Lecture 3
Statistical Process Control
Using Control Charts
Review
Review
Apakah yang anda ketahui tentang
peta kendali / Shewhart control chart?
Jelaskan pentingnya diagram pareto
Jelaskan pentingnya diagram pareto
Outline
Outline
Pengantar
Penyebab Variasi
Statistik Dasar untuk Control Charts
Statistik Dasar untuk Control Charts
Pemilihan Subkelompok rasional
Analisis Pola pada Control Chart
Pemeliharaan Control Chart
Introduction
Introduction
Proses kontrol statistik (Statistical
process control) adalah kumpulan alat
yang bila digunakan bersama-sama
dapat menghasilkan stabilitas proses
dapat menghasilkan stabilitas proses
dan pengurangan variabilitas
Control Chart: alat grafisuntuk
Introduction
Introduction
Manfaat menggunakan control chart
Ketika untuk mengambil tindakan korektif Jenis tindakan perbaikan yang diperlukan
Ketika meninggalkan proses sendirian
Ketika meninggalkan proses sendirian Kapabilitas proses
Kemungkinan sarana peningkatan kualitas Cara menetapkan spesifikasi produk
Cause of Variation
Penyebab variasi
Chance Cause
melekat proses Sesuatu (sebagai variasi alami
dalam proses ) dalam proses )
Assignable Cause
Sesuatu yang dapat diidentifikasi ditentukan Contoh: alat yang salah, kesalahan operator
Cause of Variation
Chance and Assignable Causes of Quality
Variation
Sebuah proses yang hanya disebabkan chance causes ,
proses tersebut dikatakan dalam pengendalian statistik. Alam variabilitas atau kebisingan latar belakang. Fluktuasi
15 October 2010 Materi ke-3
Alam variabilitas atau kebisingan latar belakang. Fluktuasi
Sebuah proses yang beroperasi di hadapan penyebab
dialihkan dikatakan di luar kendali. Misalnya Kesalahan operator, bahan baku yang rusak, pengaturan yang tidak tepat.
Tujuan akhirnya SPC adalah pengurangan atau
penghapusan variabilitas dalam proses identifikasi penyebab dialihkan.
A process that is operating with only chance causes of
Statistical Basis
for Control Charts
Basic Principles
Assumed to have approximately
normal distribution
normal distribution
Control limits : 99.74 % ( 3
σ
limits )
A control chart : on line process
control
Statistical Basis
for Control Charts
Selection of Control Limits
Let θ represent a quality characteristic of interest and represent an estimate of θ
( )
( )
( )
( )
( )
ˆ ˆ ˆ ˆ ˆ CL E UCL E k SD LCL E k SD θ θ θ θ θ = = + = − If k = 3 → 0.0026 of a sample statistic falling outsideStatistical Basis
for Control Charts
Errors in making inference from control chart
Type I : process is out of control when it is actually in
control
Type II : process is in control when it is really out of control
Effect of control limits on errors in making inference
Effect of control limits on errors in making inference Warning limit
Usually 2 standard deviation
Effect of sample size on control limits Influence in standard deviation
Statistical Basis
for Control Charts
Basic PrinciplesBasic Principles
A typical control chart has control limits set at values such that if the process is in control, nearly all points will lie between the upper nearly all points will lie between the upper
control limit (UCL) and the lower control limit (LCL).
Statistical Basis
for Control Charts
A control chart contains
A center line
An upper control limit A lower control limit
A point that plots within the control limits indicates the process A point that plots within the control limits indicates the process
is in control
No action is necessary
A point that plots outside the control limits is evidence that the
process is out of control
Investigation and corrective action are required to find and
eliminate assignable cause(s)
There is a close connection between control charts and hypothesis testing
Photolithography Example
Important quality characteristic in hard bake
is resist flow width
Process is monitored by average flow width
Sample of 5 wafers Sample of 5 wafers
Process mean is 1.5 microns
Process standard deviation is 0.15 microns
Note that all plotted points fall inside the
control limits
Selection of rational
Subgroups
The premise : chosen is such manner that the
variation within it is considered to due only to chance causes.
Basis : Time order
Two approaches ( Besterfield , 1990 ) Instance of time method
Period of time method
Subgroup Size ( the number of items in each
Selection of rational
Subgroups
Subgroups or samples should be selected so
that if assignable causes are present, the chance for differences between subgroups will be maximized, while the chance for
will be maximized, while the chance for
differences due to these assignable causes
Selection of rational
Subgroups
Selection of Rational Subgroups
Two general approaches to constructing rational subgroups.
Select consecutive units of production. Each Select consecutive units of production. Each
sample consists of units that were produced at the same time (or as closely together as possible)
Select a random sample over the entire
sampling interval.
Often used to make decisions about the
acceptance of all units of product that have been acceptance of all units of product that have been produced since the last sample.
Can be effective at detecting if the mean has
Analysis of Patterns
in Control Chart
Five Rules for identifying an out-of-control process 1. A single point outside the control limits
2. Two out of three consecutive points fall outside the 2σ warning limits on the same side
3. Four out of five consecutive points fall beyond the 1σ warning limits on the same side
4. Eight or more consecutive points fall to one side
5. A run of eight or more consecutive points –up, down, above or below the CL , or above or bellow the
Analysis of Patterns
in Control Chart
Nonrandom patterns can indicate out-of-control conditions
Patterns such as cycles, trends, are often of
considerable diagnostic value (more about this in Chapter 5)
Look for “runs” - this is a sequence of Look for “runs” - this is a sequence of
observations of the same type (all above the center line, or all below the center line)
Runs of say 8 observations or more could indicate an out-of-control situation.
Run up: a series of observations are increasing
An x chart with a nonrandom, up-run, down-run patterns
Analysis of Patterns
in Control Chart
Interpretation of Plots ( Non random pattern )
Determination of causes associated with out-of-control points
Require a thorough knowledge of the process and the sensitivity
of the output quality characteristic to the process parameters
The pattern and associated causes
Change in the level of the plotted pattern ( a jump ) ( change
Change in the level of the plotted pattern ( a jump ) ( change
quality raw material , change operator , failure component )
Trend in the plotted pattern ( tool wear , change in pressure ) Cyclic behavior in the plotted pattern ( seasonal effects of
quality , operator fatigue )
Concentration of points near the control limits ( two or more
operator plotted on the same chart , different production method )
Maintenance of Control Chart
Proper placement of the control
charts on the shop floor is important
→
easy to access
The control chart should draw the
attention and curiosity of everyone
involved
Quiz
Quiz
1. What are benefits of using control chart ? 2. Explain the different between chance
causes and assignable causes ? give example of each ?
example of each ?
3. How are rational subgroups selected ? explain the importance of this in the total quality systems approach?