|
Timothy G. Clapp, PhD, PE Professor NC State University Raleigh, NC
27695 tclapp@tx.ncsu.edu |
Brad A. Dickinson Design Engineer 3-TEX Incorporated Cary, NC 27511 Dickinsonb@stex.com |
Abstract
Felled seams are used
extensively in the construction of apparel and industrial textile products.
Defects can be formed in the seaming operation that cannot be detected visually.
These “internal” defects cause the product to fail when used. This paper
describes the design of an on-line monitoring system to detect internal defects
created during the felled seaming operation.
Theory of Inventive Problem
Solving (TRIZ) methods were used to efficiently design a commercially acceptable
monitoring system for the seat seaming operation in jeans production.
Introduction
Modern companies compete in a
global marketplace. Advances in information technology have made marketing
products and communicating with customers a simple task. In fact, the customer
now has a great impact on decisions regarding improvements to existing products
and the design of new products. Often, engineering design goals focus on meeting
the needs and requirements of intended customers. Companies that wish to excel
in the global market must be prepared to generate quickly products that meet the
customer's current needs and desires. They must also produce highly innovative
products to discourage competition. Thus, engineers are now faced with the need
to solve engineering design problems rapidly, cost effectively, and with a high
level of innovation. The goal of this paper is to demonstrate the utility of
TRIZ methods for rapidly creating innovative solutions at the concept generation
and prototyping phases of the engineering design process. These methods are
applied to a critical problem in the apparel manufacturing industry.
Specifically, a sensor system will be designed to monitor the quality of a
joining operation in the assembly of denim jeans.
FELLED SEAM CONSTRUCTION
|

Figure 1: Cross-sectional View of Correct
and Incorrect Felled Seatseam Constructions
|
In denim jean construction, the
felled seam comprises more seam length than any other seam type. The U.S.
government developed Federal Standard 751a (FS751a), which classifies
seams and stitches used in apparel garments. Lapped seam type C-2 (denoted LSc-2)
is generally the seam type used in jeans construction. This is the strongest of
the lapped seams since two rows of stitching are used to secure it. In this seam
“lapped bottom and top plies fold toward each other and interlock, displaying
no raw edges on either the outside or inside of the garment”[1]. Two rows of
type 401 chain stitch bind the seam. Figure 1a is a cross-sectional end view
schematic of an LSc seam.
Seam quality problems occur when
one or both of the two lapped fabric ends that are used to form the seam fail to
be formed or “stuffed” completely, such as in Figure 1b. Although a raw
fabric edge may not be visible, an “understuffed” seam exhibits less
strength than a properly formed seam. When the completed garments undergo the
post-finishing (washing) process, improperly stuffed seams tend to “blowout”
resulting in a total rejection of the garment. During the post-finishing process
it is likely to “blowout” as illustrated in Figure 1c.
In jeans, the “seatseam” has
a high instance of failure. This seam runs from the crotch to the top of the
back of the jeans. This seam was selected as the first application for the new
monitoring system.
Clapp [2] developed a method for
identifying defective seams by measuring the compressed thickness in the center
of the seam. Figure 1a shows a correct seam with four layers of fabric in the
center of the seam. Figure 1b shows a defective seam that has only three layers
of fabric in the center of the seam.
|
Table 1: Engineering
Design Specifications
for the Seatseam Monitoring System
with Completed Design Benchmark
Rating (5-highest, 1-lowest)
|
Engineering Design Factor |
Rating |
|
PHYSICAL CONSTRAINTS |
|
|
1.-Sensing Point/Location |
5 |
|
2.-Control Unit Location |
5 |
|
3.-User interface |
3 |
|
OPERATION |
|
|
4.-Environmental Conditions |
4 |
|
5.-Durability |
4 |
|
6.-Fabric Style |
4 |
|
7.-Seam Length |
5 |
|
8.-Sewing Machine Speed |
5 |
|
9.-Sewing Machine
Maintenance |
3 |
|
10.-Electrical Power |
4 |
|
QUALITY |
|
|
11.-Resolution |
4 |
|
12.-Edge Detection |
5 |
|
13.-Fault Location |
5 |
|
14.-Operator Feedback |
5 |
|
FINANCIAL |
|
|
15.-Product/Development
Costs |
4 |
|
ENGINEERING DESIGN
The process of engineering
design begins with understanding the problem and converting customer
requirements into design specifications. Table 1 summarizes the
engineering specifications that govern the design of the felled seatseam
monitoring system.
Theory of Inventive
Problem Solving (TRIZ) methods were used to accelerate and optimize the
design of the monitoring system. |
INITIAL DESIGN
|

Figure 2: Sewing Machine to be modeled
|
Three-dimensional solid
modeling is used to minimize development time and cost [3]. The first step
in the modeling process is to model the existing components. These
components establish the geometrical constraints of the sewing machine.
Several components of the Pfaff model 5489-H sewing machine, shown in
Figure 2, were modeled in the SolidWorks®. As new components are designed
and added to the existing model, the solid modeling software allows the
designer to check for mating part alignment, dynamic operation, and
interference problems. The designer reduces time and cost by not having to
physically make the parts and test them.
|
|

Figure 3: Three-dimensional Solid Model
of the Initial Mechanical Prototype.
|
The initial mechanical
design is shown in Figure 3. This design used a “ski(1)” to contact
the center of the seam. Structural components, bearings, and screws were
designed to transform rotational displacement of the ski caused by the
seam thickness into an amplified displacement. A linear variable
displacement transducer (LVDT) converted the displacement into an
electrical voltage for analysis. The LVDT is housed in bracket(11) shown
in Figure 2.
Figure 3 shows the initial
prototype that was constructed and proven to accurately measure compressed
seam thickness. Unfortunately the initial design did not meet the design
goals for cost, reliability, and jeans size variation. |
TRIZ METHODS
The Theory of Inventive Problem
Solving (TRIZ) is a collection of concepts and methods formulated by Genrich
Altshuller [4,5,6] to systematically improve the engineering designer's
ability to solve difficult problems. The theory may be best described as “a
unique way of thinking that enhances creativity by getting individuals to think
far beyond their own experience, to reach across disciplines, to resolve
problems using extracted knowledge from other areas of business, science, or
technology.”[5]
|

Figure 4: Modified Mechanical Design
after Ideality Concept was applied to the Design
|
TRIZ concepts, “Ideality”
and the “utilization of resources” were used to analyze and improve the
design of the initial monitoring system. “Ideality” is defined as the sum of
the useful functions divided by the sum of the harmful functions plus the sum of
the costs. “Ideality” approaches infinity as the design approaches the Ideal
Final Result (IFR). The ideal system meets all of the desired useful functions
with no harmful effects and at no cost. The primary function of the monitoring
system is to measure compressed seam thickness. The "Ideal" system
measures seam thickness and identifies all defects in the seam using no parts
and at no cost.
The initial prototype was
comprised of 14 parts. The function of each part was documented. The
"Ideality" method was used to systemically eliminate each part from
the system by eliminating or combining functions. This design exercise resulted
in the elimination of 6 parts from the original system. The "Ideality"
of the system was increased. The modified design reduced cost and improved
reliability of the system. Figure 4 shows the modified design.
The design of the sensor system
is improved by identifying resources that can be used to increase
"Ideality." Resources are categorized as substance, space, field,
time, functional, and informational resources[6]. Table 2 shows a partial list
of resources available to the designer.
The sensor system was optimized
using the informational resource produced by the LVDT displacement signal shown
in Figure 4. The voltage trace of the LVDT contained much more information than
just the presence of a defect in the seam. Careful analysis of the trace clearly
showed when the leading edge of the fabric was approaching the sensor. This
triggered the encoder to start counting and record the distance traveled. A
second, large dip in the signal identified the presence of a riser seam and
prevented the system from classifying this as a defect. At the end of the
fabric, a sudden rise indicated the presence of the end of the fabric. The
information contained in the voltage trace was used to eliminate external
sensors and make the sensing system independent of seam length and speed.
Table 2: Resources Available to
Increase Ideality
|
Substance Resources |
|
-Sewing machine components |
|
-Lower lever arm |
|
-Sensing ski |
|
-Lower mounting plate |
|
-LVDT mount |
|
-LVDT sensor |
|
-The fabric seam |
|
-Fabric |
|
-Lint/Dust |
|
-Thread |
|
-Thread Waste |
|
Space Resources |
|
-Sewing Machine Mounting
Locations |
|
-Empty Space Adjacent and Above
Puller Wheel |
|
Field Resources |
|
-220 VAC Electrical Supply |
|
-Compressed Air |
|
-Gravity |
|
Time Resources |
|
-Repetitive Sewing Machine
Motion |
|
-Time to Produce One Stitch |
|
Informational Resources |
|
-Fabric Sensor Signal |
|
-Encoder Signal |
|
-Linear Potentiometer Signal |
|
-User Interface |
|
-Red Indicator Lamp |
|
Functional Resources |
|
-Rotational Motion of the Lever
Arm |
|
-Vertical Motion of the Linear
Potentiometer Shaft |
|
-Rotational Motion of the
Puller Wheel |
|
-Rotational Motion of the
Sewing Machine Motor |
|
-Cyclic Motion of the Sewing
Machine Feed Dogs |
|
-Vertical Motion of the Sewing
Machine Presser Foot |
|
-Vertical Motion of the Sewing
Machine Needles |

Figure 5: Information Contained Voltage Trace
CONCLUSIONS
The use of TRIZ concepts,
"Ideality" and "utilization of resources," resulted in an
improved sensor system. The concepts directed the designer's thinking to
systematically eliminate components of the design through the use of resources
available to the designer. Design objectives were achieved using 40% fewer
components.
A production prototype was
constructed and tested. A rating system of 1-5 (5 - meets goal) was used to
evaluate the system against the original design goals. Table 1 shows the very
favorable ratings of the performance of the production prototype. A commercial
version of this system is now being used in manufacturing to eliminate the
internal defects that cannot be seen visually.
This research proves that TRIZ
methods are valuable tools for rapidly designing robust systems.
References
Hudson, Peyton B., Guide to
Apparel Manufacturing, Spark Press, Raleigh North Carolina.
Clapp, Timothy G. et al., United
States Patent Number 5,671,689, issued September 30,1997.
LaCourse, Donald E., Handbook
of Solid Modeling, McGraw-Hill 1995.
Altshuller, G. (H. Altov), And
Suddenly the Inventor Appeared TRIZ, the Theory of Inventive Problem
Solving, Technical Innovation Center, Inc., Worcester,
Massachusetts 1996.
http://www.jps.net/triz/triz0000.htm
Terninko, John, Alla, Zusman,
Zlotin, Boris, Step-by-Step TRIZ: Creating Innovative Solution Concepts,
Responsible Management Inc., Nottingham, New Hampshire 1996.