:ME .. ‘ , ‘ . , , , g u. u. .,.
3 :3. inn. . A , . s u; un._.vv..:

I. 1.:
3 .

ul‘. .4...
« I Lvhuulad

‘ 4 ‘ t I.
:jaflfinwfi

 

 

 

 

11 \.l
. :iikr
i 325.

 

 

 

 

I
xiv...

 

Ml
.

t. . .33).

 

_. Nauruwam

 

LIBRARY
Michigan State
University

This is to certify that the
thesis entitled

DETERMINATION OF CLASS 0 RADIO FREQUENCY
IDENTIFICATION TAG FAILURE MODES

presented by
Michael David Jonson

has been accepted towards fulfillment
of the requirements for the

MS. degree in Packaging

flaZé/JMQ/

l/ \ Major P ofessor’ ignature
3 / // (95"
’ /

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:/ClRC/DateDue.lndd-p.15

 

 

DETERMINATION OF CLASS 0 RADIO FREQUENCY IDENTIFICATION TAG
FAILURE MODES

BY

Michael David Jonson

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
School of Packaging

2005

ABSTRACT

DETERMINATION OF CLASS 0 RADIO FREQUENCY IDENTIFICATION TAG
FAILURE MODES

BY

Michael David Jonson

This thesis involves a study of known packaging dynamics and their destructive
forces on the readability of Class O Matn'cs® Radio Frequency Identification (RF ID) tags.
Vibration and impact procedures were used to test the effects on the RFID tags and to
evaluate the robustness of these tags as defined by the American Society for Testing and
Materials (ASTM) and the International Safe Transit Association (ISTA) as well as
exploratory testing methods.

Vibration testing was performed according to the standard tests as well as above
and beyond. This testing was done on single cases, column stacks, and palletized
configurations. Different drop and shock table heights were used to determine the impact
fiagility of these tags.

The testing revealed that shock and drop impacts damaged the tags while none of
the vibration testing affected the tags. However, tag placement was determined to be
influential in the survival of the tags during impacts. Tags which had the integrated circuits
placed over the peaks of the corrugations had the highest placement survival rate. Tag
orientation was also determined to be influential for tag survival. Tags adhered
perpendicular to the corrugated fluting survived more impact testing than those adhered

parallel to the fluting of the corrugated board.

ACKNOWLEDGEMENTS

I wish to express my deepest and most sincere appreciation to my major professor Dr.
Robb Clarke for his assistance, guidance and encouragement throughout graduate school. I
would also like to extend my gratefulness and appreciation to the other members of my
graduate committee, Dr. S. Paul Singh and Dr. Kenneth Boyer who were gracious enough
to share their knowledge and expertise as well as Dr. Arnold Stromberg, of the University

of Kentucky Statistics Department, for his help with my work.

I would like to thank the faculty and staff of the school of packaging for helping me during

the course of my graduate studies and research.

I would like to thank my parents David and Lydia and sister, Kristen. I would have never
been able to accomplish my goals without their encouragement and support. I would also
like to thank my peers for their help, support, friendship and knowledge during my

graduate studies.

TABLE OF CONTENTS

LIST OF FIGURES ................................................................................................... vii
LIST OF TABLES ................................................................................................... viii
LIST OF EQUATIONS ............................................................................................... x
GLOSSARY ............................................................................................................... xi
CHAPTER 1: INTRODUCTION

1.1 Introduction ..................................................................................................... 1
1.2 Need for Testing ............................................................................................. 5
1.3 Objectives ....................................................................................................... 6

CHAPTER 2: LITERATURE REVIEW

2.1 Dynamics ............................................................................................................... 7
2.2 Vibration ................................................................................................................ 8
2.3 Impact .................................................................................................................... 9
2.3.1 Drop .............................................................................................................. 9
2.3.2 Shock .......................................................................................................... 10
2.4 Standardized Testing ........................................................................................... 11
2.4.1 ASTM ......................................................................................................... 12
2.4.2 ISTA ........................................................................................................... 12
2.5 Automatic Identification and Data Collection .................................................... 13
2.6 Barcode Technologies ......................................................................................... 14
2.7 Electronic Article Surveillance ........................................................................... 15
2.8 Radio Frequency Identification ........................................................................... 17
2.8.1 Transponders .............................................................................................. 18
2.8.2 Integrated Circuits (IC) .............................................................................. 18
2.8.3 Power Source .............................................................................................. 19
2.8.3.1 Passive Tags ...................................................................................... 19
2.8.3.2 Active Tags ....................................................................................... 20
2.8.4 Interrogators/Reader ................................................................................... 21
2.8.4.1 Portable Interrogators ........................................................................ 21
2.8.4.2 Stationary Interrogators ..................................................................... 21
2.8.5 Electronic Product Code (EPC) ................................................................. 22
2.8.6 Standards/Regulations ................................................................................ 23
CHAPTER 3: METHODOLOGY
3.1 Introduction .......................................................................................................... 24
3.2 Vibration .............................................................................................................. 29
3.2.1 Sinusoidal Resonance of Individual Cases ................................................ 29
3.2.2 Sinusoidal Resonance of Cases Against Plywood .................................... 30
3.2.3 Random Vibration of Four Column Stacked Cases .................................. 31

iv

3.2.4 ISTA 3C Vibration and Drop Sequence .................................................... 33

3.3 Drop ..................................................................................................................... 35
3.3.1 Incremental Drop Sequence (Determination of Failure) ............................ 35
3.3.2 Drop Sequence ............................................................................................. 36

3.4 Shock ................................................................................................................... 36
3.4.1 Initial Sequence ............................................................................................ 40
3.4.2 Load Spreader .............................................................................................. 41
3.4.3 Threshold Level Determination ................................................................... 43
3.4.4 Initial PeakN alley Determination ............................................................... 43
3.4.5 Initial Random Placement Determination ................................................... 44
3.4.6 Aluminum Fixture ....................................................................................... 44
3.4.7 Wood Fixture, No Headspace ..................................................................... 45
3.4.8 IC Centered Over Peak (Tag Orientation: with Fluting) ............................ 46
3.4.9 IC Centered Over Valley (Tag Orientation: with Fluting) ......................... 46
3.4.10 IC Centered Over Peak (Tag Orientation: 90° to Fluting) ....................... 47
3.4.11 IC Centered Over Valley (Tag Orientation: 90° to F luting) .................... 47
3.4.12 IC On Slope of Fluting (Tag Orientation: with Fluting) .......................... 48
3.4.13 IC On Slope of F luting (Tag Orientation: 90° to Fluting) ........................ 48

CHAPTER 4: RESULTS

4.1 Vibration .............................................................................................................. 50

4.1.1 Sinusoidal Resonance of Individual Cases ................................................ 50
4.1.2 Sinusoidal Resonance of Individual Cases Against Plywood .................. 51
4.1.3 Random Vibration of Four Column Stacked Cases .................................. 51
4.1.4 ISTA 3C Vibration and Drop Sequence .................................................... 51

4.2 Drop ..................................................................................................................... 52
4.2.1 Incremental Drop Sequence Determination for Failure ............................. 52
4.2.2 Incremental Drop Sequence ........................................................................ 52

4.3 Shock .................................................................................................................... 53
4.3.1 Initial Sequence ............................................................................................ 53
4.3.2 Load Spreader .............................................................................................. 53
4.3.3 Threshold Level Determination ................................................................... 54
4.3.4 Initial Peak/V alley Determination ............................................................... 54
4.3.5 Initial Non-Pcak/V alley Determination ...................................................... 55
4.3.6 Aluminum Fixture ....................................................................................... 55
4.3.7 Wood Fixture, No Headspace ..................................................................... 56
4.3.8 IC Centered Over Peak (Tag Orientation: with F luting) ............................ 56
4.3.9 IC Centered Over Valley (Tag Orientation: with Fluting) ......................... 57
4.3.10 IC Centered Over Peak (Tag Orientation: 90° to Fluting) ....................... 57
4.3.11 IC Centered Over Valley (Tag Orientation: 90° to Fluting) .................... 57
4.3.12 IC On Slope of Fluting (Tag Orientation: with Fluting) .......................... 58
4.3.13 IC On Slope of F luting (Tag Orientation: 90° to Fluting) ........................ 58

CHAPTER 5: STATISTICAL ANALYSIS, CONCLUSIONS AND FUTURE TESTING

5.1 Statistical Analysis .............................................................................................. 59
5.2 Conclusion .......................................................................................................... 63
5.3 Future Testing ...................................................................................................... 65
APPENDICES

Appendix A (Vibration) ............................................................................................. 67
Appendix B (Drop) ..................................................................................................... 69
Appendix C (Shock) ................................................................................................... 70
Appendix D (Statistical Analysis) .............................................................................. 79
BIBLIOGRAPHY ...................................................................................................... 86

LIST OF FIGURES

FIGURE PAGE
1 915 MHz Matrics® X1090-LBL Class 0 Label Inlay RFID Tag ................................. 25
2 Total Thickness of Matrics® Inlay Label and Corrugated System ................................ 27
3 915 MHz Matrics® IC Protrusion from Adhesive Label ............................................... 27
4 Numbering Sequence for Weighted Case ...................................................................... 29
5 Individual Cases .............................................................................................................. 29
6 Stacked Cases Against Plywood .................................................................................... 3O
7 Four Column Stacked Cases ........................................................................................... 31
8 Drop Testing ................................................................................................................... 35
9 Accelerometer Placement on Shock Table Impact Surface .......................................... 37
10 Wooden Shock Machine Fixture with Polypr0pylene Pad ............................................ 38
11 Definition of Placement According to Corrugations ..................................................... 39
12 Load Spreaders ............................................................................................................... .41

LIST OF TABLES

TABLE PAGE
1 Structure Measurements of Label and Corrugated Board ............................................. 26
2 ASTM — 4169 Assurance Level I Demand Profile ........................................................ 32
3 ISTA 3C Drop Sequence for Products < 50 lbs. ............................................................ 33
4 ISTA 3C Test Method .................................................................................................... 34
5 Initial Drop Heights ........................................................................................................ 4O
6 Load Spreader Drop Heights .......................................................................................... 42
7 Complete Results for All Testing ................................................................................... 50
8 Drop Failure Totals ......................................................................................................... 53
9 Statistical Data Collected for Comparison ............................................................... 59, 79
A1 Sinusoidal Vibration of Individual Cases ...................................................................... 67
A2 Sinusoidal Vibration of Individual Cases Against Plywood ......................................... 67
A3 Random Vibration of Four Column Stacked Cases ....................................................... 67
A4 ISTA 3C Vibration and Drop Sequence ........................................................................ 68
BI Incremental Drop Sequence Determination for Failure ................................................. 69
B2 Incremental Drop Sequence ........................................................................................... 69
C1 IC Centered Over Peak (Tag Orientation: with Fluting) ............................................... 70
C2 IC Centered Over Valley (Tag Orientation: with Fluting) ............................................ 71
C3 IC On Slope of Fluting (Tag Orientation: with Fluting) ................................................ 72
C4 IC Centered Over Peak (Tag Orientation: 90° to Fluting) ............................................. 73
C5 IC Centered Over Valley (Tag Orientation: 90° to Fluting) .......................................... 74

viii

C6 IC On Slope of Fluting (Tag Orientation: 90° to Fluting) ............................................. 75
C7 Initial Sequence ............................................................................................................... 76
C8 Load Spreader ................................................................................................................. 77
C9 Aluminum Fixture (4 Inch Drop) ................................................................................... 77
C10 Aluminum Fixture (3 Inch Drop) ................................................................................... 78
C11 Wood Fixture, No Headspace (3 Inch Drop) ................................................................. 78
D1 Comparison 1 (A+B+C vs. D+E+F) .............................................................................. 79
D2 Comparison 2 (A+D vs. B+E) ........................................................................................ 80
D3 Comparison 3 (A vs. B) .................................................................................................. 81
D4 Comparison 4 (D vs. E) .................................................................................................. 82
D5 Comparison 5 (A vs. D) .................................................................................................. 83
D6 Comparison 6 (B vs. E) .................................................................................................. 84
D7 Comparison 7 (C vs. F) ................................................................................................... 85

LIST OF EQUATIONS

EQUATION PAGE
1 Change in Velocity for Free Fall Drop Height .............................................................. 40
2 Statistical Impact Percentage .......................................................................................... 63

GLOSSARY

 

AIDC — Automatic Identification and Data Capture

 

AIM — Association for Automatic Identification and Mobility

 

ASTM — American Society for Testing and Materials

 

CPU — Central Processing Unit

 

EAS — Electronic Article Surveillance

 

EPC — Electronic Product Code

 

EPC Global -—

 

G — Acceleration due to Gravity

 

Gm, — Root Mean Square of G (in place of Standard Deviation)

 

GMA — Grocery Manufacturers Association

 

HDPE — High Density Polyethylene

 

IC — Integrated Circuit

 

ID — Identification

 

ISTA - International Safe Transit Association

 

KE — Kinetic Energy

 

MHz — Megahertz

 

PE — Potential Energy

 

PLC — Programmable Logic Controllers

 

R/W — Read Write

 

RF — Radio Frequency

 

 

RF ID — Radio Frequency Identification

 

 

 

RO — Read Only

 

SBS — Solid Bleached Substrate

 

TP3 — Test Partner version 3

 

UCC — Universal Code Council

 

UGPIC — Universal Grocery Products Identification Code

 

UPC - Universal Product Code

 

UPC-A — Universal Product Code version A

 

 

WORM — Write Once Read Many

 

 

CHAPTER 1: INTRODUCTION

1.1 INTRODUCTION

Packaging is intimately involved in the success or failure of a product within the
supply chain. It is an area relied upon to lead innovation and technology while being
devoted to cost savings, and shields a product by providing a safety barrier. Packaging is
becoming an integrated part of the technology wave for tracking products. With the use of
barcodes, packaging is continuing to be a way to carry identification. Along with Radio
Frequency Identification (RF ID) smart labels, packaging protects and identifies products.
RFID also protects against product loss and inventory shrinkage by allowing the product
to identify and track itself within the supply chain. Packaging protects products from
physical dynamic situations that might be encountered while traveling through the supply
chain.

Packaging is a very specific combination of materials designed and used to
dissipate excessive energy changes. These energy inputs, or dynamic situations, are
encountered everyday and revolve around the buildup and release of energy. As a
package encounters movement it gains potential energy (PE) which can be destructive.
Packaging examples containing potential energy can be a palletized load, a single parcel
sitting on a table, or packages loaded on a truck.

As a product starts in motion, the contained PE is transferred into kinetic energy.
The longer the product is in motion the greater the energy gain until it returns to rest. As
the product returns to rest the energy change radiates out of the product through the easiest

means possible. During the energy transfer, fragile parts serve as the transfer points and

are the root cause for damage. One way to counteract these forces is by protecting the
product, which is done through specifically designed packaging materials.

Drop, shock, direct impacts and vibration are areas where the buildup and release
of energy are the most destructive to products. These dynamic situations will take place
through the distribution of a product. Handling, transportation and distribution dynamics
are key areas to understand as new technology is being introduced to complement current
packaging applications.

Moving from a production facility to its final destination, packaging materials
must not only make it so the product survives the trip, but must also uniquely identify each
product for tracking and inventory purposes. (Boyer, 2003) Organizations such as the
Rail Industry Forum require clear and understandable “Communication of Information”
directly in their packaging guidelines. (www.ismrif.org) Due to the similarities of many
different products each package has the task of displaying unique identifiers.

Identifiers increase the ease of distinguishing what is contained within a package,
but do not guarantee against errors. The identifiers must contain a legible language
understood by laborers or purchasers and also must be recognized by automated systems.
This level of identification is called Automatic Identification which incorporates a label, a
barcode and even a smart label.

Barcodes have been used since the mid-70’s to control inventory and track a
product’s progression into and out of grocery facilities. Developments in technology, such
as the implementation of barcodes, have helped to modernize the supply chain.
Unfortunately barcodes require a limited distance line of sight to be effectively read by

automated systems. Barcodes have reached a read rate of 99 percent according to Zebra

Technologies. (Zebra, 2004 barcode 101) This read rate is based on the integrity of the
barcode itself. Incorrect printing as well as transit damage will require manual input of the
data after the barcode becomes unreadable by an automated system. Human interaction in
the supply chain provides opportunities for products to be lost or to become transparent to
an inventory system. Zebra Technologies states in their application white paper BAR
CODING AND RFID: THE KEY TO T RACEABILITY AND SAFETY IN THE FOOD
SERVICE SUPPLY CHAI “. .. an experienced typist makes one error approximately
every 300 keystrokes” for manually entered data. Zebra also published information from
a 2003 Grocery Manufacturers Association (GMA) Logistics Study on the effects and
consequences of human mistakes “. .. errors occur in 36 percent of consumer packaged
goods orders.” (Zebra, 2004)

Barcodes eliminate errors which help companies, warehouses, and distribution
centers save billions of dollars per year, but can do nothing to prevent theft. A 2002
survey on retail theft published by the National Retail Security Survey reported retailers
lost 1.7 percent of their total annual sales due to inventory shrinkage. This survey states
that the retail economy has more than $1.845 trillion in annual transactions. The
1.7 percent loss due to inventory shrinkage makes the loss over $31.3 billion in the United
States alone. (Vargas 2004) Addressing transparent portions of a product’s life cycle can
be decreased by supply chain management.

A possible solution for tracking products and helping to eliminate theft has been a
developing technology with limited understanding of its full potential called RFID. RFID
is a wireless technology which has the ability to work without a line of sight requirement

and is able to transfer information at any point throughout the supply chain.

Commercial RFID applications began in the 1960’s with electronic article
surveillance (EAS). This single bit system was the infancy of commercial uses for RF ID.
An EAS system can only report the existence of a tag within the read field. (Landt et. al.,
2001) The application of an EAS system is used to counteract theft in the retail setting.
Applications have continued to become more apparent in everyday life since the early
1990’s. The AIMglobal.com website states that current commercial RFID and EAS
applications include: animal identification, sports timing, inventory control, airline
baggage ID, container/pallet tracking, ID badges and access control, fleet maintenance,
equipment/personnel tracking in hospitals, parking lot access and control, car tracking in
rental lots, product tracking though manufacturing as well as assembly and product
authentication. (AIM, 2004)

The EPC Global, with the strength and persuasiveness of Wal-Mart, has made a
new push for the use of RF ID for inventory tracking. EPC Global (formerly the Auto-ID
Center) began developing the current form of RFID automatic identification for automated
manufacturing line usage in the late 1990’s. The automation process allows each tagged
product to communicate with the next successive station in order for that station to prepare
for the arrival of that product to the next station. Having the ability to electronically see a
product before it is visually accessible is what is driving RFID implementation today.

Wal-Mart set a mandate for their top 100 suppliers to have RF ID technology
compliance by January 2005 for three Distribution Centers located in Texas. As Wal-
Mart suppliers scramble to understand RF ID technology and limitations, an area being

overlooked is the supply chain dynamics as an area for RFID tag failures.

The survivability of tagged packages will be great when a product is moved as a
large load. By palletizing or unitizing a load, the entire load becomes more stable. A
homogeneous pallet of products is more likely to be handled with care because of the
weight and size of the load. Individual cases on the other hand have a greater potential to
be handled more harshly. When individual cases are tagged, the level for survivability is

greatly decreased because the opportunity for handling and excessive dynamics.

1.2 NEED FOR TESTING

Much of the published information on RF ID revolves around its potential and
additional advancements in RF ID technology. Currently, research data is not readily
available as individual companies work through the shortcomings of the technology. As
information is sparingly published and distributed throughout corporate America, data
showing the robustness of these RFID tags is not included and may not be done.

Tags overwhelmed by dynamic encounters will lose functionality and be lost. The
inventory shrinkage, due to the failure of the tag to read, will cost industries billions of
dollars each year. Understanding and developing testing methods for the determination of
tag robustness is critical. Dynamic testing and the understanding of distribution hazards
will allow suppliers and users of RFID technology to eliminate vulnerable areas for tag
placement and handling procedures which in turn can reduce product loss.

The marriage of dynamics and RFID are areas that will be forced to the forefiont
of information. Dynamics is imperative to the success of a package and product. RFID
can and will affect a package and a product’s journey from the manufacturing facility to

the end user. To be aware of what effect dynamics has on RFID tags you must understand

the RF ID technology and its future benefits. Currently there are no standardized tests to
understand the dynamic situations a RFID tag will encounter in the supply chain. The
literature review will touch on dynamics, RFID and the need for development of both

areas.

1.3 OBJECTIVES
Understanding the dynamic impact and robustness on RFID tags in relation to

dynamic encounters is imperative. The overall objective of this research is to develop an
understanding of the capabilities, limitations and ruggedness of RFID tags within the
distribution channels available today. The benefits of testing help to understand the
capabilities of tags, limit orientation options of tag applications, and help provide
consistency to the companies using the technology. Immediate objectives of this study
include:

1) Determine robustness of Class 0 Matrics® RFID tags;

2) Determine the limitations and discrepancies through dynamic testing, and;

3) Raise issues which require further investigations to determine answers.

CHAPTER 2: LITERATURE REVIEW
2.1 DYNAMICS

The science of packaging deals with the protection of a product. This protection
involves the defense from dynamics or dynamic situations. Dictionary.com states the
definition of dynamics as: “The branch of mechanics that is concerned with the effects of
forces on the motion of a body or system of bodies, especially of forces that do not
originate within the system itself.” (Dictionary.com, 2004) Distribution hazards that
threaten a product once it leaves the manufacturing facility or warehouse are shock,
vibration, compression and the environment. (Polin, 2003) Studying the integral portions
of the dynamic supply chain movement a package encounters allows the development of
product and package systems that will withstand higher dynamics than a single
component.

Truck load and less than truck load palletized freight is urritized. Pallets are
moved by fork trucks and collisions along with ill-advised damage can occur when a fork
truck tine impacts a tag.

Single parcel distribution channels are treated differently than unitized load
channels. Single parcels tend to be much smaller and lighter than a pallet which allows
many different manual handling situations. Manual handling adds a window for more
damage to occur. Single parcels are not secured during movement in distribution hubs

and are subjected to impacts both air transportation and ground transportation shipment.

2.2 VIBRATION

Vibration is the cause of most of a product’s damage. At any given stage in
distribution, a package may see up to three days worth of vibration in a long haul
shipment. This vibration tests the natural frequencies of the product.

Vibration is a dynamic which causes different reactions of a product and a
package. (Clarke, 2004) Vibration acts on an internal apparatus of a product breaking
connections and loosening banding on a pallet load. The contents of the bottom box in a
stack can be damaged from the compacting forces. Resonance is a well known and
documented occurrence where components and packages vibrate violently, causing
failures.

Mechanical failures due to vibration can incorporate different areas. Cosmetic
damage, such as dents and scuffing, is a noted problem of vibration. Dr. Paul Singh
(Michigan State University) states scuffing typically occurs when vibration levels are
between two and eight times per second. (Singh et. al., 2004) This high rate of movement
can also cause flex-cracking of packages. Flex-cracking is equated with flexible films,
foils and paper but can be associated with printed circuit boards and electrical connections.

Vibration is also a likely starting component in creating static discharge of
electricity as components and packages rub. While vibration creates this electric build up,
the heat created by the flow of electricity can break connections instantly. The electrical
build up causes the surface to attract dust causing additional problems.

By palletizing or unitizing a load the entire load become more stable. A
homogeneous pallet of products is more likely to see lower levels of vibration because of

the weight and size of a load. This weight and size also lowers the overall natural

frequency of the product allowing the entire load to withstand truck and rail frequencies
equated with travel. The change in dynamics allows palletized products to survive more

transportation situations compared to single parcel shipments.

2.3 IMPACT

Impacts are inputs of energy, or concentrated dynarrric situations, encountered
everyday by packaging. Throughout the distribution of any product impacting dynamics
takes place. Handling, transportation and distribution dynamics are areas improve upon.

The area of impact encompasses both the drops and shocks a package experiences.

2.3.1 DROP

Drop impacts involve very destructive forces. Packaging is a barrier by which
products are protected from free fall drop impacts. Drop tests are an intimate part of
testing for product and package because it simulates the manual handling most products
encounter.

The size and weight of a package defines the way it is handled. As a package
becomes larger and heavier it is likely to be mechanically handled limiting the height at
which it would be dropped. “The expected drop heights for very large packages are less
than one foot because they are handled mechanically, either by fork truck, cart or dolly.”
(Singh et al, 2004) A small light parcel (1-20 lbs) can usually be moved by hand, which
has a high chance to be handled more violently. From a study of small parcel services,

data showed packages on average were subjected to 12 impacts, adequate enough to create

damage. (Singh et al, 2004)

In a drop situation the product/package system free falls at the speed of gravity to
come to rest on a solid surface. Drop testing is done with a sequence of pre-determined
drops on pre-deterrnined sides of a package. The impacting area of the package/product is
still random. With all the possible variables dealing with the package/product, the drop
machine and the seismic mass surface changes, controlling the impacting area of the

package to a specific region is almost impossible.

2.3.2 SHOCK

When discussing product handling, shock or drops are thought to be the most
extreme variable a package and product will encounter. A shock is an extreme,
concentrated variable. When a package is shipped it encounters forces in every aspect of
handling. These handling processes can be broken down into three overall categories.
The first being the macro level of handling where an entire shipping container (i.e. full
trailer of products) can be handled by a lift (i.e. a crane for cargo on a ship). The second
level of handling would be an intermediate level that refers to a pallet load of products
(usually a unitized single product load) which is handled by automated or semi-automated
equipment. The third and final level of handling is an individual package, which are
manually handled. (Polin, 2003)

In shipping situations, different dynamic forces act on everything that has energy.
Shock dynamics are some of the most devastating dynamics a product encounters. Dr.

6

Gary Burgess (Michigan State University) defines a shock as: ‘...a sudden change in

speed. It [a shock] is usually the result of an impact.” (Burgess et al, 2003)

10

Shocks can have multiple origins stemming from drops, kicks and mechanized
handling. A drop is an impact that is typically in the x-axis (or vertical) of a Cartesian
plane. It also has the characteristics of gradually building in speed as it approaches the
ground. When the package impacts the group it has a sudden change in speed over a very
short period of time (typically measured in milliseconds, 1 millisecond (ms) = 0.001
seconds). A kick, on the other hand is an impact that typically contains no or negative
readings in the vertical dimension. Kicks are generated when a package is sitting on the
ground and experiences an impact to the side of the original package. For any situation
with mechanical (mechanized) handling a machine or automated line will sort the
individual packages. These environments are most prevalent with United Parcel Service,
Federal Express and the United States Postal Service where the package will encounter

multiple impacts with sorting equipment and other packages. (Singh et. al., 2004)

2.4 STANDARDIZED TESTING

Testing a product to understand its threshold is important. Strengths and
weaknesses can be found during testing and minimum bounds can be developed to ensure
success. Testing is important and needs to be repeatable in nature. This is why standard
testing methods have been put into place. These provisions are determined through

acknowledged standardized testing by ASTM and ISTA.

11

2.4.1 ASTM

ASTM is one of the largest voluntary standard development organizations in the
world and its methods are built around many different standardized tests. These tests are
assessments of the materials that make up the entire package and its components, but are
not directed toward the package and product system. By incorporating multiple testing
methods, an inclusive test can be performed. Multiple tests allow for a better, more
proficient method for qualification.

The ASTM standard test method D-4169 is an important performance test for
shipping containers and systems. This test method is a sequence of nine individual
standardized tests GD-642, D-880, D-951, D996, D999, D-4003, D-4332, D-4728 and D-
5276). By incorporating many different packaging specific tests the result is a recognized
test sequence to qualify packaging for distribution.

All of the D-4169 standard test methods are significant, but three stand out as
being structurally important for shock impacts, those tests (D-880, D-4003 and D-5276)
are based around impacts. In the instance of any type of impact, failure is a possibility.
The wide range of possible forces a package encounters is certain and the ability of the
package to protect its contents is a fundamental need which is tested and verified though

ASTM standards.

2.4.2 ISTA
ISTA is a packaging standards organization built around developing testing
methods which helps design the correct package for the correct hazard level within the

distribution systems. The ISTA organization’s focus is developed to improve on ASTM

12

distribution standard test method D-4169. ISTA test methods are a distinct product
package system testing formula. In different supply chain dynamics each product/package
system needs to be tested and validated because not every distribution channel is the same.

As packages are evaluated the size and weight determine which procedure to use.
The ISTA 3 series testing replicates damage-producing transport hazards, while being
non-specific. The tests cover a wide collection of vehicle types, vehicle routes and
handling situations.

Procedure 3C is a general simulation test for individual packaged-products shipped
through a single parcel delivery system. The testing levels in the procedure are based on
general data while not characterizing specific distribution systems. The package and the
product are combined and thought of as a single entity for testing procedures. Some areas,
such as moisture and altitude along with any abnormal handling, are disregarded in the
testing. After the testing procedure has been completed ISTA has a specific checklist that

defines the results of the testing and determination of pass or failure for the system.

2.5 AUTOMATIC IDENTIFICATION AND DATA COLLECTION

Automatic Identification and Data Collection (AIDC) is an industrial vocabulary
term which defines any identification with or without direct collection of data. The
information is usually entered into different micro-processing units, including
programmable logic controllers (PLC) or computer systems without using any physical
input of data (i.e. a keyboard). The Association for Automatic Identification and Mobility
(AIM) describes two common goals of AIDC: (1) elimination of errors associated with

data collection and/or product identification, and (2) acceleration of products through the

13

entire manufacturing and distribution process. AIDC incorporates a family of three
distinct service groups and technologies related to RFID and how its devolvement has
occurred. The umbrella incorporates the following:

1. Barcode Technologies

2. Electronic Article Surveillance

3. Radio Frequency Identification

2.6 BARCODE TECHNOLOGIES

Barcode technology is a form of keyless information transfer. This technology has
been standardized and used commercially for almost 40 years to improve information
management, increase convenience and reduce costs.

In 1966 the first barcode was used commercially. Since there was no
standardization in the barcode area, in 1970 the Universal Grocery Products Identification
Code (UGPIC) was written by Logicon Inc. (Bellis, 2004) In 1972, a Kroger store in
Cincinnati, Ohio, used a bull’s eye code in an effort to begin automating point-of—sale use
of barcodes in their supermarket.

A committee was formed within the grocery industry to devise a standard code for
the grocery industry. On April 3, 1973, a Universal Product Code (UPC) was developed
as the standard. The UGPIC helped develop the UPC symbol standard which is still used
by the United States. (Hagey, 1998) In 1974 the first UPC scanner was implemented into
a grocery store in Troy, Ohio. (Bellis, 2004)

The UPC representation is designed to simplify the ability to be transcribed onto

many different packages, using multiple printing methods. Along with a variety of

14

printing methods all the symbols are omnidirectional, meaning they can be read in any
orientation. Universal Product Code version A (UPC-A) was the first symbol to pass a
read rate of 99 percent while using a fixed laser scanner. These figures also have a
minimal changeover error rate, which is less than one error in 10,000 scanned symbols. In
the United States any institution can obtain a unique six digit identification number by
becoming a member of the Universal Code Council (UCC). The five digit item number is
developed by the individual institution producing the product. (BarCode 1, 2004)

The development and cost of the barcodes have been a relatively low-cost task.
Through the development of barcodes each workstation at individual lines has been able to
purchase and print their own barcode labels to each product, case and pallet. Until the late
1980’s, companies hired people to manually input data to track data and information into
their systems. Now the use of computers allows this same information to be entered into
data bases without the variability of human errors. The main motivation for the
implementation of barcodes within any type of business setting is the ability for improved
precision for data collection. With the added precision, companies have the ability to
accurately report numbers and forecast future demands and operations, which can save

time and money.

2.7 ELECTRONIC ARTICLE SURVEILLANCE

Shoplifting is a major retail nightmare; approximately 60 percent of all consumers
have shoplifted. In 2002 the average amount shoplifted in one given incident was $207.18
up almost six percent from the previous year. Total inventory shrinkage worldwide is

estimated to cost retailers $60 billion a year. (Checkpoint, 2004) With multiple reports

15

showing devastating inventory loss figures, retailers and others have to protect themselves.
One form of AIDC that is used to protect retailers and service providers (i.e. libraries, etc.)
while arming them against theft is a technology called Electronic Article Surveillance
(EAS). An EAS system is built around three main components: a tag (often called a hard
tag) and label, irnmobilizers or disconnection point and detectors.

The tag is a sensor which is attached to the product or merchandise. It can be
deactivated by an immobilizer at the point of purchase. The interrogators are transmitting
and receiving readers placed in a location that gives retailers the ability to monitor product
flow at a given point (i.e. checkout lanes or exits to a store). (AIM, 2004)

An EAS system is built around technology in which a transmitter sends a specified
signal to a receiver. As with any AIDC transmission, the higher the frequency of the
reader, the longer the available read distance. Once the surveillance area has been
defined and the system is broadcasting without a tag in the available read area, there will
be no positive identifiers. If the tag has not been deactivated at a given security point for
acceptable transportation through the surveillance area, that tag will be detected. It is not
until a tag is present in the surveillance area that a positive read will occur. Tag
disturbances are built around non-naturally occurring disturbances (i.e. human interaction,
activities of doors/exits or entrances, the products themselves, etc.). (AIM, 2004) EAS is
a limited system that can only tell when something is in the read field. It cannot transfer

data and it cannot define a product.

16

2.8 RADIO FREQUENCY IDENTIFICATION

Radio Frequency Identification (RF ID) has been described as the future of
manufacturing. RFID is one of the most rapidly growing forms of AIDC and has the
ability to bring automatic identification and data capture to the forefront of technology in
and around manufacturing facilities, warehousing units and distribution centers. This
technology ultimately will increase the convenience of the consumer.

The principal purpose of RFID technology is the ability to locate, track and
identify objects remotely. Unlike barcodes, RFID tag systems do not require line of sight
to transfer the encoded data. RFID is a parallel technology with barcodes which is
designed to identify and track objects which can include products, packaging (containers),
vehicles, animals, and people. RFID can be described as the combined technologies of the
traditional barcode and the advanced EAS system. Both barcodes and EAS are the
stepping stones to RF ID, but RF ID has more capabilities than the two combined.

The RFID technologies, more specifically RFID tags, are relied upon to store,
carry and transmit distinguishable information for each product tagged. These processes
use integrated circuits to transmit, retrieve and store data over electromagnetic radio
waves.

RFID contains three primary hardware components and a specific data information
code. These items include: transponders (tags or labels), interrogators (readers which
gather information), and the central processing unit (CPU) (the mainframe which requests
the data, stores and analyzes collected information). (RF ID Wizards, 2003) All data is

stored within the tag as an electronic product code (EPC).

17

2.8.1 TRANSPONDERS

The transponder is the combined tag technology used to classify an object. The
word transponder comes from the combination of TRANSmitter and resPONDER which
illustrates its purpose. (AIM, 2004) Transponders are made up of two components: an
integrated circuit (IC) and a tag antenna. The IC stores and transmits data through an
attached antenna. A tag antenna can be an etched or printed conductor laid onto a medium
that will allow a completed circuit.

An inlay, also referred to as a transponder, is composed of an IC and an antenna on
a medium backing. In the past, the medium used primarily consisted of a thin plastic sheet
or adhesive label. New developments in technology now allow these antennas to be
printed directly to packaging material with the use of conductive inks.

Before an inlay is fully functional it must go through a conversion process. The
conversion processes include labels, badges, and production of packaging. Once these
conversions have occurred, the term “smart label” is used to refer to the transponder which

is integrated into labels.

2.8.2 INTEGRATED CIRCUIT (IC)

The IC consists of a rrricrochip segment connected to an antenna structure. The
microchip contains all of the programmable information that is contained in a tag. The
informational capacities of the ICs can be categorized into read only (RO), read/write
(R/W) and write once read many (WORM). Dependent upon the factory capacity of a
read only circuit, the IC manufacturer programs the IC permanently setting the

information encoded within it. R/W circuits, on the other hand, have the ability to be

18

programmed and reprogrammed effectively as many times as needed to update the
information on the chip by the user. This amount of use is finite dependent on the

environment in which the tag is used.

2.8.3 POWER SOURCE

Any RFID tag requires power to operate. The level of energy needed to run a tag
is constant and very small. This level can range from the microwatts to milliwatts. There
are two different methods to classify the way a RFID tag obtains its power: passively or

actively.

2.8.3.1 PASSIVE TAGS

A passive tag does not contain a battery. It receives all of the necessary energy
fiom the interrogator (reader). When the interrogator is searching for the tag within its
read field it powers the tag by the radio waves being emitted.

Passive tags can incorporate two categories: battery-less (“pure passive” or “beam
powered”) and containing a battery (“active/passive”). (Chiesa, et al., 2002) Purely
passive tags do not contain any type of onboard energy supply, such as a battery and can
be referred to as “reflective” tags because of their dependence on electromagnetic power
emitted fi'om the interrogator. The interrogator generates the energy needed to allow the
tag to transmit the data stored on its IC. These tags are simple in construction making
them easier to manufacture than other passive tags. The downside for purely passive tags

is the shorter read ranges and the need for higher powered readers compared to active tags.

19

Semi-passive or active/passive tags contain a battery which allows the tags to
function with some of the same features as an active tag (i.e. more improved speed of data
transfer). These semi-active tags communicate by using the electromagnetic power

generated by the reader in the same method as a purely passive tag.

2.8.3.2 ACTIVE TAGS
An active tag is one that has an internal power source as well as a transmitter.
These tags are usually read/write configurations. Due to the additional components, the
tag itself is larger and in some cases the size of a brick. Although active tags are more
bulky than the passive tags, active tags have a much larger effective read range. This
increased read range can be between 100 and 300 feet.
Active tags demonstrate a high power to weight ratio. This shows the reality that
a tag can gain read distances as it increases in size. With the use of low power circuitry an
active tag could have the ability to draw energy for as long as ten years. All operation
standards are dependent upon temperature, manipulation and the quantity of read/write
sequences. Also, when used to power an upper frequency response mode, active tags
boast better noise resistance and much higher data communication transmission rates.
(AIM, 2004) The tradeoff in using active tags is that they are larger in size, cost more to

manufacture and require more maintenance than passive tags.

20

2.8.4 INTERROGATORS/READER

Interrogators, also called readers, are the apparatus in which the tags are powered
and/or activated to capture the information stored on the IC of the tag. Interrogators work
in conjunction with antennas to send and receive the radio signals with information. The
interrogators are connected to a central processing unit (CPU) in order to process or

decipher data Interrogators are available in two forms, portable and stationary.

2.8.4.1 PORTABLE INTERROGATORS

A portable interrogator is usually small enough to be handheld. The antenna is
incorporated within the reader. The portable interrogators are battery powered which limit
their range from seven inches up to 10 feet. (RF ID Wizards, 2003) These handheld units

are usually equipped to read barcodes as well.

2.8.4.2 STATIONARY INTERROGATORS

Stationary interrogators are usually large, bulky items that are installed or
relocated only a handful of times. A stationary interrogator is strategically positioned at
important product movement points. These points can include portals (into and out of
dock doors, distribution areas and within warehouses), a place information is to be
gathered (i.e. on an assembly line), and even advantageously positioned within a room.

Some of the stationary interrogators can be retrofitted for wireless applications
with the addition of technology based on the 802.11 standards. By making these changes
the interrogator can communicate with a database without the need for a dedicated

computer or even network cables. (RF ID Wizards, 2003)

21

2.8.5 ELECTRONIC PRODUCT CODE (EPC)

EPCs are based on current concepts of the UPC identification code. This new
EPC system defines a way to identify all material objects. These next generation codes
are used exclusively in RF 1D. This identification system has the ability to define a
product’s manufacturer, individual product group and can assign a specific number to each
product. With an EPC identifier every individual product will have its own specific
tracking number. This character sequence can identify when the product is packaged or
produced and recognize the product’s exact location within a comprehensive supply chain.

This form of product code has the ability to incorporate a vast range of
information. The assortment of identifying codes can encompass 16 million different
products specific to one manufacturer, along with individually identifying more than one
trillion individual products in each product specific group. Unlike a UPC, the EPC
embeds its information onto an electronic tag, also referred to as a smart tag. The existing
development of these smart tags will communicate with a data base system using Radio
Frequency (RF), which allows for automatic scanning. The ability to incorporate
automatic scanning to any system eliminates manual inputs of UPC codes and allows
every product to be tracked at any point throughout the supply chain. (Mullin, 2002)

Currently, the main drawback of the implementation of these EPC enabled smart
tags is cost. The goal for each smart tag is to cost pennies on the dollar or less. This is not
currently obtainable, but with the development of new manufacturing procedures for smart
tags the goal can be reached in the near future. At that point the tags may ultimately cost

less than the projected pennies on the dollar.

22

2.8.6 STANDARDS/REGULATIONS

Current RFID standards do not encompass every aspect of the technology because
it is still deve10ping. The divisions of technology that are included comprise protocols
(communications), data transfer and frequency. These comprehensive areas are referred to
as technology standards but they neglect to incorporate the actual use of the technology or
how it fundamentally works. The standards that define actual use of the technology are
referred to as application standards. When building an infiastructure of the standards,
different organizations and geographic boundaries are incorporated. Standards
unfortunately are limited by land mass and these limitations can be broken down into

national, regional or international areas.

23

CHAPTER 3: METHODOLOGY
3.1 INTRODUCTION

Packaging protects products. RF ID testing is a necessary step in developing
packaging and technology. The testing presented in this project will show how dynamic
packaging distribution situations have an impact on Class 0 label inlay RFID tags.
Matrics® is the company that was chosen to help develop an understanding of ultra high
frequency 915 Megahertz (MHz) RFID tags and packaging distribution dynamic
interaction. Matrics® has a unique knowledge and understanding of the development for
RFID technology and RFID systems and is a leading provider of RFID products to
Wal-Mart.

Reader:

The reader used in the testing was a stationary Matrics® SR 400 reader. This
reader has the ability to work with both Class 0 (Read-Only) and Class 0+ (Read/Write)
tags. The SR 400 reader can handle up to a four antenna configuration which was used in
reading the tags. This reader was chosen because of its accuracy and the compliance to
the Wal-Mart RFID mandates for January 2005.

Antenna(s):

The antennas used in the testing were the Matrics® High Performance Area
Antennas. The polarized antennas operate with long range and large area applications
where a larger read zone is crucial. The antennas were chosen because they have an

application directly related to logistical and distribution situations.

24

RFID Tags:

The testing was completed on 915 MHz Matrics® Class 0 Inlay Labels
(X1019-LBL). The Matrics® 915 MHz inlays were converted into labels by adding an
adhesive backing and a printable top label. These tags contain an IC, antenna, adhesive
backing, poly-coated backing sheet and printable top label.

The “slap-and-ship” flexibility of these tags allows them the versatility to be put
into use by any company in the beginning RFID phase. Given the versatility of these tags,
no matter the size or volume of the product, they can be effortlessly implemented while
being run through any printing systems for tracking purposes. Figure I shows the

structure of the 915 MHz Matrics® class 0 tags.

{—— Printable Label

 

V
Poly-coated Backing Label

Figure 1: 915 MHz Matr'ics® X1090-LBL Class 0 Label Inlay RFID Tag

25

These tags were placed onto a medium (i.e. corrugated board) to complete the
RFID/packaging product system. The most common corrugated board fluting is C-Flute
which was used for the test procedures. Measurements for the structures were done in
mils. One mil is equivalent to l/ 1000 inch. The corrugated board was determined to have
an average thickness of 156.25 mils. The tag structure thicknesses were measured to
understand concentration stress points of the tag. The adhesive backing had an average
thickness of 2.84 mils. The antenna had an average thickness of 5.35 rrrils. The IC was
deterrrrined to have an average thickness of 14.34 mils. The structure measurements are

described in Table 1.

Table 1: Structure Measurements of Label and Corrugated Board

 

 

 

 

 

 

 

 

Structure Thickness (mils) Thickness (inches)
Adhesive Backing 2.84 0.00284
Antenna 5.35 0.00535
Integrated Circuit 14.34 0.01434
C-F lute Corrugated 156.25 0.15625

 

The average total thickness of the corrugated board and the tag was 177.53 rrrils thick.
Figure 2 depicts the layering from the IC to the corrugated board. As the tag and
corrugated board are placed on one another, the thickness plays a role for the protection of
the tag and added distance from the product to reduce interaction or limitations caused by
the product (e.g. metal and high water content products). Current ultra high frequency
RF ID tag technology requires a distance of 1/8 inch from any metal or high water content

product for readability.

26

 

 

lnteg rated Circuit

 

T

 

 

177.53 mils Antenna
l — Adhesive Backing
dQAng C-Flute Corrugated

 

 

 

Figure 2: Total Thickness of Matrics® Inlay Label and Corrugated System

Non-RFID pressure sensitive labels do not contain additional areas that extend
above the label surface allowing these labels to meld seamlessly to the packaging. On the
Matrics® RFID class 0 tags the IC stands above the surface of the label. This point
possibly creates an area where stress and impact can damage the operation of the tag.
Figure 3 is representative of the protrusion distance for the IC chip from the top of the

antenna.

 

T

21.28 mils

_i_

  
 

 

. if " 14.34mils

  

Figure 3: 915 MHz Matrics® IC Protrusion from Adhesive Label

Testing was developed to understand the ruggedness of the IC chip itself as well as
the antenna connection. The testing allows for determination of the effectiveness and

weakness of the RFID tags and sheds light on possible solutions to future problems.

27

By introducing the worst-case scenario for distribution, the tag’s structure will be
firlly tested allowing a higher confidence level for future use. Understanding the
ruggedness will allow more fieedom of tag placement. Although this testing is not applied
to orientation and other material issues, it will alleviate some concerns about distributional
effects on the tags.

All of the testing methods were qualified by the sole ability of the RFID inlay tag
to be read by the system. If the testing on the tag failed to kill a tag at any level, the tag
was deemed to pass that series of tests. No confirmation was done to determine whether
any of the tests affected tag read ranges or multiple read errors.

The static mass used in the testing was a Brunswick bowling ball which provides a
constant weighted mass to emulate a full case. The ball had an actual weight 15.94
pounds and was an intensity black sparkle with a model number of 60-101876-936
(lJ 97979).

The case dimensions were 13 inches by 12.25 inches by 10.25 inches with the
manufacturer’s joint on the outside of the case. The tag position was located from the
right edge, two inches left and two inches down from top of the case. The cases were
labeled with Matrics® class 0 tags. The tags were adhered to the cases facing each other to
create a friction point where the tags came into contact. Before the testing began the
weighted case was identified by labeling each of the faces. Figure 4 shows the numbering

system for the case and location of the Matr'ics® tag.

28

 

 

4 —> E

5/

   

Manufacturer’s Joint

 

 

Comer 2-3-5

w—h

Figure 4: Numbering Sequence for Weighted Case

3.2 VIBRATION

3.2.1 SINUSOIDAL RESONANCE OF INDIVIDUAL CASES

 

Figure 5: Individual Cases

29

Vibration tests were done to understand if constant IC contact rubbing can
dislodge the IC from the tag antenna to cause failure. Testing began by stacking two cases
on top of each other with the IC centered on the safety rail. The top case on one of the
stacks was weighted with a static mass (the same as described in section 3.1).

Using a Lansmont Corporation 10000-10 vibration tester, the stacked cases were
tested for natural frequency with ASTM D-999. The frequency sweep followed the range
of 3-300 Hz with an Input level of 0.5 G with a logarithmic sweep type and rate of 1.5
octaves per minute. The natural frequency or resonance fi'equency was determined to be
11.0 Hz, and the total sinusoidal vibration duration for testing was 60 minutes. The cases
ran for two 60 minute increments. In between these tests the cases were visually inspected
and checked for the readability of the tags. Failures were determined by the tag’s inability

to be read.

3.2.2 SINUSOIDAL RESONANCE OF CASES AGAINST PLYWOOD

 

Figure 6: Stacked Cases Against Plywood

30

The testing was done to determine if constant IC contact during vibration against
plywood testing can dislodge the IC. Testing began by stacking two cases. The top case
was weighted with a static mass in direct contact with an unfinished piece of plywood to
imitate the interior walls of a semi-trailer. The plywood dimensions were 30 inches by 16
inches by 0.75 inches with the grain of the wood running in a vertical direction. The mass
and the set up for both cases were the same as described in section 3.1.

The tags were adhered to the cases facing the plywood to create a friction point
between the tags and the plywood. Using a Lansmont Corporation 10000-10 vibration
tester, the stacked cases were tested for natural frequency with ASTM D-999. The sweep
followed the range of 3-300 Hz with an Input level of 0.5 G, a logarithmic sweep type and
rate of 1.5 octaves per minute. The natural frequency or resonance frequency was
determined to be 11.0 Hz. The total sinusoidal vibration duration for testing was 60
rrrinutes. After completion of the tests the cases were visually inspected for damage and
were checked for the readability of the tags. Failures were determined by the tag’s

inability to be read.

3.2.3 RANDOM VIBRATION OF FOUR COLUMN STACKED CASES

 

Figure 7: Four Column Stacked Cases

31

Vibration tests were done with eight cases to understand if constant IC contact
abrasion can dislodge the IC. The cases were stacked in two columns, with each column
four cases high. Each case contained two 2-gallon high density polyethylene (HDPE)
containers that were half full. The case dimensions were 15 inches x 10 inches x 13.5
inches with the manufacturer’s joint on the inside of the case.

The tags were adhered to the cases facing each other (on the inside of the columns)
to create a fiiction point between the tags. The tags were placed on the main display
panels. Their location was from the right edge two inches left and two inches down from
the top of the case.

The second row from the top was banded with string and a rubber band to create
tension for maximum abrasion between the cases. Using a Lansmont Corporation 10000-
10 vibration tester, the stacked cases were tested in a random vibration sequence. The
ASTM D—4169 test ran for three hours in a truck spectrum with an assurance level of one.
The spectrum demand had a Gm, value of 0.73. Table 2 shows the demand plot for the

testing.

Table 2: ASTM-4169 Assurance Level I Demand Profile.

 

 

 

 

 

 

 

Frequency PSD (GZ/Hz) Slope (db/oct)
1.0 0.000100 11.47
4.0 0.020000 0.00
16.0 0.020000 -7.54
40.0 0.002000 0.00
80.0 0.002000 -15.08
200.0 0.000020 0.00

 

 

 

 

 

32

3.2.4 ISTA 3C VIBRATION AND DROP SEQUENCE

Testing a package with an RFID tag for total single parcel distribution
environment is difficult to qualify. The ISTA 3C procedure was developed to qualify
packaged products for single parcel shipments which are less than 150 pounds. The
sequence for testing goes through a conditioning period, a drop sequence, a vibration
sequence and a final drop sequence. Vibration and drop sequences are part of the ISTA
3C procedure to qualify packaging for distribution.

Vibration and drop tests were done to understand if the Matrics® class 0 tags are
rugged enough to withstand single parcel simulation. One of the cases was weighted with
a static mass. The mass and set-up of the cases were the same as described in section 3.1.

The RFID tag was placed on face number five according to Figure 4. The drop
sequence was chosen for a product that weighs 50 pounds or less. This sequence of six
drops began from a height of 15 inches while the seventh and final drop was from a height

of 30 inches. Table 3 shows the ISTA 3C drop sequence.

Table 3: ISTA 3C Drop Sequence for Products < 50 lbs.

 

 

 

 

 

 

 

 

 

 

 

Drop Number Drop Height (<50 lbs) Specimen l

1 15 inches (380 mm) Edge 34
2 15 inches (380 mm) Edge 3-6

3 15 inches (380 mm) Edge 4-6
4 15 inches (380 mm) Corner 3-4-6
5 15 inches (380 mm) Comer 2-3-5
6 15 inches (380 mm) Face 3

7 30 inches (3 80 mm) Face 3

 

 

The vibration portion of the testing was done with two cases facing each other on

the vibration table. The set-up follows testing sequence two. This vibration sequence

33

differs from sequence two because it was completed with a random vibration spectrum
with an overall G.ms level of 0.53. The random vibration portion of the testing was done

for 60 minutes. Table 4 describes the test sequence for the ISTA 3C testing method.

Table 4: ISTA 3C Test Method

 

Sequence Description of test

1 hour of random vibration (ASTM D-4169 Assurance level I Spectrum)
Inspection of tag for failure

ISTA 3C drop sequence (7 drops)

Inspection of tag for failure

1 hour of random vibration (ASTM D-4l69 Assurance level 1 Spectrum)
Inspection of tag for failure

ISTA 3C drop sequence (7 drops)

Inspection of tag for failure

1 hour of random vibration (ASTM D-4169 Assurance level I Spectrum)
Inspection of tag for failure

 

 

 

 

 

 

 

 

 

 

 

 

 

gxoooxrcnmrswm._.

 

Once the 60 minute vibration duration was completed the case was inspected to
determine if any visible damage had occurred. The tag was then read by the RF ID system.
If the tag was still working after the random vibration portion of the testing, the first
sequence of drops was repeated. Once the second sequence of drops had been completed
the entire case was inspected. The RF ID tag was again evaluated to determine if it was
still working. Once these sequences had been completed the testing was finished. The
perimeter of the cases was guarded by safety rails to keep columns from excessive

horizontal movement.

34

3.3 DROP

3.3.1 INCREMENTAL DROP SEQUENCE (DETERMINATION OF FAILURE)

 

Figure 8: Drop Testing

The incremental drop test was done to understand direct drop impact between a
static mass and a seismic mass and to determine if a threshold drop height was apparent
for the failure of tags. The incremental drop height sequence used one case with five of
the sides having RFID tags adhered directly to the corrugated case. Drops were complete

either after tag failure or the case was no longer a usable cube.

The drop sequence was as follows:

. 18 inch drop (Face 1)
. 24 inch drop (Face 3)
. 30 inch drop (Face 4)
. 36 inch drop (Face 5)
. 42 inch drop (Face 6)

UrJr-UJNt-I

35

3.3.2 DROP SEQUENCE

The Progressive Increase Drop Sequence test was done based on the previous
understanding of the incremental drop test results. The lowest three drop heights which
caused failure were re-created with individually tagged cases for each determined drop
height. The progressive increase drop sequence used three separate cases with a RFID tag

adhered directly to face four according to Figure 4. All drops were done until failure.

The drop sequence was as follows:

1. 18 inch drop (Face 4)
2. 24 inch drop (Face 4)
3. 30 inch drop (Face 4)

3.4 SHOCK

Shock sequence tests were performed in order to manipulate a direct impact on a
specific area (the IC) of the four inch by six inch label Matrics® class 0 tag. A Lansmont
65/81 Shock Test Model shock tester was used for the testing. The impacting surface of
the shock machine measured 25 inches wide by 32 inches deep. On the surface of the
shock table, a 9.91 mv/ g piezoelectric accelerometer was mounted. Piezoelectric materials
operate by ejecting an electrical charge when stressed by a mechanical force. Some of
these materials include quartz, human skin and ceramic. (Stamer, 1996)

The quartz accelerometer contains a small steel mass attached to the quartz. As
the shock table creates a force which acts on the steel mass, the force generates a
movement of electrons. This movement creates a small electrical current through the

connecting wires at either end of the accelerometer. The accelerometer was used to

36

measure the impact signal of the table. The magnitude of the current is proportional to the
magnitude of the acceleration or shock of the table.

The accelerometer was placed on the top of the table four inches in the y-axis and
even with the base of the wooden testing fixture shown in Figure 9. The accelerometer
records the shock produced by the table when impacting the plastic programmers. The
accelerometer was mounted parallel to the impacting surface’s axis, recording the impulse
of the anticipated shock. The entire shock pulse is recorded by the accelerometer. The
generated shock pulse was analyzed by Test Partner version 3 (T P3) software developed

by the Lansmont Corporation.

 

Figure 9: Accelerometer Placement on Shock Table Impact Surface

In order to allow a direct impact of the IC, a wooden test fixture was built and
mounted to the surface of the shock table. The outside dimensions of the fixture were 10.5
inches by 10.5 inches by 10.5 inches with the total fixture being built out of 0.75 inch
plywood as shown in Figure 15. This fixture allowed the direct and repeatable centering
of the mass on the IC of the tag. On the rebounding surface (top interior panel) of the

fixture a polypropylene foam pad was added to reduce bouncing and eliminating excessive

37

multiple impacts to the tag. This foam pad was 8.00 inches by 8.00 inches by 0.25 inches

in thickness.

 

Figure 10: Wooden Shock Machine Fixture with Polypropylene Pad

The test fixture is open on two sides: the bottom panel and the panel facing the
user. The fixture was held to the shock table by two 3/8 inch diameter threaded rods. The
rods were screwed completely into the table and secured with large steel wing nuts.
Across the top of the fixture an aluminum plate measuring 30 inches by four inches by
0.50 inches with a slot running down the center of the plate where the threaded rod was
fed and secured by two large steel wing nuts.

Inside the fixture a static mass was used for the point impact testing. The static
mass used in the testing was the same Brunswick bowling ball as stated previously, which
provided a constantly weighted mass to emulate a full case.

The four inch by six inch label Matrich tag was adhered to ZOO-pound Mullen
burst strength, C-flute corrugated board. The dimensions of the pads were 9.625 inches by
8.875 inches by 0.15625 inches in thickness. The flutes of the board ran parallel to the

longest dimension of the pad. The pads were cut on an Artios sample table. The centers

38

of each pad were marked on the Artios Table (with pen) to the exact center of pad. The
marked side was transcribed to other side to understand where the center was on both
sides of the pad.

The outside paper facing the board had a higher basis weight for printing and
strength. The inner paper facing was established because of visual signs of the fluting
showing through the paper. During testing the IC was in direct contact with the face of the
shock table and the mass was placed on the opposite side of the pad centered over the IC.
When loading the pad the direction of the fluting was always parallel with the outside
walls of the wooden fixture.

The placement of the IC in the tag is one of three locations on the corrugated
board: 1) centered over the peak; 2) centered over the valley; or 3) centered over the
slope of the fluting. When stated in the testing results the four inch by six inch tags were
placed with the longest dimension running either in the same direction as the fluting (i.e.
the six inch length was in the same direction as the fluting) or 90 degrees to fluting (i.e.
the six inch length was perpendicular to the direction of the fluting). Figure 11 shows the

definition of this placement.

 

.. . —- m9, ‘m-fi-i".

 

 

 

Figure 11: Definition of Placement According to Corrugations

39

3.4.1 INITIAL SEQUENCE

The intention of this test was to develop direct impacts of the static mass and the
1C of a four inch by six inch label Matrics® class 0 tag. The mass, the wood fixture and
the corrugated pad and set-up were all the same as described in section 3.1.

As part of the ISTA 3C drop height sequence, the free fall drop heights begin at 15
inches and finish with a 30 inch drop. Free fall drop heights can be calculated with the
Lansmont Corporation 65/81 Shock Test Model machine. The shock pulse area or AV
(change in velocity) can be used to mathematically derive the equivalent acceleration due
to gravity free fall height. Equation 1 is used to derive the free fall drop heights according
to shock machine results. Table 5 shows the tag numbers and drop heights of the shock

table.

AV

. /2gh Equation 1

Table 5: Initial Drop Heights

T Number

 

1 5
2 6
3 7
4 8
5 9
6 5
7 6
8 7
9 8
10 9

40

The testing was done with 10 separate impacts to each tag. After each impact the
tag was inspected to determine if visible damage had occurred. After visual inspection
was completed the tag was read by the RFID system. Ifthe tag was still working the next
drop at the same height was completed. This sequence was repeated until the tag failed or
a total of 10 drops had occurred. Once these sequences had been completed the testing

was finished.

3.4.2 LOAD SPREADER

 

Figure 12: Load Spreaders

Shock sequence tests were performed in order to manipulate an impact on a
specific area of a product using an apparatus for spreading the total load area. The intent
of this testing was to develop an impact of a static mass with a larger bearing area or larger
static stress onto the four inch by six inch label Matrics® class 0 tag. The mass was the
same as described in section 3.1.

Two separate load Spreaders were used in the testing. The first was a square block
3.05 inches by 3.05 inches by 0.75 inches thick made from Rennwood 450 material. The

impacting surface was 9.283 square inches. The second load spreader was a circular block

41

2.3125 inches in diameter and 0.75 inches thick made from Rennwood 450. The
impacting surface was 4.2 square inches.

The wooden fixture was the same as described in section 3.1. The fixture had no
headspace between the bottom of the foam pad and the top of the mass when either load
spreader was placed on the corrugated pad. The same corrugated set-up is described in
section 3.1. Table 6 shows the tags with load Spreaders and drop heights of the shock
table.

Table 6: Load Spreader Drop Heights

#

 

|
l
l

I
I
l
|
l
I

\rjcxlmthlw N —-

l
l

l
l
l

l
1
col
l

 

 

The testing was done with 17 separate impacts to each tag. After each impact the
tag was inspected to determine if any visible damage had occurred. After visual
inspection was completed the tag was read by the RFID system. If the tag was still

working the next drop was completed. This sequence was repeated until the tag failed or a

42

total of 17 drops had occurred. Once these sequences had been completed the testing was

finished.

3.4.3 THRESHOLD LEVEL DETERMINATION

In this testing the intention was to develop a failure threshold height for impact of
a static mass on the four inch by six inch label Matrics® class 0 tag. The threshold would
determine a height where the survivability of a tag increases. The mass, the wood fixture
and the corrugated pad and set-up were all the same as described in section 3.1.

The testing began with a drop height of six inches. The table height was
sequentially lowered using drop heights of four, three and two inches. The testing was
done with a single impact to each tag. After each impact the tag was inspected to
determine if any visible damage had occurred. After visual inspection was completed the
tag was read by the RFID system. All failures were noted. Once the sequence of testing

was completed the testing was done.

3.4.4 INITIAL PEAK/V ALLEY DETERMINATION

This testing sequence was done to develop an understanding of tag placement in
accordance with the peak or valley of the fluting. The testing would determine if the
placement over the peak or valley would allow different heights where survivability of the
tag increases. The mass, the wood fixture and the corrugated pad and set-up were all the
same as described in section 3.1.

The testing began with a shock table drop height of four inches. The table height

was sequentially lowered using drop heights of three and two inches. The testing was

43

done with a single impact to each tag. After each impact the tag was inspected to
determine if any visible damage had occurred. After visual inspection was completed the
tag was read by the RF ID system. All failures were noted. Once these sequences were

completed the testing was done.

3.4.5 INITIAL RANDOM TAG PLACEMENT DETERMINATION

This testing was to develop an understanding of tag placement in accordance with
the slope of the fluting. The testing would determine if the placement of the tag over the
slope of the peak or valley would show different survivability. The mass, the wood fixture
and the corrugated pad and set-up were the same as described in section 3.1.

The testing was done with a shock table dr0p height of two inches with a single
impact to each tag. After each impact the tag was inspected to determine if any visible
damage had occurred. After visual inspection was completed the tag was read by the
RFID system. All failures were noted. Once these sequences were done the testing was
completed. It was noted some of the tags were placed 90° to the direction of the fluting

with the IC still over the slope of the flutes.

3.4.6 ALUMINUM FIXTURE

The intention of this testing was to develop an understanding about eliminating the
headspace available to determine if the headspace was a cause for failure of the tag
survivability. The mass was the same as described in section 3.1. In order to allow a
direct impact of the IC, an aluminum plate (the same as described in section 3.1) with a

centering cap was used. The centering cap was made from Rennwood 450 and had the

same dimensions as the square load spreader stated in section 3.4.2. This fixture allowed
the direct and repeatable centering of the mass onto the IC. The fixture had no headspace
between the aluminum plate and the top of the mass when placed on the corrugated pad in
the fixture. The corrugated pad and set-up was the same as described in section 3.1. The

testing was done with drop heights of three and four inches with a single impact to each
tag. After each impact the tag was inspected to determine if any visible damage had
occurred. After visual inspection was completed the tag was read by the RFID system.

All failures were noted. Once these sequences were completed the testing was done.

3.4.7 WOOD FIXTURE, NO HEADSPACE

This testing sequence was to develop an understanding about eliminating the
headspace. The testing would determine if the headspace was a cause for failure of the tag
survivability. The mass, the wood fixture and the corrugated pad and set-up were all the
same as described in section 3.1. The fixture had no headspace between the bottom of the
foam pad and the top of the mass when placed on the corrugated pad. The headspace was
filled with five untested corrugated pads with the same characteristics as in section 3.1.

The testing was done with a drop height of four inches with a single impact to each
tag. After each impact, the tag was inspected to determine if any visible damage had
occurred. After inspection was completed the tag was read by the RF ID system. All

failures were noted. Once these sequences were completed the testing was done.

45

3.4.8 IC CENTERED OVER PEAK (Tag Orientation: with Fluting)

The intention of the testing was to develop an understanding about impacting a
mass on the tag with the IC directly centered over the peak of the fluting and to determine
if this placement is a cause for failure. The mass, the wood fixture and the corrugated pad
and set-up were all the same as described in section 3.1. The testing was done with a drop
height of two inches with a single impact to each tag. After each impact the tag was
inspected to determine if any visible damage had occurred. After inspection was
completed the tag was read by the RFID system. All failures were noted. Once these

sequences were completed the testing was done.

3.4.9 IC CENTERED OVER VALLEY (Tag Orientation: with Fluting)

This testing was done to develop an understanding about impacting a mass on the
tag with the IC directly centered over the valley of the fluting and to determine if this
placement was a cause for failure. The mass, the wood fixture and the corrugated pad and
set-up were all the same as described in section 3.1. The testing was done with a drop
height of two inches and with a single impact to each tag. After each impact the tag was
inspected to determine if any visible damage had occurred. After visual inspection was
completed the tag was read by the RFID system. All failures were noted. Once these

sequences were completed the testing was done.

46

3.4.10 IC CENTERED OVER PEAK (Tag Orientation: 90° to Fluting)

This testing sequence was to develop an understanding about impacting a mass on
the tag with the IC directly centered over the peak of the fluting and determined if this
placement was a cause for failure. The mass was the same as described in section 3.1.

In order to allow a direct impact of the IC, a wooden test fixture was modified and
used. The outside dimensions of the fixture were 10.5 inches by 10.5 inches by 9.75
inches, with the total fixture being built out of 0.75 inch plywood. This fixture allowed
the direct and repeatable centering of the mass onto the IC. The inside top surface of the
fixture was lined with a polyethylene foam pad which was eight inches by eight inches
and 0.25 inches thick. The foam pad was used to dampen multiple bounces of the mass.
The fixture had no headspace between the bottom of the foam pad and the top of the mass
when placed on the corrugated pad. The corrugated pad and set-up was the same as
described in section 3.1. The testing was done with a drop height of two inches with a
single impact to each tag. After each impact, the tag was inspected to determine if any
visible damage had occurred. After visual inspection was completed the tag was read by
the RFID system. All failures were noted. Once these sequences were completed the

testing was done.

3.4.11 IC CENTERED OVER VALLEY (Tag Orientation: 90° to Fluting)

This testing sequence was to develop an understanding about impacting a mass on
the tag with the IC directly centered over the valley of the fluting and to determine if this
placement was a cause for failure. The mass and the corrugated pad and set-up were the

same as described in section 3.1. The wood fixture was the same as described in section

47

3.4. The testing was done with a drop height of two inches with a single impact to each
tag. After each impact the tag was inspected to determine if any visible damage had
occurred. After inspection was completed the tag was read by the RFID system. All

failures were noted. Once these sequences were completed the testing was done.

3.4.12 IC ON SLOPE OF FLUTING (Tag Orientation: with Fluting)

The intention of this testing was to develop an understanding about impacting a
mass on the tag with the IC over the slope of the fluting and to determine if this placement
was a cause for failure. The mass and the corrugated pad and set-up were all the same as
described in section 3.1, while the wood fixture was the same as described in section 3.4.

The testing was done with a drop height of two inches with a single impact to each
tag. After each impact, the tag was inspected to determine if any visible damage had
occurred. After inspection was completed the tag was read by the RFID system. All

failures were noted. Once these sequences were completed the testing was done.

3.4.13 IC ON SLOPE OF FLUTING (TAG ORIENTATION: 90° TO FLUTING)

In this sequence the intention was to develop an understanding of impacting a
mass on the tag with the IC over the slope of the fluting and to determine if this placement
was a cause for failure. The mass and the corrugated pad and set-up were all the same as
described in section 3.1, while the wood fixture was the same as described in section 3.4.

The testing was done with a drop height of two inches with a single impact to each

tag. After each impact the tag was inspected to determine if any visible damage had

48

occurred. After inspection was completed the tag was read by the RFID system. All

failures were noted. Once these sequences were completed the testing was done.

49

CHAPTER 4: RESULTS

Table 7: Complete Results for All Testing

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

# type Test Description Pass % Fail % Samples
1 Vibration Sinusoidal Vibe of Individual Cases 100 0 3
2 Sinusoidal Vibe of Individual Cases Against Plywood 100 0 2
3 Random Vibration of Four Column Stacked Cases 100 0 8
4 ISTA 3C Vibration and Drop Sequence 100 0 2
5 Drop Incremental Drop Sequence Determination for Failure 40 60 5
6 Incremental Drop Seflence 0 100 3
7 Shock Initial Sequence 0 100 20
8 Round Load Spreader 100 0 l6
9 Square Load Spreader 100 0 16
10 Threshold Level Determination 6" 5 95 20
ll Threshold Level Determination 4" 50 50 10
12 Threshold Level Determination 3" 50 50 10
13 Threshold Level Determination 2" 50 50 10
14 Initial Peak/V alley Determination 4" 3O 70 10
15 Initial Peak/Valley Determination 3" 45.45 54.55 ll
16 Initial Peak/Valley Determination 2" 0 100 10
17 Initial Non-Peak/Vallgy Determination 2" 47.37 52.63 l9
18 Aluminum Fixture 4" 40 60 5
19 Aluminum Fixture 3" 40 60 5
20 Wood Fixture No Headspace 40 6O 5
21 A lC Centered over peak (with fluting) 33.33 66.67 30
22 B [C Centered over valley (with flutifl) 40 60 30
23 C IC Centered over Slope (with flutifi) 40 60 30
24 D 1C Centered over peak (90 degrees to fluting) 60 40 30
25 E IC Centered over valley (90 degrees to fluting) 36.67 63.33 30
26 F [C Centered over slope (90 degrees to fluting) 53.33 46.67 30

 

4.1 VIBRATION
4.1.1 SINUSOIDAL RESONANCE OF INDIVIDUAL CASES

In the sinusoidal resonance procedure of individual cases four tags were tested.
None failed. The abrasion generated in this test was not sufficient to dislodge the IC from
the antenna. The weighted cases were not in constant contact during the testing period.

The raw data is presented in Appendix A.

50

 

4.1.2 SINUSOIDAL RESONANCE OF INDIVIDUAL CASES AGAINST PLYWOOD
Two tags were tested in the sinusoidal resonance of the individual cases against an
abrasive piece of plywood, and none failed. The abrasion between the plywood and the IC
caused by the vibration was not sufficient to dislodge the 1C from the antenna. The
plywood had a very coarse surface but the vertical distance or stroke of the case was not
sufficient to cause failure. The case was not in constant contact with the plywood during

the duration of the test. The raw data is presented in Appendix A.

4.1.3 RANDOM VIBRATION OF FOUR COLUMN STACKED CASES

In the random vibration test with four cases in a two column stacked configuration,
eight tags were used for the testing, none failed. The additional tension with the twine and
rubber band was not sufficient to dislodge the IC from the antenna. The columns did not
have enough vertical movement to cause damage. The movement needed for the cases to

cause damage would be out of the scope of conclusive laboratory testing, see Appendix A.

4.1.4 ISTA 3C VIBRATION AND DROP SEQUENCE

The ISTA 3CVibration and Drop Sequence used two tags for testing, none failed.
The drop sequence was the most likely area to show failure. Due to the numbering of the
case, in accordance to the ISTA 3C procedure, the face the tag was adhered to never saw a
direct impact during the drOp sequence. Drops directly on the chip will have an impact on

tag failure, see Appendix A.

51

4.2 DROP
4.2.1 INCREMENTAL DROP SEQUENCE DETERMINATION FOR FAILURE
The incremental drop sequence was to determine a drop height failure for the five
tags tested. Three of the five tags were tested to failure. The tag tested from 18 inches
failed, two dr0ps were proficient. The tag tested from 24 inches did not fail, six dr0ps
were completed. The tag tested from 30 inches failed, one drop was proficient. The tag
tested from 36 inches did not fail, four drops were completed. The tag tested from 42
inches failed, one drop was proficient. During this test each side was dropped until failure
or until the case was no longer in a cubical shape. All but the final impacts were not
directly on the IC because of the drop tester variability and the raw data is presented in

Appendix B.

4.2.2 INCREMENTAL DROP SEQUENCE

In the incremental drop sequence three tags were tested. All three of the tags were
tested to failure. The tag tested from 18 inches failed on drop eight, the tag tested from 24
inches failed on drop nine and the tag tested from 30 inches failed on drop three. During
this test each of the three cases was dropped until failure or until each case was no longer
in a cubical shape. All impacts before the final impact were not directly on the IC and the

raw data is presented in Appendix B.

52

4.3 SHOCK
4.3.1 INITIAL SEQUENCE

In the initial impact sequence 20 tags were tested. During this test all 20 tags were
impacted until failure. This test was to determine any correlation between drop heights
and failure points. Table 8 shows the drop heights and impacts for failure. The raw data is

presented in Appendix C.

Table 8: Drop Failure Totals (*Note Tags Failed on Drop Number Shown)

Drop Height Drop Height
in le 1n

10

11

12

1 3

14

10
11
l2
l3
l4

 

i—aNn—ai—thi—a—Nw

\DOO\)O\Ut\OOO\IO\UI
v—n—NNNu—an—ANI—si—s

1

2

3

4

5

6

7

8

9
10

4.3.2 LOAD SPREADER

The square load spreader impact sequence tested one tag. During this test the tag
was impacted one time at each table drop height to induce failure. The square load
spreaders produced no failures at any heights. The test was only performed until 18 inches
because of the relevance to real world applications. The tag was centered on the flutes
(either on the peak or valley).

The round load spreader impact sequence tested one tag. During this test the tag

was impacted one time at each table drop height trying to induce failure. The round load

53

spreaders produced no failures. The test was only done until 18 inches because of
relevance to real world applications. All tags were centered on the flutes (either on the

peak or valley).

4.3.3 THRESHOLD LEVEL DETERMINATION

The threshold level determination was done in four parts. At a table height of six
inches, 20 tags were tested. The test produced 19 failures. For the threshold level
determination sequence done at a table height of four inches, 10 tags were tested. The test
produced five failures. In the threshold level determination sequence done at a table
height of three inches, 10 tags were tested. The test produced five failures of the 10 total
tags tested. In the threshold level determination sequence done at a table height of two
inches, 10 tags were tested. The test produced five failures of the 10 total tags tested. For

all these tests the data is presented in Appendix C.

4.3.4 INITIAL PEAK/V ALLEY DETERMINATION

In the initial peak/valley determination sequence done at a table height of four
inches, 10 tags were tested. The test produced seven failures. The differentiation when
combining peak and valley showed higher failure percentages than randomly placed tags.
The results showed that there might be a difference between peak and valley placement
when adhering the tag to the corrugated board. The raw data is presented in Appendix C.

In the initial peak/valley determination sequence done at a table height of three
inches, 10 tags were tested. The test produced five failures of the 10 total tags tested. The

differentiation when combining peak and valley showed no higher failure percentages than

54

randomly placed tags. The results showed that there could be a difference for protection
of the tag because of the drop height. The raw data is presented in Appendix C.

In the initial peak/valley determination sequence done at a table height of two
inches, 10 tags were tested. The test produced 10 failures. The differentiation when
combining peak and valley showed total failure percentages. The results showed that
there was an outside factor present in the testing. The evaluation was the headspace in the

fixture needed to be eliminated. The raw data is presented in Appendix C.

4.3.5 INITIAL NON-PEAK/V ALLEY DETERMINATION

In the initial non-peak/valley determination sequence done at a table height of two
inches, 19 tags were tested. The test produced 10 failures. The differentiation when
placing tags and the IC does not fall on a peak or a valley showed higher failure
percentages. The results showed that there was an outside factor present in the testing.
The evaluation was done to eliminate the possibility that the non-peak or non-valley might

protect the IC better than peak or valley. The raw data is presented in Appendix C.

4.3.6 ALUMINUM FIXTURE

In the aluminum fixture drop sequence done at a table height of four inches, five
tags were tested. The test produced three failures. It was difficult to center the static mass
on the corrugated pads and therefore the impacts were not as calculated as with the
wooden fixture. The two tags that passed did not receive direct impacts on the [C but
impacts to the antenna. The results showed that there was a difference for protection of

the tag because of the drop height. The raw data is presented in Appendix C.

55

In the aluminum fixture drop sequence done at a table height of three inches, five
tags were tested. The test produced three failures. Using this fixture it was again difficult
to center the static mass on the corrugated pads. Because of this the impacts were not as
calculated as with the wooden fixture. The two tags that passed did not receive direct
impacts on the IC but impacts to the antenna. The results showed that there was a
difference for protection of the tag because of the drop height. The raw data is presented

in Appendix C.

4.3.7 WOOD FIXTURE NO HEADSPACE

In the wood fixture with the headspace filled with corrugated pads the drop
sequence done at a table height of three inches, five tags were tested. The test produced
three failures. The results showed that eliminating the headspace of the wooden fixture

will eliminate variables during the testing. The raw data is presented in Appendix C.

4.3.8 IC CENTERED OVER PEAK (Tag Orientation: with Fluting)

The tag orientation specific impact sequence test was done with the IC centered
over the peak of the fluting with the length of the label adhered to the corrugated pad
parallel with the fluting. The testing was completed using 30 tags. The test killed 20 of
the 30 tags. Of the 30 tags tested 21 of those tags were centered between the fluting after

impact. The raw data is presented in Appendix C.

56

4.3.9 IC CENTERED OVER VALLEY (Tag Orientation: with Fluting)

The tag orientation specific impact sequence test was done with the IC centered
over the valley of the fluting with the length of the label adhered to the corrugated pad
parallel with the fluting. The testing was completed using 30 tags. The test killed 18 of
the 30 tags. Of the 30 tags, 15 of those tags were centered between the fluting after

impact. The raw data is presented in Appendix C.

4.3.10 IC CENTERED OVER PEAK (Tag Orientation: 90° to Fluting)

The tag orientation specific impact sequence test was done with the IC centered
over the peak of the fluting with the length of the label adhered to the corrugated pad
perpendicular to the fluting. The testing was completed using 30 tags. The test killed 12
of the 30 tags. Of the 30 tags tested 18 of those were centered between the fluting after

impact. The raw data is presented in Appendix C.

4.3.11 IC CENTERED OVER VALLEY (Tag Orientation: 90° to Fluting)

The tag orientation specific impact sequence test was done with the IC centered
over the valley of the fluting with the length of the label adhered to the corrugated pad
perpendicular to the fluting. The testing was completed using 30 tags. The test killed 19
of the 30 tags. Of the 30 tags tested eight of those tags were centered between the fluting

after impact. The raw data is presented in Appendix C.

57

4.3.12 IC ON SLOPE OF FLUTING (Tag Orientation: with Fluting)

The tag orientation specific impact sequence test was done with the IC placed over
the slope of the fluting with the length of the label adhered to the corrugated pad parallel
to the fluting. The testing was completed using 30 tags. The test killed 18 of the 30 tags.
Of the 30 tested 16 of those tags were centered between the fluting after impact. The raw

data is presented in Appendix C.

4.3.13 IC ON SLOPE OF FLUTING (TAG ORIENTATION: 90° TO FLUTING)
The tag orientation specific impact sequence test was done with the IC placed over
the slope of the fluting with the length of the label adhered to the corrugated pad
perpendicular to the fluting. The testing was completed using 30 tags. The test killed 16
of the 30 tags. Of the 30 tags tested 11 of those were centered between the fluting after

impact. The raw data is presented in Appendix C.

58

CHAPTER 5: STATISTICAL ANALYSIS, CONCLUSIONS AND FUTURE
TESTING

5.1 STATISTICAL ANALYSIS

Table 9: Statistical Data Collected for Comparison

 

 

 

 

 

 

 

# Group Test Description P338 i2)“ 1.22%: d
21 A IC Centered over peak (with fluting) 33.33 66.67 30
22 B IC Centered over valley (with fluting) 40.00 60.00 30
23 C IC Centered over Slope (with fluting) 40.00 60.00 30
24 D [C Centered over peak (90 degrees to fluting) 60.00 40.00 30
25 E 1C Centered over valley (90 degrees to fluting) 36.67 63.33 30
26 F [C Centered over slope (90 degrees to fluting) 53.33 46.67 30

 

 

 

 

 

 

 

 

The calculations of the testing performed were analyzed on six different tests that
incorporated 30 samples for each test. All statistical tests preformed are one sided
comparisons of proportions used to determine the p-value, where “p” is short for
probability. This value provides a sense of the strength of the evidence against the null
hypothesis, the lower the p-value, the stronger the evidence. The p—value can be
subtracted fiom the perfect total distribution of one (or 100 percent) for a final outcome
confidence level. One way to increase survival rates was tag placement on the corrugated
board and the direction of fluting. The following seven comparisons were statistically

evaluated (tests denoted with the group letters from Table 9).

l) A+B+C vs. D+E+F
In a comparison of all tags aligned with the flutes versus tags rotated 90° to the
flutes (i.e. tests A+B+C vs. D+E+F) the following results were obtained: tags placed with

the fluting had a 62.2 percent failure rate, while the tags placed 90° to the fluting had 50.0

59

percent failure rate. Using a one sided test of proportions, it can be stated that the tags
aligned with the flutes group had a significantly greater proportion of failures with a p-value
of 0.0493 or a confidence level of 95.07 percent. Statistically this denotes tags aligned with

the flutes will fail at a higher rate than tags rotated 90° to the flutes.

2) A+D vs. B+E

In a comparison of all tags aligned over the peaks with the flutes versus tags
aligned over the valleys with the flutes (i.e. A+D vs. B+E) the following results were
obtained: tags aligned over the peaks of the flutes had a 53.3 percent failure rate, while the
tags aligned over the valleys of the flutes had a 61.2 percent failure rate. Using a one sided
test of proportions, it can be stated that the placement of the tags centered over the valleys
of the flutes did not have a significantly higher proportion of failures with a p-value of
0.1779 or a confidence level of 82.3 percent. Statistically this denotes the tags aligned
over the peaks or the valleys are not different enough to understand which manner of
placement is more beneficial. Additional testing will be needed to determine if these

placements are statistically different.

3) A vs. B

In a comparison of tags centered over the peaks with the fluting versus tags
centered over the valleys with the fluting (i.e. A vs. B) the following results were
obtained: tags centered over the peaks had a 66.7 percent failure rate, while tags centered
over the valleys with the fluting had a 60.0 percent failure rate. Using a one sided test of
proportions, it can be stated that test A did not have a significantly higher proportion of

failures with a p-value of 0.2955 or a confidence level of 70.45 percent. Statistically this

60

denotes tags centered over the peaks are not different enough from tags centered over the
valleys to understand which placement is more beneficial. Additional testing will be

needed to determine if these placements are statistically different.

4) D vs. E

In a comparison of tags centered over the peaks 90° to the fluting versus tags
centered over the valleys 90° to the fluting (i.e. D vs. E) the following results were
obtained: tags centered over the peaks 90° to the fluting had a 40.0 percent failure rate,
while tags centered over the valleys 90° to the fluting had a 63.3 percent failure rate. Using
a one sided test of proportions, it can be stated that tags centered over the valleys 90° to
the fluting had a significantly higher proportion of successes with a p-value of 0.0353 or a
confidence level of 96.47 percent. Statistically this denotes tags centered over the valleys

90° to the fluting fail with a higher rate than tags centered over the peaks 90° to the fluting.

5) A vs. D

In a comparison of tags centered over the peaks with the fluting versus tags
centered over the peaks 90° to the fluting (i.e. A vs. D) the following results were
obtained: tags centered over the peaks with the fluting had a 66.7 percent failure rate,
while tags centered over the peaks 90° to the fluting had a 40.0 percent failure rate. Using
a one sided test of proportions it can be stated that tags centered over the peaks 90° to the
fluting had a significantly higher proportion of success with a p-value of 0.0192 or a
confidence level of 98.08 percent. Statistically this denotes tags centered over the peaks

with the fluting fail with a higher rate than tags centered over the peaks 90° to the fluting.

61

6) B vs. E

In a comparison of tags centered over the valleys with the fluting versus tags
centered over the valleys 90° to the fluting (i.e. B vs. E) the following results were
obtained: tags centered over the valleys with the fluting had a 60.0 percent failure rate,
while tags centered over the valleys 90° to the fluting had a 63.3 percent failure rate. Using
a one sided test of proportions, tags centered over the valleys 90° to the fluting did not
have a significantly higher proportion of successes with a p-value of 0.3953 or a
confidence level of 60.47 percent. Statistically this denotes the tags centered over the
valleys with the fluting are not statistically different enough to understand which
placement is a more beneficial. Additional testing will be needed to determine if these

placements are statistically different.

7) C vs. F

In a comparison of tags centered over the slope with the fluting versus tags
centered over the slope 90° to the fluting (i.e. C vs. F) the following results were obtained:
tags centered over the slope with the fluting had a 60.0 percent failure rate, while tags
centered over the slope 90° to the fluting had a 46.7 percent failure rate. Using a one sided
test of proportions, it can be stated that tags centered over the slope with the fluting did not
have a significantly higher proportion of successes with a p-value of 0.1503 or a
confidence level of 84.97 percent. Statistically this denotes the difference of placement
does not show any statistical difference. Additional testing will be needed to determine if

these placements are statistically different.

62

 

5.2 CONCLUSION

Single parcel information from FedEx.com and UPS.com describe that the two
companies combined deliver more than 20 million packages per day. A five day work
week equates to 260 work days per year, and the yearly delivery approximation equates to
over 5.2 billion packages. Each package has the possibility of being dropped on one of

six flat sides, one of eight corners and one of 12 possible edges.

X%=-;—*%*$=0.001736*100=0.l736% Equation2

The chance of impacting the side where the tag is adhered is 0.1736 percent. Although
this is a small percentage, the possible number of tagged sides impacted per day is
approximately 35,000 which is 9.1 million packages per year. A study of single parcel
environments (UPS, FedEx and USPS) showed that packages, on average, receive 13
impacts sufficient enough to cause damage to the corrugated structure. (Burgess, 2004)
The goal was to complete a variety of effective distributional tests in order to
expose dynamic weaknesses in Class 0 Matrics® RFID tags. The testing covered
packaging tests of vibration and impacts. The tags were very rugged and survived much
of the testing. Once the testing was accomplished, none of the vibration tests produced
failures. Since there was no differentiation between any of the vibration tests no
additional research was done. The drop sequence showed failures of the tags but were
difficult to reproduce, as the position of the impact changed during the testing. The shock
table on the other hand showed failures that were easily reproduced. The testing revealed

that tag location and tag orientation does make a difference in the survival of the tag. The

63

 

orientation is hypothesized to have an impact on the survivability of the tags because of
the tension it exudes perpendicular to the fluting on the liner.

The failures, impact velocities, impact durations and drop height were repeatable
and consistent with the shock table. For this reason the shock table was used to
understand of how C-Flute corrugated board protected and reacted to a static load from
specific drop heights were used in protecting RFID tags. Both the drop and shock showed
high failure levels from the two inch threshold

The most significant way to increase survival rates is tag placement on the
corrugated board and the direction of fluting. When the length of the tag was
perpendicular to the fluting the survival rate has a confidence level of 95.07 percent. The
hypothesis of these results is that the tension of the label grasping the outer liner of the C-
flute board dissipated the energy release just enough to increase the survival rate.

Tags centered over the valley of the board 90° to the fluting had a better survival
rate then tags adhered over the peaks of the board 90° to the fluting with a confidence
level of 96.47 percent. In the final statically different test the comparison of tags centered
over the peaks with the fluting versus tags centered over the peaks 90° to the fluting
showed a confidence level of 98.08 percent.

During load spreading tests, results showed that if the bearing area is increased the
survivability of the tag will increase. Using additional load spreading material will
increase the chance of a tag’s survivability. An additional procedure that will effectively
help tags survive is the tag’s placement in accordance with the direction of the fluting.

The direct impact of the static mass used on the IC was of a severe nature and can

be reduced from 16 pounds in future tests. As new tags are being developed it is

64

recommended to have an understanding of the ruggedness for each tag and their variances.
Once the individual ruggedness is determined different mediums can be tested with less
variability. All testing defined the areas of vulnerability for the Matrics® tags. As
different manufacturers develop smaller, more efficient tags, variability between tags may
be finely tuned to create a higher threshold for direct impacts. Products which possess
small bearing areas and may come into contact with the outer packaging, show high
failure rates upon drops and impacts. The harder the impacting surface, the more likely it

is to show failure. Changes need to be made to the way the tags are protected on the case.

5.3 FUTURE TESTING

The testing performed in this thesis entailed tag impacts using corrugated board as
the banier between the static load and the tag. Multiple tests performed during this work
were unable to be validated with the sample size used. Supportive testing will need to be
developed in order to understand the statistical differences in the tests preformed in this

thesis as follows:

1. tags aligned over the peaks with the fluting versus tags aligned over the valleys
with the fluting;

2. tags centered over the peaks with the fluting versus tags centered over the valleys
with the fluting;

3. tags centered over the valleys with the fluting versus tags centered over the valleys

90° to the fluting; and

65

4. tags centered over the slope with the fluting versus tags centered over the slope 90°

to the fluting

Parallel testing for all the investigations performed in this thesis can be continued
on different flutes as well as multiple walled boards. In corresponding testing, additional
research should be looked into for the following tag classes: Class 0+, Class 1, and Class
1 Generation 2. Along with other classes of tags, different sized labels giving each tag a
different surface area should be tested. This area may show different advantages to the
understanding of orientation in accordance to the fluting.

Additional interior impact testing is also recommended with a smaller static weight
to incorporate different products which could jeopardize the ability of an RFID tag to
withstand further impacts. Testing revolving around drop impacts onto non-flat surfaces
where punctures and high abrasion is possible also needs to be explored. A Mullen Burst
test could provide future evidence for the effects of fiber tear and piercing contact from
external impacts.

Future testing should also look into reverse impact testing, specifically, impacting
the tag first then compressing the corrugated board. This testing will determine the
ruggedness of the individual tag with direct impacts from outside objects while
understanding the impact of a tag bunching on itself. All testing which furthers the
understanding of the ruggedness of the current RFID tag technology will be helpful to the

knowledge pool for the firture of RFID.

66

APPENDIX A (VIBRATION)

Table A1: Sinusoidal Vibration of Individual Cases

 

 

 

 

 

 

 

 

Sample Pass Failure

1 1 O

2 l 0

3 1 0

4 l 0
Average 1 0
Std Dev. 0 0
Percent 100 0

 

 

 

 

 

Table A2: Sinusoidal Vibration of Individual Cases Against Plywood

 

 

 

 

 

 

 

 

 

 

 

Sample Pass Failure 5
1 1 0
2 1 0

Averafi 1 0

Std Dev. 0 0

Percent 100 0

 

Table A3: Random Vibration of Four Column Stacked Cases

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Pass Failure
1 1 0
2 1 0
3 1 0
4 1 0
5 1 0
6 1 0
7 1 O
8 1 0
Average 1 0
Std Dev. 0 0
Percent 100 0

 

67

Table A4: ISTA 3C Vibration and Drop Sequence

 

 

 

 

 

 

 

 

 

 

Sample Pass Failure
1 1 0
2 1 0
Average 1 0
Std Dev. 0 0
Percent 100 0

 

68

 

APPENDIX B (DROP)

Table Bl: Incremental Drop Sequence Determination for Failure

 

 

 

 

 

 

 

 

 

 

Drop Height Pass Failure Drops
18 Inch Drops 0 1 2
24 Inch Drops 1 0 6
30 Inch Drops 0 l l
36 Inch Drops 1 0 4
42 Inch Drops 0 l 1

Average 0.4 0.6 2.8
Std Dev. 0.547722558 0.547722558 2.168
Percent 40 60

 

 

 

 

Table B2: Incremental Drop Sequence

 

 

 

 

 

 

 

 

Drop Height Pass Failure Drops
18 Inch Drops 0 1 8
24 Inch Drops 0 1 9
30 Inch Drops 0 l 3

Average 0 1 6.667
Std Dev. 0 0 3.215
Percent 0 100

 

 

 

 

69

 

 

 

 

APPENDIX C (SHOCK)

Table Cl: [C Centered Over Peak (Tag Orientation: with Fluting)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Peak Filter Not
Sample Delta V G Frequency Pass Failure Centered Centered

1 56.17 91.25 n/a 0 1 l 0

2 54.56 91.89 n/a 1 0 1 O
3 55.53 89.99 n/a 0 1 0 1

4 54.44 89.97 n/a 0 l 1 0
5 53.80 91.58 n/a 0 1 1 0
6 55.27 91.33 n/a 0 1 1 0
7 54.37 91.27 n/a 0 1 0 1

8 54.43 90.57 n/a 0 1 0 1

9 55.83 92.75 n/a 0 1 0 1
10 55.30 91.15 n/a 0 l l 0
11 56.32 88.96 827.81 1 0 1 0
12 55.89 88.83 909.09 0 1 0 1
13 55.91 89.14 831.95 1 0 0 1
14 55.09 87.48 900.90 1 0 1 0
15 56.98 90.65 865.05 0 1 1 0
16 56.50 92.51 902.53 0 1 1 0
17 56.04 88.99 860.59 1 0 1 0
18 57.13 90.50 862.07 1 0 1 0
19 56.16 87.93 857.63 1 0 1 0
20 56.70 89.89 868.06 0 1 1 0
21 56.66 89.69 853.24 1 0 0 1
22 55.96 89.70 905.80 1 0 1 0
23 56.65 90.97 892.86 1 0 1 0
24 55.99 89.73 886.52 0 l 1 0
25 56.25 87.67 809.06 0 1 1 0
26 57.62 91.62 856.16 0 1 0 1
27 55.30 90.40 902.53 0 1 0 1
28 56.07 89.53 825.08 0 l 1 0
29 56.14 88.57 834.72 0 1 1 0
30 55.00 88.86 884.96 0 1 l 0

Average 55.80 90.11 866.83 0.33 0.67 0.70 0.30
Std. Dev. 0.9023 1.3728 30.5290 0.4795 0.4795 0.4661 0.4661
Maximum 57.62 92.75 909.09 33.33 66.67 70.00 30.00
Minimum 53.80 87.48 809.06

 

 

 

 

 

 

 

 

70

 

Table C 2: IC Centered Over Valley (Tag Orientation: with Fluting)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Delta Peak Filter Not
Sample V G Frequency Pass Failure Centered Centered

l 56.12 91.62 n/a 0 1 O 1
2 54.75 91.23 n/a 1 0 0 1

3 55.38 90.78 n/a 1 0 0 1
4 56.05 90.76 n/a 0 1 1 0
5 54.09 89.16 n/a 0 l l 0
6 55.30 91.14 n/a 0 l l 0
7 54.93 90.40 n/a 0 1 1 0
8 54.86 90.37 n/a 0 1 0 1
9 54.30 90.49 n/a 0 1 l 0
10 55.14 91.98 n/a 0 1 0 l
11 56.33 89.90 869.37 0 l 0 1
12 56.00 89.39 894.45 1 0 0 1
13 56.67 90.10 874.13 1 0 1 0
14 57.34 88.41 815.66 1 0 0 1
15 56.31 88.72 865.05 0 1 l 0
16 56.85 89.60 823.72 1 0 0 1
17 57.28 91.36 865.05 1 0 0 l
18 57.28 91.58 874.13 1 0 0 l
19 56.65 90.16 909.09 0 1 0 1
20 56.63 89.71 874.13 1 0 l 0
21 56.29 91.94 912.41 0 1 l 0
22 56.71 88.94 863.56 0 l 1 0
23 57.12 92.41 899.28 0 1 l 0
24 56.49 91.18 862.07 1 0 l O
25 56.79 90.85 865.05 1 0 1 0
26 55.29 88.59 871.08 0 l 0 l
27 56.64 91.26 865.05 0 1 1 0
28 56.45 89.95 878.73 0 l 0 1
29 57.63 91.84 871.08 1 0 0 l
30 56.50 90.43 830.56 0 l 1 0

Avera e 56.14 90.48 869.18 0.40 0.60 0.50 0.50
Std. Dev. 0.9395 1.0852 24.8692
Maximum 57.63 92.41 912.41 40.00 60.00 50.00 50.00
Minimum 54.09 88.41 815.66

 

 

 

 

 

 

 

 

 

71

‘7'- ml 4"" -u-n

Table C 3: IC On Slope of Fluting (Tag Orientation: with Fluting)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Delta Filter Not
Sample V Peak G Frequency Pass Failure Centered Centered
1 55.72 90.06 871.08 1 0 0 l
2 56.63 91.86 892.03 1 0 l 0
3 55.17 90.60 904.16 0 l 1 0
4 56.24 91.02 900.01 0 1 1 0
5 55.83 92.06 888.1 1 0 1 0
6 56.05 90.55 899.28 0 1 0 1
7 56.85 94.93 907.44 0 l 1 0
8 57.31 92.31 862.07 1 0 0 1
9 55.56 89.62 889.68 1 0 0 1
10 56.39 90.18 857.63 1 0 O l
11 56.41 90.41 853.24 0 1 0 1
12 55.70 89.76 856.16 0 1 1 0
13 56.34 91.51 878.73 0 l O 1
14 57.22 90.61 850.34 0 l 1 0
15 55.95 89.02 840.53 1 0 0 1
16 55.87 90.77 860.59 0 l 1 0
17 56.31 89.86 838.93 1 0 1 0
18 55.62 88.86 827.81 1 0 1 0
19 57.68 92.59 853.24 1 0 0 1
20 56.57 90.69 874.13 1 0 l 0
21 55.71 90.85 869.57 0 l 0 1
22 56.21 91.72 875.66 0 l 0 l
23 56.96 90.18 859.11 1 0 1 0
24 56.72 89.40 827.81 0 l 0 1
25 56.64 90.05 841.75 0 1 1 0
26 56.71 91.64 871.08 0 1 0 1
27 56.85 91.54 868.06 0 1 1 0
28 57.32 91.59 827.81 0 l l 0
29 56.17 89.20 836.12 0 1 0 1
30 56.61 91.29 881.83 0 1 l 0
Avegge 56.38 90.82 865.47 0.40 0.60 0.53 0.47
Std. Dev. 0.5984 1.2560 23.3890
Maximum 57.68 94.93 907.44 40.00 60.00 53.33 46.67
Minimum 55.17 88.86 827.81

 

 

 

 

 

 

 

 

72

 

 

Table C4: IC Centered Over Peak (Tag Orientation: 90° to Fluting)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Filter Not
Sample Delta V Peak G Frequency Pass Failure Centered Centered

1 54.97 91.98 n/a 1 0 1 0
2 55.91 91.70 n/a 0 1 l 0
3 55.78 91.12 n/a 1 0 0 1
4 55.47 91.10 n/a 1 0 0 1
5 54.52 92.00 n/a 0 1 1 0
6 55.29 91.65 n/a 1 0 1 0
7 55.12 92.73 n/a 0 1 0 1
8 55.75 92.27 n/a 0 1 1 0
9 55.90 92.19 n/a 0 l 1 O
10 52.13 94.37 n/a 0 l 0 l
11 57.22 91.52 871.08 . 1 0 0 1
12 56.99 90.88 889.91 1 0 1 0
13 55.81 91.52 900.90 0 1 1 0
14 56.45 91.71 904.16 1 0 1 0
15 57.80 90.59 848.90 1 0 0 1
16 57.79 93.99 859.11 1 0 0 1
17 57.03 92.69 863.56 0 l 0 1
18 56.51 91.28 836.12 1 0 1 0
19 57.08 93.40 874.13 0 1 l 0
20 56.45 89.98 836.12 0 l 1 0
21 56.80 91.91 850.34 1 0 1 0
22 56.45 89.85 859.1 1 1 0 l 0
23 55.62 88.16 841.75 1 0 l 0
24 55.88 90.74 904.16 1 0 1 0
25 55.93 91.55 909.09 0 l 0 1
26 57.13 90.74 863.56 1 0 1 0
27 56.87 89.72 844.59 1 0 1 0
28 55.63 88.04 841.75 1 0 0 1
29 56.34 92.34 892.86 0 l 0 1
30 56.17 90.98 853.24 1 0 0 1

Average 56.09 91.42 867.22 0.60 0.40 0.60 0.40

Std. Dev. 1.0958 1.4100 24.6766

Maximum 57.80 94.37 909.09 60.00 40.00 60.00 40.00

Minimum 52.13 88.04 836.12

 

73

 

 

Table C5: IC Centered Over Valley (Tag Orientation: 90° to Fluting)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Filter Not
Sample Delta V Peak G Frequency Pass Failure Centered Centered
1 56.26 90.57 880.28 0 1 0 l
2 57.54 90.66 860.59 0 1 0 1
3 56.17 91.25 888.10 0 1 0 l
4 58.22 94.08 924.21 0 1 0 1
5 56.16 93.18 939.85 1 0 0 1
6 57.09 92.87 910.75 1 0 0 1
7 56.73 92.16 911.75 0 1 0 l
8 57.51 92.75 881.83 0 1 1 0
9 57.47 94.12 929.37 1 0 0 1
10 57.49 91 .89 896.06 1 0 1 0
1 l 56.97 92.27 929.37 0 1 1 0
12 56.74 91.71 874.13 0 l 0 1
13 57.70 93.51 902.53 0 l 0 l
14 57.44 93.10 934.58 1 0 0 1
15 57.00 92.86 925.93 0 l 0 l
16 59.86 96.26 883.39 1 0 0 l
17 59.39 93.44 833.33 0 1 0 1
18 56.67 91.49 878.73 1 0 1 0
19 55.46 94.29 943.40 1 0 0 1
20 57.21 91.42 863.56 0 l 0 1
21 55.74 88.24 815.66 0 l O 1
22 55.81 90.07 874.13 1 0 1 0
23 56.27 88.50 841.75 0 l 0 l
24 56.25 88.47 833.33 0 1 0 1
25 56.64 89.75 868.06 0 1 0 l
26 55.85 89.37 862.07 0 l 0 1
27 56.61 90.90 872.60 1 0 1 0
28 57.25 88.98 822.37 1 0 1 0
29 56.16 91.89 896.06 0 1 1 0
30 56.66 88.70 866.55 0 l 0 1
Average 56.94 91.63 884.81 0.37 0.63 0.27 0.73
Std. Dev. 0.9859 2.0171 35.5160
Maximum 59.86 96.26 943.40 36.67 63.33 26.67 73.33
Minimum 55.46 88.24 815.66

 

74

 

Table C6: IC On Slope of Fluting (Tag Orientation: 90° to Fluting)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Filter Not
Sample Delta V Peak G Frequency Pass Failure CenteredICentered
1 57.24 90.98 821.02 1 0 0 1
2 55.74 90.69 886.52 1 0 l 0
3 55.24 90.75 905.80 1 0 0 1
4 56.23 90.26 880.28 0 l 0 1
5 54.77 87.87 860.59 0 l 1 0
6 55.95 88.50 847.46 1 0 0 1
7 55.55 88.20 859.11 1 0 0 1
8 56.02 88.29 866.55 0 1 0 1
9 56.50 89.17 814.33 1 0 1 0
10 55.89 88.09 823.72 1 0 l 0
11 55.62 90.04 889.68 1 0 1 0
12 56.38 88.76 823.72 1 0 0 1
13 56.12 89.72 866.55 0 1 1 0
14 56.64 90.10 857.63 0 l 1 0
15 55.37 87.50 829.19 0 l 0 1
16 56.14 88.18 816.99 1 0 0 1
17 56.09 88.52 841.75 0 l 0 1
18 56.67 90.05 829.19 1 0 1 0
19 57.45 90.35 823.72 0 1 l 0
20 56.90 89.08 821 .02 0 1 O l
21 56.05 88.89 930.56 1 0 0 1
22 56.76 90.19 865.05 1 0 l 0
23 56.46 88.37 821.02 1 0 0 1
24 56.72 90.12 856.16 1 0 0 1
25 56.10 90.18 874.13 0 1 0 1
26 57.52 91.04 833.33 0 l 0 1
27 55.57 90.60 881.83 0 l 0 l
28 55.65 89.89 883.39 1 0 0 l
29 55.85 88.60 830.56 0 l 0 1
30 54.95 88.95 865.05 0 1 1 0
Avera e 56.14 89.40 853.53 0.53 0.47 0.37 0.63
Std. Dev. 0.6748 1.0391 29.7154
Maximum 57.52 91.04 930.56 53.33 46.67 36.67 63.33
Minimum 54.77 87.50 814.33

 

 

 

 

 

 

 

 

 

75

Table C7: Initial Sequence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Drop Height Pass Failure Drops
1 5 0 1 3
2 6 0 1 2
3 7 0 l 1
4 8 0 1 1
5 9 0 l 2
6 5 0 1 3
7 6 0 1 1
8 7 0 1 1
9 8 0 l 2
1 0 9 0 l 1
1 l 10 0 1 1
12 1 1 0 1 l
13 12 0 1 2
14 13 0 1 l
1 5 14 0 1 1
16 10 0 1 2
l7 1 1 0 1 2
1 8 12 0 1 2
19 1 3 0 1 l

20 14 0 1 1
Average 0 1
Std Dev. 0 0
Percent 0 100

 

 

 

 

 

76

 

 

Table C 8: Load Spreader:

 

Same for both Square and Round

 

Drop Height

Pass

Failure

Drops

 

H

l

 

 

 

 

 

 

 

 

 

 

S:E\OW\IO\M-§WN

 

l3

 

14

 

15

ml

 

 

l6

 

l7

 

l8

 

Average

 

Std Dev.

o—np—sy—s—aHHp‘p—ag—n—ay—IHp—ap—cg—AHH

 

Percent

 

 

OOH—sudp‘y—sp—sp—ap—ay—su—o—c—sy—n—s—apsy—a

e—I
O

 

OOOOOOOOOOOOOOOOOOOO

 

 

 

Table C9: Aluminum Fixture (4 Inch Drop)

 

 

 

 

 

 

 

 

 

 

 

Sample Pass Failure Drops
1 l 0 1
2 1 0 1
3 0 1 1
4 0 1 1
5 0 1 1
Average 0.4 0.6
Std Dev. 0.547722558 0.547722558
Percent 40 60

 

 

 

 

77

Table C10: Aluminum Fixture (3 Inch Drop)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Pass Failure Drops
1 1 0 l
2 1 O 1
3 0 1 1
4 0 1 1
5 0 1 1
Average 0.4 0.6
Std Dev. 0.547722558 0.547722558
Percent 40 60
Table C11: Wood Fixture, No Headspace (3 Inch Drop)
Sample Pass Failure Drops
1 0 1 l
2 1 0 1
3 1 0 l
4 0 1 l
5 O 1 1
Average 0.4 0.6
Std Dev. 0.547722558 0.547722558
Percent 40 60

 

 

 

 

 

78

APPENDIX D (STATISTICAL ANALYSIS)

Table 9: Statistical Data Collected for Comparison

 

 

 

 

 

 

 

 

 

 

 

 

 

TZSt Group Test Description P335 Fazloure 1.12%;
17 A IC Centered over peak (with fluting) 33.33 66.67 30
13 B IC Centered over valley (with fluting) 40.00 60.00 30
19 C IC Centered over Slope (with fluting) 40.00 60.00 30
20 D [C Centered over peak (90 degrees to fluting) 60.00 40.00 30
21 E IC Centered over valley (90 degrees to fluting) 36.67 63.33 30
22 F IC Centered over slope (90 degrees to fluting) 53.33 46.67 30

 

Table D1: COMPARISON 1 (A+B+C vs. D+E+F)

 

Contingency Analysis of Pass By tests

 

 

 

 

 

 

 

 

 

 

Contingency Table
tests By Pass
Count 0 1
Row %
17-18-19 56 34 90
62.22 37.78
20-21-22 45 45 90
50.00 50.00
101 79 180
Tests
Source DF -LogLike RSquare (U)
Model 1 1.36852 0.0111
Error 178 122.05016
C. Total 179 123.41868
N 180
Test ChiSquare Prob>ChiSq
Likelihood 2.737 0.0980
Ratio
Pearson 2.730 0.0985
Fisher's Exact Test Prob
Left 0.9644
Right 0.0664
2-Tail 0.1329

79

Table D2: COMPARISON 2 (A+D vs. B+E)

Contingency Analysis of Pass By tests2

 

 

 

 

 

 

 

 

 

 

Contingency Table
testsZ By Pass
Count 0 1
Row %
17,20 32 28 60

53.33 46.67
18,21 37 23 60

61.67 38.33

69 51 120

Tests
Source DF -LogLike RSquare (U)
Model 1 0.426810 0.0052
Error 118 81.395743
C. Total 119 81.822553
N 120
Test ChiSquare Prob>ChiSq
Likelihood 0.854 0.3555
Ratio
Pearson 0.853 0.3558
Fisher's Exact Test Prob
Left 0.2301
Right 0.8661
2-Tail 0.4603

80

 

Table D3: COMPARISON 3 (A vs. B)

Contingency Analysis of Pass By Test

 

 

 

 

 

 

 

 

 

 

Contingency Table
Test By Pass
Count 0 1
Row %
17 20 10 30

66.67 33.33
18 18 12 30

60.00 40.00

38 22 60

Tests
Source DF -LogLike RSquare (U)
Model 1 0.143691 0.0036
Error 58 39.285775
C. Total 59 39.429466
N 60
Test ChiSquare Prob>ChiSq
Likelihood 0.287 0.5919
Ratio
Pearson 0.287 0.5921
Fisher's Exact Test Prob
Left 0.7890
Right 0.3946
2-Tail 0.7892

81

Table D4: COMPARISON 4 (D vs. E)

Contingency Analysis of Pass By Test

 

 

 

 

 

 

 

 

 

 

Contingency Table
Test ByPass
Count 0 1
Row %
20 12 18 30

40.00 60.00
21 19 11 30

63.33 36.67

31 29 60

Tests
Source DF -LogLike RSquare (U)
Model 1 1.650408 0.0397
Error 58 39.905083
C. Total 59 41.555491
N 60
Test ChiSquare Prob>ChiSq
Likelihood 3.301 0.0692
Ratio
Pearson 3.270 0.0705
Fisher's Exact Test Prob
Left 0.0602
Right 0.9811
2-Tail 0.1205

82

Table D5: COMPARISON 5 (A vs. D)

Contingency Analysis of Pass By Test

 

 

 

 

 

 

 

 

 

 

Contingency Table
Test By Pass
Count 0 1
Row %
17 20 10 30

66.67 33.33
20 12 18 30

40.00 60.00

32 28 60

Tests
Source DF -LogLike RSquare (U)
Model 1 2.169623 0.0523
Error 58 39.285775
C. Total 59 41.455399
N 60
Test ChiSquare Prob>ChiSq
Likelihood 4.339 0.0372
Ratio
Pearson 4.286 0.0384
Fisher's Exact Test Prob
Left 0.9905
Right 0.0346
2-Tail 0.0692

83

 

Table D6: COMPARISON 6 (B vs. E)

Contingency Analysis of Pass By Test

 

 

 

 

 

 

 

 

 

 

RSquare (U)
0.0009

0.7906

0.7906

Contingency Table
Test By Pass
Count 0 1
Row %
18 18 12 30
60.00 40.00
21 19 11 30
63.33 36.67
37 23 60
Tests
Source DF -LogLike
Model 1 0.035261
Error 58 39.905083
C. Total 59 39.940344
N 60
Test ChiSquare Prob>ChiSq
Likelihood 0.071
Ratio
Pearson 0.071
Fisher's Exact Test Prob
Left 0.5000
Right 0.7020
2-Tail 1.0000

84

Table D7: COMPARISON 7 (C vs. F)

Contingency Analysis of Pass By Test

 

 

 

 

 

 

 

 

 

 

Contingency Table
Test By Pass
Count 0 1
Row %
19 18 12 30

60.00 40.00
22 14 16 30

46.67 53.33

32 28 60

Tests
Source DF -LogLike RSquare (U)
Model 1 0.537349 0.0130
Error 58 40.918049
C. Total 59 41.455399
N 60
Test ChiSquare Prob>ChiSq
Likelihood 1 .075 0.2999
Ratio
Pearson 1 .071 0.3006
Fisher's Exact Test Prob
Left 0.9023
Right 0.2189
2-Tail 0.4379

85

BIBLIOGRAPHY

AIM (Association for Automatic Identification and Mobility). 2004. AIDC
Technologies. (<http://www.aimglobal.org/technologies/>) Accessed February 19,
2004.

AIM (Association for Automatic Identification and Mobility). 2004. Data
Communications. (<http://www.aimglobal.org/technologies/>) Accessed February
19,2004.

BarCode 1. Product Code (UPC) and EAN Article Numbering Code (EAN). 2004.
Adams Communications. (<http://www.adams1.com/pub/russadam/ upccode.html>)
Accessed January 19, 2003.

Bellis, Mary. “Bar Codes - The History of Bar Code.” 2004. What you need to know
about (Inventors). (<http://inventors.about.com>) Accessed February 29, 2004.

Boyer, Dr. Ken. 2003 MSC 803 - Manufacturing Strategy. (Marketing and Supply
Chain Management) Operations Scheduling PPT presentation. Slide 4.

Burgess, Dr. Gary. (2004). PKG 805 — Advanced Packaging. (School of Packaging,
Michigan State University) Chapter 1; pgs 1.1 — 1.12.

Checkpoint Systems Inc. 2004. Electronic Article Surveillance (EAS) Overview.
Pages 1 — 4. (<http://www.checkpointsystems.com>) Accessed March I, 2004.

Chiesa, Mario; Genz, Ryan; Heubler, Franziska; Mingo, Kim; Noessel, Chris;
Sopieva, Natasha; Solcombe, Dave; Tester, Jason. 2002. White Paper. RFID — a
week long survey on the technology and its potential. (<http://www.interaction-
ivera.it>)

Clarke, Dr. Robb. 2004. Interview on Vibration. May 6, 2004. Face to Face
interview for thesis Vibration data.

Hagey, Warren. “The History of Bar Codes.” 1998.
(<http://educ.queensu.ca/~compsci/units/encoding/barcodes/history.html>) Accessed
February 29, 2004.

Landt, J. Catlin, B. (2001). Shrouds of Time — The History of RFID. (Pittsburgh,
PA, AIM Inc.) 2001: 11.

Mullin, Eileen. Electronic Product Code. 2002. Baseline Magazine Online.
(<http://www.baselinemag.com/article2/0,3 95 9,65 5991 ,00.asp>) Accessed
December 19, 2003.

86

RFID Wizards Inc. 2003. What is RFID/AutOID? (<http://www.rfidwizards.
com/rfidwizards.nsf/bca3 68ea2a1 7dac585256c6c0056701c/9ab08a7366fa62df‘85256
c6d006c8b8c!OpenDocument>) Accessed March 15, 2004.

Singh, Dr. S. Paul; Burgess, Dr. Gary (2004). PKG 410 — Distribution Packaging
Dynamics. (School of Packaging, Michigan State University) Chapter 1.

Stamer, Thad E. 1996. Piezoelectric materials. (<http://web.media.mit.
edu/~testame/TR328/node7.htrnl>). Last updated November 9, 1996. Accessed June
I, 2004.

Polin, Jenevieve Blair. 2003. Transportation Packaging “The Four Hazards of
Distribution ” (<http://www.devicelink.com/pmpn/archive/03/02/020.html>)
Accessed February 9, 2004.

Vargas, M. (2004). Retail Theft and Inventory Shrinkage. Retail Industry (Online),
A Prime Medial Company. 2004.

Zebra Technologies. 2004. “Bar Coding 101... What You Need to Know” Pages:
1-3 (<http://www.zebra.com>) Accessed March 4, 2004.

Zebra Technologies. 2004. “Bar Coding and RFID: The Key to Traceability and

Safety in the Food Service Supply Chain (Application White Paper)” Pages: 1-8
(<http://www.zebra.com>) Accessed March 4, 2004.

87