4. The Gasoline Mileage Performance of
a Car
Do you always get exactly the same mileage
performance on every tank of fuel?
Of course not - in fact, sometimes the mileage
performance varies considerably.
統計學的應用4
國立高雄師範大學軟體工程與管理學系
陳立偉老師
5. The Statistical Thinking - Sources of
Variability in a System (Factors)
This observed variability in gasoline mileage
depends on many factors, such as
the type of driving that has occurred most
recently (city vs. highway),
the changes in condition of the vehicle over time
(which could include factors such as tire
inflation, engine compression, or valve wear),
the brand and/or octane number of the gasoline
used, or
possibly even the weather conditions that have
been experienced recently.統計學的應用5
國立高雄師範大學軟體工程與管理學系
陳立偉老師
6. The Statistical Thinking - Sources of
Variability in a System (Factors)
These factors represent potential sources of
variability in the system.
Measurements also exhibit variability.
Statistics gives us a framework for describing the
variability and learning about which potential
sources of variability are most important or have
the greatest impact on the result of the problem.
統計學的應用6
國立高雄師範大學軟體工程與管理學系
陳立偉老師
7. Mechanistic and Empirical Models
The latter also uses our engineering and
scientific knowledge of the phenomenon (not
simple or well-understood), but it is directly
developed from data rather than our theoretical
or first-principles understanding of the
underlying mechanism.
統計學的應用7
國立高雄師範大學軟體工程與管理學系
陳立偉老師
8. Observing Processes over Time
The large variation displayed on the dot diagram
indicates a possible problem, but the chart does
not help explain the reason for the variation.
A possible downward shift in the process mean
level is visible in the time series plot, and an
estimate of the time of the shift can be obtained.
統計學的應用8
國立高雄師範大學軟體工程與管理學系
陳立偉老師
10. Observing Processes over Time -
Over-control or Tampering
1. He never moved the funnel. He simply dropped
one marble after another and recorded the
distance from the target.
2. He dropped the first marble and recorded its
location relative to the target. He then moved the
funnel an equal and opposite distance in an
attempt to compensate for the error. He
continued to make this type of adjustment after
each marble was dropped.
統計學的應用10
國立高雄師範大學軟體工程與管理學系
陳立偉老師
12. Observing Processes over Time -
Over-control or Tampering
Adjustments should be applied only to
compensate for a nonrandom shift in the process
- then they can help.
統計學的應用12
國立高雄師範大學軟體工程與管理學系
陳立偉老師
13. Levels of Measurement
Various scales of measurements exist:
Nominal Scale
Ordinal Scale
Interval Scale
Ratio Scale
13 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
17. Example of a Measure
0
20
40
60
80
100
0 2 4 6 8 10 12 14
Time (Hours)
Temp.
17 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
18. Example of a Metric
0
10
20
30
40
50
0 2 6 9 12 15 18
Months since release
Bugs.
18 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
19. Example of a Indicator
0
10
20
30
40
50
60
70
80
90
0 2 6 9 12 15 18
Time (Hours)
Temp.
Indicator of maximum safe temperature
19 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
21. Reliability and Validity of Measurements
Reliability – Refers to the consistency of a number of
measurements taken using the same measurement
method
Validity – Refers to whether the measurement or
metric really measures what we intend it to measure.
21 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
22. Reliability and Validity of Measurements
Reliable but not valid Valid but not reliable Reliable and Valid
22 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
32. What makes quality software?
There is not clear-cut answer
It depends on:
Stakeholders
Type of system
Type of users
…
Quality is a
multifaceted concept
Different ideas about a quality car
32 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
33. Different Quality Scenarios
Online banking system
Security
Correctness
Reliability
Air Traffic Control System
Robustness
Real Time Responses
Educational Game for Children
Userfriendliness
33 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師
34. The 3 Ps of Software Measurment
With regards to software, we can measure:
Product
Process
People
34 統計學的應用 國立高雄師範大學軟體工程與管理學系
陳立偉老師