SPC过程能力培训教材(English)

Process Capability (Cp / Cpk / Pp / Ppk)
Global Training Material
Creator Function Approver Document ID Version / Status Location : Global Mechanics Process Manager : Mechanics : Jukka Mehtonen / Global Process Team : DMT00018-EN : V.2.0 / Approved : Notes : \\…\ NMP \ DOCMANR4 \ PCP \ PC Process Library DocMan
Change History : Issue Date
1.0 2.0 21st Dec’01 17th Jun’05
Handled By
Comments
Jim Christy & S?ren Lundsfryd Approved for Global Use Yrjo Saastamoinen Approved for Global Use
NOTE – All comments and improvements should be addressed to the creator of this document.
1 ? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Contents
Section 1 2 3 4 5 6 7 Heading / Description Page Variation, Tolerances and Dimensional Control 4 Population, Sample and Normal Distribution 15 Cp and Cpk Concept 28 Confidence Interval 37 Process Capability Analysis 47 Mini Cpk Cpk Data Collection Spreadsheet DMM00024 60
53
2
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Tool Verification Process
Measurement System Analysis
? Calibration ? Variable Gauge R&R ?DMY00019, DMT00017, DMM00210
Tool Approval – at Tool Maker / Supplier
? FAI (DMY00019, DMT00014, DMM00061) ? Precontrol: Set-up Rules
5 Min Variation Short Term
Machine Approval – process capability
? 3-10hr Run ? Process Capability: Cpk, Ppk ?DMY00019, DMT00018, DMM00024
10 Hr Variation Long Term
Process Performance
? Ongoing production ? Process Stability ? Process Capability: Cpk, Ppk
3 ? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
6 Mo Variation Performance
Company Confidential

Process Capability - Evaluating Manufacturing Variation
4
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Section 1 Variation, Tolerances and Dimensional Control
5
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Two Types of Product Characteristics
Variable: A characteristic measured in physical units, e.g. millimetres, volts, amps, decibel and seconds. ly s on le ariab v with deal e i ng w in s tra i In t h Attribute: A characteristic that by comparison to some standard is judged “good” or “bad”, e.g. free from scratches (visual quality).
ON
OFF
6
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

The Sources of Process/System Variation
Equipment Methods
Environment Process
Customer Satisfaction
Operators
Material
7
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Two Types of Processes
? All processes have: –Natural (random) variability => due to common causes
–Unnatural variability
=> due to special causes
? Stable Process: A process in which variation in outcomes arises “only" from common causes
USL
? Unstable Process: A process in which variation is a result of both common and special causes
USL
Defect
nominal value nominal value
LSL
LSL
8
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

The Two Causes of Variation
USL
Nominal value LSL
? Common Causes: –Causes that are implemented in the process due to the design of the process, and affect all outcomes of the process –Identifying these types of causes requires methods such as Design of Experiment (DOE), etc.
Defect
USL
nominal value
LSL
? Special Causes: –Causes that are not present in the process all the time and do not affect all outcomes, but arise because of specific circumstances –Special causes can be identified using Statistical Process Control (SPC)
Shewhart (1931)
9
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Tolerances
A tolerance is a allowed maximum variation of a dimension.
Rejected Part Rejected Product
Acceptable part
Rejected Part
LSL (lower specification limit) 10,7
Nominal 10,8±0,1
USL (upper specification limit) 10,9
10
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Measurement Report
For Measurement report (FAI) we normally measure 1-5 parts per cavity depending on result and the type of the dimension
GO TO General Information Section Dim. Number Detailed description Location Measuring Gauge I.D. Total Gauge R&R (% Tolerance) Operator (Inspector) Type of dimension Nom. Dim. Tolerance Upper specification limit (USL) Lower specification limit (LSL) Dim. 3 Dim. 10
GO TO General Information Section
a3 b2 18-Nov-04 18-Nov-04 CMM CMM 7% 3%
white
Dim. Number Detailed description Location Measuring Nom. Dim. Tolerance Upper specification limit (USL) Lower specification limit (LSL) Part 1 Mean check on 1 part % tolerance OK / NOT OK (+/-50%)
Dim. 3 Dim. 10 snap1 a3 d2 18-Nov-04 18-Nov-04 10
±0,10
69,4
±0,20
2 10
±0,10
black
white
10,10 9,90 10,050 50,0% OK Dim. 3
69,60 69,20 69,250 -75,0% Not OK Dim. 10 69,250 69,270 69,280 69,200 69,300 69,26 -70,0% OK
69,4
±0,20
Diamond Dimensions
9,90 Dim. 3 10,05 10,03 10,1 10,06 10,08 10,06 64,0% Not OK
69,20 Dim. 10 69,340 69,370 69,410 69,320 69,310 69,35 -25,0% OK
Basic Dimensions
10,10
69,60
Part Part Part Part Part Mean check on 5 parts Mean % tolerance OK / NOT OK (+/-50%)
1 2 3 4 5
Part Part Part Part Part Mean check on 5 parts Mean % tolerance OK / NOT OK (+/-80%)
1 2 3 4 5
Range check on 5 parts Max Min Range % tolerance interval OK / NOT OK (Max = 30%)
10,100 10,030 0,070 35,0% Not OK
69,410 69,310 0,100 25,0% OK
Range check on 5 parts Max Min Range % tolerance interval OK / NOT OK (Max = 30%)
69,300 69,200 0,100 25,0% OK
Action plan. Required for each NotOK dim!
Action plan. Required for each NotOK dim!
Datum System / References used Measurement Device used
Datum System / References used Measurement Device used
11
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Example of Capability Analysis Data
? For some critical dimensions we need to measure more than 1 part ? For capability data we usually measure 5 pcs 2 times/hour=100 pcs (but sampling plan needs to be made on the basis of production quantity, run duration and cycle time)
1 s t S u b g ro u p 2 n d S u b g ro u p 3 rd S u b g ro u p 4 th S u b g ro u p 1 1 8 .5 3 1 1 8 .5 2 1 1 8 .5 4 1 1 8 .5 6 1 1 8 .5 4 1 1 8 .5 4 1 1 8 .5 2 1 1 8 .5 5 1 1 8 .5 1 1 1 8 .5 1 1 1 8 .5 0 1 1 8 .5 5 1 1 8 .5 3 1 1 8 .5 1 1 1 8 .5 2 1 1 8 .5 5 1 1 8 .5 1 1 1 8 .5 4 1 1 8 .5 4 1 1 8 .5 5 5 th S u b g ro u p 6 th S u b g ro u p 7 th S u b g ro u p 8 th S u b g ro u p 1 1 8 .5 5 1 1 8 .5 4 1 1 8 .5 7 1 1 8 .6 0 1 1 8 .5 4 1 1 8 .5 6 1 1 8 .5 6 1 1 8 .5 7 1 1 8 .5 5 1 1 8 .5 5 1 1 8 .5 7 1 1 8 .5 5 1 1 8 .5 4 1 1 8 .5 4 1 1 8 .5 5 1 1 8 .5 6 1 1 8 .5 6 1 1 8 .5 3 1 1 8 .5 4 1 1 8 .5 5 9 th S u b g ro u p 1 0 th S u b g ro u p 1 1 th S u b g ro u p 1 2 th S u b g ro u p 1 1 8 .6 0 1 1 8 .6 1 1 1 8 .5 8 1 1 8 .6 0 1 1 8 .5 9 1 1 8 .6 0 1 1 8 .6 0 1 1 8 .6 3 1 1 8 .5 8 1 1 8 .6 1 1 1 8 .6 1 1 1 8 .6 3 1 1 8 .6 0 1 1 8 .5 9 1 1 8 .6 0 1 1 8 .6 1 1 1 8 .5 9 1 1 8 .5 9 1 1 8 .5 9 1 1 8 .6 4
12 ? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Process Capability - What is it?
? Process Capability is a measure of the inherent capability of a manufacturing process to be able to consistently produce components that meet the required design specifications ? Process Capability is designated by Cp and Cpk ? Process Performance is a measure of the performance of a process to be able to consistently produce components that meet the required design specifications. Process Performance includes special causes of variation not present in Process Capability ? Process Performance is designated Pp and Ppk
13
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Why Make Process Capability Studies
These parts are out of spec and could be approved if only one good part was measured LSL (lower specification limit) 10,7 This part is within spec. The tool would be approved if only this part was measured A process capability study would reveal that the tool should not be accepted
Nominal 10,8±0,1
USL (upper specification limit) 10,9
When a dimension needs to be kept properly within spec, we must study the process capability …. but still this is no guarantee for the actual performance of the process as it is only an initial capability study
Company Confidential
14
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen

The Nokia Process Verification Process
Basic Dimensions
E1 Tolerances Tolerances applied to applied to drawing drawing
E1.5
E2
E3
E4
E5
1-5 parts/cavity measured for measurement report (see DMY00019-EN)
White White diamonds to diamonds to be discussed be discussed with supplier with supplier
5 parts/cavity measured for measurement report (see DMY00019-EN) Capability study: Requirement: Cpk >1.67 and Ppk>1.33 by E3. Quantities to be agreed with supplier. Eg. 5 parts every 1/2 hour in 10 hours measured for each cavity = 100 parts. Can vary depending on tool capacity, e.g. stamped parts (see DMY00019-EN) Ongoing Process Control (SPC) Max: 105,85 Black diamonds to Black diamonds to be fixed by E3 be fixed by E3 (often a change of (often a change of a white diamond) a white diamond)
Company Confidential
White White diamonds to diamonds to be agreed be agreed
Proposal for black Proposal for black diamonds to be diamonds to be discussed with discussed with Supplier. Supplier.
15
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen

Section 2. Population, Sample and Normal Distribution
16
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

The Bell Shaped (Normal) Distribution
? Symmetrical shape with a peak in the middle of the range of the data. ? Indicates that the input variables (X's) to the process are randomly influenced.
17
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

Population versus Sample
Population ? An entire group of objects that have been made or will be made containing a characteristic of interest
lat ion
Sample ? The group of objects actually measured in a statistical study ? A sample is usually a subset of the population of interest
Po pu
“Population Parameters” μ = Population mean σ = Population standard deviation
18 ? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Sa m e pl
“Sample Statistics” x = Sample mean s = Sample standard deviation
Company Confidential

The Normal Distribution
19
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen
Company Confidential

What Measurements Can Be Used to Describe a Process or System ?
? μ (mü), a measure of central tendency, is the mean or average of all values in the population. When only a sample of the population is being described, mean is more properly denoted as (x-bar) : x
x 1 + x 2 + . . .+ x N x = N
x ? mean (average) or distribution
describes the location of the
Example: x1 = 5
x2 = 7
x3 = 4
x4 = 2
x5 = 6
x =
20
5+7+4+2+6 24 = = 4 .8 5 5
Company Confidential
? NOKIA 2005 T0001802.PPT/ 17-Jun-2005 / Yrjo Saastamoinen