rug

J!HIWIUNle/WI‘M”!NWWIHHIIWW

_ Ea»: (“I

    

'—I
I
m

 

A.

This is to certify that the
thesis entitled

RADIO FREQUENCY IDENTIFICATION TRANSPONDER
PERFORMANCE ON REFRIGERATED AND FROZEN BEEF
LOIN MUSCLE PACKAGES

presented by
JOHN CHRISTIAN ONDERKO I

has been accepted towards fulfillment I
of the requirements for the

MS. degree in PACKAGING

 

 

pM/jfimflzg

V Major Profesiq’Lsfignature
May 11, 2004

 

 

 

 

Date

MSU IS an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

 

 

 

LIBRARY
Michigan State

University

 

 

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

 

i693 1’ Q @905

 

wmflidfl 5

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

 

RADIO FREQUENCY IDENTIFICATION TRANSPONDER PERFORMANCE ON
REFRIGERATED AND FROZEN BEEF LOIN MUSCLE PACKAGES

BY

John Christian Onderko

A THESIS

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

MASTER OF SCIENCE
School of Packaging

2004

ABSTRACT

RADIO FREQUENCY IDENTIFICATION TRANSPONDER PERFORMANCE ON
REFRIGERATED AND FROZEN BEEF LOIN MUSCLE PACKAGES

By
John Christian Onderko

Radio Frequency Identification (RFID) is a technology that is quickly
becoming an integrated tool in the field of Supply Chain Management. This
research investigated the potential use of RFID as a cold-chain management tool
for fresh meat packaging. Current RFID technology can penetrate many
materials, but the retail food supply chain presents many challenges. Beef
muscle presents a challenge due to the storage environment and moisture
content of the product.

The objective of this research was to evaluate the effects of different
storage temperatures of beef muscle on the data communication performance of
an RFID system. Two RFID systems were utilized to evaluate performance
differences between frequencies. Beef packages were tested individually and in
stacked configurations to simulate item and shelf level tracking. Packages were
frozen and refrigerated to simulate different storage conditions.

This research found that RFID systems operating in the 13.56 MHz
frequency range should experience no loss of data when transmitting through a
package of refrigerated or frozen beef. RFID systems operating in the 915 MHz
frequency range should communicate through frozen packages, but may

experience a loss of data when communicating through a refrigerated package.

Copyright by
JOHN CHRISTIAN ONDERKO
2004

TABLE OF CONTENTS

LIST OF TABLES .................................................................................................. v
LIST OF FIGURES ................................................... ' ........................................... vi
KEY TO ABBREVIATIONS ................................................................................. vii
INTRODUCTION .................................................................................................. 1
LITERATURE REVIEW ........................................................................................ 5
MATERIALS AND METHODS ............................................................................ 17
RESULTS AND DISCUSSION ............................................................................ 31
CONCLUSIONS .................................................................................................. 33
FUTURE RESEARCH ........................................................................................ 35
APPENDIX .......................................................................................................... 36
REFERENCES ................................................................................................... 42

LIST OF TABLES

Table 1. Refrigerated Read Percentages ............................................................ 32
Table 2. Frozen Read Percentages .................................................................... 32
Table 3. Depth of Reads ........................................................................... 32

LIST OF FIGURES

Figure 1. Texas Instruments S6350 system and Dell Inspiron 2500 computer..36

Figure 2. Matrics ANT-001 Antenna and Dell Inspiron 2500 computer .............. 36
Figure 3. Vacuum packaged USDA Select strip loin muscle .............................. 37
Figure 4. Face of tray packed sample ................................................................ 37
Figure 5. Sample with TI transponder label attached ......................................... 38
Figure 6. Sample with Matrics transponder label attached ................................ 38
Figure 7. Graphic of package facing toward 86350 Reader .............................. 39
Figure 8. Graphic of package facing away from S6350 Reader ......................... 39
Figure 9. Graphic of packages in a stacked configuration ................................. 40

Figure 10. Picture of single package facing toward Matrics ANT-001 antenna..40

Figure 11. Samples in a stacked configuration .................................................. 41

vi

AIDC
AMI
BSE
DC
EAN
EEPROM
EPC
EPS
F
FIFO
FMD
GHz
ISO
kHz
MHz

PVC

RF

RFID

ROM

Tl

KEY TO ABBREVIATIONS

Automatic Identification & Data Collection
American Meat Institute

Bovine Spongiform Encephalopathy
Distribution Center

European Article Numbering International
Electronically Erasable Programmable Read Only Memory
Electronic Product Code

Expanded Polystyrene

Fahrenheit

First In, First Out

Foot and Mouth Disease

Gigahertz

International Organization for Standardization
Kilohertz

Megahertz

Polyvinyl Chloride

Random Access Memory

Radio Frequency

Radio Frequency Identification

Read Only Memory

ReadNVrite

Texas Instruments

vii

UCC
UHF
USAIP
USDA

WORM

KEY TO ABBREVIATIONS

Uniform Code Council
Ultra High Frequency
United States Animal Identification Plan
United States Department of Agriculture

Write Once Read Many

viii

INTRODUCTION

Radio Frequency Identification, or RFID, is a wireless data transmission
technology that is quickly becoming an integrated tool in the field of supply chain
management. RFID can be used within a company’s supply chain for the
tracking of raw materials to finished goods. These systems can operate in many
different configurations and frequencies, depending on the application and
material through which the signals must pass. Choosing the proper frequency is
crucial to retaining visibility throughout the life of the product. While this
technology has existed for many years, little research has been conducted on the
potential and limitations of utilizing it with certain packaged products.

A typical RFID system is composed of four components: the host
computer, interrogator, antenna and transponder. The transponder, or tag, is the
wireless device that is attached to the case, item or package. This device
communicates with the system’s antenna to transmit data stored in its memory.
This data can be used to locate, track and record information associated with the
item it is attached to.

RFID systems have many benefits compared to other automatic
identification and data collection (AIDC) technologies, such as bar codes. Bar
codes require a line-of-sight between the laser and bar code. RFID systems are
not constrained by this requirement, as radio frequencies can penetrate certain
mediums, allowing items within a pallet to be recognized. Other benefits include
the ability to recognize numerous transponders at the same time, capability to

write dynamic information to the transponder, longer read ranges, and the ability

to track moving objects, just to name a few. Despite these benefits, RFID does
have its limitations. The use of RFID has been restricted by its inability to
transmit data without interference from certain materials (Albano & Engels,
2002). The largest challenge is the ability of the transponder to communicate its
data to the antenna without loss of resolution or information. Systems operating
in the ultra-high frequency (UHF) bandwidth are known to have interference
issues with moisture and metallic objects.

Current RF ID technology can penetrate many different mediums, but the
retail food supply chain presents many challenges. The transmission frequency
of an RFID system must travel through differing mediums, along with varying
environmental conditions. Beef muscle presents a challenge due to the moisture
and tissue content of the product. Prior research focused on meat packaging
has led to the development of a material testing protocol for choosing
transponders based on properties of modified atmosphere and controlled
atmosphere packages (Vorst, 2003). Vorst also evaluated the effects of
transponders on the bloom time of beef loin muscle. His research reported that
there was no appreciable consumer or product aversion in utilizing RFID
transponders as a supply chain management tool on fresh meat packaging.

Within the last year, some of the world’s largest retailers have announced
aggressive plans that detail how RFID technology will be integrated into their
supply chains. In the United States, Wal-Mart, Albertson’s and Target stores
have instituted plans that require their top suppliers to apply RFID transponders

to case and palletized goods beginning in 2005. European grocers have also

announced their own plans that will require their top suppliers to integrate RFID
transponders into shipments of cased goods. These implementation plans
forecast the application of RF ID transponders to the item level within the next 4-6
years (WaI-Mart (I), 2004).

As retailers and grocers begin to receive items that are individually
tagged, they must understand the technological and material properties that
these items present. When fresh meat packages contain transponders, the
storage environments of these products can present many challenges to the
operations of RFID systems. The affect of colder temperatures on the
performance of packages containing RFID transponders has not been studied. A
recent study found the average retail fresh meat case operates at 41° Fahrenheit
(F) (AMIR, 2004). Frozen storage conditions may also be utilized during
transportation and storage of fresh meats. These storage environments, coupled
with the moisture found in fresh meats could potentially lead to a loss of data
during the data transmission cycle.

This study will evaluate the use of RFID as a cold-chain tool by measuring
the affects of different storage temperatures of fresh meat packages on the data
transmission ability of two systems, operating at different frequencies. The goals
of this study were to understand how beef muscle interacts with data
transmission when conditioned in these operational environments. The first goal
was to understand what effects refrigerated and frozen temperatures had on the
data transmission of systems operating at 13.56 MHz and 915 MHz. RF ID

systems from Texas Instruments (TI) and Matrics Inc. were evaluated. The TI

system operated at a frequency of 13.56 MHz, and the Matrics system operated
at a frequency of 915 MHz. The second goal of this study was to measure what
effect these temperatures had on the data transmission performance of
packages in a stacked configuration.

The author hypothesizes that RFID systems operating in the 13.56 MHz
bandwidth will experience zero loss of data communication between a
transponder and interrogator while operating at both refrigerated and frozen
temperatures. It is hypothesized that systems operating in the 915 MHz
bandwidth will experience a total loss of data communication between the

transponder and interrogator at both refrigerated and frozen temperatures.

LITERATURE REVIEW

RFID Technology

Radio Frequency Identification is an AIDC technology that utilizes radio
frequency (RF) energy for the wireless transmission of data. RFID systems are
comprised of four major components, a host computer, a transponder (tag), an
interrogator (reader), and an antenna. Transponders have imbedded memory
chips to record and retrieve data from. Antennas transmit and receive the data
that is carried on the transponder and relay that information to the interrogator so
that it may be interpreted and decoded. These components can have many
different configurations, depending on the application and frequency that the
system operates at. The frequency that is used to transmit data from the tag to
the reader can very from a few hundred kilohertz (kHz), to tens of gigahertz
(GHz). Most RFID systems used for supply chain applications utilize the 13.56
MHz or 915 MHz band. These are utilized because of the cost and read range
characteristics they display (AIM, 2001).

RFID systems are classified into two main categories, active or passive.
The distinguishing factor is how the transponder receives its power to
communicate data. Transponders used in active systems carry an internal power
source such as a battery or cell to power the communication circuitry. This type
of transponder actively transmits its data and can be read at distances up to
many miles. Active transponders vary in sizes from a matchbox to a shoebox,
usually dependent on the size of the battery. Due to the size of active tags, they

are generally utilized in long-distance, capitol asset tracking scenarios.

Examples include monitoring railroad cars within switching yards, and automatic
toll collection systems.

Passive RFID systems do not have an internal source of power to
energize the transponder. The data transmission cycle of a passive system
begins when the reader’s antenna transmits an interrogation signal. As a
transponder enters this RF field, its communication circuitry is energized. The
antenna then recognizes the transponder’s signal, and the interrogator decodes
the data. The orientation between the transponder and the antenna can
influence performance of these systems, as RF fields can fluctuate. Read
ranges for passive systems vary from a few centimeters to many feet, depending
on the system/antenna/tag configuration (Kipp, 2003).

The data that are stored on transponders resides in a memory chip that
forms part of its communication circuitry. Memory “may comprise read-only
memory (ROM), random access memory (RAM) and non-volatile programmable
memory for data storage depending upon the type and sophistication of the
device (AIM, 2001).” ROM memory contains the transponder operating
instructions that allow it to properly transmit its data. Temporary data storage
during transponder field interrogation is held in RAM. Data stored on write-once,
read-many (WORM) or re-writeable (RNV) transponders is held in non-volatile
programmable memory, usually electrically erasable programmable read only
memory, or EEPROM (Clarke, 2001). Transponder manufacturers allocate

memory in many different configurations. Some transponders contain only ROM

memory, while others combine a section of RAM and a section of RNV on the

same circuit.

Data Protocol Specifications

Global standards groups such as International Organization for
Standardization (ISO) and EPCglobal, inc. have published standards for RFID
communications. These standards classify the data format and air interface
protocols for systems that operate across all bandwidths. The purpose of these
standards is to allow a commonality to exist between manufacturers and users of
RFID equipment. This allows transponders manufactured by Company A to be
read by an interrogator manufactured by Company B. The standards also define
the format that the data will be encoded in. This allows a transponder encoded
with data from User A to be decoded and recognized by User B.

In the United States, RF ID standards for interrogator to transponder
communication are being led by EPCglobal. EPCgIobaI is an international
organization that was formed between the Uniform Code Council (UCC) and the
European Article Numbering International (EAN) bodies. The goal is to drive
global adoption of its standards for the identification of individual items within the
world’s supply chains. The group has defined a new identification system that
incorporates a company’s UCC product code with new information to form the
Electronic Product Code, or EPC (EPCgIobal, 2004). The EPC is a number that
uniquely identifies an object and may be communicated using any of the defined

EPCglobal standards. Currently, these standards are being utilized by early

adopters such as Wal-Mart and Target stores for their RFID implementation
plans (Wal-Mart Stores, 2004).

The data communication standards defined by EPCglobal for passive
transponders are divided into two classes. Class 0 transponders are read-only,
factory programmed and must have the following characteristics:

- Being factory programmed with EPC, and optionally other data,

0 Being read be the reader

0 Being selected as part of a related group of transponders, and

0 Being individually destroyed (Auto-ID Center (I), 2003)
Class 1 transponders utilize WORM memory, are factory or user programmable,
and must adhere to the following characteristics:

0 Being programmed with EPC and possibly other data

0 Being read by the interrogator

. Being selected as part of a related group of labels, and

0 Being individually destroyed (Auto-ID Center (II), 2003)

Newer standards are currently being developed through EPCgIobal to
support the increasing memory capacity of newer generations of transponders.
The new Class 1.2, or C1V2, standard is currently being written with input from
both manufacturers and users of RFID equipment. This standard defines how
“data is encoded on the EPC tag itself," and defines new numbering rules for
specific values within the EPC to increase interrogator efficiency (EPCgIobal,
2004). Wal-Mart expects “to move to C1V2 protocol tags when they become

available (Wal-Mart Stores, 2004)."

Application to the Supply Chain

A company’s supply chain may see multiple benefits when RFID
technology is incorporated into its products. These benefits begin with the raw
materials and continue through to the point of retail sale of the finished goods.
“Clearly, different companies will benefit from different applications — some may
find a significant improvement in sales through improving on-shelf availability
while others may find cost savings from reducing shrinkage (Agarwal, 2001).”
Agarwal’s assessment of the benefits that RFID brings to the consumer goods
industry has identified over 8 specific benefit cases for its use. While some do
not directly apply to fresh meats, all have a strong retail application that will
benefit the supplier, retailer or consumer at some level.

Aganrval’s benefits are based on two requirements:

0 The tagging of individual items
. The infrastructure to support multiple read points in the supply
chain.

An evaluation of Aganrval’s benefits and their application to fresh meat
packaging shows many benefits. Suppliers will benefit from the ability to
automate proof of delivery. As goods reach the retailer, automated scanning of
the packages reduce human error in counting, while automatically sending
confirmation of receipt to the supplier. Retailers will benefit from an improvement
in on-shelf availability of products through a reduction in out-of-stock inventories.
As shelf inventory levels hit their reorder point, that data may be automatically

sent to the warehouse or stock room to replenish the shelf with product. This

also has the built-in benefit of eliminating stock verification by automating the
process. Retailers will also benefit from the incorporation of shelf-life dating to
their product so that stores can utilize first-in-first-out (FIFO) stock rules with
perishable products. “Goods with expiration dates can be better managed
(moved more quickly when code dates are near), reducing the need for write-offs
due to spoilage. This benefit is particularly important for perishable and date-
specific products (AT. Kearney, 2004)." RFID will also allow for increased
security of high margin products by allowing for real time locating within the store,

and conversely, signaling when a product has left the store.

Retailer Implementation of RFID Technology

Many retailers of consumer goods will soon realize the potential that RFID
brings and the benefits that may be incorporated to their existing supply chains.
Wal-Mart Stores of Bentonville, Arkansas, has recently instituted an
implementation plan that requests their top one hundred suppliers to apply
transponders at the case and pallet level. Wal-Mart expects to receive these
items beginning January, 2005 at one Regional Distribution Center (DC), one
Grocery DC and one SAM’S Club cross-dock DC. Their implementation strategy
requests all of their domestic US. business suppliers to apply transponders by
the end of 2006.

Wal-Mart has categorized the benefits of RF ID into three areas of
influence: supplier, collaborative and internal. “With respect to the collaborative

benefits we are focusing on providing you [supplier] with additional visibility into

10

our supply chain to improve Store/Club in-stock and potentially improve the
freight claims process (Wal-Mart Stores, 2004).” Each supplier is thus required
to understand and implement RFID technology into their own organization, and
provide solutions to individual issues that may arise from packaging/product
material interactions.

As of March 2004, many other national and international retailers have
announced plans to incorporate RFID into their organizations that mirror the Wal-
Mart announcement. Target Inc. has announced a pilot that requires their top
vendors to apply transponders to cased goods beginning in 2005. European
based grocers Tesco, Marks & Spencer and Metro AG have also announced
similar pilot trials that will commence in late 2004. All of these companies have
identified the use of UHF systems adhering to the EPCglobal Class 0 & 1
protocols to be acceptable solutions. Class 1.2 will be implemented when

equipment becomes available.

Animal Agriculture Tracking

The United States Animal Identification Plan (USAIP) is a recent industry-
state-federal partnership that has been formed to protect American consumers
by safeguarding and improving the security of the animals that comprise the
national herd. The USAIP “defines the standards and framework for
implementing a phased-in national animal identification plan (USAIP, 2003).”
The USAIP has been formed with the help of the United States Department of

Agriculture (USDA) and the National Institute for Animal Agriculture. Its purpose

11

is to create a plan that will “identify individual animals or groups, the premises
where they are located, and the date of entry to that premises (USAIP, 2003)."

The USAIP is divided into three phases for implementation:

0 Phase 1 - premises identification

0 Phase 2 - individual or group identification for interstate and intrastate

commerce

0 Phase 3 - retrofitting remaining processing plants and markets and other

industry segments with appropriate technology that will allow for the
tracking of animals throughout the livestock marketing chain.

The USAIP notes that “the goal of 48-hour traceback, most likely, will require
the integration of RFID technology to automate the recording of animal
movements (USAIP, 2003).” By utilizing the memory capacity of the
transponders, the necessary data may be carried on a transponder attached to
the animal. In the event of a disease outbreak which necessitates a quarantine
of specific animals, RFID will provide visibility in the supply chain to allow for the

safe removal of those animals from stock.

Cattle and Food Borne Pathogens

Bovine Spongiform Encephalopathy, or BSE, is a degenerative disease
that affects the nervous system of cattle. This disease can be spread to other
cattle through feed contaminated with meal made from BSE-infected animals.
The disease can also spread to humans who eat meat from a diseased animal

(Reagan, 2002). To prevent an outbreak in the United States, the USDA has

12

instituted three firewalls to combat its entry. The first firewall bans the import of
animals and their products from countries with confirmed cases of BSE. The
second firewall involves a “surveillance program that focused on testing high-risk
cattle for BSE”; the third firewall bans “at-risk animal protein in cattle feed
(Reagan, 2002).” Despite these precautions, the first known case within the
United States was discovered in a cow in Washington state in December, 2003
(Associated Press, 2004).

Escherichia coli O157:H7 is a foodborne bacterial pathogen that can
cause severe sickness or death in humans who consume food contaminated with
it. “Between 1993 and 1998, most (72%) of the E. coli O157:H7 outbreaks were
foodborne and of the foods implicated in the outbreaks, beef was responsible for
45% of the cases and 90% of the time the beef product was ground (Sofos et al.,
2003)”

Foot and Mouth Disease, or FMD, is a highly contagious viral disease that
can be rapidly transmitted though a population of animals. The disease primarily
affects cloven-hoofed animals, and can be harbored by certain species, such as
cattle, for over 2.5 years. Animals can become infected “through direct contact
with an infected animal or indirect contact through contaminated equipment,
facilities, people, feedstuffs and other materials (Bruning-Fann, 2002).” In the
event of an outbreak, Bruning-Fann states that “with a highly contagious, rapidly
disseminating disease, it is critical to know the livestock inventory and location of

all farms." RFID can provide this visibility into the livestock and fresh meat

13

supply chains to help quarantine animals, equipment or products that may be

carriers of these pathogens.

Tray-Packed 8. Case-Ready Meats

Fresh meats have traditionally been cut, packaged and labeled in the
store butchery before being placed on the retail shelf. According to the American
Meat Institute (AMI), “prior to the 19603, carcasses were shipped to retailers for
breaking and cutting. In the 1960s, packing plants created “boxed beef or boxed
meats,’ which provided smaller pieces called primals and sub-primals that could
be further out and wrapped at the retail store (AMI, 2004).” The traditional form
of packaging fresh meats involved placing the cut on an expanded polystyrene
(EPS) tray, and wrapping the package with polyvinyl chloride (PVC) film (Brody,
2002).

Newer meat-packing technologies have ushered in the “case-ready"
package to the grocery retailer’s shelves. A package is deemed case-ready
when its “objective is to deliver prepackaged red meat with several days of shelf
life to the retailers ready for the (display) case (Brody, 2002).” Case-ready meats
are gaining favor with retailers due to the inherent benefits associated with the
labor, packaging and distribution that the unit has to offer (AMI, 2004). With the
centralization of meat plants, retailers may see a reduction in labor costs by
eliminating the need for an in store butchery and associated personnel. Due to
their engineered atmospheres, they allow for longer shelf and product lives.

Case-ready meats will also reduce out of stocks by allowing the store to

14

purchase specific cuts of meat. The AMI reports that 1.2 billion packages of
case-ready meats were sold in the US. in 2000, with the potential to sell 9 billion

packages in the future (Mandigo, 2002).

Data Integration

The ultimate goal of RFID is to create the “perfect purchase order”:
assurance that the correct product, in the correct quantity, is delivered from
product suppliers to distribution centers to stores on schedule, with no mistakes.
RFID tracking capabilities combined with the computer database enable accurate
knowledge of product information, such as product location, quantity, and value
at all stages of the supply chain. Gaining better control over inventory would
contribute to improvement in processes, products and services, design,
methodology, and supplier relations (Clarke, 2002).

As the USAIP and retailers begin to implement RFID into their business
operations, the marriage of data from these two plans may be incorporated. The
seamless transmission of data from livestock to the retail shelf will allow the two
sources of data to be combined into a single source of storage: the RFID
transponder. The operational potential of such a data system requires the
knowledge of how the properties of fresh meats interact with RF transmissions.
This research will focus on how two RFID systems, operating at different
frequencies, affect the transmission ability of the data stored on the

transponders.

15

Radio frequency transponders are rapidly becoming a viable technology
for tracking of perishable food items through the grocery supply chain. However,
the effects of package environment on radio frequency transponders are not well
known; the impact of colder environments on transponder performance has not
been studied. Little research has been conducted on the effects of extreme
operational environments to the performance capabilities of RF ID transponders.
Recently completed research has focused on the mapping of the read-range field
during normal laboratory conditioning: 72°F, 50% relative humidity (Kipp, 2003).
These results have shown distinct field patterns that correlate to the type of
antennae used, and at what operational frequency. Another study in the field has
focused on the effects of transponders on bloom time of beef loin muscle (Vorst,
2002). This study has shown that transponders do not adversely affect the
coloring or bloom time of the muscle with the application of the transponder to

the polyvinyl chloride (PVC) ovenivrap film.

16

MATERIALS AND METHODS

This study will involve packaging individual beef loin steaks typically seen
in butcher shops and retail meat cases. Transponders will be placed on the face
of the package, located on the outer surface of the ovenrvrap film. The packages
will be conditioned at the average retail case refrigerated temperature of 41° F.
Packages will then be placed within the read field of each system’s antenna, and
read-cycled to test if a transponder’s data will be recognized. A read-cycle will
consist of one transmission of data from a transponder to the host computer.
Packages will then be stacked directly on top of another within the system’s read
field. The system will again be read-cycled to determine if all of the packages
read. If all of the packages record a read, the test will be repeated with the
addition of one package to the stack. The test will end when the system does not
record a read from at least one package in the stack. Both test procedures will
be repeated after conditioning the packages at a frozen temperature. For
consistency between experiments, all tests were conducted with the antenna
parallel to the transponder. This configuration is recommended by the

manufacturers for optimal performance.

RFID System — Texas Instruments (Plano, TX)

A Texas Instruments S6350 RF ID reader system was obtained for
purposes of testing. The system operates at a frequency of 13.56 MHz, and
houses the interrogator and antenna in a single unit (8”x 5”x 1”). See Appendix

1, Figure 1. The communication protocol for the system is proprietary to TI,

17

designated Tag-itm. The interrogator was managed with 86350 Reader Utility
software that was installed on a Dell Inspiron 2500 laptop computer. The
computer was connected to the interrogator through a serial port connection.
The operational read range for the system was determined to be approximately
five inches from the top panel on the unit. The system was installed on a
laboratory bench in a conditioned laboratory at 72° F and 50% relative humidity.

See Appendix 1, Figure 1.

RFID System - Matrics Incorporated (Columbia, MD)

A Matrics EKT-001 RFID reader system that operates at a frequency of
915 MHz was obtained for testing. The interrogator unit was model RDR-001,
attached to an ANT-001 antenna via coaxial cable. The Matrics system operates
on the EPC Class 0 data protocol. The interrogator was controlled by Matrics
Tag Tracker 3.0.1 software that was installed on the aforementioned laptop. The
computer was connected to the interrogator through a serial port connection.
The operational read range for the system was determined to be approximately
twenty feet from the face of the antenna. The system was also installed in the
conditioned laboratory at 72° F and 50% relative humidity. The antenna (29”x
13”x 2”) was mounted on a wood frame, with the interrogator was positioned 10
feet to the rear. This location was recommended by the manufacturer so that the

interrogator was not in the antenna’s RF field. See Appendix 1, Figure 2.

18

Transponder Preparation

Texas Instruments Rl-l01-112A 13.56MHz Tag-itTM transponders were
obtained from Texas Instruments. Transponders were received on roll stock and
were then individually cut to a uniform dimension of 2” by 2”. Transponders were
then individually placed within the read range of the unit. The system was read-
cycled once to record a positive read from the transponder. Once operational
status was confirmed for a transponder, they were affixed to the adhesive side of
an Avery 5163 paper label. This simulates a typical adhesive pricing label that is
applied directly to a meat package. The front of the label was sequentially
marked with a numeral starting from 1 and ending at 50. The numbers aided in
data collection and indicated the order in which the samples were tested.

Matrics DDS-001 adhesive label transponders were obtained for testing.
Transponders were received individually separated and packaged in stacks of
50. Tags were individually placed within the read range of the antenna. The
system was read-cycled once to record a positive read from the transponder.
Once operational status was confirmed, they were also sequentially numbered to

aid in data collection.

Preparation and Packaging of Tray-Packed Beef Loin Muscles

The preparation and packaging of beef loin muscle in this study was
replicated using the method developed by Vorst. For the unique purposes of this
study, certain omissions from his procedure occurred. The exact time from

slaughter to simulated distribution was not recorded, as the color of the meat was

19

not being measured. The use of high oxygen barrier nylon pouches were also
omitted because the prevention of oxygen permeation into the package was not
required (Vorst, 2003).

Two USDA Select strip loin muscles were obtained from a Lansing,
Michigan distributor. Vorst utilized this out due to the absence of connective
tissue, and the uniformity provided between cuts. The strip loin muscles were
labeled, vacuum packaged and weighed 15le and 18.85Ibs whole, respectively.
The muscles were then unpackaged and drained of all purged fluids. The
individual beef muscles were sliced to a thickness of 1/2 inch through the lateral
direction of the muscle. See Appendix 1, Figure 3.

Adams 48 trays (R.L. Adams Plastics Inc., Wyoming, MI) were used for
this study to simulate common retail packages. Trays were constructed of
expanded polystyrene (EPS) and measured 23.5cm x 18.4cm x 0.530m in length,
width and thickness. Sealed Air Corporation’s Dri-LocTM AC-25 absorbent pads
were placed on top of the trays and directly under the steak. The pad is
constructed of virgin fluff pulp, enveloped in polyethylene film. The pad
measured 10.16cm x 17.8cm and has an advertised fluid absorbency of 40-50
grams. The pad is used for the absorption of excess blood and purging from the
beef muscle. Plasticized PVC, 0.5 mil thick, was used as the ovenrvrap film. The
low barrier film was manufactured by Cryovac. This tray/pad/film combination is
commonly found in retail meat applications.

Sample packages containing one steak were constructed and hand

packed. The packaging process involved placing an absorbent pad centered in

20

the EPS tray. An individual steak was then placed centered on top of the pad. A
pre-cut piece of overwrap film was then centered over the tray, and wrapped
around the edges to form the “face” side of the package. The excess film was
wrapped and secured to itself on the underside, or “bottom” of the tray. This
produced a package with zero headspace. Packages were uniform in
appearance to typical retail meat packaging as described by Brody (Brody,
2002). The placement of the absorbent pad, steak and overwrap film were
uniform for each sample produced. After statistical consultation, a total of 50
samples were constructed and found to be a statistically significant population

size (CANR, 2003). See Appendix 1, Figure 4.

Test Procedure

The test procedure was divided into 4 consecutive experiments.

0 Experiment 1 - Refrigerated samples tested with TI system

0 Experiment 2 — Refrigerated samples tested with Matrics system

0 Experiment 3 - Frozen samples tested with Matrics system

0 Experiment 4 — Frozen samples tested with TI system

The test procedure used for these experiments was constructed to provide
consistency between them while utilizing a perishable product. The procedure
allowed the samples to maintain a refrigerated temperature during and between
each experiment. After refrigerated testing was complete, the samples were
frozen once for the remainder of the experiments. Samples were tested with one

transponder attached and conditioned for each experiment. Transponders were

21

able to be securely attached to the film and safely removed for later use.
Although the TI and Matrics systems operate at different frequencies,
transponders were never tested concurrently. This reduced the risk of false

negatives being recorded due to interference from the other transponder.

Experiment 1 - Refrigerated Samples Tested With Tl System

Step 1 - Attach Transponders

A Tl transponder was attached to the overwrap film on the “face” of a
sample package. The transponder was attached with the adhesive label on the
left side of the package. This positioning replicated the label location of a typical
retail meat package. This procedure was repeated for all 50 samples. See
Appendix 1, Figure 5.
Step 2 - Condition

Samples were conditioned at 41° F for 48 hours to equilibrate both the
package and transponder. Samples were placed in refrigeration with nothing
stacked on top of them. This eliminated the chance of a transponder failing due
to compressive forces.
Step 3 - Test Scenario

A sample was individually removed from refrigeration and placed above
the S6350 unit. The $6350’s antenna is located near the surface of the unit.
The sample was orientated with the “face” of the package parallel and facing

toward the top of the unit. The sample was located approximately five inches

22

away from the S6350. This orientation allowed for an unimpeded communication
path for the data, as only an air interface existed. The unit was read-cycled once
and a read/no read was recorded as determined by the Reader Utility software.
This procedure was repeated for all 50 samples. See Appendix 1, Figure 7.
Step 4 - Test Scenario

A sample was individually removed from refrigeration and placed above
the S6350 unit. The sample was orientated with the “bottom” of the package
parallel and facing toward the top of the unit. The sample was located
approximately five inches away from the $6350. This orientation allowed for the
meat to act as a substrate and provide a barrier for the data. The unit was read-
cycled and a read/no read was recorded as determined by the Reader Utility
software. This procedure was repeated for all 50 samples. See Appendix 1,
Figure 8.
Step 5 - Test Scenario

A sample was individually removed from refrigeration and placed with the
“bottom” of the package on the lab bench. Another random sample was removed
from refrigeration and placed directly on top of the first sample in the same
orientation. This procedure replicates typical retail meat displays where
packages are stacked in a column. The S6350 was placed above the samples
and read-cycled. A read/no read was determined for each sample. If both
samples were recognized, another sample was removed from refrigeration and
placed in the same orientation on the stack. The 86350 was read-cycled and a

read/no read was determined for each sample. This procedure was repeated

23

until a loss of one or more transponders was reported with the Reader Utility
software. The number of successful packages in a stacked configuration to be
read without loss was recorded. This test procedure was replicated 10 times
using random samples. See Appendix 1, Figure 9.
Step 6 - Remove Transponders

Tl transponder labels were removed from the sample packages without
damaging the label or overwrap film. This was accomplished because the
refrigerated temperature of the package reduced the aggressiveness of the
label’s adhesive. Transponder labels were placed in refrigeration to retain the

same conditioning as the sample packages.

Experiment 2 - Refrigerated Samples Tested With Matrics System

Step 1 - Attach Transponders

A Matrics transponder was attached to the ovenrvrap film on the “face” of a
sample package. The transponder was attached with the adhesive label on the
left side of the package. This positioning replicated the label location of a typical
retail meat package. This procedure was repeated for all 50 samples. See
Appendix 1, Figure 6.
Step 2 - Condition

Samples were conditioned at 41° F for 48 hours to equilibrate both the

package and transponder. Samples were placed in refrigeration with nothing

24

stacked on top of them. This reduced the chance of a transponder failing due to
compressive forces.
Step 3 - Test Scenario

A sample was individually removed from refrigeration and placed in front
of the ANT-001 antenna. The sample was orientated with the “face” of the
package parallel and facing toward the front of the antenna. The sample was
located approximately five inches away from the antenna. This orientation
allowed for an unimpeded communication path for the data, as only an air
interface existed. The RDR-001 interrogator was read-cycled once and a
read/no read was recorded as determined by the Tag Tracker software. This
procedure was repeated for all 50 samples. See Appendix 1, Figure 10.
Step 4 - Test Scenario

A sample was individually removed from refrigeration and placed in front
of the ANT-001 antenna. The sample was orientated with the “bottom” of the
package parallel and facing toward the front of the antenna. The sample was
located approximately five inches away from the antenna. This orientation
allowed for the meat to act as a substrate and provide a tortuous communication
path for the data. The RDR-001 interrogator was read-cycled and a read/no read
was recorded as determined by the Tag Tracker software. This procedure was
repeated for all 50 samples.
Step 5 - Test Scenario

A sample was individually removed from refrigeration and placed with the

“bottom” of the package on the lab bench. Another random sample was removed

25

from refrigeration and placed directly on top of the first sample in the same
orientation. This procedure replicates typical retail meat displays where
packages are stacked in a column. The ANT-001 antenna was placed above the
samples and read-cycled. A read/no read was determined for each sample. If
both samples were recognized, another sample was removed from refrigeration
and placed in the same orientation on the stack. The RDR-001 interrogator was
read-cycled and a read/no read was determined for each sample. This
procedure was repeated until a loss of one or more transponders was reported
with the Tag Tracker software. The number of successful packages in a stacked
configuration to be read without loss was recorded. This test procedure was

replicated 10 times using random samples. See Appendix 1, Figure 11.

Experiment 3 - Frozen Samples Tested With Matrics System

Step 1 — Condition

Samples were conditioned at 0° F for 48 hours to equilibrate both the
package and transponder. Samples were placed in a freezer with nothing
stacked on top of them. This eliminated the chance of a transponder failing due
to compressive forces.
Step 2 - Test Scenario

A sample was individually removed from the freezer and placed in front of
the ANT-001 antenna. The sample was orientated with the “face” of the package

parallel and facing toward the front of the antenna. The sample was located

26

approximately five inches away from the antenna. This orientation allowed for an
unimpeded communication path for the data, as only an air interface existed.
The RDR-001 interrogator was read-cycled once and a read/no read was
recorded as determined by the Tag Tracker software. This procedure was
repeated for all 50 samples. See Appendix 1, Figure 10.
Step 3 - Test Scenario

A sample was individually removed from the freezer and placed in front of
the ANT-001 antenna. The sample was orientated with the “bottom” of the
package parallel and facing toward the front of the antenna. The sample was
located approximately five inches away from the antenna. This orientation
allowed for the meat to act as a substrate and provide a tortuous communication
path for the data. The RDR-001 interrogator was read-cycled and a read/no read
was recorded as determined by the Tag Tracker software. This procedure was
repeated for all 50 samples.
Step 4 - Test Scenario

A sample was individually removed from the freezer and placed with the
“bottom” of the package on the lab bench. Another random sample was removed
from the freezer and placed directly on top of the first sample in the same
orientation. This procedure replicates typical retail meat displays where
packages are stacked in a column. The ANT-001 antenna was placed above the
samples and read-cycled. A read/no read was determined for each sample. If
both samples were recognized, another sample was removed from the freezer

and placed in the same orientation on the stack. The RDR-001 interrogator was

27

read-cycled and a read/no read was determined for each sample. This
procedure was repeated until a loss of one or more transponders was reported
with the Tag Tracker software. The number of successful packages in a stacked
configuration to be read without loss was recorded. This test procedure was
replicated 10 times using random samples. See Appendix 1, Figure 11.
Step 5 — Remove Transponders

Matrics transponder labels were removed from the sample packages
without damaging the label or ovenivrap film. This was accomplished because
the frozen temperature of the package reduced the aggressiveness of the label’s

adhesive.

Experiment 4 - Frozen Samples Tested With Tl System

Step 1 - Attach Transponders

A TI transponder label was then removed from refrigeration and
reattached to the ovenlvrap film on the “face” of a sample package. The
transponder was positioned in the same location as it was during Step 1. This
procedure was repeated for all 50 samples.
Step 2 - Condition

Samples were conditioned at 0° F for 48 hours to equilibrate both the
package and transponder. Samples were placed in a freezer with nothing
stacked on top of them. This eliminated the chance of a transponder failing due

to compressive forces.

28

Step 3 — Test Scenario

A sample was individually removed from the freezer and placed above the
S6350 unit. The sample was orientated with the “face” of the package parallel
and toward the top of the unit. The sample was located approximately five
inches away from the $6350. This orientation allowed for an unimpeded
communication path for the data, as only an air interface existed. The unit was
read-cycled once and a read/no read was recorded as determined by the Reader
Utility software. This procedure was repeated for all 50 samples. See Appendix
1, Figure 7.
Step 4 - Test Scenario

A sample was individually removed from the freezer and placed above the
ANT-001 antenna. The sample was orientated with the “bottom” of the package
parallel and facing toward the top of the unit. The sample was located
approximately five inches away from the $6350. This orientation allowed for the
meat to act as a substrate and provide a tortuous communication path for the
data. The 86350 was read-cycled and a read/no read was recorded as
determined by the Utility software. This procedure was repeated for all 50
samples. See Appendix 1, Figure 8.
Step 5 - Test Scenario

A sample was individually removed from the freezer and placed with the
“bottom” of the package on the lab bench. Another random sample was removed
from refrigeration and placed directly on top of the first sample in the same

orientation. This procedure replicates typical retail meat displays where

29

packages are stacked in a column. The 86350 was placed above the samples
and read-cycled. A read/no read was determined for each sample. If both
samples were recognized, another sample was removed from the freezer and
placed in the same orientation on the stack. The 86350 was read-cycled and a
read/no read was determined for each sample. This procedure was repeated
until a loss of one or more transponders was reported with the Reader Utility
software. The number of successful packages in a stacked configuration to be
read without loss was recorded. This test procedure was replicated 10 times

using random samples. See Appendix 1, Figure 9.

30

RESULTS AND DISCUSSION

The results of these experiments found that the orientation of the package
can influence the communication ability of a transponder’s data. These
experiments tested whether a transponder can communicate its data when the
package is at a refrigerated and frozen temperature. These experiments also
tested whether data could be communicated when a beef loin steak was directly
between the antenna and transponder. The test procedure measured how many
packages deep a transponder could transmit data through in a stacked
configuration.

All refrigerated and frozen samples containing a TI transponder were able
to communicate data when the transponder was facing toward and away from
the antenna. The TI system was able to communicate data when samples were
placed in a stacked configuration at both refrigerated and frozen temperatures.
The maximum number of packages was found to be four during each replication.
The maximum number of samples read can be attributed to the limited read
range of the $6350 system, which was found to be six inches. The test results
were consistent for each test scenario replication.

Refrigerated samples containing a Matrics transponder were able to
communicate data when the transponder was facing toward the antenna. When
the sample was facing away from the antenna, zero transponders communicated
data. All frozen samples were read facing toward and away from the antenna.
Refrigerated samples in a stacked configuration could only communicate a single

transponder. The sample located on the top of the stack was the only one to

31

 

communicate during each replication. Frozen samples in a stacked configuration
communicated a maximum of 9 transponders. The addition of a tenth sample to
the stack consistently saw the loss of communication for one or more

transponders. This result was consistent for each test scenario replication.

Table 1. Refrigerated Read Percentages

 

 

 

 

 

 

 

Refri erated

Toward Away

11 100% 100%
Matrics 100% 0%

 

Table 2. Frozen Read Percentages

 

 

 

 

 

 

 

Frozen
Toward Away
Tl 100% 100%
Matrics 100% 100%

 

 

 

Table 3. Depth of Reads

 

 

 

 

 

 

 

Refrigerated Frozen
TI 4 packages 4 packages
Matrics 1 package 9 packages

 

Results of this experiment show that the Matrics system could not
communicate data through a beef loin steak in common retail packaging. The
transponder could communicate when facing toward the antenna, but not away.
The stacked configuration prevented any transponders from communicating
when placed under another sample. The system was able to communicate when
samples were conditioned at a frozen temperature. This suggests that fresh
meat at a refrigerated temperature can impede the transmission of data for RFID

systems operating at 915 MHz.

32

CONCLUSIONS

The results of this experiment suggest that a loss of communication could
occur in Class 0 RFID systems operating at 915 MHz when utilized on packages
of refrigerated beef. Grocery retailers who have announced RFID deployment
strategies that adhere to UHF and EPCglobal standards need to evaluate the
potential impact this may cause to their supply chains. The loss of visibility of a
single package could nullify the advantages that have been identified by Agarwal.
This could present a potential issue for retailers that plan to implement RFID
systems into their organizations and expect to create a competitive advantage
based on efficient supply chain management.

The loss of data communication would erase all of the benefits Agarwal
has identified. Automated proof of delivery would produce erroneous results due
to incorrect product inventories. This could lead to delays in restocking,
incomplete stock verifications and a loss of virtual inventory. The ability of Wal-
Matt’s suppliers to have visibility into their supply chain could lead to incorrect
order quantities and superfluous freight claims based on these inaccuracies. As
suppliers conduct remote inventory polling, these inaccuracies will lead to
incorrect reorder quantities and shipment times.

The loss of communication leads to data that is not transmitted and
shared with the proper databases. As the USAIP is implemented, dynamic
amounts of data will be routinely added to the memory of transponders and

databases. As this data progresses down the fresh meat supply chain, a sudden

33

loss of visibility could lead to the inability to recall packages of potentially
contaminated meat.

The goals of this study were to understand how beef muscle interacts with
transponder communications when conditioned in refrigerated and frozen
environments. After completing the experiment, these goals have been
achieved. The knowledge that refrigerated temperatures adversely affect the
performance of fresh meat packages containing an RFID transponder operating
at 915 MHz has been learned. The use of RF ID systems operating at 13.56 MHz
do not appear to be affected by these temperatures. This frequency may provide

a better solution for products containing moisture in cold-chain applications.

34

 

FUTURE RESEARCH

As retailers such as Wal-Mart see their RFID implementation plans come
to fruition, suppliers of cold—case goods will need to understand the impact that
their products have on the transmission of data in the UHF bandwidth. The
performance of RFID transponders in cold-chain applications extends beyond
their use on fresh meat packaging. Items such as ice creams, frozen dinners,
pickled foods, cheeses, yogurt, dough, vegetables, creams, milks and beverages
are just a few of the products that contain moisture and require cold storage
environments. The extent to which these products and their packaging materials
inhibit or allow data communication remains to be studied.

Future research may be divided into three areas:

0 Packaging material analysis

0 Product analysis

a RF bandwidth analysis

These areas represent aspects of the packaging system that may be
investigated to understand specific variables and how they impact RFID
operations. By identifying variables that prohibit or prevent RF transmission,
packaging engineers can be proactive in their approach to package design.
Innovative packaging solutions will then allow for the seamless integration of
RFID into the package system. These are areas that need to be investigated to

overcome the limitations that have been identified in this experiment.

35

 

APPENDIX

Appendix 1. Photos and Graphics of Materials and Equipment

   

Figure 2. Matrics ANT-001 Antenna and Dell Inspiron 2500 computer

36

 

 

 

 

Figure 4. Face of tray packed sample

37

 

 

Figure 5. Sample with TI transponder label attached

 
 

Figure 6. Sample with Matrics transponder label attached

38

 

Package facing toward
S6350 Reader

 

 

 

 

 

RF Data Exchange/

Air Interface \

Tl $6350 Reader Unit

 

 

’\
I)

 

 

-->
->

 

 

 

 

l

K\\

 

 

 

Figure 7. Graphic of package facing toward $6350 Reader

 

Package facing away

from 86350 Reader
(circle denotes
transponder location
on face of package) /C )
\
RF Data Exchange/

Air Interface \

Tl S6350 Reader Unit

 

 

 

 

 

/

 

 

 

 

--9

 

\
«aw-v

 

‘

TQ

 

 

Figure 8. Graphic of package facing away from $6350 Reader

 

 

I TI S6350 Reader Unit‘l \

I RF Data Exchange/ —

Air Interface \
Packages in stacked
configuration \ @

Figure 9. Graphic of packages in a stacked configuration

 

 

 

<--->
<- ->

 

 

 

 

Figure 10. Picture of single package facing toward Matrics ANT-001 antenna

40

 

Figure 11. Samples in a stacked configuration

41

REFERENCES

Aganrval, V. 2001, September 1. Assessing the Benefits of Auto-ID Technology
in the Consumer Goods Industry. Auto-ID Center, Massachusetts Institute
of Technology, [On-line]. Available:
http:l/archive.epcgIobalinc.orglpublishedresearch/CAM-WH-003.pdf.

AIM, Inc. 2001, August 23. Radio Frequency Identification RFID A Basic Primer.
Automatic Identification Manufacturers, [On-line]. Available:

 

http:llwww.aimglobal.org/technologies/rfid/resources/RFIDprimer.pdf. F
Albano, S. and Engels, D.W. 2002, June 1. Auto-ID Center Field Trial: Phase I

Summary. Auto-ID Center, Massachusetts Institute of Technology, [On-

line]. Available: http:l/archive.epcglobalinc.org/publishedresearch/MIT-

AUTOID-TR-006.pdf. -
AMI. 2004, March. Case Ready Meats. American Meat Institute, [On-line]. I-

Available:
http:llwww.emiforg/FactsandFigures/FactSheetCaseReadyMeat.pdf.

AMIF. 1999. 1999 US. Cold Temperature Evaluation. American Meat Institute
Foundation (AMIF), [On-line]. Available:
http:llwww.amif.org/pdf/cold%20temp%20final.pdf.

Associated Press. 2004, March 24. US. Pressures Partners Over Beef Ban.
MSNBC.com, [On-line]. Available: http:llwww.msnbc.comlid/4611358/

AT. Kearney. 2004. RFID/EPC: Managing the Transition (2004-2007). AT.
Kearney, [On-line]. Available:
http:llwww.atkearney.com/shared_res/pdflRFlD-EPC_s.pdf.

Auto-ID Center (I). 2003, February 23. Draft Protocol Specification for a 900
MHz Class 0 Radio Frequency Identification Tag. Auto-ID Center,
Massachusetts Institute of Technology, [On-line]. Available:
http:llwww.epcglobalinc.org/standards_technologylSecure/v1 .OIUHF-
classO.pdf.

Auto-ID Center (II). 2003, February 1. 13.56 MHz ISM Band Class 1 Radio
Frequency Identification Tag Interface Specification: Candidate
Recommendation, Version 1.0.0. Auto-ID Center, Massachusetts
Institute of Technology, [On-line]. Available:
http:llwww.epcglobaline.org/standards_technology/Secure/v1 .0/HF-
Class1.pdf.

42

Brody, AL 2002. Case-Ready Fresh Meat: Is It Here or Not? Food Technology
56:77-78

Bruning-Fann, CS. 2002. Lessons Learned From the Foot and Mouth Disease
Outbreak in the United Kingdom and Our National Plan in the Event of
Such an Outbreak. Proceedipqs of the 55‘” Annual Reciprocal Meat
Conference. American Meat Science Association, Savoy, IL. pp. 7-9.

CANR. 2003. Staticticalconsulatation.

Clarke, R. 2001. Radio Frequency Identification: Smart or Intelligent
Packaging? Journal of Packaging Science and Technology 102233-247

Clarke, R. 2002. Radio Frequency Identification: Will It Work In Your Supply
Chain? Worldpak 2002, CRC Press, Boca Raton, FL. pp. 654-662.

EPCglobal. 2004. EPC Tag Data Standards, Version 1.1 Rev. 1.22. EPCglobaI,
[On-line]. Available:
http:l/www.epcglobalinc.org/standards_technology/EPCTagDataSpecificat
ion11rev122.pdf.

Kipp, A. 2003. A Proposed Standard for Evaluating the Performance of Radio
Frequency Identification Antennas, Tags and Readers. Master’s Thesis,
Michigan State University.

Mandigo, R.W. 2002. Ingredient Opportunities for Case-Ready Meats. Food
Product Design. February 2002.

Reagan, J. 2002. The Beef Industry Plan: Keep America Free of Bovine
Spongiform Encephalopathy. Proceedings of the 55'" Annual Reciprocal
Meat Conference, American Meat Science Association, Savoy, IL.

Ryan, M. 2002. A Model for the Implementation of A Radio Frequency
Identification System Into A Warehouse Environment. Master’s Thesis,
Michigan State University.

Sofos, J.N., Skandamis, P., Stopforth, J., and Bacon, T. 2003. Current Issues
Related to Meatbome Pathogenic Bacteria. Proceedings of the 56’"
Annua_l Reciprocal Meat Conference, American Meat Science Association,
Savoy, lL. pp. 33-37.

USAIP. 2003, September 29. Draft Documentation for 2003 USAHA
Presentation/Discussion, Version 4.0. United States Animal Identification
Plan, [On-line]. Available: http:l/usaip.infolUS_AlP_Plan
_Draft_092903_Ver_4_0_Master_.pdf.

43

 

Vorst, KL. 2002. Development of a Material Testing Protocol for Evaluation of

Radio Frequency Transponder Effect on Bloom Time of Beef Loin Muscle.

Master’s Thesis, Michigan State University.

Vorst, K.L., Clarke, R.H., Allison, GP, and Booren, AM. 2003. A Research
Note on Radio Frequency Identification Transponder Effects on Bloom of
Beef Muscle. Meat Science 672179-182

Wal-Mart Stores and Sam’s Club. 2004. RFID Guidelines and Requirements,
Document Version: 1.4.6656297.

Wal-Mart (I). 2004. Letter to author.

44

 

:1. was ' ‘

   

lililljllllllllllllllsljl