See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267994697

Finite Element Learning Module for Improving Knowledge of Fatigue using

Commercial Software

Article

CITATIONS READS

3 413

5 authors, including:

Jiancheng Liu Ashland Brown

University of the Pacific University of the Pacific
43 PUBLICATIONS   291 CITATIONS    15 PUBLICATIONS   43 CITATIONS   

SEE PROFILE SEE PROFILE

Joseph J. Rencis
Tennessee Technological University
90 PUBLICATIONS   682 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Finite Element Learning Modules for Undergraduat Engineering Students View project

Teaching Fundamentals of Finite-Element Theory and Applications Using Sophomore-Level Strength-of-Materials Theory View project

All content following this page was uploaded by Joseph J. Rencis on 10 April 2015.

The user has requested enhancement of the downloaded file.

 Finite Element Learning Module for Improving
Knowledge of Fatigue using Commercial Software

Josh Coffman1, Jiancheng Liu2, Ashland O. Brown2, Sachin S. Terdalkar1,

Joseph J. Rencis1
1
University of Arkansas, Fayetteville; 2University of the Pacific

Abstract

Finite element (FE) active learning modules have been developed for various undergraduate
engineering courses. These FE learning modules are used to introduce basic and complex
engineering problems to enhance student learning of the theory and fundamentals of the finite
element method. A review of educational literature reveals that fatigue and finite elements are
not addressed together. The fatigue FE learning modules were designed based on the Kolb Cycle
of learning experience progression. The educational value of the fatigue FE learning module is
assessed by short quizzes administered before and after students use the module. The results of
the pre-quiz and post-quiz are used to identify any Felder-Soloman learning style and/or Myers-
Briggs personality type bias in the module. Statistical study of these assessment results will
allow the content and presentation of the module to be improved to better suit engineering
students. Post-survey will be used as part of the module assessment process to include students’
opinion.

Introduction

Fatigue is a material based phenomenon that causes failure in machine parts at stress values
much lowers than static yield strength of the material. Fatigue failure is due to repeated or cyclic
loading and unloading or fluctuating reversal in loading after a large number of cycles. Fatigue
failures are estimated to occur in 80-90% of all machine component failures. Fatigue is a major
topic that is addressed in undergraduate and graduate machine design courses and textbooks. A
machine design course is required in most undergraduate mechanical engineering programs. In
academia or industry fatigue problems have traditionally been solved by hand or an in-house
computer program specialized for a particular type of fatigue application.

The finite element method (FEM) is a computational tool that has been used extensively the
past thirty years in industry and is now a standard engineering tool for both analysis and design.
When FEM first appeared in the 1960’s it was introduced into the engineering curriculum at the
graduate level. As the method and computer technology matured, FEM was introduced at the
undergraduate level in engineering and engineering technology programs, even in some two-year
engineering technology programs. Today, FEM is primarily offered as an elective undergraduate
course in mechanical, civil, and aeronautical engineering programs.

Fatigue analysis that in the past was carried out by hand and/or in-house computer programs
is now done using commercial FEM software. Fatigue design modules have recently been
integrated into commercial FEM codes that include ABAQUS®, ALGOR®, ANSYS®,
COMSOL®, COSMOSWorks ®, and Pro/ENGINEER®. The usage of FEM in fatigue analysis
does have some limitations. An absence of actual loading data throughout components life limits
the accuracy of life prediction results. A second limitation is the random variance in material
performance even in materials of the same type.

Finite element (FE) learning modules have been developed for various undergraduate
engineering courses. Modules have been developed for the following topics: curved beam, bolt
and plate stiffness, lateral frequency of a cantilever beam, lateral vibration of a tapered cantilever
beam, steady state heat transfer in a bar, transient heat transfer in a l-bar, cylindrical drag,
friction flow in a pipe, probe feed patch antenna, specific absorption rate, transmission
parameters of an infinitely long co-axial cable, and human head.1,2 These FE learning modules
are used to introduce basic and complex engineering problems to enhance student learning of the
theory and fundamentals of the finite element method (FEM). Students are also introduced to
best practices in modeling and problem solving through the use of commercial FE software. In
the development of an earlier ANSYS® based fatigue FE learning module3, a review of
educational literature revealed that fatigue and finite elements are not addressed together. The
intended usage of this fatigue FE learning module is to integrate fatigue design theory into a
FEM course or fatigue FE in a machine design course. The fatigue FE learning module will
serve as an online resource for students and a tool for effectively presenting the lecture material
for instructors.

The FE learning module considered in this paper is the fatigue loading of a stepped shaft.
COSMOSWorks®4 was selected as the commercial FE software. The design of the fatigue FE
learning module is based on student learning experience progressions using the Kolb Cycle. The
different experiences found in the module will require students to think in ways not typically
found in a traditional classroom lecture. Student assessment data will be used to evaluate and
make improvements to the FE learning module. The students’ opinion of the FE learning
module will also be evaluated using a post survey upon completion of the module. The
educational value of the FE learning module will be monitored using pre- and post-quizzes.
Additional assessment tools will be used to identify any bias in the FE learning module towards
any Felder-Soloman learning style and/or Myers-Briggs personality type. Statistical study of
these assessment results will allow the content and presentation of the module to be continuously
changed to better suit engineering students.

Learning Experience Progression

History and Overview

Experiential learning has been valued as early as the teachings of Confucius or Aristotle. At the
start of the 20th century, John Dewey5 first identified experiential education as a fundamental
foundation in formal educational. During the decades following John Dewey, many
psychologists and educators began to believe that experiential education was valuable and could
be incorporated in addition to traditional instruction methods rather than replace them.5 Building
upon earlier works by John Dewey, Jean Piaget, William James, and Kurt Lewin, David A. Kolb
determined that learning is an experienced based process.6 From this work, Kolb6 determined
that “learning is the process whereby knowledge is created through the transformation of
experience.” The theory presents a cyclical model of learning that consists of four stages.
 In developing the fatigue FE learning modules the Kolb Cycle has been selected for its
ability to reach students of all learning styles. The importance of the Kolb Cycle as a guide for
engineering education is stated in a journal paper, “The use of that model (Kolb Cycle) in the
engineering teaching assists to three main objectives: to reach all the students through the
teaching to each learning style; to stimulate the students to use all the four learning types; and, to
teach the students to complete the cycle for themselves so that they think and learn in an
independent way.”7 Learning styles will be discussed later in this paper.

Kolb Cycle
The Kolb Cycle has been proven to be an excellent technique to improve student retention of
complex numerical methods used to analyze engineering problems.6-9 The Kolb Cycle describes
a cycle through which learning is achieved by various experiences. The Kolb Cycle, shown in
Figure 1, displays four distinct stages used in the development of knowledge within an individual
through the experiences found in a stage.

Figure 1. Kolb Cycle for learning experience progression.6-9

An individual will have strengths or preferences in both vertical and horizontal dimensions
shown in Figure 1. The way this newly presented information is perceived correlates to an
individual’s learning styles and personality type.6 The Kolb Cycle creates learning independent
of how the information is perceived. Rather, the Kolb Cycle accommodates for all. Depending
on the nature of the information, presentation method, learning styles, and personality types, new
information may be difficult or easy to understand for a given individual. Within the stages of
Concrete Experience and Abstract Hypothesis and Conceptualization learning takes place
through the presentation of new factual or new theoretical information. These two vertical
stages, as shown in Figure 1, are where an individual will “Take-In Information.” The vertical
dimension within the Kolb Cycle describes how an individual will perceive this new
information.6

In the stages of Active Experimentation and Reflective Observation knowledge is gained

through the activities found in these stages of the Kolb Cycle.6 The horizontal dimension of the
Kolb Cycle describes the way an individual tries to “Process Information” previously perceived
in the vertical dimension.6 The activities found in the stage Active Experimentation are used to
investigate the validity of new information by experimental methods. This stage may or may not
match with the learning styles and personality types of an individual. Once again the Kolb Cycle
contains a contingency. Reflective Observation uses much more passive and reflective activities,
as shown in Figure 1, to verify the newly perceived information. Using the Kolb Cycle as a
guide, classroom instruction may be developed to include all stages and encompass individuals
of all types.

The inner loop of the Kolb Cycle shown in Figure 1, describes a pattern of possible thoughts
that lead to a progression from one set of experiences to new experiences. Each of the following
four questions are seen as transitional phases: “Why?”, “What?”, “How?”, and “What If?”.7
These transitional questions will tend to arise, as a natural curiosity develops in the minds of a
student.

Application of Kolb Cycle to Fatigue FE Learning Module

In a paper written by Brown8, Teaching Finite Elements using the Kolb Learning Cycle, a global
analysis of a FE course is made in regard to stages of the Kolb Cycle that are experienced in that
course. Brown states that, “ Students are provided Abstract Hypothesis/Conceptual Modules
that begin with the background of the FE method, fundamental mathematics of FE, move
through the concept of “stiffness-analysis”, one-dimensional direct stiffness analysis of various
structures, the topology of the various finite elements, error analysis of FE results, and concludes
with how to model engineering problems using this technique.”8 The Abstract Hypothesis/
Conceptual stage in Figure 1 can have experiences encompassed in the following three areas: the
modeling, analysis, and theory. One or more of these experiences may be used to engage
students in the Abstract Hypothesis/Conceptual stage. Brown then goes on to say that
experiences found in homework assignments, course projects, and the FE learning modules apply
to the Active Experimentation portion of the cycle. Additional types of Active Experimentation
classroom activities are stated in Figure 1. These activities include laboratory experiments,
product teardowns, testing using engineering tools and methods, and performing simulations.
The fatigue FE learning module focuses mainly on the simulation activity, but these other
activities could certainly be used to connect new ideas and get students involved in the learning
cycle. The problems considered in the FE course are often related to a “real-world” problem and
are an example of a Concrete Experience.8 Activities within the Concrete Experience stage
shown in Figure 1 can be used to reinforce or provide a Concrete Experience. These activities
can include dissection, reverse engineering, and case studies. In the fatigue FE learning module
the activity experienced most like a case study. After the student performs fatigue FE learning
module, they are asked to compare the FE results with the analytical solution. Most importantly,
they are asked to attempt to explain the differences between the FE and analytical results. This
requires that they engage in Reflective Observation portion of Kolb’s Cycle. Activities, shown
in Figure 1, that are found to provide a Reflective Observation type experience include: having
open discussions, keeping a journal or notebook collection, and perturbation by a course
instructor. Individual activities require inner thought and reflection which require a student to
engage in a Reflective Observation of activities or experiences recently completed. Designing
around Kolb Cycle will reach more if not all students. Brown also describes a micro learning
cycle for his FE learning modules that engages all areas of the Kolb Cycle.8 It is in this same
manner that that the fatigue FE learning module has been developed.

The fatigue FE learning module has been designed and interlaced within the four stages of
the Kolb Cycle. Prior to the introduction of the module, the students will have partially covered
the fundamentals of machine design theory. A brief introduction to FE theory may also be
provided, but will be covered as well in the fatigue FE learning module. This prior knowledge
starts the Kolb Cycle for the FE learning module at the Abstract Hypothesis and
Conceptualization stage of the cycle. In this area some of the students may begin to develop
ideas as to “How?” the theory may be applied to “real world” problems. This develops a
progression towards applying theory as is done in the Active Experimentation stage of the Kolb
Cycle.

The fatigue FE learning module is largely a listing of a step-by-step user’s guide on how to
carry out a FE analysis of a fatigue based machine design problem. In the stage of Active
Experimentation the students will be asked to perform the required steps for the FE analysis.
Later they will be asked to perform manipulations that will include changing physical geometries
and/or loading conditions. This will lead the students to form opinions as to how these changes
will affect the results, as well as reinforce guiding principles. These changes may lead the
student to draw the conclusion “What If?” while making modifications. The problem selected
for the fatigue FE learning module is a circular stepped shaft subjected to fully reversed fatigue
loading. This problem presents a simple case study that is present in many everyday
applications, such as power transmission shafts in automobiles. The example problem selected is
from Shigley9 and provides the student with a Concrete Experience as well as a reference to
applicable fatigue theory.

Reflective Observation can be achieved by asking the students to compare the results from
the FE analysis to the analytic solution from fatigue theory and compare the results match. If the
FE solution results do not match the analytical solution, the students should be asked “Why?” the
solutions are different. The instructor may prompt students with diagnostic questions to reveal
errors in steps where mistakes are commonly made. Other possible ways to invoke Reflective
Observation include group discussions and report writing. These types of assignments require
the students to reanalyze what they have done and reflect “Why?” they have done these things in
the three previous stages. Finally to complete the cycle, students will take what they have
learned from the module and want to know “What?” other problems can be modeled and solved
with FE methods. The students now have used commercial tools and developed skills to analyze
more complex problems with further practice. It is in this manner they will be able to begin
providing solutions to new problems using self conceived ideas in new areas.
Fatigue FE Learning Module

Overview
This module was integrated into the senior level MECH 125 Machine Design II course at the
University of the Pacific by Prof. Jiancheng Liu in the spring semester of 2009. The fatigue FE
learning module is designed to be used as a classroom learning tool within an undergraduate
machine design course or FE course. Very little knowledge of FE theory is required to complete
the module. However, some introductory undergraduate machine design theory is required to
understand the terminology and principles applied in the creation of the FE model. The
background required before using the module are the fatigue equations for fully reversed loading.
The fatigue problem selected is simple, so that the students may connect the solution to the
pertinent machine design theory within the FE analysis. The fatigue FE learning module will be
available in two file formats, Microsoft® Office PowerPoint® and Adobe Acrobat®. These file
formats ensure ease of use and the ability to go back and review steps in the solution
development process. An instructor can also change the PowerPoint® slides to meet his/her
needs. As mentioned in a previous paper1, certain aspects of the module will be included to
create overall uniformity. These items include module title, author, author contact information,
expected module completion time, table of contents, and references. Educational objectives
based upon Bloom’s Taxonomy10 and ABET Criteria 3 for Engineering Programs11 are stated at
the beginning of the module. A detailed problem description and relevance is included along
with the analysis objectives. A large majority of the module content will be the step-by-step
process to create a FE model and carry out a FE analysis. Portions of this guide will be directed
at properly viewing the FE results. A comparison of FE results to the analytic solution is
included to emphasize the importance of solution verification. Finally, an overall summary and
discussion section is included to review what the user has accomplished and the techniques and
underlying FE theory involved.1

Example Problem
Choices of fatigue problems that are appropriate for both introductory undergraduate machine
design and FE courses are quite limited in nature. Example 7-10 from Chapter 7 of Shigley’s
Mechanical Engineering Design was used.9 The problem selected is a circular stepped shaft with
ball bearing supports at points A and D. At each diameter change a fillet with a radius of 3 mm
is present. The shaft is subjected to a fully reversed concentrated loading. The applied load is a
non-rotational force (F) with a magnitude of 6.9 kN as shown in Figure 2. The shaft is machined
from AISI 1050 cold drawn steel with a tensile yield, Sy, of 580 MPa. The ultimate tensile
strength, Sut, is 690 MPa. The shaft is to operate at room temperature. The reliability factor is
1.0 and the fatigue endurance limit, Se is 345 MPa. The problem requires that the shaft life be
estimated for loads (F) of 1.7 kN, 3.4 kN, and 6.8 kN. Additional material properties for AISI
1050 cold drawn steel not provided by Shigley9 are required for the three-dimensional FE
analysis and they include Young’s Modulus, E = 207 GPa, Poisson’s ratio, = 0.29, and shear
modulus G = 80 GPa.
 AISI 1050 Cold Drawn Steel:
Sy = 580 MPa; Sut = 690 MPa; Se = 345 MPa; E = 205 GPa; = 0.29; G = 80 GPa.

Figure 2. Stepped circular shaft (dimensions in mm.) subjected to a fully reversed loading.9

Finite Element Model

The commercial software COSMOSWorks®4 is used for this fatigue FE learning module.
COSMOSWorks® is widely used in industry and undergraduate engineering programs, and with
the SolidWorks® three-dimensional solid modeling software. Within COSMOSWorks® there are
several analyses that can be performed. This problem requires both static and fatigue analyses.
COSMOSWorks® uses the static analysis to formulate the fatigue analysis. Essentially the
loading is considered the same as the static analysis and an event is defined for the application of
the fully reversing cyclic load with the loading ratio of (R = -1) for the defined static load for a
specified amount of cycles. The failure analysis compares the applied alternating stresses against
a fatigue strength curve (S-N curve) for the given material on the interval of the applied cycles.

The stepped shaft was modeled in SolidWorks® as a three-dimensional solid. The solid
model is meshed with ten node quadratic tetrahedral elements by the high quality automatic
mesh generator in COSMOSWorks®. The geometry, material properties, and loading are shown
in Figure 2. The FE mesh consists of 12,873 nodes and 7,940 tetrahedral elements as stated in
Figure 3. Each node has three degrees of freedom (DOF) and the mesh has a total of 38,619
DOF. The ball bearing end supports are shown in Figure 3. All DOFs were constrained on the
cylindrical surfaces of the shaft that make contact with the bearings. These constraints resemble
fixed-fixed boundary conditions. The concentrated load was defined as a normal force over a 5
 t top surfaace of the shaft in as Figuure 3. This was done to eliminate sttress
mm radiuus circle on the
concentraations in thee vicinity of the
t concentrrated load.

A B C D
OSWorks® FE
Figurre 3. COSMO F mesh of the
t stepped circular
c shafft.

Static Defeflection Anaalysis

A static deflection
d an
nalysis of thee shaft with a 6.8 kN loaad was carrieed out using
COSMOSWorks®. The T maximuum vertical deflection
d occurs 298 mm m from the leeft end of the
beam witth a magnitu ude of 0.37066 mm as shoown in Figurre 4. The maaximum defllection is to the
left of thee applied loaad.

This result was verified with mechanics of o materials principles coonsidering a fixed-fixedd
uniform circular
c shafft of 38 mm, 35 mm, andd 32 mm in diameter.
d Thhe deflectionn value is -0.258
mm for a uniform 38 8 mm shaft, -0.359
- mm for
f a uniform m 35 mm shaaft, and -0.5113 mm for a 32
mm unifoorm shaft. The
T FE soluttion of 0.37006 mm for thhe stepped shhaft is boundded between
these values for the uniform
u shaffts. The defllection may seem small, but it is actuually too largge if
the shaft included gears. The recoommended maximum
m deeflection forr a shaft withh gears is 0.1127
mm.12

A
D

B C

Figurre 4. Resulttant deflectioon (mm) anaalysis for thee 6.8 kN loadd.

Static Strress Analysiss
A static stress
s OSMOSWorkks® as show
analyssis was carrieed out in CO wn in Figure 5. 5 The higheest
stress waas found at th
he bottom suurface of the right bearinng support (ppoint D) in thhe fillet radiuus.
The magnitude of thee von-Mises stress at thaat location iss 296 MPa ass shown in Figure
F 6. Thhis
value is approximate
a ely 56% of thhe tensile yieeld strength, Sy = 530 MPa on the firrst loading cyycle.

A D

B C

Figure 5. Static von-Mises

v sttress (Pa) anaalysis for thee 6.8 kN loaad.

Figure 6.
6 Highest von-Mises
v sttress (Pa) loccation at the bottom righht bearing suupport (pointt D).
 Bending stresses were verified at the right bearing support (point D) using the mechanics of
materials solution for a fixed-fixed beam. The stress concentration values at the fillet radii were
determined from Shigley.9 The bending stress at the fillet radius of the bearing support location
was 312.30 MPa using mechanics of materials. There is a 5.2% difference in the two solutions
types for the maximum static stress. Since the educational version of COSMOSWorks® was
used, there was a limitation in obtaining a more accurate FE solution at the fillet locations,
therefore, 5% is considered acceptable in this work.

Fatigue Analysis
COSMOSWorks®4 was used to estimate the number of life cycles the shaft would survive
subjected to reapplications of the 6.8 kN load as shown in the F-t curve of Figure 2. The shaft
should be designed to withstand 106 loading cycles; however, the corrected endurance limit is
236 MPa and the highest applied static stress is 296 MPa which means that the shaft will have a
finite number of life cycles.

In COSMOSWorks® a stress-life approach is used to carry out the fatigue analysis. Stress-
life methods are commonly found in undergraduate machine design courses and textbooks. As
previously discussed in the section on the finite element model, COSMOSWorks® uses the
results from the static stress analysis to compute an alternating von-Mises stress for the defined
fatigue event. This alternating von-Mises stress is compared to the material S-N curve. The
ASME austenitic fatigue S-N curve for AISI 1045 cold drawn steel is shown in Figure 7. This
material was selected since it most closely matches AISI 1050 in the COSMOSWorks® material
library. AISI 1050 is not available in the COSMOSWorks® material library.

Figure 7. Semi-log scale S-N plot of AISI 1045 cold drawn steel from
COSMOSWorks® material library.4
 The life plot in Figure 8 shows the lowest number of cycles until failure at all locations of the
shaft. The most probable location for failure is at the bottom right bearing support of the shaft
(point D) as shown in Figures 8 and 9. The minimum number of cycles for the shaft is 99,280
until failure. The life plots in Figures 8 and 9 show a range of 99,280 to 339,500 cycles at the
bearing support. This compares well with the analytic solution of 112,000 cycles stated in
Shigley.9 Therefore, COSMOSWorks® is more conservative than the analytic solution.

See Figure 9

A B C D

Figure 8. Life plot of shaft.

Figure 9. Enlarged view reveals fillet radii at the bottom

right bearing support is the most probable failure location.
 It is important to compare these results with applicable fatigue theory found in the textbook.
This verification provides a secondary check to the FE analysis. Table 1 displays the life cycle
predictions through the analytical and FE solution. It can be observed that the values for the 6.8
kN load are within a reasonable range of values. As one can see from the Table 1, the loading
cases of 1.7 kN and 3.4 kN have an infinite life. The discussion of these loadings was not
covered in this paper; however, it discussed in the module as a modification to the FE model.
The solution from COSMOSWorks®3 by its nature is slightly on the conservative side. If the
results are not within a similar range with the analytical solution it is quite possible that an error
has been made. Stepping through the portions of the analysis and checking the results will allow
the student to develop skills on how to identify potential errors in future FE analyses.

Table 1. Comparison of solution methods for the fatigue analysis.

Loading Case Solution Type
9
(F) Analytic COSMOSWorks®
1.7 kN Infinite Life Infinite Life
3.4 kN Infinite Life Infinite Life
6.8 kN 112,000 Cycles 99,280 Cycles

Learning Styles: Felder-Soloman

History and Overview

Learning styles have only been used as an important learning tool in formal education since the
start of the 20th century. The Felder-Soloman learning style model is based on initial
psychological theory of Carl Jung13, the learning style work of David Kolb6, and the Myers-
Briggs Personality Types Indicator. In some cases, learning styles and personality types are
discussed in unison. The Myers-Briggs personality types will be discussed in-depth later in the
next sections of this paper. A large number of learning style models have been established for
various fields. A few to be mentioned are models developed by Anthony F. Gregorc14, David
Kolb6, and the Herrmann Brain Dominance.15 These learning styles may have applications in
certain educational programs; however, the work of Richard M. Felder and his associates have
focused almost entirely on engineering students. This is the reason why this learning style model
is used in this work to aide in the development and improvement of the fatigue FE learning
module.

Felder-Soloman Model
Richard M. Felder Linda K. Silverman addressed a mismatch of learning styles reached by
traditional classroom techniques and engineering student learning styles.16 This paper was based
on the prior psychological theory by Carl Jung13 and included additional learning style
information written by Kolb6, discussed earlier, for his work in the development of the
experiential learning cycle. Felder and Silverman proposed that identifying common learning
styles in engineering students would allow for the creation of new styles for presenting lecture
material that would more effectively educate students of all learning styles. Felder continued
this work and with the help of Barbara Soloman created the Felder-Soloman Index of Learning
Styles.16-18 The Felder-Soloman Index of Learning Styles is shown in Table 2 and is used to
identify the fixed learning styles present in an individual.
 Table 2. Felder-Soloman Index of Learning Styles.16-18

The Felder-Soloman Index of Learning Styles is composed of four pairs Active/Reflective,

Sensing/Intuitive, Visual/Verbal, and Sequential/Global as shown in Table 2. Felder notes
that engineering students are typically, “…Visual, Sensing, Inductive (now omitted), and Active,
and some of the most creative students are Global.”16 Felder identifies a discrepancy of
engineering student learning styles and traditional instructional methods. Felder states that
traditional instruction methods appeal to the following learning styles: “most engineering
education is auditory (Verbal), abstract (Intuitive), Deductive (now omitted), passive
(Reflective), and Sequential.” 16 In 2002 a republication of the original learning styles paper by
Richard M. Felder removed the Inductive/Deductive categories. These categories were removed
since a sampling of Felder’s students indicated that most students actually preferred the
Deductive instruction type, contrary to his personal belief that Induction methods should be used
in education until graduate school.16

Since Felder has focused specifically on engineering students, the Felder-Soloman model is
used to develop and design the FE learning module. The initial goal of our fatigue FE learning
module is to focus on designing the module to include the four typical engineering learning
styles stated above. The FE learning module will accommodate Active learners since
involvement or participation is required to complete the module during lecture/lab time periods.
Students with a preference for Sensing, “prefer concrete information such as descriptions of
physical phenomena, results from real and simulated experiments, demonstrations, and problem-
solving algorithms”.19 The concrete nature of the example problem selected for analysis will
appeal to students of the Sensing type. By knowing most engineering students have a Visual
learning preference, we created a large amount of Visual instruction through computer screen
captures of step-by-step instructions that are used to complete the FE learning module. Visual
learners will also be captivated by the presentation of FE results that include deflection, stress,
and life plots from the commercial software. Also, Visual learners will be taught how to model
the problem in SolidWorks®, which is a visually stimulating and intensive process. Furthermore,
the fatigue FE learning module is by its nature very sequential. Each step is clearly covered and
builds towards the final goal of an accurate simulation of the problem, which will make it easier
for the Sequential learner to grasp the content. Global learners may find it very easy to go
through the module once the overall problem has been solved. Global learners may be able to
avoid the step-by-step instruction methodology and can move faster through the module than
their Sequential counterparts if the overall process is quickly learned.

Index of Learning Styles On-line Assessment

The Index of Learning Styles (ILS) is available online from Richard Felder’s website at North
Carolina State University.18 The ILS provides instant results after completion of the 44 item
questionnaire. This questionnaire measures the four classifications of the Felder-Soloman Model
shown in Table 2. Each learning style classification has 11 questions. The responses of the 11
questions for each classification are used to compute the magnitude of a preference for a
particular learning style. The magnitudes of each learning style preference will be presented as
part of the assessment process for the FE learning module. The Felder-Soloman ILS may be
found at the website http://www.engr.ncsu.edu/learningstyles/ilsweb.html.18 The results of the
learning style assessments are discussed later in this paper.

Personality Types: Myers-Briggs

History and Overview

Based heavily on the psychological types of Carl Jung13, I.B. Myers and K.C. Briggs developed
their personality type paper for twenty years before releasing it in 1962. The Myers-Briggs Type
Indicator (MBTI) assessment is a psychometric questionnaire designed to measure psychological
preferences in how people perceive the world and make decisions in their life. The personality
type indicator assessment tool helped identify what kind of roles women, who were entering the
industrial workforce of war-time production jobs, would be best suited for during World War
II.20 In a related way the MBTI may be used to analyze the best instructional methods for a
range of personality types. Though the work of Carl Jung and the MBTI has no true scientific
basis, it has been one of the most popular and widely used methods to classify personality types
for the past half century.

Myers-Briggs Type Indicator

The MBTI shown in Table 3 includes four categories of personality type preferences:
Extroversion/Introversion, Sensing/Intuition, Thinking/Feeling, and Judgment/Perception.
The first pair of Extroversion vs. Introversion regards the way an individual interacts with their
environment. In the FE learning module Extroverts may find it easier to be involved and
participate if the module is completed as a group or class, whereas Introverts would prefer to
complete the module on an individual basis. The second of the four categories Sensing vs.
Intuition provides insight into how a person processes information. People who tend to process
and learn through their senses are referred to Sensors, versus people who process data based on
the view that the information is of future use are referred to as Intuitor. The Sensor vs. Intuitor
pair is seen by most researchers to be the most important of the four categories in terms of
education. A major goal of this project is to design, use, and improve the FE learning module in
ways that will be effective for students with different MBTI personality types. For example, the
module proceeds
p in a deductive manner.
m Firrst, the FE annd machine design
d theorry is presenteed,
and then module is completed. Intuitor
I typees prefer to contemplate
c theory and then
t quicklyy
implemennt the use off the theory ini an applicaation. The modules
m alsoo have contennt explicitly
addressedd to the Senssor types. Inn particular, the tremenddous visual aspects
a of thhe FE analysiis
19
results apppeal to the Sensors.
S A number of
o researcherrs have used the knowleddge of MBTII
types to enhance
e eng
gineering eduucation. Thee third pair, Thinking
T vss. Feeling, for
fo MBTI
preferencce attempts tot describe thhe manner inn which a peerson evaluaates informattion. Those who
tend to use a logical “cause
“ and effect”
e strateegy Thinkerrs versus thoose who use a hierarchy
based on values or on n the mannerr in which ann idea is com mmunicated Feelers. Thhe final MBT TI
type pairr indicates hoow a person makes decissions or com mes to concluusions. Judggers are peopple
that tend to back up their
t decisions based on evidence. Those
T who teend to wait to t be sure thaat all
data has been
b thoroug ghly consideered are known to be Perceivers.20

Personallity Types On n-line Assesssment

The assesssment of th he personalityy types will be completeed using the Jung Typoloogy Test™. A
72 item questionnaire
q e is completed to determmine the MBT TI types andd their relativve strengths..
The MBT TI on-line su
urvey providdes students with
w four lettters (either E = Extroveert or I =
Introverrt; either N= Intuitor or S = Sensor; either T = Thinker
T or F = Feeler; either P =
Perceiveer or J = Jud dger) that indicate their personality
p c
component t
types. In adddition, weigghts
or strength values forr each preferrence are proovided to thee students ass well. Fromm these strenggths
the persoonality type of
o a student may be anallyzed and used further inn the assessm ment processs to
identify any
a biases to owards any one
o personallity type. Thhe location of o the Jung Typology
T Test™
may be found
f at the website
w http://www.hum manmetrics.ccom/cgi-win//JTypes2.aspp.21 The resuults
of the perrsonality typ
pe assessmennts are discuussed in the next
n section of this paperr.

19,20
T
Table 3. Myeers-Briggs Tyype Indicatoor (MBTI) caategories of personality types.
t
Assessment Tools and Results

Overview
An assessment program is carried out for the fatigue FE learning module. The results from these
assessment tools are used for continuous improvement of the module. The four assessment tools
used are as follows:

• Post-survey. The post-survey is administered following the completion of the fatigue FE

learning module. The post-survey can be used to indicate what the students liked and
disliked about the module. The post-survey will also ask the students how much they
learned using the module in comparison to a traditional classroom approach.
• Pre- and Post-quizzes. A short quiz is administered before and after the implementation
of the fatigue FE learning module.
• Learning Styles. The Felder-Soloman learning styles of each student are identified
through an on-line questionnaire to determine whether the fatigue FE learning module is
biased towards a particular learning style based on the pre- and post-quiz results.
• Personality Types. The Myers-Briggs personality types of each student are identified
through an on-line questionnaire to determine whether the fatigue FE learning module is
biased towards a particular personality type based on the pre- and post-quiz results.

Each assessment tool above will now be discussed in-depth.

Post-survey
One assessment tool used to assess the fatigue FE learning module was the post-survey,
administered after using the module. The post-survey questions and format were developed
to follow a common template for all FE learning modules.1 This ensures present and future
FE learning modules are evaluated in a common manner to analyze the educational and
analysis objectives.1 The post-survey questions were based on the module educational
objectives and analysis objectives. The post-survey responses used a five point Likert scale.
The Likert scale used has the following five point scale: “Disagree”, “Partly Disagree”,
“Neither Agree or Disagree”, “Partly Agree”, and “Agree”. The post-survey for the fatigue
FE learning module is shown in Figure 7. Multiple questions for each educational objective
and each analytical objective were asked.

The post-survey results shown in Figure 7 were overall very positive. The results show
that over 78% of the student responses were “Partly Agree” and “Agree” and including the
“Neither Agree or Disagree” the positive response rate increased to 97%. We will now
discuss the questions with shaded rows in Figure 7. A total of eleven students responded to
the post-survey. Analyzing the 1st question of the post-survey, nine of the eleven students
found that the module helped them to better understand “fully reversed fatigue loading.”
The 4th question reveals that ten of the eleven students felt that the module improved their
understanding of static and fatigue FE analysis, as well as increased their confidence about
carrying out machine component analyses. The responses of these two questions indicate
that the students feel more confident in understanding both fatigue and FE analysis. In the 5th
question all eleven students selected either “Partly Agree” or “Agree” with a simple
conceptual question about the fatigue FE solution. The only fully negative feedback
regarding the module was in the 10th question, a student felt that the module was not helpful
in learning how to select a suitable finite element type. In the 17th question, seven of the
eleven students thought the self-learning in the module was more beneficial than an
instructor led classroom demonstration. Additionally, seven out of eleven students found the
module to be very clear in its purpose and intentions as according to the 18th question. The
19th question is of particular importance because it indicates whether students enjoyed the
module and found it to be a more effective method than traditional instruction. Only two
students were found to “Partly Disagree” that was not an effective method for presenting FE
and fatigue when compared to the traditional approach. The 20th question indicates that eight
of the eleven students would like to learn more about the FE method and how to apply it to
other mechanical engineering problems. The post-survey confirmed the perception by the
students that this module helped them understand the concept of fatigue and assisted them in
understanding FE theory.
 Figure 7. Post-survey results for the fatigue FE learning module administered
at the University of the Pacific in Spring 2009.

This survey will be used to evaluate and improve active learning activities in this class. Your
student ID is used only to match up the results of this survey with others used in the course.
Your opinions will be used to improve course learning activities. We will not correlate your
survey response with your name or the assessment of any individual. Thank you in advance for
your cooperation in our research efforts to improve learning here at the University of the Pacific
under this NSF Grant. Prof. Jiancheng Liu

Student ID: _____________________

Please put an X in the box below that corresponds to your answer.

Neither
Partly Agree Partly
# Question Disagree
Disagree nor Agree
Agree
Disagree
This activity helped me understand “fully
1. reversed fatigue loading” in a conceptual 2 9
manner?
This activity helped me to understand the
2. 1 2 7 1
assumptions of “fatigue theory?”
This activity helped me understand the
3. limitations of “finite elements and usage for 4 7
fatigue theory?”
This activity helped me understand the topic
of “static and fatigue finite element analysis,”
4. so that I have the ability to carry out finite 1 8 2
element analysis of other machine
components?”
This activity showed me that the finite
element method determines an approximate
5. 9 2
solution for the “life cycles of a rotating shaft
fatigue” problem?
Activities like this one, and similar ones done
by commercial finite element software
6. 2 7 2
vendors, are only required to understand
finite element theory?
This activity showed me that an
7. understanding of “fatigue” theory can be 1 8 2
reinforced with finite elements?
This activity helped me create the correct
8. geometry to model a “three-dimensional 2 5 4
stepped shaft?”
This activity helped me identify the material
9. properties required to model the “static and 2 5 4
fatigue finite element analysis?”
This activity helped me to select suitable
finite element type to model “the static and
10. 1 6 4
fatigue analysis of the rotating stepped
shaft?”
 Figure 7. ‘Continued’ Post-survey results for the fatigue FE learning module administered
at the University of the Pacific in Spring 2009.

Student ID:____________________

Please put an X in the box below that corresponds to your answer.

Neither
Partly Agree Partly
# Question Disagree
Disagree nor Agree
Agree
Disagree
This activity helped me understand that
11. accuracy (not the correctness) of the solution 1 6 4
is dependent on the quality of the mesh?
This activity helped me to the correct
12. boundary conditions (loads and constraints) 1 8 2
to model the “rotating shaft”?
After completing this activity, I was able to
implement a suitable finite element type and
13. 2 8 1
construct a correct finite element model
using commercial software?
This activity helped me understand why it is
14. important to check if the “applied loads” are 1 6 4
specified correctly?
This activity helped me to understand why it
15. is important to check if the “constraints” 1 6 4
were specified correctly?
This activity helped me to understand why it
is important to verify a finite element
16. solution “i.e., deflections, stresses, and 3 6 2
loading cycles” through an independent
method, e.g., hand and/ or experiment?
Personally seeing and developing the finite
17. element model on my own was better than a 4 5 2
classroom demonstration?
18. This activity was very clear? 4 3 4

This activity was more effective than using

19. 2 1 7 1
class time for lectures or board-work?
I would like to learn more on using the finite
20. element method to solve other mechanical 4 5 3
engineering problems?

Totals 1 5 42 126 46

Percentage of Students Selecting Response 0.4% 2.3% 19.1% 57.3% 20.9%

Pre- and Post-quizzes
A pre-quiz and post-quiz shown in Figure 8 was administered to the students before and after
using the fatigue FE learning module. The quiz should take no more than fifteen minutes to
complete. Table 4 presents the results of the students’ scores on the pre- and post-quizzes.
The average scores for the pre- and post-quiz is approximately 61 percent as shown in Table
4. Table 5 summarizes the statistical analysis of Table 4. Analysis reveals that the statistics of
the data was not significant. This was due to the average of the pre-quiz and post-quiz being
equal. The pre-quiz and post-quiz scores indicate that there was no overall improvement in
student learning for the course. Furthermore, some students saw individual improvement while
other students did not. This could be attributed to the quiz administered. The quiz may not be a
good assessment tool since some of the students already understood the material better than
before using the module. The authors plan to develop a new quiz that has multiple choice and
true/false question to eliminate any subjectivity in grading by the instructor. Furthermore, the
quizzes did not count as part of the course grade, therefore, the instructor will be suggested to
count the post-quiz grade as part of the course grade. The module will be evaluated and
modified before it is introduced in a future course.

Table 4. Individual student performance on the pre- and post-quiz.

Student ID Pre-quiz Results Post-quiz Results
1 70% 60%
2 60% 40%
3 50% 40%
4 55% 50%
5 65% 60%
6 45% 55%
7 50% 90%
8 85% 70%
9 85% 85%
10 40% 50%
11 65% 70%
Average Scores 60.9% 60.9%
0% Improvement

Table 5. Statistical analysis of the pre- and post-quiz results.

Quiz Mean Standard Deviation Standard Error of the Mean
Pre-quiz 60.91% 14.97% 4.51%
Post-quiz 60.91% 16.55% 4.99%
95% Lower Bound
t-value = 0 p-value = 1.0
For Mean Difference = 5.71
 Figure 8. Pre- and post-quiz administered at the University of the Pacific
in Spring 2009.

MECH 125 Machine Design II

Spring 2009

Your Student ID:______________________ Your Name:___________________________

Your responses will not be used for assessing your grade in MECH 125.

1.) The fatigue may first occur at which cross section location?
a) A b) B c) C d) D e) The cross section where the load is applied.

Answer: Prior to FE analysis: 2) Point B.

After FE analysis: 1) and 4) Points A and D.

2.) With a decrease of the external load, the shaft’s life will increase. This statement is
1) True 2) False 3) Both have no relation.

Answer: 1) True

3.) What is the difference between a static analysis and a fatigue analysis?

Answer: Static analysis estimates the stress level and compares the stress level to its
yielding or ultimate strength. Fatigue analysis has to simultaneously take the stress level
and operation time into account. The analysis procedures are also different when using
FE analysis tool.

4.) The discrepancy between the analytical results and FE analysis results is large.
Explain why?

Answer: For both methods, it is hard to get a real accurate result since there are many
assumptions when conducting hand calculations or FE analysis using computer. But, it is
clear from FE analysis results the life decreases with the increase of the load level.
 One goal of this research is to create FE learning modules that span the spectrum of learning
styles and personality types. As previously noted, we have chosen to measure learning styles
using the Felder-Solomon model and measure personality preferences using the Myers-Briggs
Type Indicator (MBTI). In order to gain insight into the effectiveness of the modules across
different learning styles and personality types, the pre-quiz and post-quiz results will be
separated based on these demographic data. Statistical analysis of these correlations will allow
us to determine if the modules are more effective for certain demographic groups than others.
This data will be used to change the modules in a closed-loop feedback manner where the goal is
serving the learning needs of students with diverse learning styles and personality types.

Table 6 shows the average pre- and post-quiz scores for each learning style pair based on
Felder-Soloman. The learning styles in Table 6 denoted by capital letters are common for
engineering students.16 The learning styles for each student was determined using the Felder-
Soloman ILS.18 The third learning style pair in Table 6 has eleven VISUAL students (N =
11) and zero Verbal students (N = 0). Most engineering students are typically VISUAL
learners; this can be seen in Table 6. No students of the Verbal learning style are present in
this course.

Table 6. Felder-Soloman learning style pairs with pre- and post-quiz percentage results.
Learning Style Pre- Post- Standard Weighted Weighted Weighted
N Delta*
Pairs quiz quiz Deviation Pre-quiz Post-quiz Delta

ACTIVE** 7 56.43 61.43 5.00 18.93 56.49 60.21 3.72

Reflective 4 68.75 60.00 -8.75 4.79 77.08 65.00 -12.08

SENSING** 4 53.75 60.00 6.25 25.62 52.88 60.00 7.12

Intuitive 7 65.00 62.14 -2.86 8.09 76.72 71.90 -4.83

VISUAL** 11 60.91 60.91 0.00 16.43 60.11 59.89 -0.21

Verbal 0

SEQUENTIAL** 7 52.14 55.00 2.86 19.55 53.29 57.14 3.85

Global 4 76.25 71.25 -5.00 9.13 75.36 67.50 -7.86

*
Delta = (Post-quiz – Pre-quiz)
**
Common engineering student Felder-Soloman learning styles.16

We are interested in determining if the “Deltas” [(post-quiz score) – (pre-quiz score)] are
statistically different between the pairs of learning styles. In order to determine this, the data
is treated as a sample of a theoretical larger population. “Student-t” distributions are used for
the statistical analysis as the sample sizes are relatively small. Note that the last three
columns in Table 6 refer to “weighted” data. The on-line learning styles survey18 returns
results indicating learning style for the individual in each of the four learning style pairs and
also includes a weight or strength for that learning style. This allows one to differentiate, for
example, between someone who is only slightly ACTIVE over Reflective in their learning
style and someone who very strongly prefers an ACTIVE over Reflective learning
environment. The data in these last three columns were weighted (using a linear
interpolation) according to the weights reported from the learning style survey for each
student.

Standard statistical “t-student” analysis is used to determine the confidence intervals that are
used that determine the likelihood that the “Deltas” for different learning styles are actually
different (in a statistically meaningful manner). Table 7 shows the confidence intervals and the
VISUAL vs. Verbal pair is missing. This is because all of the students in this data set were
determined to be all VISUAL learners as shown in Table 6. So, for example, the unweighted
confidence interval of 88.9% for ACTIVE vs. Reflective learners indicates that there is an
88.9% likelihood that there is a real (statistically speaking) difference between the Deltas for
these two opposing learning styles. It is somewhat common to set the threshold of “statistical
significance” at a confidence interval of 95%. As can be seen from Table 7, if this standard is
used, there is no statistically significant differences between effectiveness of the fatigue FE
learning module for the different learning styles for either weighted or the unweighted cases.
This would indicate that the fatigue FE learning module has relatively equal effectiveness across
the different learning styles. This is a very positive result as one goal is to avoid significant bias
toward one learning style over another.

Although the confidence interval threshold of 95% is commonly used to indicate statistical
significance, it may be informative to consider any occurrences where the confidence interval is
greater than 50%. This would indicate that there was greater than 50% likelihood that one
learning style benefited more than another from the fatigue FE learning module. If this criterion
is used, noting from Table 8 that the ACTIVE learners had a higher positive Delta than the
Reflective learners and noting from the first row of Table 7 that the confidence intervals were
88.9% and 92.6%, respectively, for the unweighted and weighted values the implication is that
the module was more helpful for ACTIVE learners than for Reflective learners. This result is
not surprising as the FE learning modules are, by nature, a very active process where the students
participate in each step of building and analyzing the computational model. This being the case,
the statistical analysis provides us with an opportunity to refine the FE learning module process
in an “active feedback loop” manner. Perhaps the Reflective learners would be more effectively
engaged in the process if, along with the step-by-step FE learning modules, reflective oriented
questions were part of the process. This will be considered before the module is integrated the
next time in the course.

Table 7. Confidence interval percentage for differences between Felder-Solomon

learning style pairs.
Learning Style Pair Unweighted Confidence Weighted Confidence
Differences Interval Interval
*
ACTIVE vs. Reflective 88.9 92.6
*
SENSING vs. Intuitive 46.1 56.9
*
SEQUENTIAL vs. Global 60.8 78.6
*
Common engineering student Felder-Soloman learning styles.16
 In a manner very similar to what was done for the learning styles, Myers-Briggs Type
Indicator (MBTI) personality type data is correlated with pre- and post-quiz scores. The goal is
the same as with the learning styles data; to determine if certain student groups (in this case
certain personality types) benefit differently from the fatigue FE learning module. Table 8 has
the pre- and post-quiz average scores as well as the Deltas (difference between the pre- and post-
quiz score) and standard deviations all separated based on MBTI pairs. In the same manner as
was done for the learning styles, Table 8 includes weighted data as well as unweighted data. The
personality types in Table 8 denoted by capital letters are common for engineering students.20
The learning style for each student was determined using the on-line MBTI survey.21

Table 8. Myers-Briggs personality type pairs pre- and post-quiz percentage results.
Personality Pre- Post- Standard Weighted Weighted Weighted
N Delta*
Type Pairs quiz quiz Deviation Pre-quiz Post-quiz Delta

Extrovert 6 58.33 58.33 0.00 11.40 57.50 57.30 -0.20

INTROVERT** 5 64.00 64.00 0.00 22.64 69.38 65.00 -4.38

SENSOR** 6 61.67 57.50 -4.17 9.70 63.06 58.33 -4.73

Intuitor 5 60.00 65.00 5.00 22.36 53.74 59.17 5.43

THINKER** 6 59.17 65.83 6.67 17.68 62.01 63.32 1.31

Feeler 5 63.00 55.00 -8.00 11.51 62.79 50.45 -12.34

JUDGER** 8 65.00 59.38 -5.62 10.16 64.20 62.60 -1.59

Perceiver 3 50.00 65.00 15.00 22.91 50.00 68.85 18.85

*
Delta = (Post-quiz – Pre-quiz)
**
Common percentage of engineering students’ Myers-Briggs personality type.20

Standard statistical “t-student” analysis is again used to determine the confidence intervals
for the four relevant Myers-Briggs personality type pairs. Table 9 displays this data. Recall that
the confidence interval is the statistical likelihood that there is a difference between the Deltas
for the different personality type pairs. For example, as can be seen in the Table 9, the likelihood
(weighted) that the Extrovert students have a statistically significant Delta than do the
INTROVERT is 27.70%. As previously mentioned, the threshold for statistical significance is
set at a confidence interval of 95%. Using this criterion there is no statistical differences,
weighted or unweighted, between the different personality type pairs. This indicates that, at least
for this fatigue FE learning module, different personality type pairs do not have significantly
more or less benefit from the module. In other words, the fatigue FE learning module is not
biased toward one student group based on a personality type. This is a very desirable result!
 Table 9. Confidence interval percentages for differences between Myers-Briggs
personality type pairs.
Personality Type Pair Unweighted Confidence Weighted Confidence
Differences Interval Interval
Extrovert vs. INTROVERT* 0 27.70
SENSOR* vs. Intuitor 49.95 56.72
THINKER* vs. Feeler 86.34 83.78
JUDGER* vs. Perciever 72.86 72.56
*
Common percentage of engineering students’ Myers-Briggs personality type.20

Conclusion

The fatigue FE learning module did not show any improvement of student learning based on
no change in the pre-quiz and post-quiz scores. Past FE learning modules1,2 have shown
improvement of student learning. The fatigue FE learning module will be modified and the quiz
will be improved before the module is implemented again into the classroom. It has been
statistically shown that the fatigue FE learning module is not biased towards a particular learning
style or personality type. Ultimately, the goal is to refine the FE learning modules and overall
modeling experience in order to remove any bias toward specific student groups and to maximize
the effectiveness of all the FE learning modules developed in this project.

Acknowledgment

This work is partially supported by a National Science Foundation three year grant through DUE
CCLI Award Number 0536197.

Bibliography

1. Brown, A., Rencis, J.J., Jensen, D., Chen, C-C., Ibrahim, E., Labay, V. and Schimpf, P., “Finite Element
Learning Modules for Undergraduate Engineering Topics using Commercial Software,” Mechanical
Engineering Division, CD-ROM Proceedings of the 2008 American Society of Engineering Education (ASEE)
Annual Conference and Exposition, Pittsburg, PA, June 22-25, 2008.

2. Brown, A., Wood, K., Kaufman, K., Jensen, D., Rencis, J.J. and White, C., “A Novel Assessment Methodology
for Active Learning Modules to Equitably Enhance Engineering Education,” CD-ROM Proceedings of the 2009
American Society for Engineering Education (ASEE) Annual Conference and Exposition, Austin, TX, June 14-
17, 2009.

3. Coffman, J., Terdalkar, S., Rencis, J., and Brown, A., “Integrating Fatigue Analysis into a Machine Design
Course or Finite Element Course”, CD-ROM Proceedings of the American Society for Engineering Education
(ASEE) Midwest Section Conference, University of Tulsa, Tulsa, OK, September 17-19, 2008.

4. COSMOSWorks, SolidWorks Corporation, 300 Baker Avenue, Concord, MA, http://www.solidworks.com/.

5. Dewey, J., Experience and Education: The 60th Anniversary Edition, Kappa Delta Pi International Honor
Society in Education, West Lafayette, IN, 1998.
6. Kolb, D.A., Experiential Learning: Experience as the Source of Learning and Development, Prentice Hall,
Englewood Cliffs, NJ, 1984.

7. Sharp, J.N., and Terry, R.E., “Combining Kolb Learning Styles and Writing to Learn in Engineering Classes,”
Journal of Engineering Education, Vol. 86, No. 2, pp. 93-101, April, 1997.

8. Brown, A.O., “Teaching Finite Elements using the Kolb Learning Cycle”, Presented at the American Society for
Engineering Education (ASEE) Pacific Southwest Section Conference, University of the Pacific, Stockton, CA,
April 1-2, 2004.

9. Shigley, J.E. and Mischke, C.R., Mechanical Engineering Design, Sixth Edition, McGraw-Hill, New York, NY,
2001, pp. 391-393, p. 1195.

10. Bloom, B.S., Taxonomy of Educational Objectives, David McKay, New York, NY, 1956.

11. “Criteria for Accrediting Engineering Programs: 2009-2010 Accreditation Cycle,” ABET Inc., Baltimore, MD,
http://www.abet.org/forms.shtml.

12. Norton, R.L., Machine Design: An Integrated Approach, Second Edition, Pearson Prentice Hall, Upper Saddle
River, NJ, 2000, pp.546-550.

13. Jung, C.G., Psychological Types, Princeton University Press, Princeton, NJ, 1971 (Originally Published in
1921).

14. Gregorc, A.F., “Mind Styles™ Model,” Gregorc Associates Inc., Columbia, CT, accessed March 2, 2009,
http://gregorc.com/.

15. Hermann International, Herrmann Brain Dominance Instrument, accessed March 2, 2009,
http://www.hbdi.com/home/index.cfm.

16. Felder, R.M., and Silverman, L.K., “Learning and Teaching Styles in Engineering Education,” Engineering
Education, Vol. 78, No. 7, pp. 674-681, 1988.

17. Felder, R. M., “Matters of Style,” ASEE Prism, pp. 18-23, December, 1996.

18. Felder, R.M., and Soloman, B.A., “Index of Learning Styles,” http://www.ncsu.edu/felder-public/ILSpage.html,
accessed February 20, 2009, http://www.engr.ncsu.edu/learningstyles/ilsweb.html.

19. Jensen, D., Wood, K., and Wood, J., “Hands-on Activities, Interactive Multimedia and Improved Team
Dynamics for Enhancing Mechanical Engineering Curricula,” International Journal of Engineering Education,
Vol. 19, No. 6, pp. 874-884, 2003.

20. Myers, I.B., and McCaulley, M.H., Manual: A Guide to the Development and Use of the Myers-Briggs Type
Indicator, Consulting Psychologists Press, Palo Alto, CA, 1985.

21. Humanmetrics.com, Jung Typology Test, accessed March 25, 2009,

http://www.humanmetrics.com/cgi-win/JTypes2.asp.

JOSH COFFMAN
Josh Coffman is a M.S. student in the Department of Mechanical Engineering at the University of Arkansas,
Fayetteville. He has worked as a civil design technician for Crafton, Tull, Sparks, and Associates in Russellville,
Arkansas. Responsibilities included design of residential subdivisions, commercial properties, and municipal water
and sewer systems. He received a B.S. in Mechanical Engineering from Arkansas Tech University in 2006.
V-mail: 479-970-7359; E-mail: [email protected].
 JIANCHENG LIU
Jiancheng Liu has been an assistant professor of the Department of Mechanical Engineering at the University of the
Pacific since 2006. Prior to joining at the University of the Pacific, he has worked in industries for many years. His
research focuses on CNC machine design and analysis, computer aided manufacturing and manufacturing system
automation. He has published more than 70 peer reviewed technical journal and conference papers. Dr. Liu was also
awarded 4 patents. He has invented many new technologies which have been practically applied in industries. He
received the Industrial LEAD Award from SME in 2001. Dr. Liu received his B.S. and M.S. degrees in mechanical
engineering in China. After receiving his Ph.D. degree in Japan, he moved to the States in 1997 and did his Post
Doctorate work at the University of California, Davis. V-mail: 209-946-3079; E-mail: [email protected].

ASHLAND O. BROWN
Ashland O. Brown is a professor of mechanical engineering at the University of the Pacific in Stockton, CA. He has
held numerous administrative, management and research positions including Program Director, Engineering
Directorate, National Science Foundation, Dean of Engineering at the University of the Pacific; Dean of Engineering
Technology at South Carolina State University; Engineering Group Manager at General Motors Corporation: and
Principal Engineering Supervisor, Ford Motor Company and Research Engineer, Eastman Kodak Company. He
received his B.S. in Mechanical Engineering from Purdue University and M.S. and Ph.D. in Mechanical Engineering
from the University of Connecticut. He has authored over 40 referred and propriety publications in automotive
design, finite element modeling of automobile body structures, and photographic film emulsion coating instabilities.
His most recent research includes development of innovative finite element tutorials for undergraduate engineering
students and vibrational analysis and measurement of human skeletal muscles under stress using laser holography.
V-mail: 209-946-3091; E-mail: [email protected].

SACHIN S. TERDALKAR
Sachin S. Terdalkar is a Ph.D. candidate in the Department of Mechanical Engineering at the University of
Arkansas, Fayetteville. His current research is mainly in using computational methods to study the mechanics of
nano structures like carbon nanotubes, graphene sheets. He has worked on molecular dynamic simulation of ion
deposition induced curvature in thin films. Currently he is working on using nudged elastic band method and
molecular mechanics to study the brittle to ductile transition in graphene sheet fracture. He has worked as senior
engineer in John Deere Technology Center, Pune INDIA. His responsibilities at John Deere included finite element
analysis and fatigue analysis to determine life of the newly designed components for new generation tractors. He
received a M.S. in Mechanical Engineering from the Worcester Polytechnic Institute in 2003. His M.S. research
formulated and developed a new algorithm for interactive stress reanalysis in early stages of design using ANSYS®.
He received his B.S. from the College of Engineering, Pune INDIA in 1999. V-mail: 479-575-6821;
E-mail: [email protected].

JOSEPH J. RENCIS
Joseph J. Rencis has been professor and Head of the Department of Mechanical Engineering at the University of
Arkansas, Fayetteville since 2004. He has held the inaugural endowed Twenty-first Century Leadership Chair in
Mechanical Engineering since 2007. From 1985 to 2004 he was professor in the Mechanical Engineering
Department at Worcester Polytechnic Institute. His research focuses on boundary element methods, finite element
methods, atomistic modeling, and engineering education. He currently serves on the editorial board of Engineering
Analysis with Boundary Elements and is associate editor of the international Series on Advances in Boundary
Elements. Currently he serves as Chair of the ASME Mechanical Engineering Department Heads Committee,
Program Chair of the ASEE Mechanical Engineering Division, and an ABET program evaluator. He currently
serves on the Academic Advisory Board of the College of Engineering at United Arab Emirates University. He
received the 2002 ASEE New England Section Teacher of Year Award, 2004 ASEE New England Section
Outstanding Leader Award, and 2006 ASEE Mechanics Division James L. Meriam Service Award. Dr. Rencis is a
fellow of ASME and ASEE. He received a B.S. from Milwaukee School of Engineering in 1980, a M.S. from
Northwestern University in 1982, and a Ph.D. from Case Western Reserve University in 1985. V-mail: 479-575-
4153; E-mail: [email protected].

View publication stats