- Yun Liu
University of Electronic Science and Technology of China, China & Sichuan Vocational and Technical College of Communications, China
- Peiji Shao
University of Electronic Science and Technology of China, China
This paper discusses how to use the RFID technology to realize the life-time traceability of animals. In order to record movements of an animal, the authors adapt the electronic pedigree designed for drugs to animals, which acts as standard data elements and is transferred between partners. Then a CIS is proposed based on the methodology which ensures movements of the animal to be recorded correctly. Results in a case study show that the solution can facilitate the life-time traceability of animals. It is reliable, convenient to query, easy to understand and use, and compatible with existing information systems.
Animal identification and tracing is so important in securing food supply and preventing disease spreading that such projects as the National Animal Identification System (NAIS) in the United States and the National Livestock Identification System (NLIS) in Australia are developed, and the National Animal Identification and Tracing (NAIT) in New Zealand is being proposed.
In NAIS, the United States Department of Agriculture (USDA) only requires the producer in a registered premise to report certain animal movements that might have a potential impact on spreading a disease (USDA, 2006). This is different from Food and Drug Administration (FDA) who want to record all movements in the pharmaceutical supply chain in order to fight counterfeit drugs.
The drug pedigree, which records the sale by a pharmaceutical manufacturer, the acquisition and sale by a wholesaler and final sale to a pharmacy (EPCGlobal, 2007), enables the life-time traceability of drugs. We borrow this idea and adapt the electronic pedigree designed for drugs to animals. The RFID technology helps read and write the electronic pedigree stored in the RFID tag.
The electronic pedigree cannot fulfill the life-time traceability of animals alone. When the animal moves on, new information must be added to the electronic pedigree to reflect its movement. Since information is produced in distributed information systems, so another problem in the life-time traceability of animals is information system integration, which requires the development of the collaborative business process model.
This paper discusses how to use the RFID technology to realize the life-time traceability of animals, which is organized as the following: first we review the application of the RFID technology in animal identification and tracing and the knowledge about the electronic pedigree. Next after putting forward existing problems, we introduce the electronic pedigree redesigned for animals and the methodology to build the collaborative business process model and develop a CIS (Collaborative Information System) which combines methods of Rajsiri, Lorre, Benaben, and Pingaud (2007) and Chaari, Binnier, Favrel, and Amar (2006). In a case study of the pet’s management, we demonstrate how to use them step by step. Finally, we discuss impacts of our solution and describe our future research direction.
The Application of RFID in Animal Identification and Tracing
In China, the RFID technology is applied first in pet identification. Increasing pets oblige the public to think about problems in the pet’s management and turn to get helps from other countries’ experiences.
Since it is the right time that market-ready RFID technologies are gradually taking place of traditional eartags, tattoos and brands in identifying animals, some Chinese cities try to utilize RFID tags to store simple information necessary for pets and build pet identification systems. Thus once a pet is lost, the person who finds it can easily figure out who it belongs to by reading the RFID tag.
RFID tags are also applied in the pork supply chain to guarantee the product’s quality. This is the first time that Chinese government requires traceability in the food supply to protect the consumer’s right to know what they will buy (Xiong, Fu, Lin, Luo, & Yang, 2009).
USDA goes further to build NAIS, which enable USDA to trace a disease back to its source in short time and limit potential harm to animal agriculture (USDA, 2008).
The premises registration component is the foundation of NAIS. When a producer registers his premise, a unique premises identification number (PIN) is assigned and his contact information is recorded, which ensures that the producer is notified quickly when an animal disease outbreaks (USDA, 2008).
The animal identification component involves assigning an animal or a group of animals a unique animal identification number (AIN), which gives animal health officials a clue for epidemiologic investigations (USDA, 2008).
The animal tracing component allows a producer to choose an animal tracing database (operated and maintained by private industry groups or States) and report certain animal movements that might pose a significant risk of disease transmission (USDA, 2008).
NAIS is a State-Federal-industry partnership, the responsibility for implementing and administering NAIS is shared among State animal health authorities, USDA, and the animal agriculture production industry (USDA, 2008).
The Electronic Pedigree
FDA specifies that the drug pedigree is the statement of a drug’s history in the pharmaceutical supply chain. According to the Prescription Drug Marketing Act (PDMA), the drug pedigree should record the drug name, dosage, container size, number of containers, drug lot or control number, names and addresses of all partners in each previous transaction, and transaction dates. But maintaining such a drug pedigree is overwhelming (Koh, Schuster, Chackrabarti, & Bellman, 2003).
EPCglobal specifies an open document model complying with PDMA, which defines two XML schemas. One is about the standard electronic pedigree format; the other is about the standard electronic envelope format to package multiple electronic pedigrees. The great benefit of these schemas is that they provide standardization for the exchange of electronic pedigrees as they pass down the supply chain (EPCGlobal, 2007).
The electronic pedigree is created using the first XML schema and the new shall always be started from an intialPedigree element, which is the innermost component of the electronic pedigree and includes serial number, product and item information. Each partner engaged in the pharmaceutical supply chain is required to provide signed electronic pedigree to the recipient who authenticates each previous transaction in the electronic pedigree and add its own certification of receipt and signature to the electronic pedigree (EPCGlobal, 2007).
The open document model does not identify exactly how electronic pedigrees must be transferred between partners, and the RFID technology is not mandatory (EPCGlobal, 2007).
The life-time traceability of animals involves collaboration among government authorities, individuals and organizations of the industry, which store information in their own databases or paper-based filing systems. So in this collaborative environment, the most significant challenge is to provide standardization for data exchange and apply modern technology to integrate information systems of partners (USDA, 2008).
RFID and the Electronic Pedigree
In this paper, we suggest that the electronic pedigree be used to record an animal’s movements as it does in the pharmaceutical supply chain. In EPCglobal’s design, parts of an electronic pedigree are distributed across different information systems; one can collect these parts into a complete electronic pedigree through a query (EPCGlobal, 2007). But FDA sees an electronic pedigree as a complete electronic legal document directly transferred from one partner to another (EPCGlobal, 2007).
There are several mechanisms which are likely to be utilized for this transfer. For example, many existing information systems could be augmented to include the electronic pedigree (EPCGlobal, 2007). In this paper, we propose that the electronic pedigree can be stored in the RFID tag and transferred with the animal.
RFID is a technology used to identify, locate and track assets. Each RFID tag has a unique number which is called EPC. EPC is an expansion of UPC-like codes, allowing for identifying a specific item (not just their broad classes) (Stapleton-Gray, 2004).
For example, a national standard of animal identification was introduced in china on December 1, 2006, in which the animal’s EPC is composed of 64 bits, and is classified into three fields: the first 16 bits are for control, then the 10 bits from 17 to 26 represent country or region, and the rest is serial number (Liu & Shao, 2007).
Now RFID tags have enough memories that allow them to store the electronic pedigree. So in this paper, we suggest that the RFID tag be used to identify an animal, and the electronic pedigree be used to trace its history. Next a CIS is developed to integrate information systems of partners and ensure the animal’s movements to be recorded correctly.
Moreover, the electronic pedigree only contains important information about an animal’s movements, while other information, for example those about vaccinations, be distributed in information systems and collected by the CIS.
Collaborative Business Process Modeling
Complex systems (such as the CIS) require the development of models as a help to better understand the business domain and as a basis for integrating information systems (Rajsiri, Lorre, Benaben, & Pingaud, 2007).
The Business Process Model is a representation that tries to capture the business process which is essential to the organization’s functions (Martins & Soares, 2006). A lot of business processes modeling techniques are developed, from data modeling to behavior modeling techniques (Martins & Soares, 2006), which are in high levels and easy to understand. At the same time there are some low-level ones, such as BPEL4WS, WS-CDL, BPML, WSCI, WSFL, XLANG and WSDL, which are XML based languages (Martins & Soares, 2006).
Business Process Modeling Notation (BPMN) positions itself as a bridge between these two groups and between people who run the business and people who implement the system (Martins & Soares, 2006). Next we will introduce a BPMN-based methodology combined by methods of Rajsiri, Lorre, Benaben, and Pingaud (2007) and Chaari, Binnier, Favrel, and Amar (2006). The objective of this methodology is to build the collaborative business process model and then to develop a CIS. This methodology includes 6 steps, which are shown below.
- Step 1: To collect the details about partners.
This step addresses to the definition of partners who collaborate around a product (an animal in this paper). Partners shall be interviewed individually to collect their information, especially their relationships with others. These relationships can be described with a matrix where the rows represent the partners and the columns also represent the same set of partners. If there is an interaction between partner i and partner j, matrix entry (i, j) is assigned a specific value that defines one kind of relationship; and 0 otherwise. It is recommended to investigate more similar products, which will help better understand the business domain.
- Step 2: To define roles and their relationships.
When analyzing relationship matrixes in step 1, we will find that many partners are similar. If two partners connect with others in the same way, we would say that they have the same role (Hanneman & Riddle, 2005). Thus we can abstract a few roles from partners.
After defining roles, we continue to analyze relationships between these roles and describe them in a new relationship matrix. Since roles are usually connected for multiple objectives simultaneously (Hanneman & Riddle, 2005), it is suggested that one matrix be created for an objective to avoid complexity.
- Step 3: To specify the collaborative business process model using the objective.
If the objective is general, it should be divided into smaller detailed ones, and for each of them a collaborative business process model will be defined. We get one matrix for a smaller detailed objective in step 2, so the collaborative business process model can be analyzed from it.
- Step 4: To analyze activities one by one in the collaborative business process model.
Activities are associated with enterprise business functionalities, and must be analyzed one by one. They can be broken down into a finer level of detail, such as tasks, using BPMN as the business processes modeling technique. In previous step, we also use BPMN.
- Step 5: To translate the collaborative business process model into a SOA model.
Activities in the collaborative business process model correspond to a set of business components which are made of business objects. Web services can be created from these business components and exposed to be invocated by roles (or partners) in the collaborative environment (Chaari, Binnier, Favrel, & Amar, 2006). A SOA model is an architecture which consists of a set of web services that can be invoked through standard protocols.
- Step 6: To use orchestration and choreography for integration.
One can orchestrate and choreograph web services to fulfill an organization’s goals (Arsanjani, 2004). Partners can continue to use their own information systems, while a CIS is used for managing the collaborative business process (Rajsiri, Lorre, Benaben, & Pingaud, 2007).
In this section, we will apply our solution to the life-time traceability of pets, since many problems in the pet’s management are associated with movements of pets, such as dishonest behaviors in transactions. First we design the electronic pedigree for pets, and then analyze the collaborative business process model and develop a CIS.
The Electronic Pedigree for Pets
Except for UID, the information recorded in the electronic pedigree for a pet includes:
- • PetInfo, such as the specie, sex, birth date and its parents;
- • RegistrationInfo, such as ID, name and the owner’s information;
- • TransactionInfo
- • ShippingInfo
- • ReceivingInfo
- • SignatureInfo
An example of such electronic pedigree is shown below (Figure 1). This example describes a scenario in which a pet is sold from a producer to a customer. Before the sale, the producer must register the pet’s birth in a police station; and after the sale the customer has to transfer the pet’s ownership in that police station. If one studies the example in detail, he will see that the electronic pedigree can record the registration and transaction, shipping, receiving information in this scenario and help us “see” movements of this pet.
|Figure 1. An electronic pedigree|
The Collaborative Business Process Model and CIS
Next we will show how to develop the collaborative business process model and CIS step by step.
Before survey, we studied Chinese laws and regulations about pets and browsed websites of animal protection associations to understand problems in the life-time traceability of pets. Then we investigated some typical pets in their experiences. Through interviews with their related people, relationship matrixes can be got, from which we can draw networks. Because these networks involve government authorities, individuals and organizations which surround pets, we call them pet-centric networks (Figure 2).
|Figure 2. A pet-centric network|
In Figure 2, partners collaborate to better achieve the common objective – the management of a pet.
Defining roles can simplify issues that we address to. We can abstract such roles as the producer (A), customer (B), animal clinic (C), animal protection association (D), police station (E), industrial and commercial administration department (F), and animal health authority (G) from partners in pet-centric networks.
Table 1 shows relationships among such roles as the producer (A), customer (B), animal protection association (D) and the police station (E).