Theseos Query Engine for Traceability Networks
A demonstration of a technology that allows supply chains to track items quickly and efficiently without compromising confidentiality and control of participating organizations.
Date Posted: January 16, 2007
|
|
 |
|
What is Theseos Query Engine for Traceability Networks?
Enabling traceability over the entire life cycle of an RFID-enabled product in a supply chain requires that the data be shared across independent (and possibly competing) enterprises. To date, a limited number of technologies exist for distributed query processing across RFID networks, and enterprises are reluctant to share data across supply chains for fear of giving away confidential information. Theseos Query Engine for Traceability Networks is general-purpose middleware technology that addresses many of these issues and can be applied for traceability applications across many industries. It includes the following:
- Theseos Discovery Service: As proposed by EPCglobal, this service in the EPCglobal network links information about RFID-enabled products as they move through the supply chain and provides trading partners with the ability to find all parties who had possession of a given product.
- Theseos Query Engine: This engine exploits the Discovery Service in order to detect remote RFID repositories that need to be contacted, and it provides the ability to retrieve information about a product from all parties in a supply chain. The engine contains functionality for mitigating access by allowing parties in a supply chain to set and manage specific data-sharing policies and to control their confidentiality requirements. Thus, the engine enables the security and privacy control of the multiple (federated) data sources.
Consider the food industry: National Produce Distributors (NPD) is a U.S.-wide fruits and vegetables wholesaler that supplies many retailers. Around the time of the Super Bowl, many spectators like to prepare homemade guacamole from California-grown avocados, which they buy from local retailers. Recently, NPD has had problems with delivering overly-ripe produce to retail stores. Retailers have become increasingly upset with NPD and are considering switching distributors. NPD's executives choose to use RFID technology to solve these problems.
Making use of Theseos technology, NPD supply chain managers can reach their objective of delivering 100% fresh produce during the avocado promotion for the Super Bowl. Throughout NPD's supply chain, RFID readers are installed and each avocado crate has a RFID tag. In addition, each container used for transportation is equipped with a TREC (Tamper-Resistant Embedded Controllers) device, which records the temperature at which the avocados were stored.
The moment the first avocado crate with overly ripe fruits is detected at a store, a troubleshooting process can be initiated dynamically across the supply chain, and specific information can be obtained about the history of that avocado crate. This information can be used to determine the cause of the problem. For example, perhaps the container in which the avocado crate was transported was stored at a higher temperature than recommended. Once this cause is isolated, additional queries can be issued in order to detect the current location of all other avocado crates in the affected container. If necessary, avocado crates can then be redirected to closer stores to ensure that all crates reach a store before they get too ripe. The success of this scenario depends on the ability to process queries across multiple enterprises. A demonstration of this scenario using Theseos is included.
This technology provides several benefits:
- Removal of the barrier to sharing information between participating entities within supply chains, resulting in improved quality of data flow within supply chains.
- Generation of pedigree information of electronically-tagged items on demand.
- Elimination of wasteful recalls and reduction of enterprises’ risks.
- Increased value of IBM middleware and representation of additional industry-specific solution opportunities.
- A compelling case of return on investment for RFID technology.
Further information about Theseos is available at IBM Almaden Research Center's Intelligent Information Systems Web site as well as in this document (also included in the download): Enabling RFID Traceability in EPCglobal Networks (PDF).
How does it work? The Theseos generic query engine is a middleware component that can be deployed and customized to suit specific industry solutions. The query engine takes queries as input and initially partitions the queries into local and remote queries. All local queries go through a data confidentiality policy check prior to being routed to the data system, which could be an arbitrary EPC-IS. The remote queries are sent to other parties via Web services after potentially being "rewritten." Why "rewrite"? For example, items might get packed or unpacked into different containers during the product flow across the supply chain. To address this problem, the query engine might need to change the item that it traces before it forwards the query to other parties.
|
|
|
| | About the technology author(s):
 Tyrone Grandison leads the Data Disclosure Management team in the Computer Science department at the IBM Almaden Research Center. The group is focused on fundamental science that is relevant and applicable to industry verticals. The team has pioneered research in Data Mining, Data Privacy (such as the Hippocratic Database technology), RFID Data Management, and Privacy-Preserving Mobile Data Management. The team is also exploring novel concepts in content management and storage systems. Dr. Grandison received his M.S. from the University of the West Indies in Mona and his Ph.D. from Imperial College in London, England.
|
 Karin Murthy is a post-doctoral researcher in the Intelligent Information Systems Research Group at the IBM Almaden Research Center. She received her Ph.D. in computer science from the University of Munich in Germany and has published several papers on databases and data mining. Her current research interests include the querying and mining of RFID data.
|
 Ralf Rantzau is a software engineer at the IBM Silicon Valley Laboratory and is a member of the product development team for IBM's WebSphere RFID Information Center. Before that, he was a post-doctoral researcher in the Intelligent Information Systems Research Group at the IBM Almaden Research Center. Dr. Rantzau received his M.S. and Ph.D. in computer science from the University of Stuttgart, Germany. His interests include databases, data mining, and security.
|
|
|
|
| |
