Seminar On Statistical Process Control On Cylinder Liners

ABSTRACT Process Capability Two concepts need to be considered: Process capability is determined by the variations that comes from common cause. Customers, internal or external, are more typically concerned with the overall output of the process.

Statistical Quality Control Software The software stores information in a database

and develops the control chart for the data. The data may be retrieved any time on user request. The multiple document interface help in the view ability and accessibility of the data. It works on set guidelines, the specification can be user defined.  

INTRODUCTION Machining Process Process control plan are made mainly on the following process such as,  Outer diameter turning (Rough)  Inner diameter turning (Rough)  Outer diameter turning (Finish)  Finish bore turning  Rough honing  CNC turning  Finish honing  Plato honing.

OVERVIEW OF SQC Statistics :

Collection, tabulation, analysis and interpretation of the data. Quality : “Fitness for Use”. Control : Directing and maintaining the quality of the thing. SQC : A tool of scientific management where the control of the quality of characteristics of a product is possible with the help of systematic measurement in terms of numbers.

STATISTICAL PROCESS CONTROL Definition : Application of statistical methods at the process

control stage is known as statistical process control. SPC is powerful collection of problem solving tools useful in achieving process stability and improving capability. The seven major SPC problem solving tools are 1. Histogram 2. Check sheet 3. Pareto Charts 4. Cause and effect diagram 5. Defect concentration diagram 6. Scatter diagram 7. Control Charts

OBJECTIVES OF QUALITY CONTROL: To reduce company cost through reduction of the

losses due to defects. To produce the optimum quality at minimum price. To make inspection prompt to ensure quality control.

BENEFITS OF SQC Reduction of scraps. Increases output. Efficient utilization of man, machines,

material & money. Reduce wasted machine and man hour. Ensures rapid and efficient inspection at a minimum cost. Easy detection of faults.

PROCESS CAPABILITY Definition :

“Minimum spread of specific measurement variation, which will improve 99.7% of the measurement from the given process”. Process capability is the measured inherent variation of the product turned out by a process. Process capability study is carried out to measure the ability of the process to meet the specified tolerances.

FACTOR INFLUENCING PROCESS CAPABILITY Condition of machines/equipment. Type of operation and operational conditions. Raw materials. Operator’s skill. Measurement instruments. Inspector’s skill.

PROCESS CAPABILITY RATIO (Cp) The ability of the process to meet the tolerances if at all strength of the output that process will produce if (6 sigma > (USL-LSL)). Cp = (USL - LSL)/6 * sigma where Sigma = standard deviation When Cp = 1.00: then the process is just about to meet desired

specification. When Cp< 1.00: then the process will not meet the desired specification and is out of control. When Cp>=1.33: then the process is under statistical control and meets desired specification.

PROCESS CAPABILITY INDEX (Cpk) Process capability indexes that measures potential for the process to generate the output relative upper or lower specification.

Z min C pk = 3

Where Zmin = the smaller of the Zupper, Zlower

Z upper

( USL − X ) = σ′

Z lower

( X − LSL ) = σ′

• When Cpk > 1.33: the process is capable • When Cpk < 1.00: process is not capable • When Cpk=Cp: the process is centered between the upper & lower specification.

STEPS INVOLED IN THE ESTIMATION OF PROCESS CAPABILITY: Using X (bar) and R chart information we determine the value of Cp & Cpk the formula for chart control limits STEP 1: The methods of computing control limits for a range chart based on 3 standard deviations are:

UCL R = D 4 * R

LCL R = D3 * R

D4 and D3 are from a table of multipliers. STEP 2: The methods of computing control limits for an x-bar range chart based on 3 deviations is: Upper control limits

Lower control limits

UCL x = X + A 2 * R

LCL x = X - A 2 * R

Where: A2 is from a table of factors of control limits. STEP 3: Standard Deviation is calculated by

σ′ =

R d2

CAPABILITY ANALYSIS Process capability is analyzed with a set of statistics that

have been calculated from a set of process data and specifications. Capability analysis compares how a process is actually

running to its specifications. A system is said to be “capable” if it is producing

approximately 100% of its output within specifications.

 Cpk

Capability Index tells how well as system can meet specification limits.  If Cpk is 1.0

The system is producing 99.73% of its output within specifications.  If Cpk is between 0 and 1.0

Not all process output meets specifications.  If Cpk is less than 1.33  If Cp is less than 1.33: Variations in the process should be reduced.  If Cp is greater than 1.33: The process should be centered within its

specifications.

 If Cpk is greater than 1.33  If Cp is less than 1.33: Not mathematically possible.  If Cp is greater than 1.33: Fine tune and improve the process continuously.

CONTROL CHARTS

 CONTROL CHART

A control chart is the graphical representation of the collected information.  THE CONCEPT OF VARIATIONS  No two items will be perfectly identical even if extreme care is taken.  Variation is a fact of nature.  The variation may are expressed in microns.  REASONS FOR VARIATION  Tools wear, Machine vibration, Loose bearings, Poor quality of raw

materials, Operator’s Carelessness, Untrained operator, Fatigue caused to operator, Change in working condition/weather, Poor maintenance, Measuring errors, etc.

 KINDS OF VARIATIONS

1. Variation Due To Assignable Causes  These can be easily traces are detected.  The variations due to assignable causes:  Difference among machines, workers, materials. 2. Chance Variation  They are difficult to trace and difficult to control.

Fig: Control Chart Patterns of Chance Variation

Assignable cause pattern of variation

 Extreme variation

Extreme variation is recognized by the points falling outside the upper and lower control limits.

 Causes:

 Errors in measurements and calculations  Wrong setting of machine, tools



Indication of Trend: The points on X or R chart tend to move steadily either towards LCL or UCL, it can be assumed that process is indicating a trend.

   

Causes: Tool wear Wear of threads on clamping devices Effects of temperature and humidity.

 Shift

When a series of consecutive points fall above or below the central line, control chart is can be assumed that shift in the process has taken place.

 Causes:  Change in materials  Change in operator, inspector, inspection equipment  Change in machine setting  New operator, carelessness of the operator, etc.

Erratic Fluctuations

Causes: Frequent adjustment of machine Different types of material being processed. Change in operator machine etc.

NAME OF THE LINER: MWM HONING

X-bar & R chart

Upper Control Limit Lower Control Limit Average Sigma Center

: : : : :

AVERAGES ---------------128.01042684 128.00407316 128.00725000 0.00211790 128.00725000

RANGE -----------------0.00994974 0.00000000 0.00436000

Control Limits based on all plotted groups Control Limits shown for complete groups of size 4. There were no observations out of control.

Process Capability chart

24.0

23.00

19.2

18.40

13.80

9.6

9.2 0

4.8

4.6 0

0.0 128.002

128.003

128.005 128.004

05/28/09 16:02 CHART1: Process Capability Chart: mwm234honning - DATA1

128.007 128.006

128.009

128.008 128.010 CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.0 35. Lack of fit is signific ant.

128.011

128.013 128.012

0.0 0 128.014

PERCENT

CELL FREQUENCY

14.4

Calculation of Cp & Cpk

NAME OF THE LINER: MWM COLLAR WIDTH

X& R chart-bar

Upper Control Limit Lower Control Limit Average Sigma Center

: : : : :

AVERAGES ---------------7.98774933 7.97679067 7.98227000 0.00365289 7.98227000

RANGE ---------------0.01716102 0.00000000 0.00752000

Control Limits based on all plotted groups Control Limits shown for complete groups of size 4. There was 1 observation out of control.

Process Capability chart

50.0 46.00 40.0 36.80

27.60

20.0

18.40

10.0

0.0 7.9 70

9.20

7.972

7.976 7.974

05/28/09 15:37 CHART1: Process Capability Chart: MWM234CW - DATA1

7.980 7.9 78

7.9 84

7.9 82 7.9 86 CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.0 00. Lack of fit is significant.

7.9 88

7.9 92 7.9 90

0.00 7.9 94

PERCENT

CELL FREQUENCY

30.0

Calculation of Cp & Cpk

NAME OF THE LINER: MWM STEP OUTER DIAMETER

X& R chart-bar

Upper Control Limit: Lower Control Limit: Average : Sigma : Center :

AVERAGES ---------------147.90180694 147.88659306 147.89420000 0.00507129 147.89420000

RANGE -----------------0.02382460 0.00000000 0.01044000

Control Limits based on all plotted groups Control Limits shown for complete groups of size 4. There were 4 observations out of control.  

Process Capability chart

35.0

34.00

28.0

27.20

20.40

PERCENT

CELL FREQUENCY

21.0

14.0

13.60

7.0

6.80

0.0

0.00 147.884

147.880

147.892 147.888

147.900 147.896

147.908 147.904

147.916 147.912

CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.008. Lack of fit is significant.

05/28/09 16:40 CHART1: Process Capability Chart: MWM234SOD - DATA1

147.924 147.920

147.928

Calculation of Cp & Cpk

NAME OF THE LINER: MWM LAND OUTER DIAMETER

X-bar & R chart

Upper Control Limit: Lower Control Limit: Average : Sigma : Center :

AVERAGES ---------------143.95145634 143.93688366 143.94417000 0.00485756 143.94417000

RANGE -----------------0.02282050 0.00000000 0.01000000

Control Limits based on all plotted groups Control Limits shown for complete groups of size 4. There was 1 observation out of control.

Process Capability chart

30.0 28.00 24.0 22.40

16.80

12.0

11.20

6.0

0.0 143.928

5.6 0

143.931

143.937 143.934

05/28/09 16:20 CHART1: Process Capability Chart: MWM234LOD - DATA1

143.943 143.940

143.949

143.946 143.952 CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.0 04. Lack of fit is significant.

143.955

143.961 143.958

0.0 0 143.964

PERCENT

CELL FREQUENCY

18.0

Calculation of Cp & Cpk

NAME OF THE LINER: MWM BODY OUTER DIAMETER

X-bar & R chart

Upper Control Limit: Lower Control Limit: Center : Sigma : Average :

AVERAGES ---------------147.00000000 146.80000000 146.90000000 0.06666667 146.90660000

RANGE ---------------0.31319563 0.00000000 0.13724311

Control Limits shown for complete groups of size 4. There were no observations out of control

Process Capability chart

50.0

40.0

32.00

25.60

20.0

19.20

12.80 10.0 6.40

0.0

0.00 146.815

146.800

146.845 146.830

146.875 146.860

146.905 146.890

146.935 146.920

CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.000. Lack of fit is significant.

05/28/09 17:10 CHART1: Process Capability Chart: MWM234PREFINISHBOD - DATA1

146.965 146.950

146.980

PERCENT

CELL FREQUENCY

30.0

Calculation of Cp & Cpk

NAME OF THE LINER: MWM TAILEND OUTER DIAMETER

X-bar & R chart

Upper Control Limit: Lower Control Limit: Average : Sigma : Center :

AVERAGES ---------------133.92175391 133.86404609 133.89290000 0.01923594 133.89290000

RANGE -----------------0.09036918 0.00000000 0.03960000

Control Limits based on all plotted groups Control Limits shown for complete groups of size 4. There were 3 observations out of control

Process Capability chart

40.0

39.00

32.0

31.20

23.40 PERCENT

CELL FREQUENCY

24.0

16.0

15.60

8.0

7.80

0.0

0.00 133.805

133.790

133.835 133.820

133.865 133.850

133.895 133.880

133.925 133.910

CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.009. Lack of fit is significant.

05/28/09 17:28 CHART1: Proces s Capability Chart: MWM234PREFINISHTOD - DATA1

133.955 133.940

133.970

Calculation of Cp & Cpk

NAME OF THE LINER: HAFL LAND OUTER DIAMETER

X-bar & R chart

Upper Control Limit: Lower Control Limit: Average 0.02080000 Sigma : Center :

AVERAGES RANGE ------------------------------110.44215559 0.04746664 110.41184441 0.00000000 : 110.42700000 0.01010373 110.42700000

Control Limits based on all plotted groups Control Limits shown for complete groups of size 4. There were no observations out of control.

Process Capability chart

35.0

34.00

27.20

21.0

20.40

14.0

13.60

7.0

6.80

0.0 110.400

110.405

110.415 110.410

05/28/09 14:55 CHART1: Process Capability Chart: HAFL912LOD - DATA1

110.425 110.420

110.435

110.430 110.440 CELL BOUNDARY Fitted curve is a Normal. K-S test: 0.000. Lack of fit is significant.

110.445

110.455 110.450

0.00 110.460

PERCENT

CELL FREQUENCY

28.0

Calculation of Cp & Cpk

Lack of control is indicated by points falling

outside the control limit either on X-bar or R chart. It means that some assignable causes of variations are present; it not constant cause system. After analyzing the collected data from CNC and conventional machine we found that in many cases the value of Cpk was less than the minimum value. The various reasons for this may be  Input of materials  Machine maintenance  Tool wear

Remedies Implementation of first piece approach that is “Doing it

right the first time”. For every material change, tool change and shift change online inspection is to be done by skill supervisors.  Implementation of preventive maintenance and predictive maintenance. For preventive maintenance check list is done in every shift. Training is provided to all workers regarding feeding of input materials, online inspection etc. Strict inspection is done for checking the hardness, chemical composition etc.