1. Check sheet 2. Pareto Diagram 3. Cause & Effect Diagram 4. Histogram 5. Stratification 6. Scatter Diagram 7. Control Chart/ Graphs
Jan 22, 2016
1. Check sheet2. Pareto Diagram3. Cause & Effect
Diagram4. Histogram5. Stratification6. Scatter Diagram7. Control Chart/ Graphs
Facts or figures fromwhich conclusions can bedrawn.
A basis for reasoning,discussion or calculation.Why Gather Data?
1. To identify problems to work on.2. To analyze the selected problem as in
- assessing extent of problem- identifying patterns- verifying probable causes
3. To prevent problems from arising or recurring.4. To select possible action plans5. To establish effectiveness of implemented solutions.
A check sheet is a form preparedto facilitate checking off or marking.
The function of a check sheet is toprovide a systematic technique forrecording observations and should be designed so that data can be posted on them and used with minimum effort.
Check sheets serve many purposes andshould be designed so data can beposted on them and used withminimum effort.
1. Agree as to what event is being observed.
2. Decide on the time period during which data will be collected.
3. Design a form which is clear complete and easy to use.
4. Collect data consistently and honestly.
A Pareto Diagram is a special form ofvertical graphs which helps us to determinewhich problems to solve in what order.
Doing a Pareto diagram based uponcheck sheets or other forms of datacollection helps us direct our attention andefforts to truly important problems.
Step 1. Select the standard for comparison, e.g., annual cost, frequency of defects
Step 2. Select the time period to be studied.
Step 3. Gather data. Use a checklist
Step 4. Transfer information from a check sheet to a column graph arranged in
descending order.
Step 5. Summarize data from the check sheet to construct the cumulative line.
1. To highlight main problems.
0102030405060708090
100M
issi
ng
Parts
Mis
orie
nted
Parts
Scra
tch
Loos
e Sc
rew
Oth
ers
2. To compare problems through the use of different measurement scales.
0
5
10
15
20
25
0
5
10
15
20
25
30
35
40
Customer complaints
No.
Wrongspec
$loss
Wrongspec
3. To aid in root cause analysis (Multi-level Pareto)
Effect Cause
05
1015202530
Frequency
Eyes Legs Hand
Types of Injury
0
5
10
15
20
Frequency
Solder Splash
Steam Dust
Causes of Injury
Multi Level Pareto Illustration
4500
35003000
25002000
0
100020003000
40005000
PPM
W/B MOLD D/A S/P TRIM
STATION
YIELD/STATION
1ST LEVEL
20001500
700300
0
500
1000
1500
2000
2500
3000
PPM
22 LDS 16 LDS 18 LDS 48 LDS
LEAD TYPE
W/B YIELD/LEAD TYPE
2ND LEVEL
Multi Level Pareto Illustration
800
500400
200100
0100200300400500600700800900
1000
PPM
Lifted ball Cratering Misplaced bond
on lead
Wrong wire size Tearing wire
Defects
Defects on 22 LD Pkg.
3RD LEVEL
4TH LEVEL
1510
5
0
10
20
30
40
50
Frequency
Machine Material Operator
Cause of Lifted Ball
Multi Level Pareto Illustration
84 3
0
10
20
30
40
50
Frequency
Bond force Time Temperature
Parameter
Adjustment on the Machine5TH LEVEL
4. To evaluate before and after corrective action.
Before
0
10
20
30
EYE LEGS HAND
Freq
uenc
y
AFTER
1510
4
0
10
20
30
EYE LEGS HAND
Effective
A cause and effect diagram is a picturecomposed of lines and arrows to representrelationships between effects and its causes.
A primary use for Cause and Effect diagramsis to analyze existing problems or situations sothat corrective measures can be taken.
The Cause and Effect diagram is a valuabletoo to use in sorting out “non-contributing”causes; leaving only the “true” causes.
Step 1. Pick a result, effect or problem to be solved. Virtually display problem statement.
Problem, Effect or Result
Step 2. Categorize causes by major elements and put boxes around them.
Start with 4Ms and 1E (Man, Machine,Materials, Method, Environment)
Problem, Effect or Result
Man Machine
Method
Materials
Environment
Step 2. Categorize causes by major elements and put boxes around them.
Start with 4Ms and 1E (Man, Machine, Materials, Method, Environment)
Problem, Effect or
Result
Man Machine
Method
Materials
Environment
Step 2. Add smaller branches to main causes and continue adding until all possible causes are exhausted.
Problem, Effect or
Result
Use brainstorming to generate a large number of specific causes in each category. Ask who, why, what, when, where, or how to stimulate thinking.
It is a creative process for generating a largequantity of ideas utilizing a group.
Compare all causes (What is) against operational standard (What should be). Circle the causes for causing “bad” effect.
The diagram can be used to evaluate if operational standard is inadequate.
Find out which of the circled causes have a significant effect on the problem.
How?
1. Use technical knowledge2. Obtain opinion3. Verify through data
gathering4. Do more analysis of data5. Possibly design an experiment
3. Construct a plan on how theproblems are to be resolved.
* the plan should include activities,timetable and person responsible.
4. Because the Cause & Effect diagram reflects graphically the results of
investigating a problem, it is agood idea to post it in the workarea so that everyone can seewhy action is necessary.
5. Continue improvements and revisionsto improvements.
A Histogram is a graph which showsthe frequency of occurrence in anumber of related measurements.
A histogram reveals how measurementsvary from one another and displaysthe distribution of data. It can beused to compare sample results withspecification.
Step 1. Gather data. Minimum of 50 observations.
Obs. # Readings Obs. # Readings Obs. # Readings Obs. # Readings Obs. # Readings
1 0.41 11 0.40 21 0.40 31 0.39 41 0.40
2 0.43 12 0.38 22 0.40 32 0.37 42 0.40
3 0.37 13 0.37 23 0.39 33 0.40 43 0.41
4 0.38 14 0.37 24 0.39 34 0.41 44 0.39
5 0.40 15 0.43 25 0.41 35 0.42 45 0.43
6 0.40 16 0.37 26 0.42 36 0.40 46 0.39
7 0.38 17 0.41 27 0.40 37 0.40 47 0.44
8 0.42 18 0.40 28 0.42 38 0.41 48 0.40
9 0.40 19 0.36 29 0.40 39 0.43 49 0.39
10 0.41 20 0.42 30 0.40 40 0.41 50 0.41
Step 2. Identify the largest and smallest measurement. Compute the range.
Range = Maximum - Minimum reading reading
Range = 0.44 - 0.36 = .08
Step 3. Determine how many classes are required to make a histogram. Compute for the class interval.
R .08Class Interval = ---- = ------
K 5Class Interval = .016 or .02
K is a constant determined from a table.
For Constant K,
DATA and Class Amounts
Number of Appropriate No.Observations of Classes
50 5 - 751 - 100 6 - 10101- 250 7 - 12Over 250 10 - 20
Step 4. Determine the boundary line between classes.
Class Class Interval 1 0.36 - 0.37 2 0.38 - 0.39 3 0.40 - 0.41 4 0.42 - 0.43 5 0.44 - 0.45
Step 5. Transfer data to a tally sheet
CLASS TOTAL
1 0.36 - 0.37 IIII - I 62 0.38 - 0.39 IIII - IIII 93 0.40 - 0.41 IIII - IIII - IIII - IIII - IIII 254 0.42 - 0.43 IIII - IIII 95 0.44 - 0.45 I 1
TOTAL 50
CLASS INTERVAL
FREQUENCY
Tally Sheet
Step 6. Transform data from tally sheet to a Histogram.
16 9
25
9
0
10
20
30
40
0.36 - 0.37 0.38 - 0.39 0.40 - 0.41 0.42 - 0.43 0.44 - 0.45
Classes
Freq
uenc
y
A Scatter Diagram is used to study thepossible relationships between one variableand another.
The Scatter Diagram is used to test forpossible cause and effect relationships.
It cannot prove that one variable causesthe other, but it does make it clearwhether a relationship exists and thestrength of that relationship.
A Scatter Diagram is set up whereby the horizontalaxis (X-axis) represents the measurement values ofone variable and the vertical axis (Y-axis) representsthe measurement of the second variable.
Variable
2
Variable 1
Step 1. Collect 50 to 100 paired samples of data that you think may be related. Construct a data sheet as follows:
Obs. # Auto (g/l)y
1 42.482 41.543 42.014 *5 *6 ** ** 53.37
Manual (g/l)
*54.21
x
***
41.3342.5342.53
Relationship between Manual & Auto Titration (Sn + 2)
Step 2. Draw the horizontal and Vertical Axes of the diagram.
Manual g/l X
40 42 43 44 45 46 47 48
50
43
42
40
Auto g/l y
Step 3. Plot the data on the diagram. If you find the values being repeated, circle that point as many times as appropriate.
4042444648505254565860
40 42 44 46 48 50 52 54
Manual g/l
Aut
o g
/l
Patterns
Positively Correlated Negatively Correlated
No Correlation
Stratification is a process of classifying data into subgroups based on categories and characteristics.
Helps analyze cases in which data actually masks the real facts.
Breaks down single numbers into meaningful categories or classifications to focus on the corrective action.
1. During data gathering to design check sheets, create checklist, scatter diagram and cause and effect diagram.
2. During data analysis when using histograms, pareto charts, scatter diagrams and cause and effect analysis.
Stratification break down single numbers intomeaningful categories or classifications to focuson corrective action.
Control Chart is a graphic representation of a process.
Sample averages are plotted on the chart.
Statistically determine the upper and lowercontrol limits drawn on either side of theprocess average.
Makes it possible to tell if a process is functioning normally and to see immediately if malfunctioning has occurred.