You are on page 1of 11

# TUM School of Management

## Production and Supply Chain Management

Prof Martin Grunow

## Statistics for Quality

Why statistics in quality?
The science of statistics provides:
Means for describing populations with variability.
Methods for estimating population quality from samples.

## 2012 from "A First Course in Quality Engineering: Integrating Statistical

and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

11

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Major statistical methods used in quality

engineering

Control charts:
used for controlling processes so they produce products of
uniform quality

Sampling plans:
used for determining acceptability of lots based on samples
taken from them

Designed experiments:
used for determining the best combination of process
parameter levels to obtain desired levels of quality
characteristics

## 2012 from "A First Course in Quality Engineering: Integrating Statistical

and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

12

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Major statistical methods used in quality

engineering (contd.)

## Regression and Correlation Analysis:

used for determining which cause variables affect, to what
extent, which quality characteristics

Reliability Engineering:
used in understanding the factors that affect the life of parts
and assemblies in order to increase their life

Tolerancing:
used for determining allowable variability in product and
process variables so the products can be produced
economically while meeting customer needs.

## 2012 from "A First Course in Quality Engineering: Integrating Statistical

and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

13

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Numerical methods for describing

populations
Population: collection of all items that are of interest in a given
situation
Sample: subset chosen from the population.
Random Sample: each item has equal chance of being included
Population

Sample

Subset

## 2012 from "A First Course in Quality Engineering: Integrating Statistical

and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

14

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Numerical methods for describing

populations
Two types of data:
Measurement data: measurements of characteristics (length,
width, strength)
"Quantitative" data
Attribute data: inspection data (small, color, taste, fit)
"Qualitative" data
Nominal
Ordinal
2012 from "A First Course in Quality Engineering: Integrating Statistical
and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

15

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Numerical methods for describing

populations
The mean () represents the center point
around which the data (population) is
distributed

Population

## The standard deviation () represents

the amount of variability (dispersion) in
the data about the center point
The square of the standard deviation is
the variance (2)

## 2012 from "A First Course in Quality Engineering: Integrating Statistical

and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

16

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Numerical methods for describing

populations
When a sample is taken, where n is the sample size, then the sample
average and sample standard deviation can be calculated:

Sample

n
S

2

S=

(Xi X )
n 1

## 2012 from "A First Course in Quality Engineering: Integrating Statistical

and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

X-bar
17

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Technische Universitt Mnchen

Population Parameters
vs. Sample Statistics
Population

## X-bar and S are sample statistics

Sample
X-bar
S
2012 from "A First Course in Quality Engineering: Integrating Statistical
and Management Methods of Quality" by K.S. Krishnamoorthi. Reproduced
by permission of Taylor and Francis Group, LLC, a division of Informa plc.
Holly Ott

18

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Who is the better marksman?

To reduce defects we need to reduce variability and target the mean
It is typically easier to change the mean than it is to reduce the
variation
Holly Ott

19

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Quality vs. Variability

Excessive variability in a population causes poor quality and
waste.

Holly Ott

20

## TUM School of Management

Production and Supply Chain Management
Prof Martin Grunow

## Technische Universitt Mnchen

Coming Up
Lecture 2.2: Random Variables and Probability
Distributions

Holly Ott

21