
and prevention
of lncjp{ent failures.
Preventive
Maintenance
can be
either scheduled
CT unscheduled
depending upon the requirements
of the m,ission.
Active Maintentice
Prediction
assesses
tbe average man-hours
of work tn
perform
the required
corrective
and preventive
maintenance
tasks.
These
active maintenance
tasks do not consider
the effects on elapsed maintenance
time due to logistics
problems
or administrative
procedures.
The active repair
time estimate
of corrective
maintenance
predictB the downttme due to active repair
which is the result of a malfunction
causing system downtime.
The preventive
maintenance
time estimate,
on the other hand, predicts
the downtime duc to
preventive
maintenance
activities.
2.1
2. 1.1
Ba8ic
Considerations
Applicable
The corrective
tasks:
(a)
and Definitions
Maintenance
maint.enancc
TaSkS
action
is divided
Localization
- Determining
the location
without using accessory
test equipment.
(b) Isolation
- Detemnintng
by the use of accessory
of a failure
- Equipment
disassembly
to the extent
to the item that is to be replaced,
(d) Intercha=
- Removing
2-8
maintenance
to the extent
to the extent
(c) Di sassemblv
gain access
the defective
corrective
necessary,
possible,
possible,
to
the J eplaccrnent.
IvIILr HDBK-472
24 May 1966
(e) Reassembly
replacement
(f)
Alignment
adjustment
made necessary
Applicable
Functional
of the equipment
my alignment,
minimum
by the repair action.
tests
after
the
and, or
Levels
(W Swe
(c) Subassembly
_- Two or more parts which form a portion of an assembly
and which is replaceable
as a whole, but having a part or parts, which
are individually
replaceable.
(Example:
IF Strip, terminal
bmrri with
mounted parts).
(d) Assembly
- A number of parts,
or subassemblies,
or any combination
thereof joined together to perform
a specific function and replaceable
as a whole.
(Example:
audio-frequency
ampjifier).
The distinction
between an assembly and a subassembly
is not always exact.
An
assembly
in one instance might be a subassembly
in another where it
forms a portion of an assembly.
(e) Unit - Any combination
of parts,
subassemblies
and assemblies
mounted together,
normally
capable of independenPoperation
in a
variety of situations.
(Example:
electronic
power supply, radio
receiver),
A unit is normal!y
directly accessible
(e. g. , mounted
2-5
MI L- HDBK- 472
24 Ma} 1966
4
Q$!W
- .4
collection
of units,
assemblies,
or
subassemblies
or system,
but which
]s not capable of performing
a complete operational
function.
(Example:
antenna group, indicator group).
A group is not
normally
a replaceable
item.
which
k)
IS a subdivision
of
an equipment
(set)
Equipment
(Set) - One or more units and necessary
assemblies,
subassemblies,
and parts connected or associated
together and
inc!udins all necessary
cabling and accessories
to perform an
radio receiving
set, sound measuroperational
function.
(Example:
An equipment is not normaliy
a replaceable
ing set, radar set).
item.
(h) Subsvstem
- A combination
of equipments,
groups,
etC. , which
perform
an operational
function within a system.
Subsystems
form
the major subdivision
of a system.
(Example:
one station of a
communications
system).
A subsystem
is usually located on one
ship, or in one geographical
location.
(i)
Svstern (Electrical
- Electronic)
- A combination
of two or more
subsystems
or equipment,
generally physically
separated
when
in operation,
and such other assemblies,
subassemblies,
and
parts as are necessary
to perform
an operational
function.
(Example:
communications
system including all stations; fire
control system Including the tracking radar,
computer
and gun
mount).
2, 1. 3 Replaceable
A unit, assembly,
during corrective
.-.6
) Principles
Item
subassembly,
or part which is normally
maintenance
upon failure of the item.
of the Maintainability
Prediction
to be replaced
Procedure
The maintainability
prediction
method described
in this
utilizing the estimated
maintenance
task times required
specific maintenance
tasks which, in total, comprise
a
These maintenance
task times may be obtained
action.
2-1o
intended
procedure
depends upon
for performing
the
repair or maintenance
from tabulations,
IbflLr HDBK-472
24 May 1966
such as Table 2-2, of Part A of this prediction
procedure
or they may be estimated
task times is then
times as described
in Part B. The sum of the maintenance
in failures per 106 hours to obtain an
multiplied
by tbe failure raw, expressed
estimate
of the number of maintenance
hours required
for that specific maintenance
or repair action.
The mean value is then detemnined
by summing the number of
maintenance
hours and dividing by the sum of the failure rates.
A similar procedure
iE followed for preventive
maintenance
except that, in this
case, instead of using failure rate parameters,
fre~~cy,
(f) is used whi~ is
a measure
of the number of preventive
maintenance
actions for a similar pertod
of 106 hours.
Hence the mean preventive
maintenance
action time is the sum of
the product of frequency and preventive
mainteriance
time divided by the sum of
the frequencies.
The symbols which are used for corrective
and active maintenance
in 3.0 which describes
the application
of this procedure,
Failure rates for use tn performing
a maintainability
prediction
from, but are not necessarily
restricted
to the following:
(a)
Handbook
Reliability,
are explained
may be obtained
for Prediction
of !%dpboard and %ore Electronic
Equipment
NAVSHIPS 93820, Naval Ships Systems Command.
(b) Reliability
Stress and Failure Rate Data for Electronic
Military Standardization
Handbook,
MIL-HDBK-217.
Equipment
APPLICATION
The application
of the techniques
outlined for both corrective
prediction
arc described
in I%ti A and Part B which follow:
3.1
can
are
PART A Corr~ctive
Majntena.nce
Prediction
and active
maintenance
Procedure
As stipulated
in 1,0, General,
this procedure
is only concerned with the final
design stage phase of the product development
cycle.
The earlier
~d in~rTht prediction
procemediary design stages are not treated in this procedure.
dures for these phases of the product development
cycle are included in
NAVSHIPS 94324 and M1bSTD-470.
2-11
MILHDBK-472
24 M8J 1966
-
3.1.1
Maintainability
Prediction
During
Final
Desdgn Stage
TMS procedure
is intended to predict malnbbabfli~
at the flna.1 design stage.
It takes into account all of the pertinent
functional level and details available,
and is applicable
when the information
lfEted tn 1.5 la available.
3.1.1.1
l%ediction
Procedure
lWs matntainabtlity
prediction
pxwcedure involves determining
the detailed functional
level breakdom
of the equipment,
eatablidhg
the functional
levels at which xnaintenance features
are effective,
and at which maintenance
tasks will be performed,
and performing
the prediction
?med on equtpment design features
and approximate
part failure
rates.
3. 1.1.2
Determining
Iknctional
tivel~
of the Equipment
or System
2-12
MILPKDBK-472
24 M:ay 1966
L
-----
~1
I.ri
I
I
I
I
i
I
I
I
I
I
I
I
I
I
I
I
I
I
:
<
.,
I
-.
I
I
I
I
1!
I
I
1
1
I
1
I
I
I
1
!
I
I
l
,,,
I
1,
11
i
I
l;-=l
I
I
I
I
I
,
I
I
I
I
Q)
i
I
i
w
(
I
I
I
I
I
I
l-!
-..
2-13
MIL- HDBK-472
24 May 1966
3. 1.1.3
Determming
Functional
Levels
at Which Maintenance
Features
Are Effective
Localization:
The functional
level to which a failure
~s~ng accessory
test equipment
is indicated by ~.
b)
Isolation
The functional Icvel to which a failure can be located using
accessory
test equipment at designed test points is indicated by ~.
If there are no designed test points below the localization
functional
level, then isolation is assu.m ed to be effectt ve at the same functional
level as lmm.lizsttiort.
Cm
be
Iocamd
without
Checkout:
The functional level at which restoration
to normaI services
is
to be verified using self-test
or other testing facilities
is indicated by ~,
2-14
MIL-HDBK-472
24 May 1966
3,1. 1,4
Prediction
be
hardware
gear trains
chassis
clutches
sure-ws
{mccnanicaij
bearings
nuts
cams
washers
dials
hardware
feed-through
terminal
Vlen
studs
terminals
rollers
tube shields
and registers
Electrical
unit,
such as:
such as:
wiring
sockets
wiring
terminal
cables
printed
harnesses
for plug- in parts
boards
wtrlng
boards
fboard only)
the worksheet
covers a replaceable
modular subassembly,
begin the list on the second row of the column.
2-15
or subassemblies
assembly,
or
MTL-HDBK-472
24 May 1966
Km
.
._
u
Q
I
I
~,li
1
I
2-16
hfI&HDBK-472
24 May 1966
Llat the identifying names or type designations
of all
step (4). Part Type
parts opposite their re8pecti ve circuit &slgnationa.
men the worksheet
covers
a replaceable
modular
mkssembly,
assembly,
or unit, enter the designation
of the replaceable
item LOthe top row of the column followed by the designation
of each pm.
Step (5). Failure
Rate
List the failure mte
~, of each P*
opposite tbe
appropriate
part designation.
The fafiun
rate for each pafi and tube tn the
equipment
can be determined
by use of standard
reliability
prediction
methods
wtth applicable
electrical
and thermal
stress condltiona
considered.
Failure
rate data should be taken from sources
which condder
only tie
random
catastrophic
failures
asaocinted
with normal equipment operation.
%Itable
references
for failure rate data were listed in 2.2.
When using these references
care should be exercised
in not mixtng fathare
rate data of one with the other.
Once a selection has been made this same
reference
should be used throughout
uxdeas other failure rate data not included
theretn iR required.
in thlE case other sources may be used if justified and
approved.
IMermi.ne
the sum of the ~ values listed
space at the bottum of the A mlumn.
and record
~p
(6). Maintenance
Taak Times
(Columns 6, 7, 8, 9, 10, 11 and 12).
For individually
replaced
parts detexnnine the maintenance
task times through
the use of Table 2-2,
For each part, record theue task times in the appropriate
columns oppos!te the respective
part destgnationm
For replaceable
modular
subassemblies,
assemblies,
or units, determine
a single time interval for each
maintenance
task for the overall replaceable
item and record each Ibne interval
in the top row of the appropriate
maintenance
kmk column.
Table 2-2 provides
charts for determining
maintenance
task times for the localization,
isolation,
disassemble y, reaasembl
y, alignment,
and checkout tasks.
Interchange
time
is determined
through the use of work factor data aa shovm in Tables 2-3 and 2-4.
The use of the charts tn Table 2-2 requires
determining
the specific functional
level within the equipment
at which all maintenance
features
are effective.
As used in this procedure,
functional
level denotes the physical subdivisions
of electronic
equipment
at which diagnostic,
replacement,
and test features
we effective.
Tbe functional
levels under consideration
are:
1) part.
2) stage,
5)
unit,
6)
group,
7)
equipment,
8) subgystem,
3) subassembly,
4) assembly,
and 9) Bywtem.
2-17
MIL-HDBK-472
24 MaY 1966
TABLE 2-1
AVERAGE PARTS FAILURES PER PART PER 106 HOURS BY PART
Part
CATEGORY~/
Failures/106
Category
Ftilures/i06
Part
Calegory
Hours
Hour~
Capacltora
Rectifier.
2.96
Selenium)
(Power
FIud
Cermmic,
Feed-Thru
Ceramic,
(Otbcr
2.06
Relays
o. s?
Re~latorfi
2.48
Fixed
2.06
Than
Feed Thtu)
EkctroIytlc
Mica
0.46
Carbon
Composition
P@er
0.50
Carbon
Film
Variable
Air
0.13
Ceramic
O. OB7
P18st\c
0.00045
Vacwm
3.08
Metal Film
0.36
0.39
0.51
Other
0.60
Film
Wi r?would
1.40
Variable
0.42
Compocltlon
Wire !Vound
O. 84
0.058
Connector
25.47
Resolvers
Counters,
(TaUy
Mechanical
Register
Cryatale,
(Signal
4.54
Type)
2.98
Dmde
socket8
0.042
SWilches
0.48
Detector and
Synchrm
Miser)
C~stala,
1.36
Frequency
Hardw~
0,033
0.092
Electrical
Mechanical
Tranafomners
1.16
Trandntoro
1.03
Ttia,
Eiectron
24.66
Crt
Inductors
Focus,
R.
Pulse,
(including
ReceiVi~
Filter,
F. (MIs)
9,42
VR T*s)
Special
Pupose
Electrical
1.36
Tranmntttitg
c] U&B ATRO
Mechanical
2.19
Klyntron,
Metera
Vlbratoro
Motors
Reproduced
and
(inTR,
42.88
ThyrUron)
32.18
(Chopper)
2. 8s
25.47
1.80
S.84
Blower
Reao)ver
Synchro
Other
~/
544.54
Magnetrons
0.28
(Deflection,
1.80
from
Handbook
for
the
Prediction
of
Shipboard
and Shore
Electronic
Equ~pmeM
MIL- HDBK-472
24 May 1966
The charts
in Table
manner:
The appropriate
chart,
TUBES or P.4RTS. is selected depending
The application
notes for the chafis define
item to be replaced.
types of items each chart applies to,
on the t.vpe of
the specified
To ohin
localization
md isolation time. the chart is entered using the FUNCTIONAL
LEVEIS column headed by the type of replacement
that will be made.
For example,
replacement
of a part would indicate Column 1, a stage;
Column 2, a subassembly;
Column 3, etc.
To obtain disassembly,
reassembly,
alignment,
and checkout
times, the chart is entered using Column I only.
The appropriate
of maintenance
corrective
maintenance
task being considered.
task
column
to
.nrl
~~~~ing~.
n-IeChfJAid
Examples
The following
examples
are presented
to demonstrate
2-2:
in the Functions:
Level Diagram
(1) The Modulator Power Supplv. illustrated
in this example because of its apparent
of Figure 2-1. is demonstrated
localization
and isolation features.
From the Functional
Level Diagram.
localization
for the power supply is determined
to be effective at the
Equipment
Level.
Considering
the Power Supply to be a plug-in chassis,
then the Tubes Chart of Table 2-2 is used.
Since the Power Supply is
located at the Unit Level where failure can be removed and localization
is effective at the Equipment
Level, tbcn the chart is entered from the
.
unit Column (5) down to the intersection
of the Equipment row,
This row
is +&en carried over ta the intersection
with the localization
column where
the maintenance
task time is read as 0.037 hours.
2-19
.
L
.
J-
2-20
!mt
,-.
MIL-HDBK-472
24 May 1966
TABLE
2-2 (Coetl~d)
A.PPLICATYON
NO~S
wernrrlpl
(me
\eI1*r* on ske[
1.
IOLd
stm-
ar,
(rea Lslors,
vere#e
%eclf~c
mcbnlquea
include
Add
be ~Pl~ed
{eseltii~
~iel
to front
D@
pup-
Iuees,
ttifi).
treosforene?s.
etc. ).
10 all
lime
However.
time, $nlercho~e
t!me snd reeeaembly
flm*
11UUI08B Is dlfflcutt
to @Natn, 11 Imo not been Imtuded m
muet be dewrmiaed
lrom TdAc 1-3-13.
lube Iatlumc.
The
dest~r
corrsldcr
must
Ihes?
In
[a)
(b)
boerever, this Ie
TM ~verue
ttme (ntervale
in thIo cha~ ~ UII Include dm~rdetrettvt
time.
A&nlolstreilve
ewendcd
In Pam procurement ltlme ~ot
by malntertame
pereonoel
10 ObWnlog ~lecement
iechnlcal
nhtpboed
rmut Ine (!nepectlon,
eating,
04her
mllltary
rbAteo,
etc. ).
quipment
When
matntenenee
=mowd
@te tm
this functlond
level,
To determiw
uoc column
To
Iechnlques
R. Ior tubes.
ttetemnlnlng
properly
mad.
Irerrwotors,
elerlmnlc
dl~ueembly
cppllceble
httemkeage
ttme.
leek
DE ~ItWTIONS
chtrt
should be mmtled to IUUD nol w the Iront panel,
klyctrww,
lc. ). wl~d
In non-waled
moduleo,
wd
ell
Ihetors,
tn? tcbuktied
fn the PARTS
lmagnclnmB,
c~acttora.
AND
1)
t!meo
ttiree
NOTES
the time
Wproxlmste
.< stu@
!ndmaled
made
Lhu
the
LEVELS
checkou!
by the number
ltK
only
e-e
Federe!
SO Miums
locallxetlon
to Ike futvotmttal
level
!hrottgh
ehmrn
wh!ch Iallure
In
is bel~
bows.
I of WNCTfONAL
and multlply
Ieeturec
of FUNCTIONAL
instead use 0.000
twueemt#y,
in the ~roprlate
time,
enter
of @erU!onel
Elcclr,:
Cowratlon
(7%} could Poestbly
m)lfrmtent
nw
ettd ebeckout
M wftkeh tb
the &stgoer
tacks,
ehould mdy
teak ia per{omned.
Is bel~
modes
2-21
tn one type
vl~u~
of
mc~nt
h,phoe.rd
!ransmltl?r
MI L- HDBK-472
24 May 1966
(2)
In@rchange
Time
This ttme can be obtained from Table 2-3 for
of tubes and replaceable
itemg.
For items having mounting methods
not conforming
to the description
given in Table 2-3, the interchange
times can be
determined
from Table 2-4.
most
types
In determining
the interchange
time from Table 2-4, each deta[led step (unsolder
joint, remove unit, etc. ) tnvolved h removtng a failed ftem and in positioning
and attaching a new item must be accounted for with a specific element time
interval,
The same ttrne interval for an attachtng etep can be used as an appNMmation for a corre apondtng detaching mep.
The interchange
tune for an hem is
equal to the sum of all ttme intervals.
An example of determining
interchange
time for a part {B shown in Table 2-5.
Step (8).
Rp
Determine
the repair time ( Rp ) rewired
for Perform:llg
corrective
maintenance
action in the caee of failure of each replaceable
adding the values recorded
on the worksheet
in Cohm.ns 6 through 12.
For individually
and is recorded
designation.
For replaceable
modular eubaseembliee,
aeeemblies,
once for the overall replaceable
item, and is recorded
top row of the Rp coluznn.
step
(9).
part failure
~RP
When individual
rate
{ ~ ) by the reepecttve
~RP column
a o~~e
item by
oppoeite
the respective
parts
or uniti,
Rp is ~te~~~
on the worksheet in the
are replaced,
multiply
value of Rp, -d
record
part
2-22
designation.
each recorded
tie pr~ct
~ the
TABLE
MIb HDBK-472
24 May 1966
2-3
INTERCHANGE
TIME
AVERAGE
Plug-i n tubes
TIllE
(HOURS)
0.015
Wired tubes
(4 wires)
0.149
U:red tubes
(more
than 4 v.nres)
with shield
with C~iUllJl
with C~
All tubes
ADD
ADD
O. 007
O. 027
ADD
0.007
fuses
Plug-in
Screw-in
fuses
All fuses
with screw
0.010
0.015
cap
0.014
PARTS OTHER
PART TYPE
-.
IJzrts with 2 vnrcs or 2 tabs
to be soldered
ADD
Parts
nuts,
ADD
For attachments
change time by
TIME (HOURS)
0.081
NOTE:
AVERAGE
0.081
0.027
+ 0.034
not conforming
to the above type deterrnine
using
work factor dat% Table 2-13
nut and
the inter-
MIL-HDBK-472
24 May 1966
TABLE 2-4
ELEMENT TIME9
ELEMENT
BMED
ON THE WORK
FACK)R
SYSTEM
ELEhfENT TIWE
{Hours)
DESCRLPTION
tubes.
Tube
or
cw
plug-in
parts,
etc.
o.007s
0.003s
shield
Fume
loseR
Inseul
Intohorizontal
2. WRING
holdrr
into
vetiicsl
A-
SOLDERING
o.00s0
0.0075
bolder
wire
(2) Ends
md csble leads
0.0150
0.0237
End
(1)
0.0134
0.0265
(2)Ends
wllh
(1) Eod
urts.1 leads
Put
(Includes
part
)mtdllrg)
0.0176
(2) Ends
0.0289
Per
0.0058
tihkrt~
3,
REPLACE
)olnt
M E Nl WITH
Rcplsce
RepIsce
Replsce
scnw
screw
w8s&r
IWplsce
Rtp\ace
ad
stop nut
HARDWARE
lato t-d
through
bole
clesrsnce
0.0093
O.0023
o.Oolfl
0.0071
hole
0.0210
0.0075
Replue
6et ucrmw
Apply glyptol
Ocrew
4.
PART
Pull
5.
HANDLING
up
PRINTED
T&
0.0018
pm
and ponltion
CiRCUIT
in chsa~ia
for
0.0025
saeembly
WIRING
Replacc (Inseti)
o.oo33/End
Solder
O. 0056/End
ioterchmge
occurriog
I rrefpdarly
tuk
motlon8:
occurring
abosrd
ship rhea mt
Invoive
the repetitive
element
motions.
2-24
times
(cyclic)
factory
type of regulsrly
and
hflL-HI)BK-472
24 May 1966
-.
u
tc
w
DC-O
00
Ceoo
1(
,,
*K
KKVW
2-25
XXX
MIL-HDBK-472
24 May1966
When subassemblies,
assemblies,
or units are replaced,
sum of ~ 8 (from space 17) by the recorded
value of Rp.
Determine
the sum o; the ,\Rp values
at the bottom of the
kRp column,
listed
and record
multiply
tie
the recorded
in the sum
space
of each recorded
value of Rp m the
,,S&!p (1o). log Rp List the logarithms
log Rp column oppdte
the respective
Rp value.
Where values of Rp are less
than one, the logarithms
should be expressed
as negative numbers.
(For example,
log 0.25 = 9.3979 - 10 = -0.6021).
step (l]). ~ log Rp When indltidual
parts are replaced,
multiply each recorded
part fidlure rate ( ~ ] by the respective
value of log Rp for that part, record the
product in the ~ log Rp column opposite the appropriate
part &signation.
When subassemblies,
assemblies,
or units are replaced,
multiply
value of log Rp.
sum of ~~s (from space ]7) by the recorded
Determine
the sum of the log Rp values
at tbe bottom of the~ log Ru column.
3.1, 1.5
After
Obtaining
all worksheets
(0
(b)
Maintainability
are completed,
and record
~ARp
space
Parameter
be consolidated
to determine:
(c) ~~
3.1.1.6
Predicted
listed
the recorded
mme
of
kRp
(space
sums
18).
19).
Calculations
(a) Mean time to repair
by the calculation:
(MTTR).
The predicted
~hRp
MTTR
XA
MI L- HDBK-472
24 May 1966
MTTRG
3.2
PART B Active
{ )
,.-.
1 corrective
= ontllog
Ma]ntenmce
hllin:en~n,:e
x
[
L
(A
log
F@
v\
LA
Prediction
Procedure
Prediction
part numbers.
names
Norkshem A - Coiumn
List the circuit designaB,
Circuit
Designationtion of each part in the product except for the following categories
for
which no circuit designation
shall be Iisted:
1.
Mechanical
2.
Electrical
hardware
hardware
teT7??in21S
~e~~-ilh~l~~h
Terminai
such as:
stucis
Wiring
Cables
Wiring harnesses
Sockets for plug-in parts of subassemblies
Terminal
boiirds
Printed wiring boards (board only)
Circuit designations
of subassemblies
or assemblies
(e, g. , modular
assemblies)
that normaIly will be replaced upon failure of any part
contained therein shaf I be listed separately,
followed immediately
by
the circuit designations
of all parts included in that subassembly
or
Part listings shaIl be grouped on worksheets
according
assembly.
to product subdivisions
(e. g. , unit, equipment.
set, system,
etc).
(c) Column C, Failure Rate:
( ~):
The failure rate of each part shall
The failure rate for
be determined
in accordance
with MI L- STD-756.
each part, in part failures per 106 hours, shafl be listed in Column C
2-27
MIbMDBK-4?2
24 May 1966
opposite the respective
part fctentlftcatton.
For parts with life
characteristics
measured
in units other than time, an operational
analysis
shall be performed
and a ~ in part failures per 106 hours
The sum of all part failure rates {Z ~) on each
shall be computed.
sheet shall be determined
and recorded
in the space at the bottom of
Column C.
F, D{saseemblv:
G, Interchange:
(d) Column D, Locaiizatton:
E, Isolation:
H, Reassembly;
J, All gnment: and K, Checkout: List the estimated
average time required
to perform each of the listed corrective
maintenance
tasks corresponding
to tbe respective
part identlflcation,
This may be done
restricted
by foIlowing a sequence of steps, similar
to, but not necessarily
to the
following:
(1) Assuming that each component fails in Its most likely mode, note
the fault localization
features and determine
the necessary
steps
to localize the fault to the module or function. Estimate
the
average localization
time.
(2) Determine
the average isolation time by referring
to schematics,
noting the loation
of test points and esttrnate the observations
and steps required
to tsolate the trouble to tbe failed component,
aasembly
details
(3) Note mechanical
assembly
drawings,
as required,
(4) Observe the method
surface and perform
by referring
to mechanical
or
to estimate
disassembly
time.
to
is requtred,
note the characteristics
the time required to perform the task.
(7) Estimate
the check-out
time by reviewing
performance
procedures.
the minimum
established
2-28
.-
hfIb HDBK-472
24 May 1966
(e)
column
recorded
in Column
D, E, F, G,
The value
of each
Mc multiplied
by the respective
value
Worksheet
B: F@ure 2-4 shows a copy of Worksheet
B. This is used to
record the results of the eurnmdfon of ( ~ k)ad ( x ~Mc 1gho~ at ~tbm
of Worksheet
A. In essence then, the sum of all the en[ries Ln Column 2
of Worksheet
B, is the sum of all the falkre ratea (x ~) and represents
tie fti)ure of the equipment
or product and hence is called the Product
appearing
Similarly,
the sum of all the( z ~MC)V~UeS
Failure Rate.
at the bottom of Column 3, Worksheet
B is the total corrective
maintenance
do~mtime per 106 hours.
Hence by substituting
in Equation 2-6 the mean
man-hours
for corrective
maintenance
EC is obtatned.
This equatjon ia
repeated
for immediate
reference.
Tip
.-
3.2.2
Preventive
Maintenance
X(AM,)
Prediction
Worksheets
C and D are used to record the data required
to calculate
the mean manhours of preventive
action Hp. The forma are illustrated
by Pigure8 2-5 and 2-6.
columns
References
such as Column A, Column B, etc. , refer to the respective
of Worksheet
C. References
such as Column 1, Column 2, refer to the respective
columns of Worksheet
D.
(a) Column A, Description
of Preventive
Maintenance
Ta8k8: Determine
all
during the
of the preventive
maintmance
tasks that must be perfomned
A detafled description
of each task shall be
service
use of the product.
presented
in a separate
report.
Determine
those preventive
maintenance
tasks which can be performed
while the product is opcratlng.
Determine
the frequency
of occurrence
and man-hours
required
to perform
each task.
man-hour6
required,
and a statement
that
The frequency
of occurrence,
while the equipment
is operating
the tasks can or cannot be performed
of each task.
A brief descripshall be included in the detaihxi description
tion of each task that cannot be perfomed
while the product is operating
shall & listed in Column A.
2-29
Column B, Frequency
of(kcurrence(
f): The frequency of occurrence
of
each task, listed in Column A, shall be determined
using existing data.
The frequency
of occurrence
for each task, in events per 106 hours of
calendar
time shall be listed in Column B opposite the respective
task.
The sum of all frequencies
of occurrence
on each sheet shall be determined
and recorded
in the space at the bottom of Column B.
D, fMp:
of occurrence
multiplied
by the respective
value of man-hours
required
shall be recor&d
in
The sum of the values appearing
Column D opposite the respective
task.
in Column D shall be recorded
in the space provided at the bttom
of
Column D.
D: The Worksheet
C sheet nurntn?r shall be recorded in
(e) Worksheet
C shall be
Column 1. The sum of each Column B on each Worksheet
recorded
in Column 2 opposite the respective
Worksheet
C sheet number.
The sum of each Column D on each Worksheet
C shall be recorded tn
Worksheet
C sheet number.
The sum
Column 3 opposite the respective
in the space marked
of all values in Column 2 shall be recorded
Frequency
of preventive
matntensnce
tasks at the bottom of the last
Worksheet
D. The sum of aIl values in Column 3 shall be recorded tn
the space marked Total preventive
maintenance
time per 106 hours of
operation
at the bottom of the last Worksheet
D.
(f)
Calculating
tip:
The mean
preventive
main%nance
time,
calculated
by substituting
the data recorded
in Worksheet
(2-7) which is repeated for immediate
reference,
namely:
is
~p=
Xp in man-hours
D in equation
~ (fMp)
x,
Q)
2-30
which term
is determined
from
maintenance
ML-H DBK-472
24 May 1966
(b) Calculating
~bstituting
..
2-31
Index) shall
be calculated
by
bULEDE#K-472
24 May
1966
.
C9
i2
u
I
..-
2-32
MnrEDBK-472
24 May 1*6
Workabeet
Contractor
Contract
Date
No.
Sheet
A
II
Worksheet
A
Sheet Number
Subtotale
roduct
Total
repeir
time
failure
per
106
Column
Total
rale.
zA = Total of column
hou~e,~ ~Mc= Total of column
r-igure
2-4
Worksheet
2-33
-...
iMc
Cohnn n C Column
Workeheet
A
Sheet
Number [
Total
Total
i
of
~iMc
Column
Total
I
2 subtotals
3 subtotals
MIL-HD3K-472
24 May 1966
-.
2-34
MIL-HDBK-472
24 ?da~ 1966
Worksheel
D
Date
Contractor
Contract
No.
Sheet of
1
I Worksheet
~ Sheet Number
Total
Total
I ! Sheet N~ber
Tot~
Tot~
1
i
1
[
I
~
I
I
t
t
I
I
1
I
i
. .
I
I
Frequency
of preventive
maintenance
tasks,
f = Tot~ of col~n
Total preventive
maintenance
time per million hrs, fMp = Total
Figure
2-6
Worksheet
2-35
~ SUb-tOt~S_
of column 3
sub-totals
MIL-HDBK-472
24 May 1966
PROCEDURE
1.0
Ill
GENERAL
This maintainability
procedure
prediction
of ground electronic
principles
of random sampling.
describes
systems
a method of performing
a maintainability
and equipment by utilizing the basic
Philosophy,
Assumptions
and Summary
The underlying
philosophy of this procedure
is that system failures
are principally
due t n the malfunction
of replaceable
items and therefore,
the time cycle for the
var. -w. steps required
to replace these items is a measure of downtime which is
The duration of this downtime is assumed
a puameter
of system maintainability.
to be a function of specific design parameters
which relate to: the physical
the facilities
provided for maintenance
by the design;
configuration
of the system;
and the degree of maintenance
skills required
of personnel
charged with the repair
responsibility.
Accordingly,
it is assumed that a similar type of maintenance
activity is required
to repair a part of the same class when repair by replacement
is used because
the same steps are followed.
These include the time required
for preparation,
fault
location,
fault correction,
adjustment
and calibration
and final test.
On this basis,
it is inferred
that the analysis of the times required to perform
each listed step
comprise
maintenance
tasks which can be evaluated in terms of elapsed time.
such
The various classes
of replaceable
items are subdivided into categories
as transistors,
receiving
tubes, transmitting
tubes, resistors,
capacitors,
etc.
3-1
ML- HDBK-472
24 May 1966
The procedure
also assumes that because of a basic uniformity
of design, a mdom
selection of replaceable
items by class will provide a representative
sample of
maintenance
tasks whose time of performance
can be established
by stmuhtion
in a manner representative
of system characteristics
in actual operation.
The
method of selection of this sample is described in detafl in 6ections 2.0 and 3.0.
The ~signment
of the times of performance
for each of the steps involved
in the maintenance
cycle, commonly referred
to as maintenance
tasks, is
determined
by using three types of check lists.
These are intended to provide
a uniform method of scoring the various maintenance
tasks and are labelled
Check Lists A, B and C respectively.
Check List A is used for scoring physical
Check List B scores design dic~tes-faciltttes
and Check LJst C
deEign factors,
skiIls.
The theory 1s employed that
is used to score design dictates-maintenance
by using these check lists which include uniform scoring and scoring criteria,
variations
due tn indltidual
appraisers
are minimized
and the resulting
scores
A re~SSiOII
equation @WdOn
3-7)
can then b correlated
with actual downtime.
for this purpooe which provides a corresponding
eattmate
of downtime
is provided
when the numerical,
A, B and C scores are substituted
therein.
1.2
Applicability
This maintainability
prediction
procedure
is used to predict the mean and maximum
corrective
maintenance
downtime for ground electronic
systems
and equipment.
1.3
Point of Application
The procedure
is adaptable for perfonntng
maintainability
predictions
Desi@ and Development
Stage.
Usually prior to completion
of design,
estimate
is performed
as a first step.
This is later followed by more
estimates
as development
progresses.
1.4
Basic
The basic
Parameters
parameters
nc~ =
mean
of Measure
of measure
are:
corrective
maintenance
time
mean preventive
maintenance
time
ii@
El
= mean downtime
3-2
during the
a gross
detailed
MIL-HDBK-472
24 May 1966
max
maximum corrective
maintenance
The mathematical
in 2.6.
In this application
preventive
1.4, 1 Preventive
Maintenance
The maintenance
1. 4.2
performed
Corrective
and corrective
Information
maintenance
are shown
Maintenance
of measure
as a result
of a failure,
to restore
an item to a
Required
In order to accomplish the task predictions the evaluator must have detailed
information and have accessibility to, and be familiar with at least the following:
(a) Schematic diagrams
.-
(d) Description
(e) Maintenance
(f) A description
1.6
nf the equipment
aids to be incorporated
of the
operational
and maintenance
environment
Data Basis
The data utilized for the development of this prediction procedure were obtained
during the surveillance of three equipments of varying complexity, use, maintenance
and packaging concepts, and the nature of the circuitry.
The three equipments were:
(a) AN/ FPS2:
Imng range search radar, two channels.
Average complexity is 10,976 parts.
hlaintensnce is performed at the part level.
3-3
MIL-HDBK-472
24 May 1966
(b) ANi FST-2: T~o-chael
data processor which converts analog radar
Average
complexi~ is 114,500 pwts.
returns to digital form.
hlalntenance is performed at the moduIe level.
(c] .%h/GK4-5: Time-di\fision data link transmitting equipment.
Contains
both digital and radio frequency sections.
Its average complexig is
44, 520 parts.
The dimtal section uses boards and maintenance is
provided by modular replacement,
that is replacing the defective
boards at the board level.
The radio frequency section is comprised
of individual parts and maintenance is performed by replacing defecttve
parts.
1.7
Correlation
Between Predicted
fhe correlation between predicted and observed values can be good provided
that adequate information is available and mature experienced analysts are used.
The follo~ring is a tabulation of some typical results.
PREDICTED VERSUS ACTUAL TIMES
Equipment
AN/ FP&6
(Radar)
AN/GRT-3/GRR-7
Type of Data
67.26
Actual
94.03
51.90
Actual
63.31
Although it appears that in the case of the Radar AN/ FPS-6, the actual measure
of maintainability does not correlate with the preliminary
and final design prediction, it has been verified, that this has been due to a nonrepresentative
sample
of maintenance tasks. This was demonstrated by selecting maintenance tasks
for simulation from actual failures in the field, the analysis of which, resulted
in good correlation,
The recorded data also show a high degree of correlation
between the actual and predicted values of maintainability for the AN/GR>3/GRR - 7.
3-4
MIL-HDBK-472
24 May 1966
2.0
ANA L}TJC
FOUNDATION
Determination
of
Sample Size N
by substituting
3-5
MIPHDBK-472
24 May 1966
(3-1 )
v=
X
k=
2.2
Derivation
confidence level
population mean
the desired accuracy of the prediction
percent of the mean
given as a
(3-2)
.
where:
WY =
3-6
MIL-HDBK-472
24 May 1966
r=
standard
deviation of population
N =
sample size
should be obvious:
(3-3)
k~ = # W:
where:
confidence to be applied to the measurement
(~ = 1.645 for 95% confidence)
accuracy
*betItutlng
vaiue of
~W
(3-4)
+
[1
confidence
equality.
If
we label
cx=~
(3-5)
k~
wrttten as:
it is necessarY
N=
($),
k
[1
N=C
3-7
-.
(3-6)
Ml bHDBK-472
24 May 1966
Figure 3-1.
Comparison
3-8
-
MIL-
HDBK-472
24 May 1966
This equation has been solved for several values of k accuracy for 95% confidence.
The results are shown in Figure 3-2, Sample Size Nomography,
The use of the nomograph will be illustrated
I&le
3-1
Cx+
50
= 0,5
100
of 10%is desired.
the sample
Taak Sample
Task Sampling
in the tnterest of clarity it should be stated that Task Sampling involves the
sampltng of replaceable parts for evaluation of hypothetical part failures.
The results of the evaluation will, on the average, be repre.sentati ve of the
number of maintenance tasks which can be e xpected to occur, due to ~
faflure, under operational conditions.
Task Sampling becomes a necessi& when the complexity of. equipments now
For example, the AN/ FPS- 20 radar has
appearing in the field is considered.
over 10, 000 active electronic parts and hence evaluation of each part with respect
3-9
MIL-HDBK-472
24 May 1966
I000
600
300
Id
100
60
30
10
.I
,2
.3
.4
.5
COEFFICIENT
Figure
3-2.
.6
.7
OF VARIATION
Sample
3-10
.8
.9
Cx (&#)
Size Nomograph
LO
1.1
1.2
M!LrHDBK-472
24 May 1966
to tw contribution to maintenance time would be both impractical and unnecessary
This is particularly
true, when it !s reallzed that the physical arrangement and
function of many of these parts are similar with reswct ti a maintenance task.
selected parts, wltht.n each Taak Sample
Therefore,
the evaluation of randomly
will re~ult @ azcurate predictions of maintainability.
2,3,2
Technique
{a)
General .
COnslderatlons:
Through the conrdderation of vnriou8 f8ctors
associated with the failure and replacement of partB, the maintenance
capab!l!ty of the equtpment can be predicted.
The maintenance time
thus derived is an estimate of the aver~e time to accompl!rh a
maintenance task under actual oixwsting conditions.
(b) ..
Pmcees:
The procem of tagk ~election is illustrated by Example
3-3 which le ehown in 3.2. Reference to Table 3-1 shows that the
parts are llBted by part class, quantity and average part failure
rati per 106 hours. The expected number of failures per 106 hours
per clam is computed as the product of the quantity of parts and the
average clam fatlure rate.
The ratio of the expecbd number of
failures per 10s hours for a clam to the total of the expected number of
failureB for all claases is the percent contribution to fatlure of that
class.
The product of the percent contri-button of each ciasu and the
sample size N is the claae sub-uample,
n, which giveB the number of
replaceable items to be anaiyzed for that claB9. The last column of
Table 3-1 Bh0W8the sub-sample sizes , n, rounded off to the nearest
whole number.
2.4
Application
of Check LteJte
MIL-HDBK-472
.-
2.5
Linear Regression
the predicted
downtime
(3-7)
The coefficients of this equation were derived from 101 corrective maintenance
ta8kB appearing on appropriate check Iieti for tbe AN/ FP%20 long range
AN/ FST- 2 ho channel data processor
and AN/GKS
S data link
sealch r M&,
tranmnftting equipment.
The symbol Met means corrective maintenance ttme
represent the scores obtatned by the
and tbe letters A, B and C, respectively,
use of the respective applicable check lists.
2.6
MathematlcaJ
Relationships
Gf Maintenance
Times
for calculating
(3-8)
N
where:
&t
N=
= mean corrective
maintenance
Met = corrective
maintenance
3-12
ttme
maintenance
taaka
tasks
MIL-HDBK-472
24 May 1966
similarly:
(3- 9)
where:
~mox
~pt
MPt
= preventive
tasks
is expressed
mean preventive
maintenance
maintenance
time
as:
Mmo~ = ontilog
where:
~ct +1.645C
(3-1 o)
IOQMet
Nt
met =
~
1:1
log
Mcti
:
NC
NC-l
(3-11)
where:
Kt
= mean dowmtime
3-13
maintenance
MXLHDBK-472
24 May 1966
3.0 APPLICATION
3.1 General Approach
The maintainability prediction technique for evaluation of electronic
is accomplished in four steps as foIlows:
sample size,
equipment
(a)
Determination
of
(b)
Determination
of @k sub-samples,
(c) Prediction
(d) Calculation of maintenance
indices
3. 1.1
Determination
of
Sample Size
The sample size to be used in the prediction is dependent upon the statistical
With stated accuracy requirements
(k) and desired confidence
accuracy desired.
level,
(~) , the s~ple
size (N) which satisfies these requirements
is computed
2
as shown in Equation (3-l) which is repeated here for convenience:
w
N:
#
[1
kx
3. 1.2
Step-by-Step
Procedure
for Calculating
Sample Size
(a) Decide on the accuracy (k) or precision of prediction desired; that is,
decide how large a confidence interval is adequate for prediction
purposes and express that interval as a percent of the expected MTTR
(the latter is to be expressed
in minutes).
3-14
-
MIL-HDBK-472
24 May 1966
(W
@tide
interval
on be
~=el
determined
Of confidence
(~)
IO be -ociati
wjth the
in (a) above.
where:
u
deviation of MTTR
(3-13)
where:
c~
the coefficient
of variation
3-15
of the prediction,
given as a percent
MfbHDBK-472
24 May 1966
Example (3-2~
Assume that it is desired to state the MT TR with an accuracy of t25 percent
wjth a confidence of 95 percent.
What sample size 1s required?
Substituting
in Equation (3-13) we obtain
Figure 3-2 is a nomograph which relates the sample size to Cx for a number
of common accuracy values.
To solve umder the same assumptlorm ae in
the precedtng example find the value of Cx (= 1. 07) on the abscissa, follow
up to the curve Iabelled 25%, and across to the ordinate to read N (= 50).
3.2
Step-by-Step
Procedure
for Calculating
Task Sample
(a) Determine the failure rate contribution of each part type in the
system or equipment to the ove~ll failure rate,
That is, determine
what percent of the failurws will be attributable to tubes, to resistors,
etc.
to transistors,
to capacttore,
(b) Ustng the percentages computed shove, calculate how many of the N
items (determined from (a) a&ove) will be apportioned to tubes, to
traneistora,
to re8i8tort3, etc. , 1.e. , calculate the class sub-sample
size, n.
{c) Select randomly from each part type enough item8 to meet the requirements set up in (b).
Example
(3-3~
Assume an equipment with parts as shown !n the first column of Table 3-1 and
that the number of each type of part is aa Ehown in Column 2. Given the part
hilure
rates
as in Column
3, the numixsr
of failures
expectad
per
10* hours
-n Column
4.
Expressing
the
3-16
MIL HDBK-472
24 May 1966
>
5
w
N
t-
3-17
MIL-HDBK-472
24 h!ay 1966
-
evaluator,
ew.
hlaintcnance
Analysis:
Pr]or to task scoring,
it is necessar~ that,
for each task a maintenance
analysis he performed.
This analysis
entails a step-by-step
accounting of the rnrmt likely chagnostic
procedure.
Beginning uit.h the symptoms of malfunction,
each step
requi red in locating the defective part is recorded.
Complementary
to each step, notations regarding
access problems,
test equipment
requirements,
and related information
which is important to determining
the task scores,
are made.
Figure 3-3 the Maintenance
Analysis
3- -! tk
ccmth-luatim Shcat, illustrate
a f017J-t&t uSd
for~t &-Id T~~uiE
The form is divided inti ti$o columns,
The left
for this analysis.
column !abelled Maintenance
Steps, is used to record each test
or step that a technician
should make,
Scoring comments associated
with each step are entered in the column on the right.
Completion of
the maintenance
analysis provides n firm basis for the scoring.
The
full scope of a m,alntmance
situation is realized
through this process.
S~stem hlaintenance
Di~ram:
This analysis may be facilitated
by the
preparation
of a System hl~nt.enance
Diagram,
which would clearly
detail the system functional block diagram,
with main signal paths,
major test points and other diagnostic
aids shown.
An illustration
of such a diagram is sho~vn in Figure 3-3, Maintenance
Dia.gran~
AN/ GRT-3.
This diagram assists
in the determination
of the
malfunction
symptoms
and in selecting
steps to isolate the malfunction
to a functional area.
It is necessary
to have a schematic
diagram for
each block to troubleshoot
within a block and to determine
the effect
nf an assumed failure on the output (s) of other blocks.
The illustrated
diagram is representative
of the minimum requirements
for such a
diagram and ma! be expanded to [ arying dcg-rees depending cm the
complexi~
of the equipment and the information
available.
MI L- HDBK-4?2
24 May 1966
.-
.
Task No.
Equip.
Primary
Date
D}
Assy
function
failed
unit,ipart
Mode of failure
Ma.lfurfc:[on symptoms
Nlaintenance
!iIai ntenance
Steps
Scoring
Checklist
Predicted
.4nalvsis
Comments
Scores
downtime
Figure
3-3
Maintainability
:{- ~9
Prediction
Form
Min.
M!L- KDBK-472
24 May 1966
\I.QSTEN.iYCE
Equip.
ASALYSIS
COXTISLAnOS
Par!
\lalntcrlancc
SHEET
Task So.
Steps
scoring
Comments
Analysis
:]-2(,
Continuation
Sheet
MX~KDBK-472
g
~g____________________
1~
+
F-----
24 May 1$66
I
I
--
I
L----
TW
g 1>G------w>--
~
-v
SE
-------=--
vi
J---m
J-;
I@
+----
J-
-----------
l--------
---
@@e@
____
I
I
I
I
I
I
1
1
3-21
II
I
I
MIb HDBK-472
24 May 1966
ExarnPle
(3-4)
illustrates
the procedure
Evaluation
of
Resistor
The resistor
failure was assumed to have nccurred
by opening.
Following
maintenance
analysis
wm made, drawing
a step-by-step
&his assumption,
from both general maintenance
experience
and techntca.1 data.
This procedure
is illustra&d
in Table 3-2 Maintenance
Analysis, R- 7801 /FPS-20.
Here
the demiled steps necessary to isolate the defective resistor are listed.
For each step, uomnients rcgdrding availability of test indicators, need for
test equipment, access problems, and related inform ation needed to effectively
Figure 3-6, AN/FPS-20, Transmitting System,
score this task were listed.
illustrates functionally the major circuits associated with the transmitter
section.
Within the diagram each step has been numerically
identified.
It will be noted
that the troubleshooting
pa~h chosen is one of several possible routes.
The
of the check and the e-e with
route estahl i shed IV3S based on the importance
witlch ;t crmid be taken:
For example,
in Step 2, the trigger input which is
vital for the proper operation of three portions of the transmitter,
was tested.
The third step was selected because of the ease with which the information
could be secured
(built-in metering) . Such choices are generally consistent
with procedures
employed by electronic
technicians.
Fig_urc :-7 Arnplificr
hlixcr AM-1347, FPS-20, illustrates
the mechanical
layout of tic section of the equipment wlthln uhich the defective part was
located,
,Is indicated in the maintainability
analysis,
the plate power supply
sub-chassis
had to be removed for further testing.
Figure 3-8, Expanded
liclv, Plate Pn\vcr Supply, shows the underside
of the power supply.
Here,
thr terminal
board cm uh]ch resistor
R-7801 is located has been identified.
Functionally.
the use of reslstnr
R-7801 is illustrated
in Figure 3-9, Plate
.Sllpply Block I)lagrm
The rcsi stor provides continuity between the rectifier
and series dropping electron tubes.
Its opening caused loss of plate supply
volta~e 10 the buffer arnplifler,
thus preventing
operation
of the transmitter.
These illustrations,
in assoclaticm
with other technical
material
secured
applicable
technical orclcr, provided the basis for the scoring comments
Nla[nlenancc
.4nalvsis Cont]nuation Sheet (Table 3-2).
from the
in the
T.4SKPRED1~TIOY,
31~T~m]LITY
Equip.
.4,,, ~FPS- 20
Primary
function
output network.
Llm$? of failure
Series
failed unit/part
Resistor
symptoms
Maintenance
\o.
Date
remstor
in 620 V. D. C. power
opened
l~o return
on radar
indicator
.4nalys1s
Scoring
Sleps
Checklist
Predicted
Task
R-7901
Maintenance
I
FC)R?,l
Ilv
A9sy.
MaJfunct\on
R-7901
PREDICTION
lnlt ~Pari
MIL-HDBK-472
24 May 1966
downtime . . . . . .
Scores
. . . . . . . . . ...1....
TABLE
Comments
3-2
. . LO \!in.
supply
h!IL- HDBK-47Q
24 May 1966
TABLE
M.41STEN.4!!CE
ANALYSIS.
>faintpnanc~~ An~ysis
Equipment
J>art R-7801j
AX, FPS-20
Slai nlenmce
3-3
Steps
R-7901, JFPS-20
Continuatlon
Sheet
FPS-20
Task
Scoring
No.
Comments
1.
Equipment malfunction
is initially
indicated t.)}no target returns on
Maintenance
action iJroIndicator.
ceeds to isolate trouble to major
equipment functions.
Pover output check at bidirectional
coupler
(CU-516) isolates trouble to transmittcr function.
2.
oscilloscope
is used to check trigger
pulse at J-1405 on front panel of
modulalor
MD-276 (I PA modulator).
Oscilloscope
set-up and adjustments
Proper reading listed in
requi red,
T C).
3.
3-24
JLI]L- HD13K-472
24 hlay 1966
TABLE
\Iaintenance
Equl]wnent
llai ntenance
I 4.
~
5.
Analysis
Part
AN /FPS-20
3-3 (Continued)
Steps
Continuation
Shce~
R-7901 /FPS-2O
Scor~ng Comments
-.
f-
1
I
I
I
I
II
If
I
I
I
1
1:
1;
1
I
I
I
I
I
I
I
I
I
I
I
I
I
!
I
1
I
I
I
I
!
I
I
!
I
I
I
la
Im
I
I
I
le
0
i
II-
!L
3-26
MI~ HDBK-472
24 May 1966
u%
M7T03
M7702
I IST BuFFER
FILAMENT
Bzmimlm
q
J772
1!
F7701 F7702
F7703
m
J7724
FRON7 VIEW
Figure
3-7.
~plifier-lvlixer,
3-27
MI~HDBK-472
24 May 1966
R-7801
BOARD
IIS
./
Figure
:1-x.
Ewanderi
lIeN,
3-28
Plate
Power
Supply
MIL-HD33K-472
24 May 1966
R-?ou
~
120 VACJPLATE
VOLTA6E
TRAN9F
m-
Ac- 7?001
t
v78a ~
RECTIFIER
TUBE
OROPPWG
TUBE
TUBE
%L\
V7804
VOLTAOE
~ ~
V7805
VOLTAOf
REFERENCE
TUBE
5
1
%
V7@06
vOLTAGE
REFERENCE
TUOE
{
t
e
PLATE
vOLTAGE
TRANSF
T7803
RECTIFIER
TUBE
b
?+
DROPPINfl
+360
7{
780
UIFF
AMP
nl[
Wel I
Figure
3-9.
VOLTA@E
REFERENCE
TUBE
Plate
Supply,
3-29
.
w
vOLTAOI
REFERENC
~
JOE
-#
Block Diagram
AA
Voc
M13.rHD13K-472
24 May 1966
3. -! Calculation
of hlalntenance
lndlces
to maintenance.
To illustrate
further hour check list scores are obtained.
some of the
in Check List A, question
specific scores in Table 3-3 tii!l be examined.
receiied a score of 2 (external latches and fasteners
meet two of the
criteria
that the~ are capti!e. need no special tools, and require only
In Check L!st B question 1 received
a fraction of turn for r~!ease)
are needed).
Examination
a score OJ 1 (: or :; p~eces of test equipment
,;- :](J
MIL-HDBK-472
24 May 1966
of Table 3-2 indicates
that an oscilloscope,
multimeter,
and tube checker
For Check List C, quetiion 5
were used to accompl~sh this task.
received a score of 1. This score was assigned
because the initial
symptoms
gave very Iittle indication as to the cause of malfunction
@) ~wntime
UC,
antilog
(3. 54651
0. 02512A
0.03055B
O. O1O93C)
To facilitate this calculation a nomograph was developed for the prediction equation and is shown in Figure 3-10, Nomograph - Downtime.
The use of this nomograph permits the determination
of downtime
directiy in real time (instead of log values).
All instructions for use of
to calculate
maintenance
tndices are contained in
the nomo~aph
Figure 3-10; and are the mean active corrective maintenance downtime fie t,
and the maximum active corrective maintenance downtime Mmox . The
meti corrective maintenance tire+, wet, is Ubt&iirIadby substituting Lxi
Equation (3-8). The ma.xtmurn corrective maintenance time, MmoX,
la found by substituting in Equation (3-10).
3-31
d\
\
\
\l
\!
3-32
hlIb HD131i-472
24 hlay 1966
DESIGN CHECK
The three
scoring crlterla
follow:
CHECK LIST A - SCORING PHYSICAL DESIGN FACTORS
Detemnines if the exiernal access is adequate for
visua.! :rtspecbon and manipulative
actions.
Scoring will apply to external
packaging as related to maintainability
design concepts for ease of
maintenance.
This item is concerned
with the design for external
\isua.1 and man]puiat]ve
actions which would precede
interns! maintenance
actions,
The following scores and scoring criteria
will apply:
Scores
(a) Access adequate
(c) Access
tasks,
(electrical
Scoring Criteria
An explanation of the factors pertaining to the above scores is consecutively
shown . This procedure is followed throughout for other scores and scoring
criteria.
(a) To be scored
(b) To be scored
{or
visual
inspection.
h~lt nnt
A:)- 3:{
access
for
is adequate
n3n~p~ll~ti~:e
of
(no delay)
actions.
MI~HDBK-472
24 May 1966
External
covers,
panels, etc. , can be located visually:
external packaging or obstructions
hinders mani-
screws,
however,
pulative actions (removal,
tightening,
replacement,
etc. ).
(d) To be scored
@) latches
and Fasteners (External): Determines if tie ecrews. C1lPS,
latches, or fasteners outside the assembly require special tools,
or if significant time was consumed in the removal of such items.
Scoring will relate external equipment packaging and hardware to
Time consumed with preliminary
maintainability
design concepts.
external disassembly will be proportional to the type of hardware
and tools needed to release them and wiH be evaluated accordingly.
Scores
Crlterta
fastmers
(1)
screws,
are:
Captive
special
tools
A3-34
latches,
and
MIL- HDBK-472
24 hfay 1966
f:\I
Releasing
Can be
released
a DztSI fastener
with a fraction
}vhich requires
a 90
of a turn
de~ee
(c) To he scored
meet
when external
screws,
latches,
only one or none of the three conditions
and fasleners
stated in (a) above.
(lnterna~:
Determines
if the internal screws.
clips, fasteners
or latches within the unit require
special tools,
time was consumed
in the removal of such items.
or if significant
Scoring will relate internal equipment hardware
to maintainabilit-v
ciesign concepts.
The types of latches and fasteners in the equ ipmcnt
and standardization
of these throughout the equipment shall tend to
affect the task by reducin~ or increasing
required
time to remove and
replace them.
Consider internal
latches and fasteners
to be !vithin
the inkrior
of the asscmbl~.
Scores
(a) Internal latches and.for fasteners
tools, and require rmly a fraction
Criteria
when internal
screws,
(1) CYtptive
(1) M not require
special
tools
A3-35
latches
and fasteners
are:
MIL-HDBK-472
24 May 1966
(3) Can be released
with a fraction
of a turn
Releasing
a D2US fastener
which requires
a 9(1-degree turn using a standard
screw clrlver u)ould be an example of all three conditions.
when mtcrna] screus,
latches, and fasteners
meet NO of the three conditions stated in (a) above. A screw
which is captive can be removed witi a standard or Phi Iiips
(b) To be scored
screw
driver,
but requires
se~eral
full turns
for release.
as meeting
require-
(4) Access
visual
Scores
(a)
or
visual
actions
access.
while performin~
in a sub-as sembl~ or unit,
because of thr internal construc-
construction
(testing,
(c)
hut not
for
manipulative
removal,
is adequate
access
manipulative
actions
etc. ).
inspection.
during
the maintenance
action.
Determines
the access (within the subassembly) to components
or parts requiring mechanical disassembly.
This question concerns
itself with the internal packaging of parts relative to the maintenance
on module
action. Current design concepts have been concentrated
.. 11..
L
14 .
..A.1
:
even Lhese va4j hi k mwh.ritc~llj
,Ielu
~hlvuL&b,
type packaging; however,
while others are plug-in type only. This item deals with the mechanical
problems involved in gaining access to failed components or parts.
(5) Packatin&
Scores
and parts can be
(a) Internal access to components
made with no mechanical
disassembly.
. . . . . . 4
(b) Little
disassembly
required
A3-37
MIGfiDBK-472
24 May 1966
(b) To be scored when less than three minutes is expended
gaining access to the failed component or part.
in
Determines
the manner m which umts
Units - Parts (Faiied):
or parts are removed or replaced
during the maintenance action.
secured
such items
Mechanically
held items include tubes prntected
varies considerably,
from vibrations
b~ speclai shields or clamps,
printed boards clipped
into their sockets,
and parts and components held by brackets.
soldered
items include resistors,
capacitors,
etc.
(a) Units or parts
of plug-in
nature
. . . . . . . . . . . . . . .
of plug-in
nature
and mechanically
held.
. .
. . . . . . . . . . . . . . . . . . 2
Scoring Criteria
(a) To be scored when units or parts are plug-in types requiring
Plug-in t+ype parts such as tubes. some
only to be pulled out.
rela~s, crystals,
etc. , would be included in this category.
(b) To be scored when units or parts are plug-in types, but are
mechanically
held by clips, shields,
clamps,
etc. Also applies
to maintenance
requiring
the removal of a tube having external
anti-vibration
shields,
etc.
grid or plate connections,
(c)
such as
requires
MIL-
HDBK-472
24 May 1966
(7)
Visual Displays:
Determines
if sufficient vlsua! information
pertaining
is displayed within one area or unit.
to the equipment malfunction
Cirruit indicators
and meters provide, to some extent, symptom
it is important
that these indications
be disp)ayed
analysis.
Therefore,
within one area to ensure rapid analysis and action.
If several areas
before
a qualifled estimation of the difficulty can
must be consulted
be made, much time is required
Scores
Sufficient visual information
on the equipment ik given within
one display area . . . . . . . . . . . . . . . . . . . . . . . .
be wnsulted
to obtain sufficient
. . . . . . . . . . . . . . . . . . . . . .
Two
dIsp)3y
visua]
More
VI Su:l
areas
must
information
than two areas
in formation
Scoring
-.
Criteria
(8)
Fault
and
Operation
Indicators
(Built-b
Test
Equipment~:
Determines
if an equipment malfunction
or fault is clearly discemtble
via audible
alarms,
indicators,
etc. , and that such information
is clear!y presented
for rapid maintenance
action,
The use of indicators is increased as
A3-39
M?LHDBK-472
24 hflly 1966
complexity
Score6
(a) Fault or malfunction
clearly
and forrapld
@) Fauit or malfunction
but requires
operator
interpretation,
and requires
information
is provided
action . . . . . . . . . . . . . .
information
clearly presented,
interpretation
. . . . . . . . .
information
requires
but is not clearIy
information
operator
not clearly
Intevretation
no
presented.
pre.
. . .
Scoring Criteria
(a) To be scored when an equipment fault or malfunction
occurs and is evidenced by alarms, indicators, etc. ,
which provide for rapid diagnosis and maintenance action.
An example of this would be when a power supply failure
occurs because of an open fuse which is pointed out by an
indicator
b)
or alarm.
hfIL-KDBX-472
24 May 1966
(d) To be scored when an equipment fault or malfunction occurs and
is not clearly discernible,
and which requires s~ptorn
interpreTesting
is
also
necessary
to
determine
the
equipment
tati on.
status and cause of failures.
(9) Test Pointi (Availabili~~:
Determine
needed tests pertaining
to maintenance
if test points
action.
for
are available
A test
point
shall
be
as aoy test
scores
(a)
. . . , . . . . , . .
. . , . . . . . . .
. . . . . , . .
Scoring Criteria
(a) To be scored when the maintmance action did net require the use
can be diagnosed
of test points, but when, instead, the malfunction
and repaired via built-in test equipment.
(b) To be scored when all needed tests were accomplished at test points.
Sufficient information to diagnose and repair the trouble was available
at test points.
(c) To be scored when at least 51% of the required tests were accomplished
at test potnta. Troubleshooting required that several separate tests,
most of which made use of test points, had to be made.
when the majority of needed tests were not accomplished
at test points.
Malfunction diagnosis and repair requtred the making
of tests for which few or no test points were available.
(d) To be scored
A3-41
~HDBK-472
2$ May 1966
8uftabIy marked
. . . . . . . . . . . . .
(c) @tits are not marbd aud test ch.ta are not
given . . . . . . . . . . . . . . . . . . . . . . ...0
Scor @z Criteria
(a) To be aoorgd ~
all test points needed for task &nnpletion
are kkkified
(ctiouit symbol), wtth required readings gi%n
(+8VDC. -18VDC, 115VAC, etc.).
This is indtcatlve of a best
maintainable condition.
(m To be scored when the majortty of test points required
completion are sdltably Identffkl.
for task
oompiettom
(c) To be scored when last points required for ~
are not suitably identified. -Troubleshooting
at test points is a
cause for delay because required voltage read&gs, signal
characteristics,
etc. , are not specified.
This would indibate
that a least
(11)
maintainable
condition
exfsts.
Labelltn&
Determines if pruds associated with the maintenance actions
mw identified with respect to N rctiit symbol and part identification.
Proper iden~fication
of parts can be an important asset to the maintenance
if part circuit number is emitted from the equipment,
the circuit
to identify
it.
time could be wasted tracing
if information is hidden, requiring rem.ovtd of other parts
it. much time u_il\ be consumed.
task in that.
considerable
~imilarly,
to read
fiIIL. HDBK-472
24 May 1966
lniorma~ion.
but
(b) Aii parts iaheiled with full identifying
some inform ation hidden..
. . . . . . . . . . . . . . . .
. . . . . . . . .
Scoring
. . . . . . . . . . . . . . . .()
Criteria
identified,
Applies
uhich
Information
is hidden
b} {~i]structlons.
To he
task
scored
contain
(12) AcJustmcnts:
are
required,
operate
hidden
Determines
after
according
if adjustments
a maintenance
to specifications.
artion,
to make
and ;lllumment
the equipment
An adjustment
will be any action
cmmprmcmts such :1s polcntinmtitcrs,
A?-l?
such as tuning
etc. ,
whcreb~
the nperatlon
of the
MlbfiDBK-472
24 hlay 1966
_,
systcn~
assemb
upon their
I!
or
suh:issembiy
critica!in
15 affected,
uill
an{! frequency,
affect
time.
Scores
(a) 1% ad.iustments
or reiilignment
are necessary
to place
equipment back in operation
. . . . . . .
. . . . . . .
h}
A few adjustments.
required
. . ...<...
or major realignments
must be
(c!) Man! adjustments
made. . . . . . . . . . .
. . . . . . . . . . . . . . . .~
.Scoring Criteria
(a J To be scored when no adjustments
are required to bring the
Appl ics
equipment back to normal operating
specifications.
of the malfunction,
if the equipment
need only be
to repair
turned on.
(l]) To be scored when a few adjustments of a minor nature are
required
to place equipment back in operation
according to
specifications.
(c) To be scored
tuning
or
operating
(13)
TestinE
alignment
specific
(In Circuit):
can be tested
on the nature
is required
equipment
or a major
back
to normal
attmm.
Determines
removal
of the equipment
without
(time-consuming)
to place
design.
Scores
removal
(b) Testing
requires
A3-44
. . . . . . . . . .
. . . . . . . . . . . . .
4
@
h!l L- HDBK-472
24 May 1966
Scoring
Criteria
b)
(14)
the circuit,
Protective
Devices:
Encompasses
equipment design provisions
for self-protect
ion against damage to components
or parts after a
malfunction
has occurred.
If a system has protection
devices such
as fuses, circuit breakers,
etc. , then the equipment can be protected
(c) Noprovisions
have becnrnade
has occurred
. . . . . . .
. . . . . . . . . . . . . . . . .
Scoring Criteria
(a) To be scored when automatic shut-off devices protected parts
or components from further damage after a malfunction
occurred
in a critical area. A typical example of such a malfunction would
be if the Bias supply fails and B+ voltage is automatically cut off
by circuit breakers,
fuses, or relay action.
A3-45
MIL24
HDBK-472
May 1966
when automatic shut-off-devices
do not protect
parts or components from further damage, but when visual
indicators or audible alarms warn personnel of the situation.
(b) To be scored
(c) To be scored
arid parts or
components are not protected by automatic shut-off devices,
indicators, or alarms.
Involves malfunction which damages
parts or components because automatic shut-off devices or
alarms were not provided.
when a critical
malfunction
occurs
. . . . . . . . . . . . . . . . . . . . . . ...4
Criteria
A3-46
-
MIL-HDE3K-472
24 May 1966
{c) Tobe scored when precautions
taken because of hazardous
caused
a considerable
delay
to the maintenance
action.
conditions
Maintenance
required
that testing be done in close proximit}
to high voltage where extreme caution was necessary,
or the
motors.
assistance
from
other
maintenance
Determines
personnel,
if external
super
~isor,
test equipment 1s
Scores
(a)
Task
accomplishment
external
b)
needed.
does
equipment..
not
require
the use
of
. . . . . . . . . . . . .
is needed
of test equipment
One piece
(c) Several
(d)
test
. . . . . . .
pieces
Four or more
. . . . . . . . . 0
ScorinK Criteria
(a)
A3-47
-.
etc.
MIb~BK-472
24 May 1966
b) To be scored
equipment
was required
Sufficient information
was avaiiable through the use of one piece of external test
equipment for adequate repair of the malfunction.
to complete
(c) To be scored
test equipment
action. Th]s type
of areas
with different
testing
in a
test equipments.
if supplementary
Determines
condition exists.
test
equipment
requires
special
tools,
or adapters
to adequate] y perform
special fittings,
tests on the electronic system or sub-system.
During troubleshooting
of electronic
systems,
the minimum need for test equipment adapters
or connectors
indicates
that a better maintainable
condition exists.
Scores
.
(a) Connectors
fittings,
to test
or
m) Connectors
tools,
(c)
fittings,
Connectors
tools,
fittings,
equipment
adapters
to test
or
to test
require
no special
tools,
. . . . . . . . . . . . . . . ...4
equipment
adapters
equipment
and adapters
require
(less
some
than
special
. . . . .
require
two)
special
Scoring Criteria
(a) To be scored when special fittings
tools are not required
for testing.
or adapters
and special
MXL-HDBK-472
24 May 1966
(b) Applies when one special fitting, adapter or tool is required for
testing.
An example would be if testing had to be accomplished
using
a 10 &
attenuator
pad in series
set.
Scores
(a) No supplementary
materials are needed to perform
tat3k . . . . . . . . . . . . . . . . . . . . . . . ...4
(b) No more than one piece of supplementary
needed
to perform
task....
matertal
is
. . . . . . . . . ...2
hfxb HDBK-472
24 May 1966
The activities
of each member are always visible
tothe other member . . . . . . . . . . . . . ...4
one member
can see
. . .
,Scoring Criteria
(a) Applies when the team
the entire maintenance
members
action.
are visible
to each other
during
ontactw asrequired
(c) Considerable
personnel
. .
. . . . . . . . . . . . . . . ...2
coordination
required
. . . . . . . . . . . . . .
Scoring Criteria
(a) To be scored when the maintenance action does not require the
assistance of operations personnel.
This would apply if physical
or verbal aid to the technical personnel was not required.
(Jam
than one minute. )
A3-50
MIL- HDBK-472
24 hfay 1966
(b) To be scored when the maintenance action requires a small
amount of assistance from operations personnel.
(One to
five minutes.
(c) To be scored
(G) Assistance
personnel
(Technical Personnel\:
Determine
tke number of technical
required to complete the maintenance
action, not including
adrninistmtive
or operations
type personnel.
Scores
.
{a) Task required
(b) Two technicians
(c) Overtwower
were required
euse
d...
for completion
. . . .
. . . . . . . . . . - . .
. . . . . . . . . . . . ...0
Scoring Criteria
(a) To he scored When oniv one technician vv&srequired
the maintenance action.
(b) To be scored when two technicians
m aintenanee action.
(c)
were required
To be scored
tc complete
to complete the
were required
to
supervisor
A3-51
MIL-HDBK-472
24 May 1966
fb)Some
(c)
belp needed
Considerable
. . . . . . .
assistance
needed
. . . . . . ...2
. . .
. . . . .
Scoring Criteria
(a) To be scored when no supervisor
are consulted
during
or contractor
the maintenance
action.
personnel
or contractor
action.
CHECK LIST C - SCORINC DESIGN DICTATES-MAINTENANCE
This check list evaluates
and attitude characteristics,
SKTLLS
the personnel
requirements
relating to physical,
as imposed by the maintenance
task.
mental,
Evaiuniion procedure for ibis check iist can best beexpiaineci by way of severai
Consider first question which deals with arm, leg and back st.rcngth.
exampies.
Should a particular task require the removal of an equipment drawer weighing
100 pounds, this would im~se a severe requirement on this characteristic.
Hence, in this case the question would be given a low score (Oto 1). Assume
another task which, due to small size and dclicatc construction,
required
extremely
careful handling.
Here question 1 would be given a high score (4),
but the question dealing witi eye-hand coordination and dexterity would be given
a low score.
Other questions in the check Ii st relate to various persomel
characteristics
important to maintenance task accomplishment.
in completing
the check list, the task requirements
for each of these characteristics
should be
viewed with respect to average technician capabilities,
Scores
Score
1.
Arm.
kg,
.>
-,
Endurance
and Ener~
A3-52
MIL-IIDBK-472
24 May 1966
score
3.
Eye-Htmd Coordination,
and Neatness
4.
Visual
5,
Logical
6.
Memory
7.
Playfulness
8.
Alertness,
9.
Concentration,
10.
Initiative
hlanual
Dexterity,
Acuity
Analysis
- Things
and Ideas
and Resourcefulness
Cautiousness,
and Accuracy
Persistence
and Patience
and Incisiveness
Scoring
Quantitative
evaluation
the following manner:
of these
items
Criteria
range
from
4.
The maintenance
the technician.
action
requires
a minimum
3.
The maintenance
action
part of the technician.
requires
a below average
2.
The maintenance
the technician.
action
requires
an average
1.
The maintenance
action
part of the technician.
requires
an above average
effort
O.
The maintenance
of the technician.
requires
a maximum
on the part
action
These criteria
will be used in scoring
mental,
and motor requirements.
the following
A3- 53
effort
effort
effort
specific
in
on the part of
effort
on the
on the part of
divisions
on the
of physical,
MIL-HDBK-472
24 May 1966
(1)
(2)
(3)
Eye-hand coordination,
manual dexterity,
and neatness:
Determines
the degree of eye-hand coordination
required
to complete the maintenance
action.
Refers to any act involving the use of the eves while manipulating
the hands to accomplish
the same action.
This type of action would be
applicable
mostly in testing and measuring
actlv~ties: however,
it is not
inconceivable
that this item would also be appI]caMe m other areas of
Scoring shaJl be proportional
to the degree or
the maintenance
action.
the intensity of the requirements
of the task.
Determines
the degree of manual dexterity
required to complete the
maintenmce
action.
When the skillful use of the hands is required to
accomplish
the task, appropriate
degrees of necessity
shall be established.
Those type actions invoiving manual dexterity
would more naturally
apply
to the repair,
assembly,
or disassembly
of equipments
rather than the
troubleshooting
processes.
Also determines
the defrree of neatness
required
by the maintenance
action.
Applies specifically
to the requirement
of the actual repair where tidiness
Since
is of prime importance
to accomplish
the task adequately.
equipment is designed and constructed
in accordance
with quality control
specifications,
it is importint
to consider
the care which has to be
exercised
during a particular
repair.
Determines
the degree of visual acui~ required to complete
When the maintenance
action is such that the
task.
of the technician
is required
to accomplish
the task,
A3-54
MIb
KDBK-472
24 May ]966
----
a degree of requirement
shall be established.
Such actions shaIl
include the need for accurate
and prec!se visual activity in finding
indications
of trouble,
fault~ components,
or the visual sensitivity
sometimes
necessaw
in reading certain oscilloscope
presentations.
(5) bgical
Analysis:
Determines
the degree of Iogica! analysis required
to comple~
the maintenance
action.
Refers to the need for involved
logical analysis or for extensive mental reasoning
to determine
the
If
the
problem
is
such
that
it
requires
origin of the fault or malfunction.
orientation
on the logical signal sequence,
then this shall also be
considered
as part of this question.
(6) Memory:
Also determines
the degree to which the maintenance
action requires
a previous knowledge of the equipment.
Refers to the degree that
the task requires
recall of concepts or principles
of operation,
function
and operation of circuits
and parts,
or electronic
theory and maintenance
proWlllres.
v)
Planfuiness
and resourcefulness:
Determines
the degree of plsmning
required to complete the maintenance
action successfully.
Refers to
the extent to which the task requires
a planned and methodical
approach
to assure rapid diagnosis
and repair of the equipment fault or malfunction.
Also determines
(8) Alertness,
cautiousness,
and
promptness
in comprehending
all events or factors affecting
the exercise
of forethou~ht.
srI
accuracv:
Alertness is a readiness
or
and a keen awareness
and knowledge of
the maintenance
action.
Cautiousness
is
that risks mav be avoided or minimized
A3-55
MI LHDBK-472
24 May 1966
during the maintenance
action,
(A purveyance
of all possible consequence
before making a decision. ) Accuracy
is attained by the exercise
of care
by showing close attention to the details of the maintenance
taak and
The design requirements
for these
cautioutmesa
in avoiding errors.
characteristics
are to be assessed.
(9) Concentration,
p ersiatence,
and patience;
Concentration
is the clone
task and
mental application
or exclusive
attention to tbe maintenance
tbe direct focusing of tbe mind upon one thing to the exclusion of
everything
else,
Persistence
refers
to maintenance
taeke with the
implication
of being able to carry performance
to a successful
conclusion.
Patience
is tbe quiet persever~ce,
calmness
in working, and being
undisturbed
by obetacles,
delays,
or failures which might occur during
the maintenance
task.
(10) Initiative
and incisiveness:
Idtiattve
is the ener~
or aptitude displayed
in the initiation of action and the ability or power to introduce a new
measure or course of action.
Incisiveness
is the keenness of mind wd
acuteness
of understanding
the task at hand,
A3- 56
?QL-HDBK-472
24 May 1966
PROCEDURE
1.0
?\
GENERAL
This procedure
is based on the use of historical
experience,
subjective
evaluation,
expert judgement
and selective
measurement
for predicting
the downtime of a
system/ equipment.
The procedure
uses existing data to the extent available.
It provides an orderiy process by which the prediction
can be made aod integrates
Task times to perform variou8
preventive
and corrective
maintenance.
maintenance
actions are estimated
and then combined to predict overalI syetem/
equipment
maintainability.
1.1
Philosophy,
Assumptions
and Summary
This procedure
recognizes that throughout a mission,
various operational
functions and that the maintenance
specific operational
function which is in process.
To
meaning of an operational
function we wrill discuss the
airliner
as an example.
a system/equipment
performs
time depends upon the
clearly understand
tie
mission of a commercial
Mlb HDBK-472
24 May 1966
1.2
Applicability
1.3
techniques,
this procedure
is
Point of Application
Basic
Parameters
of Measurement
The intrinsic
maintainability
of the syst.emiequipment
asmrnption
of optimum utilization
of specified support
The intrinsic
matntatnabtltty
1s given by the following
Mean Corrective
Downtime
- MCDT
Mean Preventive
Downtime
- MPDT
Total
1. S Information
is predicted
under the
equipment and personnel.
parameters.
Mean Downtime
- TMD1
Required
System
Block Diagram
b.
Functional
Flow Diagrams
c.
Shbsystem
Block Diagrams
d.
Subsystem
Flow Diagrams
e.
4-2
-
hl[L- HDBK-472
2!4 May 1966
1.6
f.
~.
hltiintenance
h.
lda~ntainabilitj
i.
Operational
j.
A detailed
k.
Imcatiori
1.
Environmental
Rates
Concept
Goals
Resources
definition
(facilities,
personnel,
supp{>rt equinment,
etc. )
is being performed
Constraints
Data Basis
This procedure
utilizes expert judgment and existing data snurres on maintenance
The applicability
task time, but the procedure
does not rely solely on existing data.
of the data is decided by the anslyst and is supplemented
h~ his expert judgment
in estimating
maintenance
task time uhcn such information
is not available.
1.7
Correlation
Between
Predicted
and Ohsrrved
Vtilues
Procedures
similar to this one have been applied selectively
.md verified to a
This accurac~ is npplicahle
only ~vhere specific
high degree of accuracy.
maintenance
actions and equipment end Items arc being anal}zed,
and is
dependent upon the qualifications
of the persrmnel
performing
the evaluation.
2.0
ANALYTIC
2.1
Gencraf
FOUNDATION
The analytical
foundation of the task analysis procedure
integrates
the development
of task performance
time for preventive
and corrective
maintenance
actions.
A maintenance
action is defined as the exclusive
maintenance
task ~vhich occurs
at a specific location and within a specific set of conceptional
and physical
constraints.
This maintenance
action permits
the logical development
of elapsed
times,
subsystem
equipment levels.
The mean corrective
maintenance
time for the system equipment wil] vaq for
each individual scheduled preventive
maintenance
action appllcahle
to a specific
4-3
MIL- HDBK-472
24 May 1966
The task zma.lysis procedure
permits
the evaluation of
operational
function.
these times from the end item, up to the system level.
The products of the
procedure
are:
a.
b.
c.
The distribution
of corrective
maintenance
times for detectable
malfunctioning
end Items for each preventive
maintenance
action
of an operational
function.
d.
e.
T%e distribution
and subsystems.
f.
The preventive
downtime
Z?ec!ficti caleni?ai time.
g.
h.
of corrective
action,
malfunctioning
end items detected
action of an operational
function.
maintenance
task times
assuming
during
and subsystems
for a
time.
These maintenance
downtimes
relate only to the inherent maintainability
of the
equipment,
since administrative
and other delays are not normaHy definable
The estimated
elapsed time required to
during tie deai~ of the equipment.
perform
matnt.enance
on a system will vary as a function of tbe conceptional
and physical conetralnts
within which the esttrnation
was made.
These constraints
consist of the availability
of physical resources
(i. e. , personneA, spares and
con sum akdes, support equipment,
and facilities)
and applicable
maintenance
testing
concept,
level
of
repair,
m~sion deand operational
concepts
(i. e. ,
scriptions,
etc, ). The applicabili@
of specific constraints
must be documented
if a given time estimate
is to be meaningful.
Only single elapsed ttmes are
estimated
for each maintenance
task.
This number should approximate
the mean
time required
to perform
the task under actual conditions.
The correlation
4-4
MIbHDBK-472
24 hhy 1966
Wwen
estimated
and verified task times described
tn paragraph
the use of single value~ for the purposes
of this procedure.
1.7 juattfiem
A series of mission/maintemmce
profiles will be established
based on the
system operational
requirements.
The8e profiles *ail specify the echeduies
of operational
functions and preventive
maintenance
actions for a given calendar
time.
The mean commtive
downtime and preventive
downtime for the system
are calculated
tn sequence by function,
mission/m~nmn~ce
profile * -d
A procedural
flow block diagram of the procedure
is shown
complete system.
tn Figure 4-1. lb explanation of each block in the diagram follows:
The end itenm (1) of tbe system are identified and categorized
under the appropriate
headings as: system, subsystem,
assembiy,
etc. , down to the smallest piece of
The
equipment on which a specific maintenance
action will be accomplished.
failure rate 1s identified for each end item.
The preventive
maintenance
actions
of an operational
function (2) to be performed
on tbe categorized
end items are
cieflned (e. g. , check out, servicing,
adjustment,
etc.).
The physical and
conceptual
constraints
previously
described
mu~t be defined and documented for
each function.
Ihe corrective
maintenance
actions (3) to be performed
on
appropriate
cate~rized
end items are deff.ned.
Tbeae actions wtll inclu*.
tit
are not necwmaril y limited to the maintenance
actions of @@t, remo V% ~P~~ei
epec!fled by the applicable
constraints.
The phystcd
nd
adjust, repatr,
etc.,
conceptual
constratnt8
previoudy
duacxibed defined s set cf corrective
nctton,
which can be undertaken.
The detectable
end ttem malfunctions
for each preventive
matntensnce
action of an operational
function are defined,
(4). Thoee end items
which can be detected as malfunctioned,
but which cannot be corrected
(within the
Constratnta
of the location at which the maintenance
is occurring),
are grouped.
No troubleshooting
witl be cimducted wkhtn these groups of end items since,
by definition,
no corrective
action can be umiertaken.
A task analysis is con~cted
of end {tern task times
for each preventive
maintenance
action (5). A distribution
and a total time for each nperatlonal
function iE generated.
The total task time
for the operational
function is compared
to the allocated time to determine
if the
maintatnabillt
y design of the equipment
ts adequate.
Xfnot, the dhtrfhtton
of
end item task times pennitg
McnMfication of critical
desi~
points.
A tad
analysis k! conducted for corrective
matnt.enance
(6) associated
with eaoh of the
prevent{ ve matntenence
of an operational
function.
Thts ansJysis is corductd
b
deriving the t roubleshoottng,
repair,
and verification time for each end item
defined a+ a detectable
and repairable
malfunction.
These times
previously
are described
in terms of a distribution
of end item corrective
maintenance
ttmes
versus frequency
of occurrence and by a mean corrective
downtime ( MCDT)
for the spectfied operational
function.
are
As before,
the MCDT and distributions
The preventive
nnd corrective
maintenance
used to identify critical
design points.
4-5
MIIJ-HDBK-472
24 May 1966
-1~
1%1
1.
.- m
lx
I
4-6
II
MIL-HDBK-472
24 May 1966
.-
2.2
Theoretical
Considerations
A parametric
description
of the previously
described
procedure
is provided
m the following paragraphs
and is referenced
to the numbered blocks in
Figure 4-1.
2. 2.1
is described
Function
The preventive
maintenance
actions of an operational
function associated
with
the system are given by Pj such that PI) P2 . . . . Pm -1, and Pm are lnciusive
of all preventive
maintenance
actions and where each action is defined by type
(e. g. , inspection,
servicing,
etc. ) and hy physical and conceptual
constraints
(e. g. , personnel,
spares and consumables,
support equipment.
facilities,
testin~ concepts,
etc. ). A new function must be defined if the type or constraints
are modified.
A subset of Ii is associated
with each function pj. The operational
functions of
the system are given by Or such that O,, 02 . . . . Or-t, Or are inclusive of all
operational
functions,
where each function is defined by the end items of the
An operational
function must be defined for each different
system being used.
subset of Xi utilized durin~ the operation
of the system.
By definition,
all end
items Xl, 12 . . . . In-,, I ~ will appear in at least one subset of Ii. associated
with the preventive
2. 2.3
Corrective
The corrective
c,,
C2
where
maintenance
Maintenance
action
Action
of an operational
function.
Definition
maintenance
actions associated
with the system are given by
and Cg are inclusive of all corrective
maintenance
actions
. . . CQ-l,
each action
is a maintenance
action
4-7
taken to correct
a detected
malfunction
MJL HDBK-472
24 May 1966
2.2.4
Malfunction
Il?tection
Analysis
Preventive
Maintenance
for preventive
oTm
where:
action
PDTm
Pm
T 1-
Task Time
maintenance
Analysis
actions
f
= i:,Ti
m
maintenance
performance
A distribution
of the individual task times within each action
to identify critical
design potnts as previously
described.
4-8
time for
can be developed
MiLr HDBK--li~
24 May 1966
2. 2.6
Corrective
hlaintenance
T ~m (~T$l~)
+ ~cim +
V,m
item 2:
T2m
Item n:
T nrn = (~T~nm ) +
(~l,am)
TC2m
h2m
TCnm
v.m
Item ith:
L.
where:
Tim
end
The troubleshooting
test times
item xi during action pm
end
Tcim
For function
-.
1:
to isolate
Llm
Item
required
1,,
Or:
:
(~~slf)
+ Tel,
+ G,,
4-9
MIL- HDBK-472
24 May 1966
:
Item 2:
(~Ts2
) + Tc2f + ,2
r
Item n:
Item iw:
(XT,,,)
Tc,,
+ TV,,
where:
T:*?
S1,
tn carrcci
tnahnctiming
required
Tcf
T vi,
In addition,
the time to isolate
action Pm is gtven by:
end item
to Isolate
end item
adjuat, calibrate,
end item I i
the non-repairable
given
during
Tjm = ~T~j~
where:
The total time required
to isola~ he )th gTOUP during
action Pm of an operational
function
TIur
Tsjm
The troubleshooting
during action Pm
to isolate
time
the non-repairable
= &jr
4-1o
required
to isolate
tie
jl~
during
grouP
function
Mlb HDBK-472
24 May 1966
The troubleshooting
during function Or
downttie
maintenance
to isolate
time
required
@ isolate
the
jt~
tbe
!PW
where:
MCDTm
Aim
Ai,m
The failure rate of the it~ end item in the jt~ non-repairable
group
which can be isolated during action pm
The mean-corrective-dowtime
to shift to another maintenance
is given by:
of the system
or operational
malfunctioning
or identification
function during
during
end item
of the requirement
function Or
where:
MCDTr =
Aijf
The mean-corrective-downtime
function Or
durtrig
M.Ib HDBK-472
24 May 1966
2.2.7
Total
Maintenance
Task Ttme
AnalyBis
A total maintenance
time analysis 1s conducted to define the total time requtmd
to perform
preventive
maintenance,
and tbe total meam-comective-downtime,
Tbe total time for preventive
matntfmance
for maintenance
of the eyetem.
IS given by:
PDTt
= Xem
PDTm
where:
PDTt
Total praventtv-downtime
calendar ttme
Frequency of occurrence
of the mt~ pmventtve
time
actton during the peclfied calen~r
The mean-corrective-downttme
xnalnterumce profiles.
durtng
the 8pectfied
is dertved
from
maintenance
tbe mtnslon/
wbe re:
MCDT~
Applying
The mean-correctlv-downtime
gtven mi8aion/m8intanence
the equation
to a hypothetical
mtsalon/nmintemace
profile
for the
IIMUlte in:
MW HDBK-472
24 May 1966
-.
em,
+Q1]
(Ai
of the smm
MCDTL = f(MCDTJ
where :
MCDTt
of the system
PDTm
faihmw
with a specified
of tbe system
occurring
for tbe
dur!ng the
mtssion/maintemnce
+ MCDTt
Tp
am
where:
profiles
4-13
of tie action
~flx
tie
gives a total
MIL-HDBK-472
24 May 1966
where:
3.0
o?
maintenance
profile
APPIJCATION
The development
of the systa!n maintenance
times is initiated by establishing
and grouping the physical and conceptual
constralnte
existing wiWn the
maintenance
environment.
The aLlowable corretive
maintenance
actions
(e. g., remove/replace,
repair,
test--troubleshoot--,
adjust, etc. ) are specified
for each end item for each Bet of constraints.
Step 1 - The
constraints
applicable
to each preventive
operational
function,
and to tbe corrective
maintenance
the use of the matrix shown in Figure 4-2.
CORRECTIVE
k
PI
Figure
ACTIONS
Cq
c~
x
x
f2
Pm
MAINTENANCE
1
c1
~w
2:*
l-az
220
ww>++
UJZQ
a-d
~;
maintenance
action of an
action are related through
4-2.
Related
Constraint
An X at a row/column
junction of the matrix
istralnts to the actions Pm of an Or wII1 permit
corrective
maintenance
actions (C ~).
Matrix
con-
hfI&KDBK-472
24 May 1966
A ttme wlue at the matrix junction of an end item row and an action column
indicates that this end item is acted upon or utilized during that action.
For
example,
in Figure 4-3 end item (1J is acted upon during corrective
maintenance
action (cl ).
END
iTEM
CORRECTWE
1,
A,
c,
c~
I,
h,
T,,
1~
k~
13
k3
T3,
MAINTENANCE
ACTIONS
c~
C9
T 1,
TI
?22
72,
I
I
T3,
I
0
I
t
I
I
I
I
i
I
x~
Figure
undertaken
oonstrainta
function.
)S~
4-3.
Tn,
----
----
End item/Corrective
------
Maintenance
Trig
Acdon
Matrix
4-J 5
MILHDBK-472
24 May 1966
END ITEM
PREvENTIVE
P,
Ii
hi
Prcv.
1,
AI
1,
X2
AZ
12,
Is
As
P,
Prav.
Corr.
Pm
Pm
Prwt.
Corr.
I
I
0
T3*
#
I
I
I
I
I
1
t
I
I
I
I
Tp,
I
I
I
I
IJ
I
I
I
X*
ACTIONS
P2
P*
corT.
12,
MAINTENANCE
1
o
A*
TR,
Tot
(in-l)
PDT,
/
MCDT2
PDTm
Figure
~Tim
4-4.
End M6m@reventive
Matnmce
Action ~x
Each weventive
maintenance
action colxmn is divided tnto two parts; the fird
for the times required
to perform the specified action on the affected end
items of the sWem;
and the second part for the times required to troubleshoot,
repair,
and verify detectable malfunctioning
end items.
The end item /corrective
maintenance
action matrix is used to establish
those
items which can & corrected
if malfunctioning
within constrain
specifjed
4-16
end
MIL HDBK-472
24 May 1966
OPERATIONAL
ITEM
FUNCTIONS
-----
or
Ii
Ai
o,
o~
03
1,
A,
1,,
T,r
1~
X2
Tz,
Tat
13
AS
T3a
An
Tnt
Tn,
T ~,
I
I
I
I
In
MCDT,
Figure
4-5.
MCDT2
4-17
MCOTr
MCDT s
Function
Matrix
MIL- EDBK-472
24 M8J 1966
h a manner
ctions.
mabtenance
time, tbe total mean-correctlvedmmtlme,
for maintenance
are cald.ated
for each miaai~nWnin Pamgm@
2.2.7,
CwItodhuls:
~fig
Army - MI
Navy - AS
Project No. MXSC-0327
Navy - AS
Air Force -11
Review
ldentia.1
activities:
Amy
- EL, Ml
Navy - AS, EC, 0S, SH
Air Force -11,
13, 14, 17
4-18
activity:
MIL-HDBK-472
24 May 1966
--
BIBLIOGRAPHY
Theoretical
Basis
and Practical
Maintainability
- Prediction:
1.
of, LSAF Aeronautical
Systems
Division,
- Prediction:
2. Maintainability
of, USAF Aeronautical
Systems
Theoretical
Basi ~ and practiral
Division,
31 Dec 1963.
Approach
15 Feb 1962.
Approach
of Shipboard
4, A Research
Plan for Ikveloping
Methods of Maintainability
Measurement
and Prediction,
Rome Air Development
Center Tech. Note No, RADC-TN-60AD-231 090, 6 Jan 1960.
5,
6.
MSintainability
Electronic
15 -p
Measurement
Equipments,
Phase
and Prediction
Methods
RADCTN-60-221,
for A. F. Ground
II Report
AD-247 155,
1960.
7. Maintainability
Measurement
and Prediction
Methods for A, F. Ground
Phase Ill Report RADC-TN-61-141,
AD-262 454,
Electronic
Equipments,
15 Jun 1961.
8. Maintainability
- Prediction
AD-277 199, 15 Mar 1962.
9. Maintainability
Engineering,
AD-404 898, 5 Feb 1963.
10. Maintainability
RADC-TDR-63-85,
Technique,
Phase
Phase
1 Report
IV Report
RADC- TDR-62-156,
RADC-TDR-63-
Report,
85, Vol. 2,
Phase
V Report
11. Maintainability
and Supportability
Evaluation Technique,
Wright-Air
Development
Division,
Part I, Part II, Vols. 1 and 2, May and Jun 1860.
5-1
NUTE nib
apDcifkatm
to mm
form
my
nqwwmmb
Dot
Oa
my portwm of th
M-d
to
rwqll=t
mpa
of rhcunmta,
Crumat CQIlhcts.~nu
nferwmd
dtxua=atfd
or
Subdwd
to
~d
aor w mquut
dm@t30m,
mmtrutual
or
CiUifiitmn
of
or tnlpJy $uLhorUdJon
roquirwmenh.
,. ....
(Fold dons Mb
DEPARTMENT
Ufv)
OF THE NAVY
111111
OFF)CiAL USINE=
~P4ALT~
FOR RIVATE
USE t=
BUSINESS
REPLY MAIL t
CLASS
FfRST
PERM}T
NO. $2603
WASHINGTON
D C
&
Commander
Naval Air
Washington,
OF THE NAVV
Systems
Command (AIR 51 122)
DC 20361
.-.
..
2 00 CUMEWT TITLE
MUMOC n
In
Ruomm.nO.d
c.
haaonfR*tmulo
OTHER (hpuity)
WO-dIDW
for
Ruomwundstton
!
i
1
I
I
I
I
I
I. NAME of
SLOMITTER
fL~I,
Ftrtf. ,wif -
b WORK TELEPHONE
OpImOI
Co&l
MAILING
400RCSS
ISIWI
CIIY. Slink
8 0A7E
ml
FORM
2 MAR
1426
IS 0f150LETf
NUMOER
(hCk+d#
optional
OF SUSMISS!ON
(V YMMDD)
Afsa
I
1
I
I