CHAPTER 1

THE ROLES OF STATISTICS

IN ENGINEERING
TEAM MEMBERS
1. LÊ NGUYỄN HẢI NHƯ
2. TÔ TRIỀU VỸ
3. ĐÀO BẢO KHA
4. HÀ THÁI HOÀNG
5. PHẠM CHU CHÍ KHANG
6. NGUYỄN TRUNG TRỰC
 SUMMARY
1 The Engineering Method and Statistical Thinking

2 Collecting Engineering Data

3 Mechanistic and Empirical Models

4 Probability and Probability Models

 THE ENGINEERING METHOD AND
STATISTICAL THINKING
- Statistics is a science that helps us make
decisions and draw conclusions in the presence
of variability, it is the science of data, focusing
on collecting, presenting, analyzing, and using
data for decision-making, problem-solving, and
designing products and processes. Statistical
techniques play a vital role in designing new
products, improving existing designs, and
enhancing production processes.
 THE ENGINEERING METHOD AND
STATISTICAL THINKING
Probability and Statistics in Engineering:
Probability and statistics are powerful tools in engineering
problem-solving.
Widely used statistical technique: Regression analysis
(Chapters 11 and 12).
Probability models (Chapters 2 through 5) provide precise
estimates for various engineering problems.
Concepts contribute extensively to solving problems in the
transportation and healthcare domains.
 THE ENGINEERING METHOD AND
STATISTICAL THINKING
Transportation Field and Trip-Generation Model:
Civil engineers in transportation focus on regional highway system
capacity.
Key data includes nonwork, home-based trips, persons per household,
and vehicles per household.
Objective: Create a trip-generation model linking trips to persons and
vehicles per household.
Statistical technique: Regression analysis is used to construct the model.
Importance: Trip-generation model aids in transportation systems
planning.
 THE ENGINEERING METHOD AND
STATISTICAL THINKING
Hospital Emergency Department (ED):
Arrival process at ED is variable, influenced by the hour, day, and
longer-term cycles.
Service process varies based on patient needs, ED capacity, staffing,
and organization.
Capacity limitations lead to some patients experiencing long waiting
times.
Critical Questions:
+ Average waiting time for patients is crucial for healthcare providers.
+ Proportion of patients who leave without treatment (LWOT) is a
significant concern.
 THE ENGINEERING METHOD AND
STATISTICAL THINKING

Solution Approach: Employ probability

models (discussed in Chapters 2 through 5)
for precise estimates of waiting times and
LWOT occurrences
Definition of an Engineer:
+ Problem-solving for societal interest
+ Efficient application of scientific principles
 THE ENGINEERING METHOD AND
STATISTICAL THINKING
Dual approaches: Refining existing or designing new products/processes
Steps:
1. Develop a clear and concise description of the problem.
2. Identify, at least tentatively, the important factors that affect this problem or that may play a
role in its solution.
3. Propose a model for the problem, using scientific or engineering knowledge of the
phenomenon being studied. State any limitations or assumptions of the model.
4. Conduct appropriate experiments and collect data to test or validate the tentative model or
conclusions made in steps 2 and 3.
5. Refine the model on the basis of the observed data.
6. Manipulate the model to assist in developing a solution to the problem.
7. Conduct an appropriate experiment to confirm that the proposed solution to the problem is
both effective and efficient.
8. Draw conclusions or make recommendations based on the problem solution.
 THE ENGINEERING METHOD AND
STATISTICAL THINKING
Variability:
Variability in Observations:
+ Successive observations of a system or phenomenon do not produce identical
results.
+ Statistical methods help describe and understand this variability.
Everyday Application:
+ Varied occurrences of variability are encountered in everyday life.
+ Statistical thinking provides a useful approach to incorporate variability into
decision-making.
Statistics as a Framework:
+ Describing variability in systems.
+ Learning about the significance of potential sources of variability.
+ Understanding the impact of factors on gasoline mileage performance.
 THE ENGINEERING METHOD AND
STATISTICAL THINKING
Example of variability in dealing with engineering problems.
An engineer is designing a nylon connector for an automotive engine, considering a
3/32-inch wall thickness. Uncertain about its impact on pull-off force, eight
prototypes were tested, yielding varied pull-off forces (12.6 to 13.6 pounds). The
concern is low pull-off force, risking connector failure during engine installation.
Addressing this variability is vital for determining reliable design specifications.
Issues: Because the pull-off forces measurements varies, the pull-off force is
considered as a random variable.
Pull-off Force Variability:
+ Treated as a random variable (X).
+ Model: X = μ + ϵ, where μ is a constant, and ϵ is a random disturbance.
+ Constant μ remains the same in every measurement.
 THE ENGINEERING METHOD AND
Source of Variability (ϵ):
STATISTICAL THINKING
Small changes in the environment.
Variance in test equipment.
Differences in individual parts.
Ideal Scenario vs. Real World:
In an ideal scenario, ϵ would always be zero, making X consistently equal to μ.
In the real world, disturbances are unavoidable, leading to variability in X.
Objective:
Describe, quantify, and reduce variability in pull-off force measurements.
 COLLECTING
ENGINEERING DATA
Effective data-collection procedure can greatly simplify the analysis, leading to
improved understanding of the population or process that is being studied.

There are 3 basic methods:

Retrospective study: using historical data

Observational study: observes the process or population

Designed experiment: the engineer makes deliberate in the controllable

variables of the system, observes the resulting system output data
 COLLECTING ENGINEERING DATA
Retrospective study: using historical data

A retrospective study would use either all or a sample of the historical

process data archived over some period of time. The study objective might
be to discover the relationships among the two temperatures and the reflux
rate on the acetone concentration in the output product stream.

A retrospective study may involve a significant amount of data, but those

data may contain relatively little useful information about the problem.
Furthermore, some of the relevant data may be missing, there may be
transcription or recording errors resulting in outliers (or unusual values), or
data on other important factors may not have been collected and archived.

Retrospective study may identify interesting phenomena, but obtaining

solid and reliable explanations proves challenging in such cases.
 COLLECTING ENGINEERING DATA
Observational study: observes the process or population

Observational studies involve engineers observing a process without

significant interference and recording relevant quantities of interest. These
studies, conducted over a short period, can include variables not routinely
measured. In the context of a distillation column, engineers might design a
form to record temperatures and reflux rates alongside acetone
concentration measurements, and even measure input feed stream
concentrations for a comprehensive analysis.

In general, an observational study can solve several problems and obtain

accurate and reliable data. However, certain problems cannot be solved
 COLLECTING ENGINEERING DATA
Designed experiment: the engineer makes deliberate in the controllable
variables of the system, observes the resulting system output data

Experimental testing is crucial in engineering and physical-chemical

sciences, even when underlying scientific theories exist. Statistical methods
play a vital role in planning, conducting, and analyzing engineering
experiments.

The advantages of factorial experiments include their ability to detect

interactions between factors and their extension to include multiple factors.
The section introduces the concept of fractional factorial experiments to
reduce the number of trials required when dealing with multiple factors.
These experimental designs play a crucial role in industrial research and
development for designing, developing, and improving technology,
products, and processes.
 COLLECTING ENGINEERING DATA

Observing Processes Over Time

Whenever data is collected over time it is important to plot the data over
time. Phenomena that might affect the system or process often become
more visible in a time-oriented plot and the concept of stability can be
better judged.
 MECHANISTIC AND EMPIRICAL MODELS
A mechanistic model is built from our
underlying knowledge of the basic
physical mechanism that relates several
variables.

An empirical model is built from our

engineering and scientific
knowledge of the phenomenon, but
is not directly developed from our
theoretical or first-principles
understanding of the underlying
mechanism.
 PROBABILITY AND PROBABILITY MODELS
Introduction to Statistical Inference Quantifying Risks in Sample Decisions
and Sample Representation Focus on detecting defective
Statistical inference involves items in samples and quantifying
reasoning from a sample to the criterion for "detecting well".
conclusions about a larger Utilizing probability models to
population. quantify risks associated with
Importance of analyzing how decisions based on samples.

well a sample represents a

population for making sound
decisions.
PROBABILITY AND PROBABILITY MODELS

PLOT OF THE PREDICTED VALUES OF PULL STRENGTH

FROM THE EMPIRICAL MODEL.
 PROBABILITY AND PROBABILITY MODELS
Probability Models in Defective Probability Models for Continuous
Wafer Detection Measurements
Using probability models to Transition to continuous
assess the probability of measurements and process
detecting defective wafers in changes.
production lots. Discussing the application of
Analyzing sample sizes and their probability models, especially the
central limit theorem and normal
relation to effectively detecting
distribution, in quantifying the
defects, emphasizing practicality
probability of detecting process
over sampling infinite series of
changes based on continuous
lots.
measurements.
PROBABILITY AND PROBABILITY MODELS
 THE END

THANK YOU FOR

READING
MAS291 - THANGLD31 - NET1810