1 SPC Material

5/19/2018 1 SPC Material

1/76

TRAINING MATERIALON

STATISTICAL PROCESS CONTROL

(SPC)

116-A, JWALAHERI MARKET, 2nd FLOOR,PASCHIM VIHAR, NEW-DELHI-63

Ph: 011- 22!"# M$%&'(: )"10!0)))E-*+&' :'$%+'(./&(d('h&+h$$4$*

%+'+5.&hn+n%+%h$7*+&'4$*


2/76

D(7(7&$n: A past oriented strategy that attempts to

identify unacceptable output after it has beenproduced and then separate it from the good output.

D(8(7 D(7(7&$n:Is reactionary

Tolerates wasteRelies on inspection, audits, or checks of large samplesof output

Reacts to all defects indiscriminatelyFocuses on conformance to specicationsInoles action only on outputRelies on delayed feedback for defect detectionIs not cost e!ectie

TOOLS FOR PROCESS CONTROL


3/76

Prevention :- A future oriented strategy that improves quality and

productivity by directing analysis and action toward correcting the

process itself so that unacceptable parts will not be produced.

Defect Prevention:

Is pro-active

Avoids wasteUses small samples of product and process information

Is analytically based

Discriminates between potential defects based on causes

Involves action on the process or process parameters

Provides timely feedbacIt is cost effective


4/76

14 M&7+5( P.$$9n :" In this techni#ue $%%& process

control is achieed by preenting all types of failures byusing modern techni#ues to get defect free product. 'erecauses are preented from making the e!ect.

(. 100 In;(7&$n: In this techni#ue $%%& checking of allthe parameters of all products has been done to get defect

free product. 'ere only defects are detected.34 S7+7&7&+' P.$( C$n7.$': In this )tatistical techni#ue

such as *ontrol *hart, 'istogram etc. are used so as toanalyses the process and achiee and maintain state ofstatistical control to get defect free product. *auses are

detected and prompting correctie action before defectoccurs.

WHY S.P.C. IS REQUIRED ? +!ectieness of any actiity inan rgani-ation is measured with respect to time and costinoled in it.

TECHNIQUES FOR PROCESS CONTROL


5/76

Mistake Proofing 100% InspectionStatistical ProcessControl

In this method

more advancedand moderntechniques areused whichrequire substantial

investment duringits installation andmaintenance.

As it is detection type

of technique it can!tavoid failure butre"ects defectiveproducts.#equires more

inspectors$ moreinspection times andin turn more cost.

%or this technique

investment is very lessand process iscontrolled on eachworstation thereforedefective components is

not forwarded to ne&toperation. Predictabilityreduces frequentad"ustments ' in turnincreases productivity$reduces inspection cost

at station ' at finalinspection

From above we can observe that S.P.C. is the economical way of controlling

the process in comparison with Mistake Proofing and 100% inspection.


6/76

WHAT IS S.P.C. ?14 S7+7&7& :" A alue calculated from or based uponsample data e.g. a subgroup aerage or range/ used tomake inferences about the process that produced theoutput from which the sample comes.

24 S7+7&7&+' C$n7.$' :" The condition describing a processfrom which all special causes of ariation hae beeneliminated and only common causes remain.

34 S7+7&7&+' P.$( C$n7.$' :" The use of )tatisticaltechni#ues such as control charts to analy-e a process or

it0s outputs so as to take appropriate actions to achiee andmaintain a state of statistical control and to improe theprocess capability.


7/76

Population: A part of the universe$ which is under study for gathering

information$ is called population. (he concept would be clearer from

the following e&amples:). (he number of persons out of the total employees of a company$ who

are engineers.

*. (he total number of households in a city who have computers in

house.

+. (he crop production of a particular variety in a district in the currentyear.

,. (he number of devotees visiting a shrine in a year.

any events and activities are occurring in the world

universe/. 0y maing the definition of 1population! we clearly mar the

scope of the study2 information gathering process.

It is generally neither practical nor necessary to observe the

entire population. #ather$ we observe only a small subset of it at a

time$ often doing so periodically. 3hen we do it this way$ we are said

to be sampling the population.

Populatio a! Sa"pl#


8/76

Sample:

(o find out how the population is behaving$ we will observe only a

part of the population and gather data. 3e will then use this data to

infer something about the population. 4ample is a sub-set of thepopulation.

For example

(he number of engineers within a department$ in a company.

(he number of household in a bloc of a city$ who have computers.

(he crop production of a particular variety in a small village of thedistrict.

(he number of devotees visiting a shrine per month.


9/76

UN$ERSTAN$ING %ARIATION

No two things are totall alike or i!entical" 5ence variation is a

natural phenomenon and is universal.

6o two products or characteristics are e&actly alie$ because any

process contains many sources of variability. (he differences among

products may be large$ or they may be almost e&tremely small$ but they

are always present.

(he diameter of a machined shaft$ for instance$ would be

affected by$ Machineclearances$ bearing wear/ #ool strength$ rate of wear/

Materialchemical composition$ hardness/ $perator part feed$ accuracy of centering/ Maintenancelubrication$ replacement of worn parts/ nvironmenttemperature$ constancy of power supply/


10/76

4ome sources of variation in the process cause very short-run piece-to-piece

differences 7

e.g.$ baclash$ clearances within a machine and its fi&tures$

8ther sources of variation tend to cause changes in the output only over a

longer period of time$ either gradually as with tool or machine wear$ step-wise as

with procedural changes$ or irregularly$ as with environmental changes such as

power surges .

(herefore$ the time period and conditions over which measurements are

made will affect the amount of the total variation that will be present

%rom the standpoint of minimum requirements$ the issue of variation is often

simplified - parts within specification tolerances are acceptable$ parts beyond

specification tolerances are not acceptable9 reports on time are acceptable$ latereports are not acceptable. 5owever$ to manage any process and reduce

variation$ the variation must be traced bac to its sources. (he first step is to

mae the distinction between causes of variation.


11/76

Cau& o' %aiatio

T*## a# t+o ,au& o' -aiatio&Common Causesalso called as hance auses or 6on-controllable

auses/Special Causesalso called as Assignable auses or ontrollable

auses/

C$MM$N C&'SS

Arise from causes that are inherent in the process 4ome degree affect all the output of the process ;&ist even as the process is statistically stable and behave lie a

constant system. 3hile individual measured values are different as a group they tend

to form a distribution pattern that can be described as predictable in terms of


12/76

(he e&tent of ommon auses of variation can be indicated bysimple statistical techniques$ but the causes themselves need more

detailed analysis to isolate.

(hese common causes of variation are usually the responsibility of

management to correct$ although other people directly connected withthe operation sometimes are in a better position to identify these

causes and pass them on to management for correction.

8verall$ though$ the resolution of common causes of variation usually

requires actions on the system.


13/76

SPCI&( C&'SS

Affect only some output of the process and are not inherent in the

process. 4pecial auses refer to any factors causing variation that cannot be

adequately e&plained by any single distribution of the process output$

as would be the case if the process were in statistical control. Unless all the special causes of variation are identified and corrected$

they will continue to affect the process output in unpredictable ways.

Examples of Sp#,ial Cau&of Variation includePeople= %atigue$ illness$ state of health.=


14/76

Plant)Machines= #otation of machines= Differences in test or measuring devices

= 3orn out machinery or tools=


15/76

4pecial causes of variation result in output behaviour that is usually erratic

and unpredictable.

(he discovery of a special cause of variation$ and its removal$ are usually

the responsibility of someone who is directly connected with the operation$

although management sometimes is in a better position to correct.

(he resolution of a special cause of variation$ then$ usually requires localaction. harting highlights occurrence of special causes.

4pecial auses of ?ariation= 5ave frequency distributions that are unstable over time

#esult in unpredictable outcomes= #eappear local unless action is taen

an only be eliminated by actions 1on the shop floor!


16/76

Ra!o" %aiatio NoRa!o" %aiatio

$nl common cause are

present

Common * &ssigna+le cause

are presentCommon causes are more innos"

&ssigna+le causes are ver fewin nos"

Common causes are part ofprocess

,isitor to the process

Contri+utes to constantvariation

-ighl fluctuating variation

Pre!icta+le 'npre!icta+le

Statistics &ppl Statistics shall not appl

Management controlla+le $perating personnelcontrolla+le


17/76

8nly a relatively small proportion of all process troubles industrial

e&perience suggests about )@/ is correctable locally by people directly

connected with the operation9 the ma"ority 7

the other B@ - is correctable only by management action on the system.

onfusion about the type of action to tae is very costly to the organiCation$

in terms of wasted effort$ delayed resolution of trouble$ and aggravated

problems. It would be wrong9 for e&ample$ to tae local action e.g.$ad"usting a machine/ when management action on the system was required

e.g.$ selecting suppliers that provide consistent input materials/.

A(I864 taen to combat the above types of variation include$


18/76

Actions on the 4ystem are those that almost always require management

action for correction and are usually required for eliminating common

causes of variation. (hese actions can correct about B@ of the process

problems.

%ARIA/LE 0 ATTRI/UTE ;very process generates data that can be categoriCed in 1?ariables

data! ' 1Attributes data!.

,aria+le !ata: ?ariables data relates to what can be measured and

e&pressed quantitatively in specific units of measurements as

dimensions$ volume$ temperature$ pressure$ time$ strength$ etc.

(he eamplesinclude: Diameter of a bearing

?olume of the cylinder (emperature of an oven 3eld strength of a location apsule filling time 5ardness of a gear


19/76

&ttri+utes !ata: Attributes data are qualitative data where the results

are recorded in terms good or bad$ defective or defect free$ pass or fail$

yes or no$ acceptable or not acceptable$ conforming or non-conforming$

presence or absence of a desired characteristic.

!amplesinclude sticer printing acceptable or not acceptable an item meets the gauge or not Dents on body pass or fail 0raCing "oint good or bad


20/76

SELECTING SAMPLE SI1E

8ne of the first basic questions which comes in our mind before

carrying out a survey or statistical study is 5ow many samples

should I selectEF

>enerally we select the samples as per our convenience or

without any particular logic e&cept cost ' time.

sampling is time consuming and costly$ our ob"ective in selecting a

sample is to obtain a specified amount of information about a

population at a minimum cost.

Samples ) su+groups for control charts

#ational subgroups or samples are collection of individual

measurements$ whose variation is attributable only to a constant

system of common causes. In the development ' continuing use ofcontrol charts$ subgroups or samples should be chosen in a way that

provides the ma&imum opportunity for measurements within each

subgroup to be alie sub"ect only to forces of common cause

variation/ and ma&imum chance for subgroup to differ from one

another if special cause arisebetween subgroups.


21/76

3ithin sample or subgroup only common cause variation should be

present.

Sample si.e consi!erations

(he siCe of a rational sample is governed by the following

considerations. 4ubgroups should be sub"ect to common cause variation. 4ubgroup should ensure the presence of normal distribution for the

sample means 4ubgroups should ensure good sensitivity to the detection of special

causes. 4ubgroups should be small enough to be economically appealing

from a collection ' measurement standpoint

Generally 4 to 6 sample / subgroup is commonly used

considering the above factors.


22/76

"ote #

4hewhart suggests / as the ideal subgroup siCe In

industrial use of control chart @ is most common ;ssential ideaof control chart is to select sub groups in such away that it gives

minimum opportunity for variation within sub group$ it is

desirable to be as small as possible for economic purpose/.

A siCe of /is better than * or + on statistical grounds9the distribution of B is nearly normal for sub groups of four or

more even though the samples are taen from non normal

universe.

4ub groups of * or + are used 86


23/76

(his write up is for assignment any one identify Assignablecauses. As you are aware the success of any 4P program is not in ourability to collect data$ draw charts etc.$ but in effectively identifying andeliminating assignable causes. Assignable causes are those causes thatdo not allow one to predict the behavior of processes. (here is nomeaning in calculating Process apability without having a predictableprocess.

any companies have initiated 4P charts. 0ut the charts do notbenefit them. 8ne of the main reason for this is that they have notstopped the process when an assignable cause is indicated andeliminated the cause. (his is not done because no body is aware on howto do it. any e&perts only say that the cause is to be eliminated but noone is able to assist a company in doing this. 3e are sharing with you ourapproach for doing this cause elimination.

METHO$OLOG2 TO I$ENTIF2 ASSIGNA/LE CAUSES


24/76

0efore starting the 4P data collection$ let us do the following steps:

). Identify the characteristic for which 4P is to be done.

*. 5ave a brainstorming to list all the causes that may influence the

variation in this characteristic

+. Prepare a ause ' ;ffect Diagram

,. Prepare a aster ause Analysis (able Anne&ure )/

@. Prepare a 3hy-3hy Analysis (able Anne&ure */

H. Identify factors that may affect Average and those that may affect#ange

After completion of the above$ plan for data collection$ calculationof preliminary limits$ etc. (hen use the chart for 8n -


25/76

Sl"No Cause Is thereanSpec"

If sowhat istheSpecn"

2asis3or theSpecn

Is itchecke!* how

4hat istheactual

Diff" inSpecfn"

,s&ctual

&ctionplan

ANNE3URE 4 5MASTER CAUSE ANAL2SIS TA/LE


26/76

Sl"No Cause 4h 4h 4h 4h 4h

ANNE3URE 4 6

WH2 4 WH2 ANAL2SIS TA/LE


27/76

). ;nter a serial number

*. ;nter the cause from the cause and effect diagram. All the

causes from the cause and effect diagram must be covered

+. %or each cause as the uestion: Is there a 4pecificationE Please

note that the specification is for the cause. (he answer can be

Ges or 6o.

,. If there is a 4pecification$ write the actual value of the

specification. If there is no specification$ ;nter in the Action Plan

column 4pecification is to be establishedF.

@. >ive the basis for the specification mentioned in olumn 6o.,.

4ometimes the 4pecification may be based on the drawing$machine manufacturer!s catalogue$ wor instruction$ Past

;&perience$ etc. Do not write your e&pectations. 8nly enter what

is actually e&isting. #emember that there has to be some basis.

GUI$ELINES FOR USING ANNE3URE 4 5


28/76

H. Is this specification being checed. If yes$ write the actual method used

for checing. If it is not being checed$ then enter 6o. It may be possible

that there are methods for checing but not done here$ in which case the

answer is 6o. If the answer is 6o$ then enter in the Action Plan olumnethod of checing is to be establishedF.

J. ;nter here the actual value of the cause by using the method of

checing. 4ometimes it may be the range of variation ;&: Input material

condition/ or it may be 8ne ?alue ;&: (aper in the fi&ture/. (his is the

actual value and not a guess. (ime may be required to complete thiscolumn.

B. If there is a difference between the actual value and the specification$

then e&amine how important based on technical nowledge. If the

difference is not ma"or$ then mention 6o. 8therwise mention Ges. If the

answer is Ges$ then enter in the Action Plan olumn that further analysis

is needed lie 3hy-3hy Analysis or correction to eliminate the variation.

K. Under this column enter the specific Action Plan needed as already

mentioned above.


29/76

(his table can be used for all the causes identified in the

ause and Analysis (able with top priority for the auses found to

have variation from the aster ause Analysis (able.

). ;nter the running serial number

*. ;nter the cause to be studied

+. ;nter 3hy this cause should vary. (here may be more than one

reason. ;nter all the reasons one below the other.

,. %or each of the 3hy identified in olumn +$ write the possible

causes. 6ote that the cause is to be identified only for column +

and not bacward.

@.Proceed in the same manner as olumn ,. ;nsure that each timethe focus is only on the previous column.

H. Proceed in the same manner as olumn ,. ;nsure that each time

the focus is only on the previous column.

GUI$ELINES FOR USING ANNE3URE 4 6


30/76

J. 1roceed in the same manner as *olumn 2. +nsure thateach time the focus is only on the preious column.

*ontinue in this manner, till any of the following happen:

a. 3o further 4hy can be answered

b. The system cause has been identied +5: 3o systemfor checking, erication, control, etc./

c. The reerse is the solution

6. 7ased on the listed why8s, deelop the action plan forimplementation.

R(*(*%(.:

. I7 & 7h( 7(*,


31/76

SPC PROCE$URE FLOW CHARTMS&

D& C$((C#I$N

Is ProcessPre!icat+l

5I"e" In Control

N$

3in! out assigna+le

cause an! eliminateit"

6esN$

Is process No

Capa+le

Improve the

process

6es

sta+lish the Control

(imits

Prepare 7eaction

Plan

$n going

Process Control


32/76

Measures of central ten!enc * !ispersion

(here are various measures of central tendency ' dispersion$ the most

commonly used are as follows

Mo!e: It is the most often occurring observation. It may be noted that

more than one mode may e&ist in a population.

7ange: It is the midpoint between the highest ' lowest observation.

Me!ian:It is the middle observation when all observations are arranged

in the order of magnitude.

Mean: It is arithmetic average of observations

Example: onsider a sample of )L observations as follows :

*.)$ *.@$ *.+$ *.*$ *.,$ *.,$ *.+$ *.+$ *.+$ *.)

%ind out ode$ #ange$ edian$ ean


33/76

CONTROL CHARTS TOOL FOR SPC

Dr. 3alter 4hewhart of the 0ell


34/76

;'ID(INS 3$7 S(C#I$N $3 &PP7$P7I C-&7#

?arious types of ontrol harts are employed based on the type of data

generated. (he chart below provides information about selection criteria of

control charts used in various situations.

< * M 7

n = 1

< * 7 o r < * s

n > 1

I s n = o r n !

, a r i a + l e

u

' n i t

c

P o r t i o n

" n i t o r # o r t io n

n = 1

n p

C o n s t a n t

p

, a r i n g

$ o n s t a n t o r % a r y i n g

n > 1

I s n = o r n !

& t t r i + u t e

# p e o f D a t a


35/76

PROCESS CONTROL :-

A process is said to be operating instate of statistical control when the

only source of ariation is common causes.PROCESS STABILIITY: -

The process is said to be stable when the process is in control and

ariation is constant with respect to time i.e. 7eing in statistical control.

PROCESS CAPABILITY: -

The measure of inherent ariation of the process i.e )i5 )igma ;" < =/

when it is in stable condition is called as process capability.

OVER ADJUSTMENT: -

It is the practice of ad>usting each deiation from the target as if it were

due to a special cause of ariation in the process. If stable process is ad>ustedon the basis of each measurement made, then ad>ustment becomes an

additional source of ariation and inturn it increases the total ariation.


36/76

C-C?IN; 3$7 P7$CSS P7DIC#I$N :8

Process is said to be predictable when it is in control and stable i.e. when

all 4pecial causes are removed from the process. (he process can be

checed from ontrol hart and 5istogram.

ontrol hart: - 3hen all points are within control limits or there is no

obvious run or non-random pattern of points with in the control limits.

5istogram: - 3hen bell shape is observed on 5istogram.

#;8?I6> A44I>6A0


37/76

C&(C'(IN; P7$CSS C&P&2I(I#6: 8

After removing all special causes from the process calculate the capability

indices pand p .

If pand p is greater than ).++ then process is said to be within acceptable

capability. 0ased on the priority mae improvement plan for the process.

If pand p is less than ).++$ then find out ma"or common causes and remove

it.

S#&2(IS-IN; C$N#7$( (IMI#S: 8

3hen p and p is greater than ).++ $ (hen $ ;stablish U


38/76

$N ;$IN; P7$CSS C$N#7$( :8

ontinuous Periodical review of control chart and recorded process

events to identify the preventive action and revise the control limits.

8P;#A(8#M4 #8#8UP.

Data Collection

Plot $n Chart

Is process in Control

I"e" no special cause

No

7efer 7eaction Plan

6es

#ake Corrective &ction

#ake Disposition

&ction 5 If 7e@!"A

&'(E )

1"&fter Corrective action taken the Imme!iate su+group shall +e measure! an! plotte!"

B"7ecor! in !etail the causes corrective action an! !isposition action taken for ever out ofcontrol con!ition"


39/76

FACTORS FOR COMPUTING LIMITS

n !B &B D D/ B

B 1"1B 1"0 0 "BE B"EE

1"EF 1"0B 0 B"GH/ 1"HH

/ B"0GF 0"HBF 0 B"BB 1"/E

G B"BE 0"GHH 0 B"11/ 1"BF

E B"G/ 0"/ 0 B"00/ 1"1


40/76

1" D& C$((C#I$N

8 7 C-&7#= :

87 chart is !evelope! from measurements of a particular=characteristic of a process output" 87 chart eplains process !ata in=terms of +oth its sprea! 5piece to piece varia+ilitA an! its location5process averageA"

D& C$((C#I$N:

=>Measure of (ocation

7>Measure of Sprea!

Scan the plot points confirm that the calculations an! plots arecorrect" Make sure that the plot points for the correspon!ing an! 7 is=verticall in line"

Initial stu! charts use! for first time capa+ilit or for stu!ies afterprocess improvements)changes shoul! +e the onl process control

charts allowe! on the pro!uction floor which !o not have control limitsplace! on them"


41/76

24 CALC@LATE CONTROL LIMITS :

R ? R$ R( @@@ R / ;

B5? C$ C( @@@ C/ ; 4here

is the number of subgroups.

R$is the range of the rst subgroup.

=5 is the aerage of the rst subgroup.

S(7; $n7.$' h+.7 :

= and R charts are normally drawn with the= chart aboe the Rchart.

Data block should include spare for each indiidual reading, aerage= /, Range R / and the date;time or other identication of thesubgroup.


42/76

*haracteristics to be plotted are the sample aerage = / andthe sample si-e R / for each subgroup, collectiely these reEect

the oerall process aerage and its ariability.Aerage =/ ? C$ C( @@@. Rn / ; n

4here n subgroup sample si-e.

Range R / ? 'ighest Gowest

S('(7 7h( S+'( 8$. $n7.$' h+.7 :)ome general guidelines for determining the scales may be

helpful, although they may hae to be modied in particularcircumstances.

F$. = Ch+.7 :The di!erence between the highest and the

lowest alues on the scale should be at least two times thedi!erence between the highest and the lowest of the subgroupaerages =/.


43/76

F$. R Ch+.7 :

Halue e5tends from -ero to an upper alue about two times the largest range.

F$. R Ch+.7 :

*GR? D2RJJ

G*GR ? D< RJJ

F$. B Ch+.7 :

*G5 ? CK A( RJJ

G*G5 ? CK " A( RJJ

4here D2, D


44/76

3. INTERPRETATION FOR PROCESS CONTROL

)ince the ability to interpret either the subgroup ranges or

subgroup aerages depends on the estimate of piece to pieceariability, the R chart is analy-ed rst. The data points arecompared with the control limits, for points out of control or forunusual patterns or trends.

For Range *hart :

a/ 1oints beyond the control limits are primary eidence of non"control of that point. Any point beyond a control limit is the signalfor immediate analysis of the operation for the special cause.

A point aboe the control limit is generally due to

$/ 1lot point may be miscalculated.

(/ 1iece to piece ariations has increased.


45/76

A point below the control limit is generally due to

$/ 1lot point is in error.

(/ 1iece to piece ariation has decreased.


46/76

%ind and Address 4pecial auses

%or each indication of special cause in the range data$ conduct an

analysis of the operation of the process to determine the cause and toimprove the process.

A process log may also be a helpful source of information in terms of

identifying special causes of variation. 4ingle point out of control is reason to

begin an immediate analysis of the process.

#ecalculate ontrol


47/76

4


48/76

AnalyCe the data on the A?;#A>; 5A#(

3hen the ranges are in statistical control$ the process spread 7 the

within subgroup variation is considered to be stable. (he averages canthen be analyCed to see if the process location is changing over time.

ontrol limits for &O

are based upon the variation in the ranges. (hen ifxxxxxxxxxxx

the averages are in statistical control$ their variation is related to the

amount of variation seen in the ranges common cause variation of the

system/. If the averages are not in control$ some special causes ofvariation are maing the process location unstable.

Points beyond control limits indicates that there is

)/ shift in process

*/ Plot points are in error.

%ind and address the special causes and then recalculate the control

limits after eliminating the special causes.


49/76

#4 INTERPRET FOR PROCESS CAPAILIT

Interpretation process capability is to be carried out only under thefollowing assumptions:

$/ 1rocess is statistically stable.

(/ Indiidual measurements from the process conform to normaldistribution.


50/76

*alculate 1rocess )tandard Deiation:

)ince within subgroup process ariability is reEected in thesubgroup aerages, the estimate of the process standard

deiation N=N can be based on the aerage range R/. =? RJ ; d(

4here

RJ the aerage of the subgroup ranges.

d(the constant arying by sample si-e.*apability can be described in terms of the distance of the

process aerage from the specication limits in standarddeiation units, O.

For unilateral tolerance

O ? )G CK/ ; = or/ O ? 5K "G)G / ; =


51/76

%or bilateral tolerance

U4

1 SPC Material

Documents

process control5192018

process informationis

lessand process

spc material 276d77

defect free product

defect prevention

specicationsinoles action

defect detectionis