Case Study: Michigan Department of Agriculture

A Case Study on Tracking and Eliminating Tuberculosis (TB) from Animal

Population with RFID Implementation

Courtesy: Michigan Department of Agriculture

The Client

The Animal Industry Division of the Michigan Department of Agriculture (MDA) is responsible for working with a statewide clientele of more than 15,000 beef and dairy producers (the producers) that raise approximately 1.1 million livestock animals. The livestock is worth close to $800 million.

The Challenge

Free movement and trade of healthy animals is the key to the profit and survival of the livestock business. If a cow cannot be traded in an auction market or sent to a slaughterhouse, the producer is unable to recoup the investment made in the material and infrastructure needed to raise and feed the cow. A widespread disease outbreak that kills the animals or requires the producer to depopulate his herd can have devastating consequences for an individual producer as well as other businesses that depend on healthy livestock industry for their income.

A few years back, the producers and the state of Michigan were looking at this stark scenario. After initial discovery of Tuberculosis (TB) in the tissue samples of a deer herd in Northern Michigan, the Michigan Department of Agriculture (MDA) found the disease present in beef herds in 13 northern counties of the lower peninsula of the state. This led to depopulation of several herds. The neighboring counties and states were also weary about the spread of the disease inside their borders over time. The disease spreads when a healthy animal comes in contact with the saliva or exhaled air of an infected animala situation common on farms as well as auction markets where animals share feed and drinking stations. In 2000, the United States Department of Agriculture (USDA), which was monitoring the situation and saw the growing spread of the disease, revoked Michigan's status as a TB-free state (for animals), triggering various restrictions on animal movements in and out of the Michigan counties. The restrictions included extra TB tests required for the animals, and added overhead per animal in terms of dollars and time spent compared to animals in TB-free states. Clearly, this was not conducive to business for the state's livestock industry.

To regain the TB-free status, the officials from MDA continued to test the livestock for TB in the affected counties, but they faced several issues. Because the number of animals testing positive was small compared to the number of total animals, it didn't make sense to depopulate them all or restrict movement of (ability to buy and sell animals) animals from all farms. To complicate matters, TB can remain dormant for a while before surfacing. An animal that tested healthy could, over time, develop TB. This meant uniquely identifying every animal and keeping a very accurate record of its movement. A typical bovine can change hands three times from birth to slaughterhouse. With the large number of animals involved, the paper-based tracking was turning out to be costly, time consuming, and prone to errors.

Faced with these challenges, the MDA looked at deployment of the National Farm Animal Identification Records (FAIR) program. RFID technology was seen as an enabler for this project.

Scope of the Project

The scope of the initial project, limited to the 13 affected counties, was to develop an Electronic Animal Identification (EID) system with the goal to identify all bovine (cattle) in the affected areas, and track and record their movement to eliminate infected animals. Successful completion of the project would restore Michigan's status as a TB-free state (for animals) over five years,^[1] provided no more infected animals were found. Of the 15,000+ farms in the state, approximately 1,600 farms located in the affected 13 counties participated in the project. The farms contained approximately 70,000 animals. The project was implemented in four phases:

^[1] The TB-free status is not awarded right away because TB can remain dormant for a while. A 5-year period ensures that multiple generations of animals are tested and the ones with dormant TB are identified and eliminated.

• Phase 1: Obtain and load animal farm location (premises) ID information into FAIR database.

• Phase 2: Develop an electronic data recording system, incorporating data from the old paper-based system.

• Phase 3: Record animal movement to and from markets and processing plants.

• Phase 4: Implement movement permit system to track animals as they moved from one place to another.

The USDA provided a grant for this project, which totaled $1.5 million over three years from 2001 to 2004. Most of this money was used to set up the requisite infrastructure for the project. Additionally, the MDA used its own people resources to test and record animals as well as manage the project.

Hardware and Software Products Used

• National Farm Animal Identification Records (FAIR) program: Holstein Association, a non-profit membership organization has tracked animal identification and pedigree information for decades and made it available to its members. It maintains a National FAIR program to enable animal identification and tracking. MDA partnered with National FAIR to use its model to develop the state's Electronic Animal Identification (EID) tracking system. This central database stores all the information about the animals, including their unique ID number, pedigree, TB test date, and test results.

• Tags: Close to 180,000 RFID tags made by Allflex were issued to producers at the cost of $400,000.

• Handhelds, computers, and printers: Fifty Psion handhelds were used as mobile readers at the cost of $120,000. Fifty computers were used at the cost of $50,000, and 30 printers were used at the cost of $9,000.

• Stationary readers: Readers and antennae made by Allflex were installed at various animal markets and slaughterhouses, averaging $10,000 per facility.

• Movement Permit Application: A Web-based application was developed, which, based on the animal ID number and certain qualifying questions, can check the FAIR database and, if appropriate, generate a movement permit for an animal that the producer can print on his computer. The permit is required at all animal markets and slaughterhouses in Michigan and surrounding areas.

The Solution: How It Works

The EID (Electronic Animal Identification) system consists of four components:

1. An RFID ear tag per animal, which sends the specific animal ID number to the reader.

2. Readers, which receive the signal from the tag and convert it into a unique number.

3. Host computers, which receive information from the reader and deliver to appropriate software program.

4. Software, which collects and analyzes the data.

For more information about how host computer and software interact, see Chapter 3, "Components of RFID Systems."

All animals are assigned a unique number using the American Identification Numbering (AIN) system. This number and the corresponding information about the animal are loaded in the FAIR database on an on-going basis. This data synchronization step is critical to the proper execution of the system. MDA officials armed with hand-held readers and computers ensure that the correct data is recorded. Farmers are also issued certain tags (with IDs) to be used on newborns. To date, approximately 180,000 RFID tags have been issued to animals on 1,600 farms in the affected counties.

To identify animals as they change hands, RFID readers are installed in 12 live-stock auction markets in Michigan along with seven slaughter plants in Pennsylvania, Wisconsin, Illinois, and Michigan. The majority of Michigan's livestock passes through these facilities. As animals pass through the narrow animal alleys in these locations, the stationary readers mounted on the side read their IDs from the tags in their ears. These stationary readers can read tags up to 36" away (although the hand-held readers can read tags only 6" away). More than 33,000 animal identifications have been made at these facilities.

To further safeguard the system, the MDA requires movement permits for all animals, moving to any destinations. The system allows farmers to print the permits from their computers using a Web-based application. MDA can also fax a permit to a farmer who doesn't have access to the Internet. If an animal meets certain criteria, it is issued a movement permit. An animal without a permit is not allowed to be unloaded at the auction markets or slaughterhouses. After an animal is identified at these locations, the computers at these facilities update the animal's profile in the FAIR database with a new location. After an animal is slaughtered, the corresponding entry is removed from the database.

Results and Lessons Learned

• More than 33,000 animal identifications have been made since the start of the project. The accuracy of the handheld readers has been approximately 99% and tag retention has been over 96% (that is, the RFID tag stayed in the ear and was recovered at the market or slaughterhouses), higher than the retention rate for the typical metal tags. Approximately 1,600 farms in the affected counties (out of more than 15,000 farms in the state) have adopted RFID (a little over 10%). A total of 180,000 (estimated) tags have been issued so far.

• Twelve trace-backs have been achieved (that is, animals were traced back to their farm of origin). In May of 2003, an animal tested positive for TB outside of the affected counties. The entire history of the animal, including the farm of origin was traced within 15 minutes. In the past, such tracing would have been either impossible due to incomplete records or taken weeks to complete. The current process is fast enough to stop an infected animal from leaving the restricted area and pose a health hazard to other animals or humans.

• The computer and RFID-based process is more accurate, simpler to use, and easier to maintain than the paper-based system. As a result, 3.5 data entry positions have been saved. The current solution has also reduced the time it takes to retest a herd by 50%. It prevents wrong animals from being tested and possibly killed.

• The resulting superior tracking and elimination of infected animals has enabled Michigan to make progress towards regaining its TB-free status. The progress has ensured continued access to markets for Michigan producers. Over time, such traceable meat can even be marketed at a premium, increasing profits for the producers.

• Successful execution of this project has enabled MDA to lay the groundwork for deploying EID in the rest of the state. The blueprints of this project can also help other states replicate Michigan's success. A cooperative effort is already underway between the government and the industry in the form of The U.S. Animal Identification Plan (USAIP). It is chartered to define the standards and framework for implementing and maintaining a national animal identification system for the U.S. As part of the project, USDA has established a National Animal Identification Development Team with participation from more than 70 animal associations, organizations, and government agencies. A fully implemented plan could identify all premises that had contact with a foreign animal disease within 48 hours after discovery.