Statistical Process Control

(SPC)
Kalevi Aaltonen, Aalto University
SPC
1. Introduction
2. Standard deviation, normal
distribution, six sigma
3. Control cards
4. Applications and experiences
5. Summary
SPC Normal distribution


Point of Inflection

1 p(d)

T USL
3
© Professori Kalevi Aaltonen, Aalto-yliopisto
3.11.2016
3
SPC
In statistics, the standard deviation (SD,
Greek letter sigma σ or s) is a measure
that is used to quantify the amount of
variation or dispersion of a set of data
values.
A low standard deviation indicates that
the data points tend to be close to the
mean (also called the expected value) of
the set, while a high standard deviation
indicates that the data points are spread
out over a wider range of values.
SPC
Standard deviation (σ) formula

Where xi is the i:th observed value

and x is the mean value of these
observations.
SPC
In the formula remainer is calculated
between every observation and the
mean value
The remainer is raised to the second power.

Summation of i…n values.

SPC
Summation is diveded by the number of
observation (n) and finaly extracted
square root. The final result is the
standard deviation.
Above-mentioned standard deviation formula fits for
whole population calculations.
When only a sample of data from a population is
available, the term standard deviation of the sample
or sample standard deviation can refer to either the
above-mentioned quantity as applied to those data or
to a modified quantity that is a better estimate of the
population standard deviation
Standard deviation and variance are
connected to each other. Variance is the
second power of stadard deviation.
Standard deviation is the square root of
the variance.
Normal distribution
Normal distribution – Six Sigma
 Normal distribution – Six Sigma
( Cpk and PPM; Motorola 1,5 sigma fluctuation included PPM values)

Sigma ? Cpk PPM

1 0,33 691 462
2 0,66 308 537
3 1,00 66 807
4 1,33 6 210
5 1,67 233
6 2,00 3,4
 What is SPC?

• Quality control and assurance tool to prevent

poor quality
• Quality errors are prevented by studing
production processes and by defining process
performance
• Claryfy and prevent production problems
• Stable processes
• Continuous process development and
improvment
 SPC principles

• Principles and theory of SPC was developed by W. A.

Shewhart years 1925-31 in Bell Telephone laboratory
USA.
• During the second world war great development leaps
• Official study programme in GB by Demingin ja
Shewhart.
• The actions fade out after the war.
• Japanese adopted the philosophy from Deming road
show in Japan year 1950. Nationwide quality
programme was established.
 SPC principles

• In 1970 japanese business success forced western

countries to addopt SPC philosophy.
• In Finland a renaissance of the method started in 20th
century.
• Automobile industry is the forerunner of SPC
• For examble Ford is not only utilising SPC but
requires the business partners to adopt the common
quality tools in their production
• Statistical: information based on collected
data
• In SPC statistical stands for collecting and
analysing data and evaluation of test results
by scientific means.
• Statistical methods offer the possibility to
analyse the risks on grounds of collected
information and to forecast the future trends
and results.
SPC key performance indicators (KPI):

• mean: sample measurement average

• range: biggest value – smallest value
• standard deviation: a measure that is used to
quantify the amount of variation or dispersion of a
set of data values
SPC
Process = set of actions

• Production process includes methods, labour force,

materials, environment and machinery to produce a specific
component.
• Process stads for a set of actions to produce and evaluate a
measurable product (shaft or bearing)
• Many processes are difficult to measure and evaluate (for
examble how to hire new staff)
SPC
Control

• The process is controlled by measured

statistical performance indicators
• Variation of nominal value is minimized
• Number of defective components is
supposed to approach zero (principle of
continuous improvment).
SPC – control card and diagram

• Control cards are the main tools of SPC

• Control cards are suitable tools to visualize
production processes; any change in
process and control card gives warning or
allert.
• There are two control cards: variable cards
and on/off –cards.
SPC– control card and diagram
• There are three parts in control card:
administrative part, control part and data
collecting part.
• Data collecting part consists information
conserning the process.
• Collected data is refined to key performance
indicators like mean and distribution; time
scale figures are plotted on control part and
individual data points are connected to form a
control diagram.
SPC - control card and diagram
SPC–

• Cotrol diagrams visualize the state of the

processes.
• Key indicators are the mean and deviation
• Before the control cards are in the daily use
the process must be stable and well known.
• Implementation of the control cards must be
well organized.
• Basic information and control limits are
labelled on the control card
SPC
• The aim is to keep the process performance
on a good level and to enable zero defect
production by controlling the state of the
process.
• Continuous objective is to develop and
improve the production process.
• Control target in SPC is the nominal
distribution of the process parametres.
• Both the movement of the mean and the
distribution changes are recorded.
SPC
• The aim of the SPC is to define the
performance and stability of the
production process and maintain the
controlled production
• SPC helps to study and analyse
production processes, it reveals the
changes in production and the root
causes behind the product and
production variation.
 SPC – is the process under control?
• The process is under control when the only factor inflecting the
production results is the normal distribution. No systematic
distraction impacts the process.

• The result of the process is steady. The only variation is the

normal distribution.

• Process control tries to observe systematically the trends in

production, to guide the mean value to the mid tolerance zone
and to minimize the variation

• If changes occur the root cause is found out by analyzing

parameter’s control chart alerts (overrun the control limits,
trends …).
SPC mottos
• Is the process stable – eliminate the disturbance
• Define the performance indicators
• Quality is not equal to inspection
• Keep the process simple
• Worker is reponsible for quality – not the inspector
• Disturbances are predictable
• Control with samples – not with 100% inspection
• SPC is not the core issue but the quality and good
results are
SPC – positive feedback
• Better understanding of the production process
• Control charts give facts instead of fiction
• Long term production trends turn visible with
control cards and charts
• Process parametres are indentified and
development actions and projects are well defined
and focused
• Introduction of systematic problem solving tools
SPC – negative feedback
• Updating control charts is tedious and hard;
difficulties to solve problems indicated by the
control charts
• Too many control charts in one workplace
• If the source of error is human behaviour it is very
hard to solve the problem
• Control charts do not solve the problems but the
improvement in management, better work
instructions, new working procedures, education
and training do
SPC – passport to success
• Simple measuring devices
• Measurement and quality control are important duties
• Training and informing are important tasks
• SPC’s benefits must be verified; only € counts
• SPC should be part of the total quality management
• Respect to the control charts
• The idea is not only to draw the control diagrams but to
solve the quality problems and challenges!
SPC – control charts
• Example: mean – range –control card
• Decide the sample (n=5)
• Fill in the identification information
• Decide the sample procedures
• Collect the data
• Evaluate the results using histogram
• Calculate the mean and the range
SPC – control charts
• Draw x- diagram
• Draw R- diagram
• Control limits for x-diagram
• Control limits for R-diagram
• Interpretation of the control chart
• Calculating the performance indicators
• Improvement of the process
SPC
SPC – identification information
SPC - histogram
SPC – process’s performance indicators
SPC – calculate the means
SPC – calculating x
SPC – drawing the mean values
SPC – calculating the range
SPC calculating R
SPC – drawing range values
SPC calculating the control limits
SPC
SPC drawing the control limits
SPC – calculating R- control limits
SPC
SPC – drawing R –control limits
SPC
SPC
SPC
SPC
SPC
SPC
SPC - histogram
SPC – repetitio mater studiorum est

x-R – control chart

Kalevi Aaltonen
Aalto University
SPC
Point beyond the control limit
SPC

Seven points in a row

SPC

Path with 7 points above the mean

SPC

Path with most points above mean value

SPC

Rare to have two points out of three

near the control limit
SPC

Too many points at the mean value

SPC

Level up or level down

SPC

Random results do not form a pattern

SPC
LESSONS TO LEARN:

 SPC is an exellent tool to improve production

performance
 Target is to control and improve the production
process
 SPC garantees high quality production processes 
stable processes enables high quality products and
exellent operation  good customer service and
customer satisfaction!
 SPC supports production machinery proactive
maintenance and operational repair and service
LESSONS TO LEARN:

 Facts instead of fiction and belief

 SPC is a good framework to train and learn quality
related issues and topics

Vita, si scias uti, longa est , Seneca

(Life, if well lived, is long enough)
 Acta est fabula.
Plaudite!
(Play is over. Applaud!)
(emperor Augustus)

© Professor Kalevi Aaltonen

3.11.2016 Aalto University 68